Home

COPYRIGHT NOTICE for An Enhanced

Contents

1. 152 6 49 DIM POSITIVE 152 6 50 DIMAG IMAGINARY 152 65 1 DIENT 153 6 52 DLOG10 LOGARITHM BASE 10 153 6 53 DLOG 2 22 2 153 6 54 DMAXI MAXIMUM 2 154 6 55 DMINI MINIMUM 020222 154 6 56 REMAINDER 154 6 57 DNINT NEAREST 8 154 6 58 DPROD PRODUCT mut nte he LL 155 6 59 RT Ei n 155 6 60 HYPERBOLIC SINE eae eee ve ev ee 156 6 61 DSIN SINE 2 RES OE RE RRS 156 6 62 DSORT Leia repre eh reete ect WI OT a E AER En REAR 156 6 03 DT ANEIE HYPERBOLEIC TANGENT eee eee Dein Eee ee exe eee Poen Beles 156 6 04 DTAN DANGENIT e ex ee Dein IE 157 6 605 EXIT STOP PROGRAM EXECUTION eee E eo ese eee voie eie sevev 157 6 06 EXP EXPONENTIATL REE OE 157 6 07 MOA T REAL VALUE 158 6 68 GET CL GET COMMAND LINE ete e ege t
2. REAL FUNCTION REAL INTEGER REAL FUNCTION REAL COMPLEX REAL FUNCTION REAL COMPLEX 16 Description REAL a is a generic function that performs type conversion and returns a real result The argument can be integer real double precision or complex For a complex argument ar ai the result is real ar See also FLOAT and SNGL 6 104 SIGN Transfer of Sign 170 PromulaFortran Compiler User s Manual Synopsis INTEGER 2 FUNCTION SIGN INTEGER 2 INTEGER 2 INTEGER FUNCTION SIGN INTEGER INTEGER REAL FUNCTION SIGN REAL REAL REAL 8 FUNCTION SIGN REAL 8 REAL 8 REAL 8 FUNCTION SIGN REAL REAL 8 REAL 8 FUNCTION SIGN REAL 8 REAL Description SIGN 1 2 is a generic function whose result is the same type as its first argument The result reflects the magnitude of the first value combined with the sign of the second value In particular if al 0 0 and a2 0 0 return al or if al lt 0 0 and a2 gt 0 0 return 1 else return al See also DSIGN ISIGN and I2SIGN 6 105 SINH Hyperbolic Sine Synopsis REAL FUNCTION SINH REAL REAL 8 FUNCTION SINH REAL 8 Description SINH a is a generic function that returns a hyperbol
3. dota enn ex E RAE ORT 147 6 36 5 148 6 37 DACOS ARGCOSINE cede oi ce inei EE e aeons 148 The Promula Fortran Compiler User s Manual 6 38 IDASIN CARGSINE bretoia reoeo 148 6 39 DATE CURRENT DATE eerte teorie io eRe Bee 149 6 40 DATAN2 ARCTANGENT OF 2 22 2 411 00 000001 000000000000000000000000000000005 ERE ER SPEER 149 641 DATAN ARGTANGENT e bre eate 149 642 DBEE DOUBLE PRECISION VALUE euin eere ttt ord nocet eode 149 6 43 DCMPEX DOUBLE COMPLEX VALUE tabe oder 150 6 44 DCONJIG dtes dtes ure Eu ore Aube eas esses 150 6 45 DEOSH HYPERBOLIC 0 eere ied re tet 151 6 16 DE OS COSINE 5 eher ed esee ru 151 6 47 DDIM POSITIVE DIFFERENCE ccccccccecececececececcescecececececececececscecececececscecececececececececececececececscecececececececececececcs 151 6 48 DEXP EXPONENTIAL
4. 75 PromulaFortran Compiler User s Manual As a second example consider that CHARACTER AR 2 10 DATA AR 10X 212 1 0X F6 2 READ 5 AR I J X is equivalent to READ 5 100 I J X 10 FORMAT 10X 212 10X F6 2 If a format specification is contained in a character array the specification may cross element boundaries Only the array name need be specified in the input output statement all information up to the closing parenthesis is considered to be part of the format specification 5 27 2 Noncharacter FORMAT Specifications Format specifications can be contained in a noncharacter array The rules for noncharacter format specifications are the same as for character format specifications 5 27 3 Edit Descriptors Edit descriptors specify the data conversions to be performed Below are the repeatable edit descriptors Type Descriptor Description Character Character with data dependent length Aw Character with specified length Numeric Dw Double precision floating point with exponent Single precision floating point with exponent Ew dEe Single precision floating point with exponent length Fw d Single precision floating point without exponent Gw d Single precision floating point with or without exponent Gw dEe Single precision floating point with or without explicitly specified exponent length Iw Decimal integer Iw m Decimal integer with minimum number of digits Logical Lw Logical Typeless
5. INTEGER 2 FUNCTION INT2 REAL 8 INTEGER 2 FUNCTION INT2 COMPLEX INTEGER 2 FUNCTION INT2 COMPLEX 16 Description INT2 a is a generic function that performs conversion to short integer and returns a short integer result The conversion to integer for noncomplex arguments is performed via simple truncation For complex arguments only the real part is used and truncated See also IDINT IFIX INT and INT4 6 78 INT4 Integer Value Synopsis INTEGER FUNCTION INT4 INTEGER 2 161 PromulaFortran Compiler User s Manual INTEGER FUNCTION INT4 COMPLEX 16 Description INT4 a is generic function that performs conversion to integer and returns an integer result The conversion to integer for noncomplex arguments is performed via simple truncation For complex arguments only the real part is used and truncated See also IDINT IFIX INT and INT2 6 79 INT Integer Value Synopsis INTEGER FUNCTION INT INTEGER 2 INTEGER FUNCTION I COMPLEX 16 Description INT a is a generic function that performs conversion to integer and returns an integer result The conversion to integer for noncomplex
6. REAL 8 FUNCTION ABS REAL 8 REAL FUNCTION ABS COMPLEX REAL 8 FUNCTION ABS COMPLEX 16 Description ABS a is a generic function that returns an absolute value of its argument The result is integer real or double precision depending on the argument type For noncomplex arguments the result is a if a is greater than or equal to zero else it is a For a complex argument the result is the noncomplex square root of the sum of the squares of the two components of the complex number See Also IABS DABS CABS CDABS and I2ABS 6 2 ACOS Arccosine Synopsis REAL 8 FUNCTION ACOS REAL 8 REAL FUNCTION ACOS REAL Description ACOS a is a generic function that returns an arccosine The result is expressed in radians The result is real or double precision depending on the argument type See also DACOS 6 3 AIMAG Imaginary Part Synopsis REAL FUNCTION AIMAG COMPLEX REAL 8 FUNCTION AIMAG COMPLEX 16 Description 137 PromulaFortran Compiler User s Manual a returns the imaginary part of a complex argument The result is real if the argument is complex and double precision if the argument is double complex See also DIMAG 6 4 AINT Truncation Synopsis REAL FUNCTION AINT REAL REAL 8 FUNCTION AINT REAL 8 Description AINT a is a generic function that returns
7. Where name has one of the forms var len c array d d len clist len specifies the integer subtype and can be an unsigned nonzero integer constant whose value is 1 2 or 4 var is a variable function name symbolic constant or dummy procedure array isan array name d specifies the bounds of a dimension clist isa list of constants or symbolic constants specifying the initial values Each item in the list can take the form 101 PromulaFortran Compiler User s Manual is a constant or symbolic constant isa repeat count that is an unsigned nonzero integer constant or the symbolic name of such constant Description The INTEGER statement is used to define a variable array symbolic constant function name or dummy procedure name as type integer short integer or byte A length specification immediately following the word INTEGER applies to each entity not having its own length specification A length specification immediately following an entity is the length specification only for that entity If the length specification for a given entity is 4 or if it is omitted then the type defined is integer If the value is 2 the type is short integer and if the value is 1 the type is byte See also See the discussion of the DIMENSION statement for a description of how dimension bounds are defined See the discussion of the DATA statement for a description of how ini
8. EENE iege 100 3 33 INCLUDE STATEMENT ier rr tenter 101 3 94 INTEGER STATEMENT po tete merae e Ee ete tete dee Posee 101 3 39 INQUIRE STATEMENT deter tee ee cedes uec iet ete toes 102 3 30 INTRINSIC STATEMENT 5 rrr iet tie rH RN rore ON MPH IF e rb USER Re ER 105 5947 LOGICAL STATEMENT nte FO n eror EUR REO HIFI EI NEEE REIR HEP E ERE 106 5 26 NAMELIST STATEMENT RE IR YER EPA E E RER OT ER 106 5 36 1 NAMELIST READ Statement 00 108 The Promula Fortran Compiler User s Manual 5 38 2 NAMELIST WRITE PRINT PUNCH Statements 109 DISS Arrays IN NAMELIS T ei critt p eri e e se Shes petet e eerte teet 110 3 39 OPEN STATEMPBNTE oerte robe tte 111 23 39 1 Direct ACCESS 113 3 40 55 114 SAT PAUSE STATEMENT 3 eere pe tds fo euer er eR keene 115 542 PRINESTATEMENT he a 116 53 43 PROGRAM STATEMENT 3 55
9. Description 140 PromulaFortran Compiler User s Manual ANINT a is a generic function that returns a real or double precision result from a real or double precision argument It computes the nearest integer to its argument In particular the result is dint 0 5 if a gt 0 0 and dint a 0 5 if a lt 0 0 where dint is the FORTRAN intrinsic function DINT See also DNINT IDNINT I2NINT and NINT 6 14 ASIN Arcsine Synopsis REAL FUNCTION ASIN REAL REAL 8 FUNCTION ASIN REAL 8 Description ASIN a is a generic function that returns an arcsine The result is expressed in radians The result is real or double precision depending on the argument type See also DASIN 6 15 ATAN2 Arctangent of Quotient Synopsis REAL FUNCTION ATAN2 REAL REAL RE D 8 FUNCTION ATAN2 REAL 8 REAL 8 RE D 8 FUNCTION ATAN2 REAL 8 REAL REAL 8 FUNCTION ATAN2 REAL REAL 8 Description ATAN2 1 2 is a generic function that returns the arctangent of a2 divided by a1 The result is expressed in radians The result is real or double precision depending on the type of the arguments If both arguments are zero the result is zero See also DATAN2 6 16 ATAN Arctangent Synopsis 141 PromulaFortran Compiler User s Manual REAL FUNCTION ATAN REAL RE
10. Table 6 1 Intrinsic Functions Name 1 AMINO AMINI AMOD AND ANINT ASIN ATAN2 ATAN CABS CCOS CDABS CDCOS CDEXP CDLOGIO CDLOG CDSIN CDSQRT CEXP CHAR CLOGIO CLOG CMPLX CONJG COSH COS CSIN CSQRT DABS DACOS Type REAL REAL REAL REAL REAL LOGICAL 2 LOGICAL REAL REAL 8 REAL REAL 8 REAL REAL 8 REAL 8 REAL 8 REAL REAL 8 REAL COMPLEX REAL 8 COMPLEX 16 COMPLEX 16 COMPLEX 16 COMPLEX 16 COMPLEX 16 COMPLEX 16 COMPLEX CHARACTER 1 CHARACTER 1 COMPLEX COMPLEX COMPLEX COMPLEX COMPLEX COMPLEX COMPLEX COMPLEX COMPLEX COMPLEX COMPLEX COMPLEX COMPLEX COMPLEX 16 REAL REAL 8 REAL REAL 8 COMPLEX COMPLEX 16 COMPLEX COMPLEX REAL 8 REAL 8 Arguments INTEGER REAL INTEGER REAL REAL REAL LOGICAL 2 LOGICAL REAL REAL 8 REAL REAL 8 REAL REAL REAL 8 REAL 8 REAL 8 REAL REAL REAL 8 REAL REAL 8 COMPLEX COMPLEX COMPLEX 16 COMPLEX 16 COMPLEX 16 COMPLEX 16 COMPLEX 16 COMPLEX 16 COMPLEX 16 COMPLEX NTEGER 2 INTEGER COMPLEX COMPLEX INTEGER 2 INTEGER REAL REAL 8 COMPLEX COMPLEX 16 REAL REAL REAL 8 REAL 8 REAL 8 REAL REAL REAL 8 COMPLEX COMPLEX 16 REAL REAL 8 REAL REAL 8 COMPLEX COMPLEX 16 COMPLEX COMPLEX REAL 8 REAL 8 Description Maximum value Maximum value Minimum value Minimum value Remaindering Logical and Nearest whole number Arcsine Arctangent of quotient
11. cee 116 5 5 sige 117 5 45 READ STATEMBNI 118 5 446 REAT STATEMENTS 5 tito temer mee eA e 122 5 47 RETURN 5 cor tror ER ee teri teet re DR ee Tei tei ve EO ee tin cree De es 123 5 48 RECORD STATEMBENT 123 5 49 REWIND 5 11 124 2 90 SAV ES TATBMENIT 9 5 ale sat ueteres e A Bat Uode at Mee ier E 124 5 9 1 solos hse tere triremes ree te tet cat eee bees seek hs ket 125 5 52 S LRUGTURE STATEMENT ot rr rette rn D Ori Eis 125 5 53 SUBROUTINE STATEMENT ccccccccccccecscecececececscecececscecececececscecececscecececeeeeecececececececececececececececececececececececesess 127 2 94 EX PE STATEMENT 5 2 ic bosses stat ute oe Nh ke Bet cosh rire ELA Ue S ve Rk 128 5 55 ERE ia Rishi Rss 128 FORTRAN INTRINSIC FUNCTIONS cccssssscsssssccsssssccesssscccssseccscsscccscsscccsssssccsessscccessaceccssnsccescsscesessececcssnee 132 6 1 ABS
12. Eee 13 25 7 Selecting Dialect Conventiornis aiite eee tn eie etr e e a pe dea Het ree tte e eg 14 2 6 STORAGE QUANTITY VALUES i sett eet needed lesse etr oc ete Tent Et A Eve SCOPE E Dee e ee eo 14 2 7 MISCELLANEOUS OPTIONS et o itp e a ee Tee NER deste cde Een EE eret ERR IERI ENS 14 2 74 The Compile Only Option e eet e Rc eet Ba er Pe E Hs PUR Re Ree Posta oc p ee eee 14 2 7 2 The Debugging Flag e ee he etr eie hende iad oder eie bee 15 2 7 3 Compiler Include File Patliname eae c e Re ette eee de a e ide Fase 15 2 7 4 Library File Name or Pathname for Linker eee 15 2 7 5 NAME of executable 4 e d e fed eee Ree vet ee t ee fee e ee meu 15 2 7 6 Location of FORTRAN Files Be Included 15 2 8 PROTOTYPEEIEES ioa e rene te e e eO te e e D RES 15 2 6 1 Formation of External Symbols e d e be eiie etr Fee io eive Dine 16 2 6 2 F nction PrototypeAS yx sd eU ROTER er eO aa ais avi e beo reped 16 252 Value Parameters as i e eo ceres Nee este de ieu ee E Ede eet ne uve erede Pee a eoe beg 17 ZSA External Name Clash EE We ee eO ERE ehe bc Mt NER 17 2 6 2 MUltiDle EOTIRS es eue o UU e ie eene ee gendum dedii 18 2 6 6 Global Symbols and PFOIODypes 1 RU oe eire e gegen SE ey 19 2 6 7 Renaming Identifiers Only AR UR E
13. INTEGER 2 FUNCTION MOD INTEGER 2 INTEGER 2 INTEGER FUNCTION MOD INTEGER INTEGER REAL FUNCTION MOD REAL REAL REAL 8 FUNCTION MOD REAL 8 REAL 8 REAL 8 FUNCTION MOD REAL REAL 8 REAL 8 FUNCTION MOD REAL 8 REAL INTEGER FUNCTION MOD INTEGER INTEGER Description MOD al a2 is a generic function that returns the remainder of a1 divided by a2 The result is integer real or double precision depending on the argument type The result is al int al a2 a2 If a2 equals zero then the result is zero See also and I2MOD 169 PromulaFortran Compiler User s Manual 6 102 NINT Nearest Integer Synopsis INTEGER FUNCTION NINT INTEGER 2 INTEGER FUNCTION NINT INTEGER INTEGER FUNCTION NINT REAL INTEGER FUNCTION NINT REAL 8 Description NINT a is a generic function that returns an integer result from an arithmetic noncomplex argument It computes the nearest integer to its argument In particular the result is int a 0 5 if a gt 0 0 and int a 0 5 ifa 0 0 where int is the FORTRAN intrinsic function INT See also ANINT DNINT IDNINT and IZ2NINT 6 103 REAL Real Value or Part Synopsis REAL FUNCTION REAL INTEGER 2
14. ei te RI dte ener 27 34 VARIABLES d 28 SAL ANGE ET it d eet nd erede tei ess 28 34 2 Short Integer Variable tette Rd nitet d 28 343 Byte Variable a ente eR ette e deett Rees 28 3 4 4 s inasre iiir 29 3 4 5 Double Precision Variable ei e et t ed eee ee bietet ee sedes 29 3 4 0 Complex Variable 4 nemo entender Od ener eee 29 3 4 7 Double Complex Variable ee etae RI re ERA der ERU FEE 29 3 4 9 LOGICAL Variable io aro te due ER PARC Ee 29 3 4 9 Short Logical Variable eiie ete etre erar Rar rere bet IEEE Eee 29 3 410 Logical Byte Variable iei pecore eir Rip ren e per ERU ed HERE ERR Ep t 29 S4 I Character Variable RE RE REEF RARE 29 BRUDE I IS E tt RU e LEBER PRESE ORE REM 30 3 2 1 Aff y SOTA BE 1 eerte en pe bep dr ER PR e e dre e Eee EH 31 3 5 2 References denoted er OU PER e e pb PPP eate 32 3 6 CHARACTER SUBSTRINGS DURO o HI NORRIS E REPRE RE 33 3 0 1 Substring Refetences ise aet ro ne e ER EP PU OH HERR ERE EUR 33 3 6 2 Substrings and Arrays epu ROO RHEIN rir e PERI PU o eR 33 3 7 STATEMENT ORDER EE rh rr EUR GER e RR URL I E PARE RES 34 3 8 USER WRITTEN PROGRAM UNITS eese A 34 3 6 1 Program Unit and
15. 2 MAXO and MAXI 6 55 DMINI Minimum Value Synopsis REAL 8 FUNCTION DMIN1 REAL 8 Description DMIN1 al a2 an is a specific function that returns the value of the smallest argument All arguments are double precision and the result is double precision See also AMINO AMINI I2MINO MINO MINI and MIN 6 56 DMOD Remainder Synopsis REAL 8 FUNCTION DMOD REAL 8 REAL 8 Description a1 a2 is a specific function that returns the remainder of al divided by a2 The result and both arguments are double precision The result is al int al a2 a2 If a2 equals zero then the result is zero See also AMOD I2MOD and MOD 6 57 DNINT Nearest Integer Synopsis 154 PromulaFortran Compiler User s Manual REAL 8 FUNCTION DNINT REAL 8 Description DNINT a is a specific function that returns a double precision result from a double precision argument It computes the nearest integer to its double precision argument In particular the result is dint 0 5 if a gt and 0 0 dint a 0 5 if a lt 0 0 where dint is the FORTRAN intrinsic function DINT See also ANINT IDNINT I2NINT and NINT 6 58 DPROD Product Synopsis REAL 8 FUNCTION DPROD REAL REAL Description DPROD a1 a2 is a specific function that returns a double precision product The arguments are real and the result is double precision The
16. Arctangent Absolute value Cosine Absolute value Cosine Exponential Logarithm base 10 Natural logarithm Sine Square root Exponential Character value Logarithm base 10 Natural logarithm Complex value Conjugate Hyperbolic cosine Cosine Sine Square root Absolute value Arccosine 133 PromulaFortran Compiler User s Manual Table 6 1 Intrinsic Functions Name DASIN DATE DATAN2 DATAN DBLE DCMPLX DCONJG DCOSH DCOS DDIM DEXP DIM DIMAG DINT DLOG10 DLOG DMAXI DMINI DMOD DNINT DPROD DSIGN DSINH DSIN DSQRT DTANH DTAN EXIT EXP FLOAT GETCL IABS Type REAL 8 REAL 8 REAL 8 REAL 8 REAL 8 REAL 8 REAL 8 REAL 8 REAL 8 COMPLEX 16 COMPLEX 16 COMPLEX 16 COMPLEX 16 COMPLEX 16 COMPLEX 16 COMPLEX 16 COMPLEX 16 REAL 8 REAL 8 REAL 8 REAL 8 REAL REAL 8 REAL 8 REAL 8 INTEGER 2 INTEGER INTEGER REAL 8 REAL 8 REAL 8 REAL 8 REAL 8 REAL 8 REAL 8 REAL 8 REAL 8 REAL 8 REAL 8 REAL 8 REAL 8 REAL 8 REAL 8 REAL REAL 8 COMPLEX COMPLEX 16 REAL REAL INTEGER Arguments Description REAL 8 Arcsin CHARACTER Current date REAL 8 REAL 8 Arctangent of quotient REAL 8 Arctangent INTEGER 2 Double precision value INTEGER REAL REAL 8 COMPLEX COMPLEX 16 INTEGER 2 Double complex value INTEGER REAL REAL 8 COMPLEX 16 COMPLEX REAL 8 REAL 8 COMPLEX 16 Conjugate REAL 8 Hyperbolic cosine REAL 8 Cosine REAL 8 REAL 8 Po
17. Octal integer Ow m Octal integer with leading zeros and minimum number of digits Zw Hexadecimal integer Zw m Hexadecimal with leading zeros and minimum number of digits Below are the nonrepeatable edit descriptors Type Descriptor Description Input control BN Blanks ignored in numerics BZ Blanks treated as zeros in numerics Scale factor kP Scaling for numeric editing 76 PromulaFortran Compiler User s Manual Hollerith nH Output Hollerith string Character output Output character string Output character string Skip spaces nX Position forward Numeric output SP Plus signs produced SS Plus signs suppressed S Plus signs suppressed Tabulation Tn Position forward or backward TRn Position forward TLn Position backward Format control Terminate format control End of record Indicates end of current input or output record Output control Suppress end of record on output In both tables uppercase letters indicate the type of conversion Lowercase letters indicate user supplied information that has the following meaning w Nonzero unsigned integer constant specifying the field width in number of character positions in the external record This width includes any leading blanks or signs decimal point and exponent d Unsigned integer constant specifying the number of digits to the right of the decimal point within the field On output all numbers are rounded e Nonzero unsigned integer constant specifying
18. SUBROUTINE W897 2 CALL DEMO A B 3 CALL DEMO A 3 warn897 for C897 The actual argument list contains less than the 2 argument s used when the subroutine DEMO was previously referenced or defined 4 END Comment 898 Message The statement labels on elseif else or endif may not produce the desired effect Example of message 898 If Line Nl Source 1 SUBROUTINE 11898 2 DATA A 2 0 3 IF A GE 2 THEN 4 I 3 5 GOTO 10 30 40 1 6 10 ELSEIF A LT 2 THEN 6 warn898 for C898 The statement labels on elseif else or endif may not produce the desired effect 7 2 8 30 ELSE 8 warn898 for C898 The statement labels on elseif else or endif may not produce the desired effect 9 4 10 40 ENDIF 10 warn898 for C898 The statement labels on elseif else or endif may not produce the desired effect TI END 296 PromulaFortran Compiler User s Manual Comment 899 Message The subprogram ident is not an intrinsic function Example of message 899 If Line Nl Source 2 warn899 for 2 warn899 for 3 4 5 Warning 900 Message SUBROUTINE W899 INTRINSIC INT IFIX IDINT XREAL FLOAT SNGL DBLE CMPLX ECHAR C899 The subprogram XREAL is not an intrinsic function C899 The subprogram ECHAR is not an intrinsic function 3 1415 WRITE END An implied do must be used for the assumed size array ident Example of message 900 If Line Nl Source 4 w
19. Use of the asterisk form of the adjustable dimension prevents subscript checking for the array so the user must be careful not to reference outside the array bounds Use of this form is preferable to the common practice of declaring arrays to have dimension 1 37 PromulaFortran Compiler User s Manual 3 8 3 Using COMMON Blocks Common blocks can be used to transfer values between a referencing program unit and a subprogram Common blocks can reduce the number of storage units required for a program by enabling two or more subprograms to share some of the same storage units The variables and arrays in a common block can be defined and referenced in all subprograms that contain a declaration of that common block The names of the variables and arrays in the common block can be different for each subprogram The association is by storage and not by name Common blocks cannot be used to pass data to intrinsic functions or statement functions the method used to pass data to these procedures is through an argument list A reference to data in a common block is valid if the data is defined and is the same type as the type of the name used in the main program or subprogram The exceptions to agreement between the type in common and the type of the reference are 1 either part of a complex entity can be referenced as real 2 character arrays may have different lengths and or dimensionality 38 PromulaFortran Compiler User s Manua
20. 02 First 647E 01 is read converted and placed in location second specification 7 2 exceeds the width of the second field by two characters The number 2 36 5 is read instead of 2 36 The specification error E7 2 instead of E5 2 caused the two extra characters to be read The number read 2 36 5 is a legitimate input number Since the second specification incorrectly took two digits from this number the specification for the third number is now incorrect The field 321 0277 is read The OPEN statement specifies that trailing blanks are to be treated as zeros therefore the number 321E 0200 is read converted and placed in location c Here again this is a legitimate input number which is converted and stored even though it is not the number desired Some additional examples of Ew d input specifications are shown below Input Field Specification Value Remarks 143 26 03 112 0 14326 subfields present 327 625 E7 3 327 625 No exponent subfield 000362745 17 36 27 Integer subfield only a minus sign and a plus sign appears instead of E 0003627E5 1 17 36 27 Integer subfield left of decimal contains minus sign only PANDAAN E4 1 0 All subfields empty E 06 E10 6 0 No integer or fraction subfield zero stored regardless of exponent field contents On output the width w must be sufficient to contain digits plus or minus signs decimal point E the exponent and bla
21. 20 PromulaFortran Compiler User s Manual A FORTRAN statement is composed of four different types of fields 1 A statement label field 2 A statement content field 3 A comment field 4 A continuation indicator field where the sequence of content comment continuation fields may repeat The line containing the statement label of the first statement field 1s the initial line and all other lines are continuation lines There are four distinctive formats that may be used to enter FORTRAN programs 1 Standard fixed format lines 2 Tab format lines 3 Normal free format lines 4 Continuation free format lines Blank lines are the same for all formats They may appear anywhere and are always ignored Comment lines have slightly different formats depending upon the format type however they may appear anywhere within a program unit including before or within a continuation line sequence For all formats the maximum length of the concatenation of the content fields for a given statement is 4096 characters excluding blanks and inline comments 3 1 1 Standard Fixed Format Lines The typical FORTRAN program uses standard fixed format Under this format a comment line is any line with a nonblank non numeric character in column 1 Statement lines have fields as follows Field Column s Statement label 1 5 Continuation 6 Content 7 endcol Comment endcol 1 The only variable aspect of this format is the value of endcol
22. 5 I W828C J 1 6 CALL W828C J I 6 warn828 for C828 The subroutine W828C has been previously referenced or defined as a function 7 END Note 829 Message The array ident has more than 7 dimensions Example of message 829 If Line Nl Source 1 SUBROUTINE W829 2 DIMENSION A 2 2 2 2 2 2 2 2 2 warn829 for N829 The array A has more than 7 dimensions 3 A 1 2 1 2 1 2 1 2 3 3 4 WRITE A 1 2 1 2 1 2 1 2 5 END 271 PromulaFortran Compiler User s Manual Note 830 Message An entry point is being defined within an if or do range Example of message 830 If Line Nl Source 1 SUBROUTINE W830 2 DO 10 I 1 3 Bs d ENTRY GEORGE 3 warn830 for N830 An entry point is being defined within an if or do range 4 1 10 CONTINUE END Warning 831 Message The variable ident subscript is of type type rather than integer Example of message 831 If Line Nl Source 1 SUBROUTINE W831 2 DIMENSION B 0 1 3 D B 0 0 3 warn831 for W831 The variable D subscript is of type real 4 rather than integer 4 END Note 832 Message The use of hexidecimal octal or binary constants is nonstandard Example of message 832 If Line Nl Source ah SUBROUTINE W832 2 IT20 77 2 warn832 for N832 The use of hexidecimal octal or binary constants is nonstandard 3 IP H FF 3 warn832 for N832 The use of hexidecimal octal or binary constants is nonstandard 4 WRITE IT IP 5
23. 6 STOP 7 END Warning 904 Message The formal argument list contains more than the number arguments used when this subprogram was previously referenced or defined Example of message 904 If Line Nl Source al PROGRAM W904 2 CALL SUB904 A B 3 STOP 4 END 5 SUBROUTINE SUB904 A B C D 5 warn904 for W904 The formal argument list contains more than the 2 arguments used when this subprogram was previously referenced or defined 6 RETURN 298 PromulaFortran Compiler User s Manual END 9 2 2 Fatal Preprocessor Errors Fatal preprocessor errors occur when requested or needed files cannot be opened or when available memory has been exhausted When a fatal error occurs the preprocessor exits to the control program No intermediate C files are produced File problems cannot be avoided but are usually simple to correct The preprocessor always gives the name of the file it is trying to open when such a problem occurs The preprocessor is as conservative as possible in its memory utilization Extremely large source files can be processed Typically other FORTRAN processors have much tighter memory constraints There are several command line options which allow control of memory allocation for systems with limited memory Refer to the chapter on using the compiler for a discussion of these options should the need arise The fatal errors have the same general form as the syntax messages except that the identify
24. Example of message 314 If Line Nl Source 1 FUNCTION ERR314 2 COMMON J 3 EQUIVALENCE J N 4 PARAMETER 1 4 err314 for E314 The equivalenced variable N may not be redefined as a parameter 5 ERR314 K J N 6 END 260 PromulaFortran Compiler User s Manual Error 315 Message The object ident may not be redefined as being external Example of message 315 If Line Nl Source 1 SUBROUTINE ERR315 2 PARAMETER SINE 3 3 3 EXTERNAL SINE 3 err315 for E315 The parameter SINE may not be redefined as being external 4 WRITE SINE K 4 err315 for E105 The parameter identifier SINE may not be used to represent a function 5 END Error 316 Message The common variable ident may not be redefined as being external Example of message 316 If Line Nl Source 1 SUBROUTINE ERR316 2 COMMON SINE 3 EXTERNAL SINE 3 err316 for E316 The common variable SINE may not be redefined as being external 4 WRITE SINE 5 END Note 801 Message The statement statement is nonstandard Example of message 801 If Line Nl Source 1 PROGRAM W801 2 1 FORMAT I5 3 ACCEPT 1 IVAL 3 warn801 for N801 The accept statement is nonstandard 4 PRINT IVAL 4 warn801 for N801 The print statement is nonstandard 5 5 6 END 261 PromulaFortran Compiler User s Manual Note 802 Message Omitting the period from the token format specification
25. INTEGER INTEGER INTEGER REAL INTEGER REAL 8 INTEGER COMPLEX INTEGER COMPLEX 16 INT INTEGER INTEGER 2 Integer value INTEGER INTEGER INTEGER REAL INTEGER REAL 8 INTEGER COMPLEX INTEGER COMPLEX 16 ISIGN INTEGER INTEGER INTEGER Transfer of sign DABS INTEGER 2 INTEGER 2 Absolute value DDIM INTEGER 2 INTEGER 2 INTEGER 2 Positive difference I2MAX0 INTEGER 2 INTEGER 2 Maximum value I2MINO INTEGER 2 INTEGER 2 Minimum value DMOD INTEGER 2 INTEGER 2 INTEGER 2 Remainder DNINT INTEGER 2 REAL Nearest integer DSIGN INTEGER 2 INTEGER 2 INTEGER 2 Transfer of sign LEN INTEGER CHARACTER Number of characters LGE LOGICAL 2 CHARACTER CHARACTER Lexically greater or equal LGT LOGICAL 2 CHARACTER CHARACTER Lexically greater LLE LOGICAL 2 CHARACTER CHARACTER less or equal LLT LOGICAL 2 CHARACTER CHARACTER Lexically less LOGIO REAL REAL Logarithm base 10 REAL 8 REAL 8 COMPLEX COMPLEX COMPLEX 16 COMPLEX 16 LOG REAL REAL Natural log REAL 8 REAL 8 COMPLEX COMPLEX COMPLEX 16 COMPLEX 16 135 PromulaFortran Compiler User s Manual Table 6 1 Intrinsic Functions Name 1 MINO MIN1 MIN MOD NINT REAL SIGN SINH SIN SNGL SQRT TANH TAN TIME Type INTEGER INTEGER 2 INTEGER INTEGER 2 INTEGER REAL REAL 8 INTEGER INTEGER 2 INTEGER INTEGER 2 INTEGER REAL REAL 8 INTEGER 2 INTEGER REAL REAL
26. If Line Nl Source 1 PROGRAM ERR286 251 PromulaFortran Compiler User s Manual 2 REAL METERS 3 REAL PARIMETER INCHES PER METER 39 37 3 err286 for E286 The attribute specifier PARIMETER is not defined 4 READ METERS 5 PRINT METERS meters METERS INCHES PER METER inch 6 END PROGRAM ERR286 Error 287 Message The attribute specifier token is defined as part of the standard but is not implemented in this dialect Example of message 287 If Line Nl Source 1 PROGRAM ERR287 2 REAL METERS 3 REAL PARAMETER PRIVATE INCHES_PER_METER 39 37 3 err287 for E287 The attribute specifier PRIVATE is defined as part of the standard but is not implemented in this dialect 4 READ METERS 5 PRINT METERS meters METERS INCHES_PER_METER inche 6 END PROGRAM ERR287 Error 288 Message The dimensioned variable ident may not be redefined as being external Example of message 288 If Line Nl Source 1 SUBROUTINE ERR288 2 DIMENSION SINE 100 3 EXTERNAL SINE 3 err288 for E288 The dimensioned variable SINE may not be redefined as being external WRITE SINE 5 END Error 289 Message The type of object ident is undefined Example of message 289 If Line Nl Source 1 PROGRAM ERR289 2 IMPLICIT NONE 3 INTEGER IVAL JVAL 4 IVAL 100 5 JVAL 200 252 PromulaFortran Compiler User s Manual 6 KVAL IVAL JVAL 6 err28
27. PromulaFortran Compiler User s Manual 6 STOP 7 END Error 107 Message The second quantity token in the complex constant is not numeric Example of message 107 If Line Nl Source 1 PROGRAM ERR107 2 COMPLEX A 3 0 0 6 1 3 errl07 for E107 The second quantity I in the complex constant is not numeric 4 END Error 108 Message A complex constant is terminated by the symbol token rather than a right parenthesis Example of message 108 If Line Nl Source 1 PROGRAM ERR108 2 COMPLEX A 3 0 0 6 71 3 errl108 for E108 A complex constant is terminated by the symbol I rather than a right parenthesis 4 END Error 109 Message There are too few right parentheses in the expression terminated by the symbol token Example of message 109 If Line Nl Source 1 PROGRAM ERR109 2 COMPLEX A 3 0 0 6 7 5 5 3 errl09 for E109 There are too few right parentheses in the expression terminated by the symbol END STATEMENT 4 END 193 PromulaFortran Compiler User s Manual Error 110 Message An improper symbol token is being used to introduce a factor Example of message 110 If Line Nl Source Jd PROGRAM ERR110 2 LOGICAL FINISH STARTED 3 FINISH NAT STARTED 3 errl110 for E110 An improper symbol is being used to introduce factor 4 END Error 111 Message The binary type type cannot be combined with the binary type type in this
28. WRITE fn iolist As a simple example consider PROGRAM PRINT CHARACTER B 3 A 1 2 B YES N 19 PRINT 4 A B N 4 FORMAT G20 6 A 15 PRINT 50 50 FORMAT END OF FILE STOP END which will convert the values of A B and N into a coded record using format 4 and will then write that record to the OUTPUT file followed by a record containing END OF FILE See also The WRITE statement describes format specifications and iolists in detail See that statement for details on these 5 43 PROGRAM Statement The PROGRAM statement defines the program name and begins a new program definition Syntax 116 PromulaFortran Compiler User s Manual PROGRAM name fpar fpar Where name is a symbolic name fpar isa token of any type Description The PROGRAM statement defines a program name and begins a new program definition The name of the program and any parameters associated with it are purely for documentation and are not used by the system in any way 5 44 PUNCH Statement The PUNCH statement transfers formatted information from the values named in the input output list to the PUNCH file Syntax PUNCH fn iolist Where fn is a format specification iolist is an output list Notes The PUNCH statement is a special case of the formatted WRITE statement Description The PUNCH statement transfers formatted information from the values specified in the input output list to t
29. INTEGER 2 Description 2 al a2 an is a specific function that returns the value of the largest argument All arguments are short integer and the result is short integer See also AMAX1 DMAX1 and MAX1 163 PromulaFortran Compiler User s Manual 6 84 I2MIN0 Minimum Value Synopsis INTEGER 2 FUNCTION I2MINO INTEGER 2 Description I2MINO al a2 an is a specific function that returns the value of the smallest argument All arguments are short integer and the result is short integer See also AMINO AMINI DMINI MINO MINI and MIN 6 85 RMOD Remainder Synopsis INTEGER 2 FUNCTION I2MOD INTEGER 2 INTEGER 2 Description I2MOD num dem computes the value of the remainder of num divided by dem If dem is zero then the result is zero Both arguments are the result are short integer See also AMOD DMOD and MOD 6 86 IZNINT Nearest Integer Synopsis INTEGER 2 FUNCTION I2NINT REAL Description I2NINT a is a specific function that returns a short integer result from a real argument It computes the nearest integer to its real argument In particular the result is int 0 5 if a gt 0 0 and int a 0 5 if a lt 0 0 where int is the FORTRAN intrinsic function INT See also ANINT DNINT IDNINT and NINT 6 87 I2SIGN
30. Message The common block ident has a type variable ident and a type variable ident Example of message 823 If Line Nl Source PROGRAM W823 COMMON ALPHA IVAL RVAL SVAL INTEGER IVAL REAL RVAL CHARACTER 10 SVAL READ IVAL RVAL SVAL CALL HELPER STOP 9 END 9 warn823 for N823 The common block ALPHA has a real 4 variable RVAL and a character 10 variable SVAL o ES Warning 824 Message The variable ident substring subscript number is not in the range of number to number Example of message 824 If Line Nl Source all PROGRAM W824 2 CHARACTER 10 NAME HOME SON 3 0 9 Freddy 3 warn824 for W824 The variable NAME substring subscript 0 is not in the 269 PromulaFortran Compiler User s Manual range of 1 to 10 4 HOME 1 13 Columbus Ohio 4 warn824 for W824 The variable HOME substring subscript 13 is not in the range of 1 to 10 5 SON 5 1 5 warn824 for W824 The variable SON substring subscript 1 is not in the range of 5 to 10 6 STOP 7 END Warning 825 Message The subscript value number for subscript number of array ident is not in the range of number to number Example of message 825 If Line Nl Source 1 SUBROUTINE W825 2 PARAMETER IA 3 3 REAL A 0 3 4 4 A 1 4 0 4 warn825 for W825 The subscript value 1 for subscript 1 of array A is not in the range of 0 to 3 5 A 3 IA 2 0 5 warn825 fo
31. R EQUIVALENCE 5 is not legal Second an EQUIVALENCE statement must not cause a common block to be extended by adding storage words before the first storage word of the common block On the other hand common storage words are added at the end of the common block For example COMMON X A REAL B 5 EQUIVALENCE A B 4 is not legal whereas COMMON X A REAL B 5 EQUIVALENCE B 1 be used to extend the common block 5 26 EXTERNAL Statement The EXTERNAL statement defines a symbolic identifier as referring to an external procedure Syntax EXTERNAL proc proc Where proc isasymbolic name Description The EXTERNAL statement is used to identify a symbolic name as representing an external procedure and to permit such a name to be used as an actual argument The following shows an example of this SUBROUTINE CHECK EXTERNAL LOW HIGH CALL AR LOW VAL CALL AR HIGH VAL UBROUTINE AR FUNC VAL AL FUNC VAL Uu utuc ut zZ D EAL FUNCTION LOW X AL FUNCTION HIGH X 73 PromulaFortran Compiler User s Manual The names Low and HIGH are declared as external In the first call to subroutine AR LOW is passed as an actual argument and the function reference FUNC VAL is equivalent to LOW VAL In the second call to subroutine AR the function reference
32. err152 for E152 The positive if branch is IJK rather than an integer constant 207 PromulaFortran Compiler User s Manual OY 4 W Error 153 Message 10 PRINT Hello 15 PRINT Goodbye 20 STOP END The assign statement keyword is followed by the symbol token rather than an integer statement number Example of message 153 If Line Nl Source C ho 1 SUBROUTINE ERR153 WHICH INTEGER 2 LABEL WHICH 1 FORMAT Hello World 15 2 FORMAT Hello America I5 IF WHICH GT 0 THE ASSIGN 90001 TO ELSE ASSIGN 90002 TO ENDIF GOTO LABEL 90001 ASSIGN WHICH TO LABE LABEL LABEL Li 11 err153 for E153 The assign statement keyword is followed by the symbol gt WN Error 154 Message WHICHTOLABEL rat GOTO 90003 90002 ASSIGN 2 TO LABEL 90003 WRITE LABEL WHICH RETURN END her than an integer statement number A complex constant is terminated by the symbol token rather than a right parenthesis Example of message 154 If Line Nl Source 3 errl54 for 4 5 6 Error 155 Message PROGRAM ERR154 COMPLEX A 3 DATA A 0 0 1 0 2 0 4 0 0 0 0 0 E154 A complex constant is terminated by the symbol rather than a right parenthesis WRITE A STOP END The constant token of type type is being read into variable ident of type type 208 PromulaFortran Compiler User s
33. is a object and not a variable Example of message 305 If Line Nl Source 1 SUBROUTINE ERR305 2 A I 3 7 1 3 ENTRY SUM A T 3 err305 for E305 The entry parameter identifier A is a statement function and not a variable 4 ALPHA A I 5 END Error 306 Message The value token is not a valid repeat count Example of message 306 If Line Nl Source 1 SUBROUTINE ERR306 2 REAL A 10 B 10 3 DATA A 0 0 0 10 0 0 3 err306 for E306 The value 0 is not a valid repeat count 4 DATA B I I 1 10 0 1 0 4 err306 for E306 The value 0 is not a valid repeat count 5 WRITE A B 6 END Error 307 Message The illegal character token is in the binary constant token Example of message 307 If Line Nl Source 1 SUBROUTINE ERR307 2 IT B 0101 d IP B 0121 3 err307 for E307 The illegal character 2 is in the binary constant B 0121 4 WRITE IT IP 5 END Error 308 Message The illegal character token is in the octal constant token 258 PromulaFortran Compiler User s Manual Example of message 308 If Line Nl Source 1 SUBROUTINE ERR308 2 IT O 0101 3 IP O 0191 3 err308 for E308 The illegal character 9 is in the octal constant 0701915 4 5 END Error 309 Message The illegal character token is in the hexidecimal constant token Example of message 309 If Line Nl Source 1 SUBROUTINE ERR309 2 IT H 0101 3
34. iteration count does not equal 0 its control variable retains its most recent value unless it has become undefined Transfer into the range of an inactive DO loop from outside the range is strongly discouraged 64 PromulaFortran Compiler User s Manual 5 14 2 Nested DO Loops When a DO loop entirely contains another DO loop the grouping is called a DO nest The range of a DO statement can include other DO statements providing the range of each inner DO is entirely within the range of the containing DO statement The last statement of an inner DO loop must be either the same as the last statement of the outer DO loop or must occur before it A terminal statement that is shared by more than one DO loop can be referenced in a GO TO or IF statement from within the range of the innermost loop only If the terminal statement is referenced from any loop other than the innermost loop results are undefined 5 15 DOUBLE COMPLEX Statement The DOUBLE COMPLEX statement defines some user defined entity to be of type double complex Syntax DOUBLE COMPLEX name name Where name has one of the forms var c array d d clist var is a variable function name symbolic constant or dummy procedure array isan array name d specifies the bounds of a dimension clist isa list of constants or symbolic constants specifying the initial values Each item in the list can take the form r c c i
35. may be backspaced When the file is positioned at beginning of information this statement acts as a do nothing statement If an error occurs as a result of the backspace and if neither err nor status are supplied then execution will terminate with an error code set If either or both are supplied the execution continues despite any errors If supplied the status lvalue receives the runtime error code for the error condition encountered or a zero if no error occurred If err is supplied then execution will branch to the statement labeled by it if an error occurs Examples The files associated with units 1 through 4 backspaced one record via the following If any errors occur the program will terminate abnormally lun 1 4 1 BACKSPACE lun The following also attempts to backspace units 1 through 4 however an error reports the error code and the unit number Note that the unit number need not be the first specification when it is preceded by the UNIT symbol DO 2 lun 1 4 2 BACKSPACE IOSTAT UNIT lun ERR 15 WRITE Backspace successful STOP 15 WRITE Backspace error j on unit lun See also The discussion of the OPEN statement describes the FORTRAN file system in general including the different file types 5 3 BLOCK DATA Statement 49 PromulaFortran Compiler User s Manual The BLOCK DATA statement introduces a BLOCKDATA subprogram Syntax BLOCK
36. 1 4 51 31 52 32 1 53 33 2 1 54 394 n3 n2 n1 5 51 31 52 32 1 53 33 2 1 54 34 3 2 1 55 15 4 3 2 1 6 51 31 52 32 1 53 33 2 1 54 34 3 2 1 s5 j5 n4 n3 n2 n1 56 36 5 4 3 2 1 7 51 31 52 32 1 53 33 2 1 54 34 3 2 1 55 15 4 3 2 1 s6 j6 n5 n4 n3 n2 nl s7 j7 n6 n5 n4 n3 n2 n1 Where ji Lower bound of dimension i ki Upper bound of dimension i ni Size of dimension i ni ki ji 1 32 PromulaFortran Compiler User s Manual si Value of the subscript expression specified for dimensioni 3 6 Character Substrings When a character variable or character entity is declared the entire character string can be defined and referenced Specific parts of the character string can also be defined or referenced with character substring references A character entity must be declared with the CHARACTER statement The declaration of a character entity specifies the length in characters 3 6 1 Substring References If the name of a character entity is used in a reference the value is the current value of the entire string A reference to part of a string is written as a character substring whose syntax is shown below char first last Where char 15 the name of a character variable or array and can be an array element reference first specifies an integer real or double precision exp
37. 163 If Line Nl Source 1 PROGRAM ERR163 2 READ I 3 GO TO 10 20 I 3 errl63 for E163 The computed goto is opened by the symbol rather than a left parenthesis 4 10 PRINT Hello 5 20 STOP 6 END Error 164 Message An element in the computed goto statement list token is not a statement label Example of message 164 If Line Nl Source 211 PromulaFortran Compiler User s Manual 1 PROGRAM ERR164 2 READ I 3 GO TO 10 20 I 3 errl64 for E164 An element in the computed goto statement list 70 is not a statement label 4 10 PRINT Hello 5 20 STOP 6 END Error 165 Message The computed goto is closed by the symbol token rather than a right parenthesis Example of message 165 If Line Nl Source 1 PROGRAM ERR165 2 READ I 3 GO TO 10 20 I 3 errl 65 for E165 The computed goto is closed by the symbol I rather than a right parenthesis 4 10 PRINT Hello 5 20 STOP 6 END Error 166 Message The statement introduced by the keyword token is not a valid statement type Example of message 166 If Line Nl Source 1 PROMGRA ERR166 1 errl66 for E166 The statement introduced by the keyword PROMGRAERR166 is not a valid statement type 2 DIMENSION AMAT 10 IMAT 5 3 EQUIVALENCE IMAT 2 4 4 STOP 5 END Error 167 Message The call statement subroutine name is token and not an identifier Example of message 167 If Line Nl Sour
38. 2 INCLUDE 238DDL 2 err238 for E238 The include statement filename 238 is not a valid string constant or identifier symbol 3 STOP 4 END Error 239 Message The include statement file token cannot be opened Example of message 239 If Line Nl Source ai PROGRAM ERR239 2 INCLUDE DDL239 INC 2 err239 for E239 The include statement file DDL239 INC cannot be opened 3 STOP 4 END Error 240 Message The length specification number associated with ident is not available for the type type Example of message 240 If Line Nl Source al PROGRAM ERR240 2 LOGICAL 4 ALPHA BETA 3 2 err240 for E240 The length specification 3 associated with BETA is not available for the logical type 3 STOP 4 END Error 241 Message The type statement length specification token is not available for the type type 236 PromulaFortran Compiler User s Manual Example of message 241 If Line Nl Source 1 PROGRAM ERR241 2 LOGICAL 3 ALPHA BETA 2 2 err241 for E241 The type statement length specification 3 is not available for the logical 4 type 3 STOP 4 END Error 242 Message The character statement variable length string token is not a constant Example of message 242 If Line Nl Source i PROGRAM ERR242 2 CHARACTER N ALPHA BETA 2 err242 for E242 The character statement variable length string N is not a constant 3 STOP 4 END Error 243 Message The character statement vari
39. 3 VAL I J 99 5 3 err282 for E282 The left hand side of the assignment statement begins with the symbol which is not a valid identifier symbol 4 STOP 5 END Error 283 Message The right hand side of the assignment is preceded by the symbol token rather than an equals sign 250 PromulaFortran Compiler User s Manual Example of message 283 If Line Nl Source 1 PROGRAM ERR283 2 C 2 err283 for E283 right hand side of the assignment is preceded by the symbol rather than an equals sign 3 STOP 4 END Error 284 Message The equivalenced variable ident may not be redefined as being external Example of message 284 If Line Nl Source 1 SUBROUTINE ERR284 K 2 EQUIVALENCE ISINE SINE 3 EXTERNAL SINE 3 err284 for E284 The equivalenced variable SINE may not be redefined as being external WRITE SINE 5 END Error 285 Message The attribute specifier list contains the symbol token where a comma or is expected Example of message 285 If Line Nl Source 1 PROGRAM ERR285 2 REAL METERS 3 REAL PARAMETER INCHES PER METER 39 37 3 err285 for E285 The attribute specifier list contains the symbol where a comma or is expected 4 READ METERS 5 PRINT METERS meters METERS INCHES PER METER inche 6 END PROGRAM ERR285 Error 286 Message The attribute specifier token is not defined Example of message 286
40. 30 1 Unconditional GOTO The unconditional GOTO branches directly to a statement whose label is specified Syntax GOTO slab Where slab isthe label of an executable statement Description As an example in the following 10 AAA B Z B X Y IF 20 20 30 20 Z A GOTO 10 30 2 control transfers to statement 10 when the GOTO statement executes 5 30 2 Computed GOTO Statement The computed GOTO statement transfers control to the statement identified by one of the specified labels Syntax GOTO slab slab lexp Where slab is the label of an executable statement 94 PromulaFortran Compiler User s Manual exp isan arithmetic expression Description The label selected is determined by the value of the expression If exp has a value of 1 control transfers to the statement identified by the first label in the list if exp has a value of 4 control transfers to the statement identified by the fourth label in the list and so forth The value of exp is truncated and converted to integer if necessary If the value of exp is less than 1 or greater than the number of labels in the list execution continues with the statement following the computed GOTO As an example in the following GOTO 10 20 30 20 L the next statement executed is 10 if L equals 1 20 if L equals 2 or 4 and 30 if L equals 3 5 30 3 Assigned GOTO Statement The assigned GOTO statement transfers control to the executable statement las
41. 5 S1 Value STRING Note that the substring reference 51 has the same effect as the reference 51 since all characters in the string are referenced 3 6 2 Substrings and Arrays 33 PromulaFortran Compiler User s Manual If a substring reference is used to select a substring from an array element of a character array the combined reference includes specification of the array element followed by specification of the substring For example CHARACTER 8 75 5 CHARACTER 4 RSEN ZS 4 5 6 FG RSEN ZS 1 4 The first reference refers to characters 5 and 6 in element 4 of array 25 The second reference refers to the first four characters of the first element of array ZS 3 7 Statement Order The order of various statements within the program unit is relatively free given the rules of FORTRAN A PROGRAM statement can appear only as the first statement in a main program The first statement of a subroutine function or block data subroutine is respectively a SUBROUTINE statement FUNCTION statement or BLOCK DATA statement The END statement is the last statement of each of the preceding program units If a variable is to be explicitly typed then that typing must precede the first reference to that variable either in a DATA statement or in an executable statement Statement function definitions PARAMETER definitions and NAMELIST specifications must precede their first reference Comments c
42. 5 B 5 8 B 6 9 B 7 10 B 8 11 5 B 9 12 B 10 13 B 11 277 PromulaFortran Compiler User s Manual 14 15 16 17 18 19 20 21 22 23 24 25 26 27 3 warn847 for 28 29 Comment 848 Message The variable ident in Aou Oe oO ur N847 More than 19 continuation lines in one statement is nonstandard WRITE A END the assign is either an argument or is in common or is equivalenced Example of message 848 If Line Nl Source 4 warn848 for 5 6 6 warn848 for 7 8 8 warn848 for 10 11 Note 849 Message SUBROUTINE W848 IOTA COMMON JOT JOT EQUIVALENCE JOT NAM ASSIGN 10 TO IOTA C848 The variable IOTA in the assign is either an argument or is in common or is equivalenced GOTO IOTA 10 ASSIGN 20 TO JOT C848 The variable JOT in the assign is either an argument or is in common or is equivalenced GOTO JOT 20 ASSIGN 30 TO NAM C848 The variable NAM in the assign is either an argument or is in common or is equivalenced GOTO NAM 30 CONTINUE END The use of debugging comments is nonstandard Example of message 849 If Line Nl Source 2 warn849 for PROGRAM W849 I 100 2 N849 The use of debugging comments is nonstandard 278 PromulaFortran Compiler User s Manual 3 D WRITE Ay CEX4 IT5 I 4 STOP 5 END Note 850 Message The use
43. 6 39 DATE Current Date Synopsis SUBROUTINE DATE CHARACTER Description DATE a obtains the current date as set within the system The date is returned as an 8 character character string of the form mm dd yy If the argument is too short to receive eight characters the return value is truncated If the argument is longer that eight then trailing positions are padded with blanks See also TIME 6 40 DATAN2 Arctangent of Quotient Synopsis REAL 8 FUNCTION DATAN2 REAL 8 REAL 8 Description DATAN2 al a2 is a specific function that returns the arctangent of a2 divided by a1 The result is expressed in radians The result and argument are both double precision If both arguments are zero the result is zero See also ATAN2 6 41 DATAN Arctangent Synopsis REAL 8 FUNCTION DATAN REAL 8 Description DATAN a is a specific function that returns an arctangent The result is expressed in radians The result and argument are both double precision See also ATAN 6 42 DBLE Double Precision Value Synopsis 149 PromulaFortran Compiler User s Manual REAL 8 FUNCTION DBLE INTEGER 2 REAL 8 FUNCTION DBLE REAL REAL 8 FUNCTION DBLE REAL 8 REAL 8 FUNCTION DBLE COMPLEX REAL 8 FUNCTION DBLE COMPLEX 16 Descr
44. ABSOLUTE V ALBUE n 137 6 2 TT 137 6 3 AMAG IMAGINARY PARE rrt eere see oreste tune 137 GAAN rete eee 138 65 ArOGIO LOGARTEHM BASE 10 dd deese ee Eu eror 138 66 AL OG e a 138 6 7 AMAX0 MAXIMUM V ADLUE odere ted tore ese ore eee se eee eee ore deed eve dee ee oe eee ee 139 6 8 AMA X 1 MAXIMUM VALUE eene eub eored eee eed eee eee oe beo eee ee deed eese bee beo ed te eee nbn 139 6 9 AMINO MINIMUM WALUE odere teer ever ore oe tee tb ere dd ota dosed eure eee e ee eu eere rece soe tee 139 6 MINIMUM M ALUE k odere tete eroe rette ee e Peta eese 139 GH AMOD REMAINDERING edes tes dessus 140 6 12 140 6 13 ANINT NEAREST WHOLE 140 Oo Pa ASINSARCSINES 21 heh Dei ies 141 6 15 ATAN ARCTANGENT OF QUOTIENT obe
45. ASSIGN 105 23 90 GOTO IOT 105 65 21 warn894 for C894 The do loop ends on a transfer 22 105 DO 110 I 1 6 23 A I 24 110 DO 115 J 1 5 i Z9 2 WRITE J 26 2 115 CONTINUE 26 warn894 for C894 The do loop ends on a transfer 27 END Comment 895 Message The character token has already been assigned type type Example of message 895 If Line Nl Source 1 SUBROUTINE W895 2 IMPLICIT INTEGER 2 REAL I J 2 warn895 for C895 The character I has already been assigned type integer 4 2 warn895 for C895 The character J has already been assigned type integer 4 3 A 1 4 END Note 896 Message References to nonconstants like object ident in parameter statements is nonstandard Example of message 896 statement statement statement 295 PromulaFortran Compiler User s Manual If Line Nl Source 1 SUBROUTINE W896 2 PARAMETER 1 3 2 warn896 for N896 References to nonconstants like variable J in parameter statements is nonstandard 3 PARAMETER L J 3 3 warn896 for N896 References to nonconstants like variable J in parameter statements is nonstandard 4 PARAMETER A 1 3 B 1 A 5 J 99 6 WRITE I K L A B Z W Y 7 END Comment 897 Message The actual argument list contains less than the number argument s used when the object ident was previously referenced or defined Example of message 897 If Line Nl Source
46. Compiler User s Manual An equivalence relation is being assigned to the subprogram argument ident Example of message 214 If Line Nl Source 1 SUBROUTINE ERR214 AMAT 2 DIMENSION AMAT 10 IMAT 5 3 EQUIVALENCE 2 4 3 err214 for E214 An equivalence relation is being assigned to the subprogram argument AMAT 4 STOP 5 END Error 215 Message The size of dimension number ident of the equivalence element ident has not yet been resolved Example of message 215 If Line Nl Source 1 PROGRAM ERR215 2 DIMENSION AMAT N IMAT 5 3 EQUIVALENCE IMAT 2 AMAT 4 3 err215 for E215 The size of dimension 1 N of the equivalence element AMAT has not yet been resolved 4 STOP 5 END Error 216 Message The equivalence element ident is a object rather than a variable Example of message 216 If Line Nl Source 1 SUBROUTINE ERR216 2 EQUIVALENCE E 1 F 3 COMMON D D 4 EQUIVALENCE A B 5 DIMENSION B 20 6 2 1 1 2 7 WRITE B D E 8 END 8 err216 for E216 The equivalence element A is a function rather than a variable Error 217 Message 228 PromulaFortran Compiler User s Manual The equivalence element subscript specification token for ident is not a constant Example of message 217 If Line Nl Source 1 PROGRAM ERR217 2 DIMENSION AMAT 10 IMAT 5 3 EQUIVALENCE IMAT 2 AMAT N 3 err217 for E217 The eq
47. DBA variables Again we compile and link this code to produce a new executable EXPO EXE The size of this file is smaller than the one we produced earlier by about 2400 bytes the number of bytes needed to store the 12 vector variables of the program each storing 50 values or 200 bytes which are now stored in the EXPO DBA file instead see Lines 11 22 of the EXPO PRM datafile description 8 6 Syntax of the Globals File In most other FORTRAN host systems the instructions to access information from a database must be placed directly in the FORTRAN source code To add value to the FORTRAN application it must be rewritten and typically be re engineered The philosophy behind the design of the PROMULA host system is that the original FORTRAN source code not be changed rather the compiler itself is instructed to make certain variables accessible to the database manager during the compilation step It is the PROMULA FORTRAN compiler that adds value to the application not the programmer Any variable in a FORTRAN program can be uniquely defined in terms of its identifier and the global symbol dominating it Here global symbols are either common area names or subprogram names Given this observation the syntax of the globals file such as the one shown above is straight forward The first line of the globals file simply contains the name of the PROMULA database to be used to contain the program values Remember that the compiler does not actual
48. END 8 3 The Initial Compilation The filename for this program is EXPO FOR We compile it by simply entering the following at the operating system command level pf expo for o expo This produces an executable file When we run this program it reads the input data from an input text file does some calculations and writes the output report on an output text file EXPO DAT expo EXPO RPT The input and output datafiles for this program EXPO DAT and EXPO RPT are shown in Figures 8 3 and 8 4 respectively ENTIAL AREA INHALATION ADULT CHILE 14 DAY 50 PPM 100M DOWNWIND METER MONTH B DH RH DN ee DN Bj DOanrPFNNNNNWWNHNNE rn U 1 1 5 1 01 01 01 01 01 01 01 01 01 01 01 01 01 01 01 01 01 Or e e 1 ION o Figure 8 3 Input Data for EXPO HK KKK HK KKK HK HK EXPOSURE ANALYSIS MODEL VERSION 1 Ok kk kk ek RESIDENTIAL AREA INHALATION ADULT CHILD 14 DAY 50 PPM 100M DOWNWIND MASS M UNITS UG LENGTH L UNITS METER TIME T UNITS MONTH INHALATION EXPOSURE ROUTE 182 PromulaFortran Compiler User s Manual MONTH AMBIENT YEAR CONCENTRATION M L 3 C
49. Fs72 which is that good old format referred to above Ft Selects tab format which comes from VAX FORTRANS Ff Selects the free form format which is relatively typical of those FORTRANS that accepted terminal input Fv Selects the VS FORTRAN free form format F9 Selects the Fortran 90 free format 2 5 4 Default Local Variable Storage Type There are two fundamentally different ways in which local subprogram variables can be allocated to memory First there is static storage This is the default storage for local subprogram variables It is selected via the Ss option It is equivalent to the normal FORTRAN memory allocation It is wasteful in that local variables all have unique memory locations but simple to implement Also static storage has memory between calls This means that local variables can retain their values between calls to their code Second there is auto storage This storage is selected vi the Sa option Auto storage is allocated to the stack each time a subprogram is called When the subprogram exits the storage is returned to the stack for use by other subprograms The advantages of auto storage are that it is fast easy to use and economizes on total storage used The disadvantages are that it has no memory and that it is very limited on PC platforms where program stacks are typically quite short and always less than 64K Note that local variables that are initialized via a DATA statement have static storage as their
50. IF statements are arithmetic IF logical IF and block IF See also The ELSEIF ELSE and ENDIF statements which discuss the other components of the block IF statement 5 31 1 Arithmetic IF Statement The arithmetic IF statement transfers control to one of three labeled statements depending on the value of an expression Syntax IF exp slabl slab2 slab3 Where exp is an arithmetic expression 1 1 labels of executable statements slab2 slab3 Description If the value of exp is negative control transfers to the first statement label if exp is 0 control transfers to the second statement label if exp is greater than 0 control transfers to the third statement label 96 PromulaFortran Compiler User s Manual As an example in the following READ 5 10 I J K N 10 FORMAT 10X 414 IF I N 3 4 6 3 ISUM J K 6 CALL ERROR1 WRITE 6 2 ISUM 2 FORMAT 110 4 END If 1 is less than N control transfers to statement 3 else if 1 equals N control transfers to statement 4 else control transfers to statement 6 5 31 2 Logical IF Statement The logical IF statement allows for conditional execution of a statement Syntax IF exp stat Where exp is a logical expression stat is any executable statement except a DO block IF ELSE ELSE IF END END IF or another logical IF statement Description If the value of exp is true statement stat is executed If the value of exp is false sta
51. IOPT EQ 0 READ LUN1 1020 VBL FREQ DUR F IOPT EQ 1 READ LUN1 1020 VBL 6 WRITE OUTPUT HEADER FOR SELECTED EXPOSURE ROUTE 180 PromulaFortran Compiler User s Manual WRITE LUN2 2001 WRITE LUN2 2010 WRITE LUN2 2011 WRITE LUN2 2020 READ NUMBER OF ENTRIES IN TIMESERIES READ LUN1 1010 NUM DO 30 I 1 NU IF IOPT EQ 0 READ LUN1 1030 IDAY I IMNTH I IYR I CONC 1 IF IOPT EQ 1 READ LUN1 1030 IDAY I IMNTH I IYR I CONC I FREQ DUR IF IOPT EQ 2 READ LUN1 1030 IDAY I IMNTH I IYR I CONC I VBL FREQ DUR Continued on next page CONVERT UNITS OF INPUT CONCENTRATION CONC I CONC I XMULT COMPUTE SEVERITY PERIODICITY EXPOSURE AND INTAKE SEVER I CONC I VBL PERD I FREQ DUR EXPOS I SEVER I PERD 1 1 EXPOS I ABS COMPUTE CUMULATIVE AND AVERAGE VALUES CUMEV CUMEV EXPOS 1 1 CUMEV AVGE I CUMEV I CUMIV CUMIV TAKE I CUMI I CUMIV AVGI I CUMIV I WRITE LUN2 2030 IDAY I IMNTH I IYR I CONC I EXPOS 1 I AVGE I I CUMI I
52. Manual Example of message 155 If Line Nl Source 1 PROGRAM ERR155 2 DIMENSION A 3 3 DATA A 0 0 1 0 2 0 4 0 0 0 0 0 3 errl55 for E155 The constant 0 0 1 0 of type complex 8 is being read into variable A of type real 4 4 WRITE A 5 STOP 6 END Error 156 Message The do statement dummy variable identifier is the symbol token rather than an identifier Example of message 156 If Line Nl Source 1 PROGRAM ERR156 2 DO 10 1 5 2 errl56 for E156 The do statement dummy variable identifier is the symbol 5 rather than an identifier 3 PRINT I 4 10 CONTINUE 5 STOP 6 END Error 157 Message The do statement dummy ident is a variable of type type Example of message 157 If Line Nl Source 1 PROGRAM ERR157 2 CHARACTER 4 I 3 DO 10 I 1 5 3 errl57 for E157 The do statement dummy I is a variable of type character 4 4 PRINT I 5 10 CONTINUE 6 STOP 7 END Error 158 Message The do statement dummy is followed by the symbol token rather than an equals sign 209 PromulaFortran Compiler User s Manual Example of message 158 If Line Nl Source 1 PROGRAM ERR158 2 INTEGER 4 I 3 10 I 1 5 3 err158 for E158 The do statement dummy is followed by the symbol rather than an equals sign PRINT I 10 CONTINUE STOP END d Error 159 Message The object ident being used to control the data range is not an integer consta
53. Message A value of one will be assumed for the zero format count field Example of message 882 If Line Nl Source 1 SUBROUTINE W882 2 WRITE 6 3 6 FORMAT 3X 0F10 6 3 warn882 for C882 A value of one will be assumed for the zero format count field 4 WRITE 8 5 8 FORMAT 3X 0 ABC 3X 5 warn882 for C882 A value of one will be assumed for the zero format count field 6 END Comment 883 Message There is a T format specification within a repeated parenthetical group Example of message 883 If Line Nl Source 1 SUBROUTINE W883 2 INTEGER K 6 3 DATA J K 0 1 2 3 4 5 6 4 WRITE 10 J K N N 1 6 5 WRITE 20 J K N N 1 2 6 10 FORMAT 1X 110 3 T5 I3 I5 6 warn883 for C883 There is a T format specification within a repeated parenthetical group 7 20 FORMAT 1X 110 T5 I3 15 8 END Note 884 Message The use of zero length character constants is nonstandard 290 PromulaFortran Compiler User s Manual Example of message 884 If Line Nl Source 1 SUBROUTINE W884 2 10 FORMAT 2 warn884 for N884 The use of zero length character constants nonstandard 3 20 FORMAT 3 warn884 for N884 The use of zero length character constants nonstandard 4 30 FORMAT 4 warn884 for N884 The use of zero length character constants nonstandard 5 CHARACTER 2 A B 6 6 warn884 for N884 The use of zero leng
54. OP L ROA BA TRDE DSO E AN H ME ET 173 Oo ELTIME CURRENT TIME e met 173 CONTROLLING RUNTIME BEHAVIOR e e ees sees ense esas seen sete ao esto P eaae e seno eese e seno sane seo esto oes 174 7 1 INTERPRETING CARRIAGE CONTROL TO 2 0 0000000000000000000000000000000000 0 174 7 2 CHECKING SUBSTRING LENGTHS FOR OVERFLOW cccccccscssssseecececeeseueaeceeececseaaececccecsessaaeceeccecseaaeeeeeceseensaaeeeeees 175 7 3 EXECUTING AN EXPLICIT PAUSE ccccccccccccccssssssncccececeesssnnecececeensnsnaececcesesessaaecececsenssaaecececeecessaaecececsesensaaeaeeess 175 7 4 USING VAX FORTRAN RUNTIME 175 7 5 ASSIGNING A STANDARD INPUT UNIT cccccccccccsesessscecececcesenseaecceeceeseseaeceeececnsensaeeeeececsesuaaeceeececseseaaaeeeeceeeensaaeeeeess 175 7 6 ASSIGNING A STANDARD OUTPUT UNIT 175 7 7 ASSIGNING A STANDARD TERMINAL 175 7 8 SPECIEYING A VIRTUAL A E NA ME eter t toe F r PE UE E EYE Pe rr Ee 176 7 9 SPECIBEYING A VIRTUAL FILE SIZE idee e Ee ever ee Ep tities eere ree e epe Ue e 176 7 10 SPECIFYING A VIRTUAL SHEET COUNT cccccccccccscssssscesececeecesseaececececsensnaecesececsesussceee
55. PARAMETER 5 247 PromulaFortran Compiler User s Manual 3 CHARACTER AMAT 10 IMAT 5 4 EQUIVALENCE IMAT 2 5 AMAT N N 3 4 err274 for E274 The equivalence element substring specification is closed by the symbol rather than a right parenthesis 5 STOP 6 END Error 275 Message The kind parameter is the symbol token rather than an integer constant or the identifier of a constant integer parameter Example of message 275 If Line Nl Source 1 PROGRAM ERR275 2 INTEGER SHORT DOUBLE QUAD 3 PARAMETER SHORT 2 DOUBLE 8 QUAD 16 4 PRINT 23 0 1234567 42 1 42_SHORT 5 PRINT 13 5 0 1234567 123 45678 00 30 DOUBLE 6 PRINT 3 0 3 12345 0 1234567 QUAD 1 PRINT 1 1 3 14 1 0 3 14 7 1 0 1E 27 QUAD 7 err275 for E275 The kind parameter is the symbol rather than an integer constant or the identifier of a constant integer parameter 9 END Error 276 Message The kind identifier token has not yet been defined as a constant integer parameter Example of message 276 If Line Nl Source 1 PROGRAM ERR276 2 INTEGER SHORT DOUBLE QUAD 3 PARAMETER SHORT 2 DOUBLE 8 QUAD 16 4 PRINT 23 0 1234567 42 1 42 SHORT 5 PRINT 13 5 0 1234567 123 45678 00 30 DOUBLE 6 PRINT 3 0 3 12345 0 1234567 QUAD 7 PRINT 1 1 3 14 1 0 3 14 DUBLE 7 8 1 0 1E 27_QUAD 7 err276 for E276 The kind iden
56. PromulaFortran Compiler User s Manual The binary type type is being used where type type is expected Example of message 816 If Line Nl Source 1 PROGRAM W816 2 INTEGER IVAL 3 IVAL TRUE 3 warn816 for C816 The binary type logical 2 is being used where type integer 4 is expected 4 WRITE 1X I5 IVAL 5 STOP 6 END Note 817 Message If the option unit is omitted from the unit specifier then it must be first Example of message 817 If Line Nl Source 1 SUBROUTINE W817 2 READ FMT 6 3 A 2 warn817 for N817 If the option unit is omitted from the unit specifier then it must be first 3 6 FORMAT 1X F10 2 4 30 FORMAT 1X G20 10 5 5 4 6 WRITE ERR 20 6 30 B 6 warn817 for N817 If the option unit is omitted from the unit specifier then it must be first 7 20 CONTINUE 8 END Warning 818 Message The argument token is being defined with type type when it has been passed an argument of type type Example of message 818 If Line Nl Source 1 PROGRAM W818 2 CALL 8818 34 5 56 8 7 6 5 FALSE F 3 STOP 4 END 5 SUBROUTINE S818 I R C L S 6 REAL R 6 warn818 for W818 The argument R is being defined with type real 4 when it has been passed an argument of type integer 4 7 COMPLEX C 8 CHARACTER 5 9 LOGICAL L 267 PromulaFortran Compiler User s Manual 10 10 warn818 for W818 The argument I is being defined with type integer 4 when it
57. Source ah PROGRAM W809 2 1 FORMAT 1X Hello World 2 warn809 for N809 A comma in a format preceding a right parenthesis is nonstandard 3 WRITE 6 1 4 WRITE 6 1X 8HGood Bye 4 warn809 for N809 A comma in a format preceding a right parenthesis is nonstandard 5 STOP 6 END Note 810 Message A sequence of more than one comma in a format is nonstandard Example of message 810 If Line Nl Source 264 PromulaFortran Compiler User s Manual 1 PROGRAM W810 2 1 FORMAT 1X Hello World 2 warn810 for N810 A sequence of more than one comma in a format is nonstandard 3 WRITE 6 1 4 WRITE 6 1X 8HGood Bye 4 warn810 for N810 A sequence of more than one comma in a format is nonstandard 5 STOP 6 END Note 811 Message The use of variable expressions within a format is nonstandard Example of message 811 If Line Nl Source 1 PROGRAM W811 2 DIMENSION 5 3 DATA A 1 2 3 4 5 4 DO 10 I 1 10 5 WRITE 6 100 I 6 0 CONTINUE 7 DO 20 I 1 5 8 WRITE 6 101 A I J 1 I 9 20 CONTINUE 10 00 FORMAT 1 lt 1 5 gt 10 warn811 for N811 The use of variable expressions within a format is nonstandard 11 01 FORMAT lt I gt F10 lt I 1 gt 11 warn811 for N811 The use of variable expressions within a format is nonstandard 11 warn811 for N811 The use of variable expressions within a format is nonstandard 12 END Comment 812 Message A floating point
58. The SUBROUTINE statement can also have dummy arguments for statement labels these arguments are represented by asterisks 127 PromulaFortran Compiler User s Manual Dummy arguments that represent array names must be dimensioned by a DIMENSION or type statement Adjustable dimensions are permitted in subroutine subprograms 5 54 TYPE Statement The TYPE statement transfers formatted information from the values named in the input output list to the OUTPUT file Syntax TYPE fn iolist Where fn is a format specification iolist is an output list Notes The TYPE statement is a special case of the formatted WRITE statement Description The TYPE statement transfers formatted information from the values specified in the output item list to the OUTPUT file according to the specified format This statement is identical to the statement WRITE fn iolist As a simple example consider PROGRAM TYPE CHARACTER B 3 A 1 2 B YES N 19 YPE 4 A B N 4 FORMAT G20 6 A 15 YPE 50 50 FORMA END OF FILE STOP END which will convert the values of A B and N into a coded record using format 4 and will then write that record to the OUTPUT file followed by a record containing END OF FILE See also The WRITE statement describes format specifications and output lists in detail See that statement for details on these 5 55 WRITE Statement The WRITE statement transmits data
59. The only extension is that pseudo name FILL can be specified in place of a field name to create empty space in a record for purposes such as alignment This creates an unnamed field 2 Substructure declarations A field within a structure can be a substructure composed of atomic fields and or other substructures These substructures are exactly like STRUCTURE declaration blocks except that they include a list of field names 3 Union declarations A union declaration declares groups of fields that logically share a common location within a structure Each group of fields within a union declaration is declared by a map declaration with one or more fields per map declaration Union declarations use the same area of memory to alternately contain two or more groups of fields The names specified in these statements are not the names of variables and the statements in a structure declaration do not create variables The names are field names and the information provided in the statements describes the layout or form of the structure The ordering of both the statements and the field names within the statements is important because this ordering determines the order of the fields in records See also The explicit TYPE statements BYTE INTEGER DOUBLECOMPLEX DOUBLEPRECISION REAL LOGICAL and CHARACTER which describe the actual meanings of the various types 5 53 SUBROUTINE Statement The SUBROUTINE statement introduces and specifies the name
60. User s Manual For G editing the effect of the scale factor is nullified unless the magnitude of the number to be output is outside the range that permits effective use of conversion namely unless the number N is less than 10 or greater than or equal to 104 In these cases the scale factor has the same effect as described for Ew and Dw a scaling For example the numbers 3 1415926536 and 00031415926536 are represented on output under the indicated Gw a scaling as shown below 3PG20 6 3 14159 1PG20 6 3 14159 G20 6 3 14159 1PG20 6 3 14159 3PG20 6 3 14159 5PG20 6 3 14159 7PG20 6 3 14159 3PG20 6 000314E 00 1PG20 6 031416E 02 G20 6 314159 03 1PG20 6 3 141593E 04 3PG20 6 314 1593E 06 5PG20 6 31415 93E 08 7PG20 6 3141593 E 10 5 27 21 S SP SS Plus Sign Control The nonrepeatable S SP and SS edit descriptors can be used on output with the repeatable D E F G and I edit descriptors to control the printing of plus characters S SP and SS have no effect on input Normally FORTRAN does not precede positive numbers by a plus sign on output If an SP descriptor is encountered in a format specification all succeeding positive numeric fields will contain the plus sign w must be of sufficient length to include the sign If an SS or S descriptor is encountered the optional plus signs will not appear 5 SP and SS no effect on plus sign
61. a digit or a blank and it can contain a string of digits The integer field is terminated by a decimal point E or the end of the input field The fraction subfield begins with a decimal point and terminates with an E or the end of the input field It can contain a string of digits The exponent subfield can begin with E or When it begins with E the is optional between E and the string of digits in the subfield For example the following are valid equivalent forms for the exponent 3 E 03 E 03 E03 E3 3 Valid subfield combinations are as follows 1 6327 04 Integer fraction exponent 32 7216 Integer fraction 1328 45 Integer exponent 629E 1 Fraction exponent 136 Integer only 07628431 Fraction only E 06 Exponent only If the field length specified by w in Ew d is not the same as the length of the field containing the input number incorrect numbers might be read converted and stored The example below illustrates a situation where numbers are read incorrectly converted and stored yet there is no immediate indication that an error has occurred OPEN 3 BLANK ZERO READ 3 20 A B C 20 FORMAT E9 3 E7 2 E10 3 82 PromulaFortran Compiler User s Manual On the input record quantities are in three adjacent fields columns through 24 6 47 01 2 36 5 321E 02 9 5 10 would be read as 9 710 6 47 01 2 36 5 321
62. an integer after truncation The result is real or double precision For a real or double precision argument the result is 0 if the absolute value of a is less than 1 If the absolute value of a is greater than or equal to 1 the result is the largest integer with the same sign as argument a that does not exceed the magnitude of a See also DINT 6 5 ALOG10 Logarithm Base 10 Synopsis REAL FUNCTION ALOG10 REAL Description ALOG10 a is a specific function that returns the logarithm base 10 of the argument The argument is real and the result is real See also CDLOG10 CLOG10 DLOG10 and LOG10 6 6 ALOG Natural Logarithm Synopsis REAL FUNCTION ALOG REAL Description ALOG a isa specific function that returns the natural logarithm of the argument The argument is real and the result is real See also CDLOG CLOG DLOG LOG 138 PromulaFortran Compiler User s Manual 6 7 AMAX0 Maximum Value Synopsis REAL FUNCTION INTEGER Description al a2 an is a specific function that returns the value of the largest argument All arguments are integer and the result is real See also AMAXI MAX DMAXI 2 MAXO and MAXI 6 8 AMAXI Maximum Value Synopsis REAL FUNCTION AMAX1 REAL Description AMAX1 al a2 an isa specific function that returns the value of the largest argument All arguments are real
63. and the result is real See also DMAXI 2 MAXO and MAXI 6 9 AMINO Minimum Value Synopsis REAL FUNCTION AMINO INTEGER Description AMINO al a2 an is a specific function that returns the value of the smallest argument All arguments are integer and the result is real See also AMINI DMINI I2MINO MINO MINI and MIN 6 10 AMINI Minimum Value Synopsis REAL FUNCTION AMIN1 REAL Description 139 PromulaFortran Compiler User s Manual AMIN1 al a2 an is a specific function that returns the value of the smallest argument All arguments are real and the result is real See also AMINO DMINI I2MINO MINO MINI and MIN 6 11 AMOD Remaindering Synopsis REAL FUNCTION AMOD REAL REAL Description AMOD al a2 is a specific function that returns the remainder of a1 divided by a2 Both arguments and the result are real The result is al int al a2 a2 If a2 equals zero then the result is zero See also DMOD I2MOD and MOD 6 12 AND Logical And Synopsis OGICAL 2 FUNCTION AND LOGICAL 2 OGICAL FUNCTION AND LOGICAL Description AND al a2 an isa generic function that returns the logical product of its arguments See also None 6 13 ANINT Nearest Whole Number Synopsis REAL FUNCTION ANINT REAL REAL 8 FUNCTION ANINT REAL 8
64. arguments is performed via simple truncation For complex arguments only the real part is used and truncated See also IDINT IFIX INT2 and INT4 6 80 ISIGN Transfer of Sign Synopsis INTEGER FUNCTION ISIGN INTEGER INTEGER Description 5 1 2 is a specific function that returns an integer result from two integers The result reflects the magnitude of the first value combined with the sign of the second value In particular ifal gt 0 0 and a2 lt 0 0 return a1 or ifal lt 0 0 and a2 0 0 return a1 else return al 162 PromulaFortran Compiler User s Manual See also DSIGN I2SIGN and SIGN 6 81 DABS Absolute Value Synopsis INTEGER 2 FUNCTION I2ABS INTEGER 2 Description I2ABS a returns the short integer absolute value of its short integer argument The result is a if a is greater than or equal to zero else itis See Also ABS IABS DABS CABS and CDABS 6 82 RDIM Positive Difference Synopsis INTEGER 2 FUNCTION I2DIM INTEGER 2 INTEGER 2 Description I2DIM al a2 isa specific function that returns a short integer positive difference between its short integer arguments If al is greater that a2 then the value of al a2 is returned else zero is returned See also DDIM DIM and IDIM 6 83 IZMAX0 Maximum Value Synopsis INTEGER 2 FUNCTION I2MAX0
65. array size For double complex the number of storage words per element is 8 For short arrays logical or integer two elements occupy a storage word while for character arrays and byte arrays four elements occupy a storage word Though assumptions about the relative locations of the standard numeric types can be made based upon the above notion of storage unit in a transportable manner great care must be taken when making assumptions about the sizes of boundaries between these types especially in COMMON blocks 31 PromulaFortran Compiler User s Manual Storage patterns for 1 dimensional 2 dimensional and 3 dimensional arrays were shown above In general array elements are stored in ascending locations by columns The first subscript value increases most rapidly and the last subscript value increases least rapidly 3 5 2 Array References Array references can be references to complete arrays or to specific array elements A reference to a complete array is simply the array name A reference to a specific element involves the array name followed by a subscript specification An array element reference is also called a subscripted array name A reference to the complete array references all elements of the array in the order in which they are stored For example DIMENSION XT 3 DATA XT 1 2 3 CALL CALC uses the array reference xT in the DATA statement and the CALL statement A reference to an array eleme
66. branches to an END statement in a main program execution terminates If control flows into or branches to an END statement in a function or subroutine it is treated as if a RETURN statement has preceded the END statement An END statement cannot be continued it must be completely contained on an initial line A line following an END statement is considered to be the first line of the next program unit even if it has a continuation indicator 5 21 END DO Statement The END DO statement terminates the range of a DO loop Syntax END DO Description The END DO statement terminates the range of a DO or DO WHILE statement It must be used to terminate such a statement if the statement did not contain a terminal statement label See also See the DO and DO WHILE statements for a detailed discussion of loops 5 22 The ENDFILE Statement The ENDFILE statement writes an end of file to the designated unit Syntax ENDFILE unit or ENDFILE UNIT unit IOSTAT status ERR err Where unit is an integer expression status is an integer lvalue err the label of an executable statement Notes 69 PromulaFortran Compiler User s Manual The simplest form of the ENDFILE statement consists of a single unit specification not enclosed in parentheses With this form no additional parameters can be specified If the parenthetical form of ENDFILE is used then the UNIT specification is o
67. code processing time however one cannot Under VAX runtime conventions file record lengths are expressed in long words rather than in bytes The VAX FORTRAN runtime convention flag informs the runtime system that lengths are to be multiplied times the size of a long word before they are used 7 5 Assigning a Standard Input Unit Some FORTRAN runtime systems assume that a certain unit is to be assigned to the standard input file The Inumber command line switch assigns the unit number specified to the standard input file 7 6 Assigning a Standard Output Unit Some FORTRAN runtime systems assume that a certain unit is to be assigned to the standard output file The Onumber command line switch assigns the unit number specified to the standard output file 7 7 Assigning a Standard Terminal Unit Some FORTRAN runtime systems assume that a certain unit is to be assigned to standard input for reads and to standard output for writes Such files are referred to as terminal files The Tnumber command line switch assigns the unit number specified to the terminal file 175 PromulaFortran Compiler User s Manual 7 8 Specifying a Virtual Filename The Vname command line switch specifies the name of a file which is to be used as the virtual disk file This may be an existing PROMULA array datafile or it may be a file to be created during the execution of the application program 7 9 Specifying a Virtual File Size The Svalue command line s
68. compiler for a discussion of the valid command line options Having checked the command line the control program processes each FORTRAN source file specified along with any prototype and global files to form intermediate C source files These C source files are then converted to compiled form via the C compiler During this conversion you may get some messages from the host C compiler These depend upon the particular platform The only messages which the control program might issue are insufficient memory or File formation error The insufficient memory message means that the control program itself is unable to satisfy its modest memory requirements The file formation error message means that the temporary files formed during processing cannot be formed into the final object or executable files needed Either of these messages indicates a serious problem beyond the domain of this compiler Finally if an executable is to be formed all compiled files formed and any additional compiled and library files included on the command line are passed to the linker along with the specified linker options 189 PromulaFortran Compiler User s Manual 9 2 FORTRAN Preprocessor Errors By far the bulk of the messages issued by the PROMULA FORTRAN compiler are issued by the preprocessor which converts the FORTRAN source to C source The messages are divided into three general groups 1 Syntax errors 2 Warnings comments and notes 3
69. context Example of message 111 If Line Nl Source 1 PROGRAM ERR111 2 LOGICAL 4 TEST 3 REAL 4 ALPHA BETA 4 TEST TRUE 5 ALPHA BETA TEST 5 errlll for E111 The binary type real 4 cannot be combined with the binary type logical 4 in this context 6 END Error 112 Message The variable ident has already been allocated to the common block ident Example of message 112 If Line Nl Source 1 SUBROUTINE ERR112 2 COMMON A JOE FRED FRANK 10 3 COMMON BBB FRED 3 err112 for E112 The variable FRED has already been allocated to the common block A 4 COMMON C FRANK 4 errll2 for E112 The variable FRANK has already been allocated to the common block A 5 JOE 5 6 WRITE JOE 7 END 194 PromulaFortran Compiler User s Manual Error 113 Message The entry name token is not unique Example of message 113 If Line Nl Source 1 FUNCTION ERR113 A 2 ERR113 A 3 ENTRY ERR113 3 err113 for E113 The entry name ERR113 is not unique 4 ERR113 B 5 END Error 114 Message An illegal binary combination of type with type is being made Example of message 114 If Line Nl Source I PROGRAM ERR114 2 LOGICAL TEST 3 INTEGER 4 ALPHA 4 ALPHA TEST ALPHA 4 errll4 for E114 An illegal binary combination of integer 4 with integer 4 is being made 5 END Error 115 Message An expression of type type is being used in the type context Example of mess
70. direct access file is specified in words If the iolist for an unformatted WRITE contains character data the record length to be written must still be specified in words For all systems integer logical and real numeric items count as one word Double precision and complex count as two words and double complex count as four words The method for counting character items and short or byte length items unfortunately varies from system to system You should consult with your system manager if you need to write records containing such items to an unformatted direct access file A record number identifies a record in a direct access file The record number is a positive decimal integer that is assigned when the record is written Once a record number is assigned to a record the record can always be accessed by that record number The order of record numbers is the order of records on a direct access file Records can be written rewritten or read by specifying the record number in a READ or WRITE statement Records can be read or written in any order they need not be referenced in the order of their record numbers The number of the record to be read or written is specified ina READ or WRITE statement with the REC parameter The REC parameter on a direct access READ statement must not be set to a record number greater than the highest record number written in the file An attempt to read record numbers greater than the highest in the file can return
71. display becomes too wide 2 5 FORTRAN Dialect Convention Flags Though the FORTRAN language or Fortran as it is called in the newest standard has been three times standardized in 1966 1977 and 1990 it remains a highly nonstandard language The dialect convention flags allow you to specify the characteristics of your particular version of FORTRAN 2 5 1 The FORTRAN Integer Type There is variation between FORTRAN compilers as to whether the default type of the INTEGER specification should be INTEGER 2 or INTEGER 4 In fact an interesting aspect of many modern FORTRAN compilers is that the user may specify whether the default integer type is to be a short 16 bit representation or a long 32 bit representation on the command line The FIs and command line options allow for this specification The FIs specification says that the default integer type is short INTEGER 2 while specifies that it is long INTEGER 4 The default setting for this switch is FII Note that this switch also effects the LOGICAL type Thus if default integers are short then so are default logicals 2 5 2 The Treatment of Short Integer Arithmetic In moving from one environment to another one must always be concerned with the accuracy of floating point arithmetic however there is also a real problem with fixed point arithmetic even when dealing with identical word sizes This section concerns itself with short integer arithmetic in non short integer e
72. edit descriptor corresponds to exactly one list item Complex editing requires two D E F G descriptors the two descriptors can be different The type of conversion should correspond to the type of the variable in the input output list The FORMAT statement specifies the type of conversion for the input data with no regard to the type of the variable which receives the value when reading is complete For example the statements INTEGER N READ 5 10 N 10 FORMAT F10 2 will assign a floating point number to the variable which could cause unpredictable results if is referenced later as an integer 5 27 1 Character FORMAT Specifications A format specification can also be specified as a character expression or as the name of a character variable or array containing a format specification The form of these format specifications is the same as for FORMAT statements without the keyword FORMAT Any character information beyond the terminating parenthesis is ignored The initial left parenthesis can be preceded by blanks For example CHARACTER FORM 11 DATA FORM I3 2E14 4 READ 2 FMT FORM END 50 N A B is equivalent to READ 2 FMT 10 END 50 N A B 10 FORMAT I3 2E14 4 The examples above can also be expressed as READ 92 FMT I3 2E14 4 END 50 N A B or CHARACTER FORM PARAMETER FORM I3 2E14 4 READ 2 FMT FORM END 50 N A B Y
73. enclosed in parentheses With this form no additional parameters can be specified If the parenthetical form of OPEN is used then the UNIT specification is optional however when omitted the unit specification must be the first specification Other than the above the order of the parameters within the parenthetical version of the OPEN statement is free Description The OPEN statement can be used to associate an existing file with a unit number to create a new file and associate it with a unit number or to change certain attributes of an existing file The UNIT parameter is required all other parameters are optional except for the RECL parameter which must be specified if a file is being opened for direct access If an error occurs as a result of the open and if neither ERR nor STATUS are supplied then execution will terminate with an error code set If either or both are supplied the execution continues despite any errors If supplied the STATUS sta lvalue receives the runtime error code for the error condition encountered or a zero if no error occurred If ERR s1 is supplied then execution will branch to 51 the statement labeled by it if an error occurs The meaning of the other individual parameters on the OPEN statement is as given below UNIT u specifies the unit number of the file to be opened FILE fin is a character expression whose value is the name of the file to be opened Trailing blanks are removed This file be
74. error condition is encountered during output processing IOSTAT ios specifies an integer variable into which an error code is stored as follows 0 means that the output operation completed normally and a value greater than zero indicates some other error condition olist is an output item list described below Regardless of the type of the write if an error occurs as a result and if neither ERR s1 nor IOSTAT i0s is supplied then execution will terminate with an error code set If either or both are supplied the execution continues despite any errors If supplied the IOSTAT ios lvalue receives the runtime error code for the error condition encountered or a zero if no error occurred If ERR s1 is supplied then execution will branch to the statement s1 if an error occurs The output item list portion of a WRITE statement specifies the items and or values to be written and the order of transmission The output item list can contain any number of items List items are written sequentially from left to right If no list appears one or more empty records are written A list item consists of a variable name an array name an array element name a character substring name an rvalue of any type or an implied DO list List items are separated by commas Subscripts in an output item list can be written as any valid subscript form An array name without subscripts in an output list specifies the entire array in the order in which it is stored The entir
75. executed In general block IF structures provide for alternative execution of blocks of statements A block IF structure begins with a block IF statement ends with an END IF statement and optionally includes one ELSE or one or more ELSE IF statements Each block IF ELSE and ELSE IF statement is followed by an associated block of executable statements called if block The simplest form of a block IF structure is shown below IF exp THEN if block ENDIF If expression exp is true execution continues with the first statement in the if block If exp is false control transfers to the statement following the END IF statement The if block can contain any number of executable statements including block IF statements An actual example of this structure is as follows IF I EQ 0 THEN X X DX Y Y DY END IF If I is zero the subsequent statements are executed If not control transfers to the statement following END IF Control can be transferred out of an if block from inside the if block Control can be transferred into an if block from outside the if block however this practice is not recommended It is not permissible to branch directly to an ELSE ELSE IF or END IF statement However it is permissible to branch directly to a block IF statement When execution of the statements in an if block has completed and if control has not been transferred outside an if block execution continues with the statement followin
76. for N886 Including a repeat count before the format Specification is nonstandard 5 warn886 for N886 Including a repeat count before the T format Specification is nonstandard 5 warn886 for N886 Including a repeat count before the format Specification is nonstandard 6 WRITE 30 7 30 FORMAT 1X E10 2 G20 10 8 READ 40 I 9 40 FORMAT 2BZ 12 2 A 3BN 9 warn886 for N886 Including a repeat count before the format specification is nonstandard 10 END Comment 887 Message Output format contains a superfluous bn or bz specification Example of message 887 If Line Nl Source all SUBROUTINE W887 2 1 3 WRITE 6 10 4 10 FORMAT 2 1 110 4 warn887 for C887 Output format contains a superfluous bn or bz specification 5 20 FORMAT 1X BZ 110 6 WRITE 6 20 IA 6 warn887 for C887 Output format contains a superfluous bn or bz specification F END Note 888 Message The block data subprogram contained no data statements Example of message 888 If Line Nl Source 1 BLOCK DATA W888 2 COMMON PI PI 3 COMMON D D 10 4 END 4 warn888 for N888 The block data subprogram contained no data statements 292 PromulaFortran Compiler User s Manual Note 889 Message The use of zero width or free form format value editing specifications is nonstandard Example of message 889 If Line Nl Source 1 SUBROUTINE W889 2 10 FORMAT F0 5 2 warn889 for N889 The use of zer
77. from memory storage location or values of expressions to an external or internal unit Syntax WRITE fn olist WRITE olist 128 PromulaFortran Compiler User s Manual WRITE UNIT u FMT fn REC rn IOSTAT ios ERR sl END sl olist WRITE UNIT u FMT nl REC rn IOSTAT ios ERR sl1 END s1 WRITE UNIT u FMT2 IOSTAT ios 51 51 olist WRITE UNIT u IOSTAT ios REC rn ERR sl END sl olist Where u is an integer expression or a character variable identifier or an asterisk fn is an integer variable the label of a FORMAT statement a character expression or an array identifier rn is an integer expression nl is the identifier of a NAMELIST group ios is an integer lvalue sl is the label of an executable statement olist isan output item list Notes The simplest form of the WRITE statement consists of a single format specification not enclosed in parentheses possibly followed by an output item list With this form no additional parameters can be specified If one of the parenthetical forms of WRITE is used then the UNIT and FMT specifications are optional however when omitted the unit specification must be the first specification and the fn specification must be second Other than the above the order of the parameters within the parenthetical versions of the W
78. good programming practice to avoid naming conflicts by assigning unique names to program entities Certain of these conflicts are illegal and are diagnosed 3 3 Constants A constant is a fixed quantity There are nine types of constants supported 1 integer 2 real 3 double precision 4 complex 5 double complex 6 logical 7 character 8 Hollerith 9 exact representation Integer real double precision complex and double complex constants are considered arithmetic constants 3 3 1 Integer Constant 22 PromulaFortran Compiler User s Manual An integer constant is a string of 1 to 16 decimal digits written without a decimal point It can be positive negative or zero If the integer is positive the plus sign can be omitted if it is negative the minus sign must be present An integer constant must not contain a comma Its syntax is as follows did Where a isa decimal digit The range of an integer constant is 2147483648 to 2147483647 Examples of valid integer constants are as follows 237 74 136772 0024 Examples of invalid integer constants are as follows 46 Decimal point not allowed 23A Letter not allowed 7 200 Comma not allowed 3 3 2 Real Constant A real constant consists of a string of decimal digits coefficient written with a decimal point or with an exponent or with both Commas are not allowed The plus sign can be omitted from either the coefficient or the expo
79. have simple returns A simple return always begins execution in the calling program unit at the statement immediately following the CALL statement A multiple return exists when the subprogram has more than one RETURN statement or when a single RETURN statement is separated from the END statement by other statements An alternate return is used only within a subroutine subprogram It returns control to the referencing program unit at a place other than the next executable statement after the CALL statement The RETURN statement in the form RETURN exp is used for an alternate return Control is returned to a specified point in the referencing program unit The specified point is a statement label in the referencing program unit The statement labels must be included in the actual argument list each preceded by an asterisk Control returns to the statement label determined by the integer value of the alternate return expression If the value of the expression is less than one or greater than the number of asterisks in the SUBROUTINE statement or ENTRY statement that is the current entry point control returns to the statement following the CALL statement For example if a CALL statement contains five statement labels and if the alternate return expression evaluates to three control returns to the third statement label specified in the actual argument in the alternate return list 5 48 RECORD Statement The RECORD statement creates a record of the fo
80. ident Example of message 103 If Line Nl Source 1 SUBROUTINE ERR103 I B 2 C I C 1 1 B 1 C 1 1 2 errl03 for E103 The statement function either calls itself or there is a missing dimension for C 3 A C 3 4 WRITE A I 5 END 191 PromulaFortran Compiler User s Manual Error 104 Message The substring expression for ident is terminated by the symbol token rather than a right parenthesis Example of message 104 If Line Nl Source 1 PROGRAM ERR104 2 CHARACTER 10 ERR_MES 3 ERR MES 1 3 5 Bad Version 3 errl04 for E104 The substring expression for ERR MES is terminated by the symbol rather than a right parenthesis STOP 5 END Error 105 Message The object identifier ident may not be used to represent a function Example of message 105 If Line Nl Source al PROGRAM ERR105 2 STRUCTURE SHIP CHARACTER 10 OWNER 4 1 CHARACTER 10 DESTINATION 5 END 6 RECORD SHIP NINA PINTA SANTA MARIA 7 CHARACTER 10 WHERE WHERE NINA 3 8 errl05 for E105 The record identifier NINA may not be used to represent a function 9 STOP 10 END Error 106 Message The object identifier ident cannot contain a value Example of message 106 If Line Nl Source 1 SUBROUTINE ERR106 2 DIMENSION A 10 20 3 A 2 3 0 0 4 ENTRY BVALUE 5 A 2 3 A 2 3 BVALUE 5 errl06 for E106 The entry point identifier BVALUE cannot contain a value 192
81. iei 141 6 16 788 141 6 17 CABS ABSOLUTE 202 142 6 18 CEOS S COSINE 142 6 19 CDABS ABSOLUTE 2 142 0 207 C DC OS AE TEE 143 6 21 CDEXP 9 143 6 22 CDLOG10 LOGARITHM BASE 10 en een 143 6 23 CDEOG NATURALE LOGARITHM 143 624 CDSN 144 6 25 CDSOR T SQUARE ROOT ern esee Tbe E edere redes en 144 6 26 EXPONENTIAL Saba 144 6 21 CHAR CHARACTER VALU Ern a E 145 6 28 CLOGIO EOGARITHM BASE 10 oco eee ee oen 145 629 CLOG NATURAL LiOGA RITHM 145 6 30 4C MPEX COMPLEX V ABUE 146 6 31 CONTIG Go UC WIDE C EP 146 6 147 OH GONE v RP P PPP 147 O34 Red EMI Soe Sh I EE ioe EN IEE A RS 147 6 35 CSOR T SQUARE ROOT ws 4
82. length specification for a given entity is 4 or if it is omitted then the type defined is logical If the value is 2 the type is short logical and if the value is 1 the type is logical byte See also See the discussion of the DIMENSION statement for a description of how dimension bounds are defined See the discussion of the DATA statement for a description of how initial values are defined 5 38 NAMELIST Statement The NAMELIST statement defines a group of variables and or arrays for later input or output Syntax 106 PromulaFortran Compiler User s Manual NAMELIST name a a name a a Where name is a symbolic name a is a variable or array name Description The NAMELIST statement permits the input and output of groups of variables and arrays with an identifying name No format specification is used The NAMELIST statement is a nonexecutable statement that appears in the program following the declarative portion The symbolic group name must be enclosed in slashes and must be unique within the program unit The namelist group name identifies the succeeding list of variables or array names It must be declared in a NAMELIST statement before it is used in an input output statement The group name can be declared only once and it cannot be used for any purpose other than a namelist name in the program unit It can appear in READ WRITE PRINT and PUNCH statements in place of the format designator When a
83. loop must be terminated with an END DO statement Example The loop below DO WHILE I GT J A I J 1 0 1 END DO is equivalent to DO 10 WHILE I GT J A I J 10 1 10 CONTINUE 5 18 ELSE Statement The ELSE statement specifies an optional statement block to be executed within a block IF statement Syntax ELSE IF 1 THEN 1 block END IF 67 PromulaFortran Compiler User s Manual Where el is a logical expression block isa block of executable statements Description If no preceding logical expression with a block IF statement has been true and if the expression e1 is true or if e1 is omitted then the executable statements in the block are executed and control is transferred to the first statement immediately after the END IF statement If the e1 expression is omitted then it must be the last ELSE within the block IF See also See the block IF statement for a complete discussion of the various component statements within the block IF construct 5 19 ENCODE Statement The ENCODE statement is an internal file output statement Syntax ENCODE c fn u iolist Where is an unsigned integer constant or variable having a value greater than zero fn is a statement label of a FORMAT statement or a character expression whose value is a format specification u is an extended internal file variable array element or array name in which t
84. not have a file connected to it has no effect After a unit has been disconnected by a CLOSE statement it can be connected again within the same program to the same file or to a different file A file connected to a unit specified in a CLOSE statement can be connected again to the same or to another unit provided the file still exists The cissta variable is a character expression that determines the disposition of the file associated with the specified unit Valid values are as follows KEEP The file is kept after execution of the CLOSE statement DELETE The file is unloaded after execution of the CLOSE statement The default if c1ssta is not specified is STATUS DELETE if the file status was SCRATCH when it was opened otherwise the default is STATUS KEEP If an error occurs as a result of the close and if neither err nor status are supplied then execution will terminate with an error code set If either or both are supplied the execution continues despite any errors If supplied the status lvalue receives the runtime error code for the error condition encountered or a zero if no error occurred If err is supplied then execution will branch to the statement labeled by it if an error occurs 52 PromulaFortran Compiler User s Manual Example CLOSE 2 ERR 25 STATUS DELETE When this statement is executed the file connected to unit 2 will be closed and disconnected If an error occur
85. of comment characters other than or C is nonstandard Example of message 850 If Line Nl Source SUBROUTINE W850 A 10 0 C THIS IS A COMMENT B 15 0 50 The use of comment characters other than or C is nonstandard THIS IS A COMMENT 20 0 THIS IS A COMMENT RETURN END 4 warn850 for N8 01 Note 851 Message The use of inline comments is nonstandard Example of message 851 If Line Nl Source 1 PROGRAM W851 2 DIMENSION IA 10 This is the key vector 2 warn851 for 851 The use of inline comments is nonstandard 3 WRITE 1 IA 4 STOP 5 END Note 852 Message The use of multiple statements on a single line is nonstandard Example of message 852 If Line Nl Source al PROGRAM W852 2 IA 10 IB 15 2 warn852 for N852 The use of multiple statements on a single line is nonstandard 3 WRITE 1X 215 IA IB 4 STOP 5 END 279 PromulaFortran Compiler User s Manual Note 853 Message The standard delimeter for a character constant is the single quote Example of message 853 If Line Nl Source L SUBROUTINE W853 2 CHARACTER 9 A 3 A ABCDEF 3 warn853 for N853 The standard delimeter for a character constant is the single quote 4 END Note 854 Message The use of r or 1 character constants is nonstandard Example of message 854 If Line Nl Source al PROGRAM W854 2 CHARACTER 10 IA 3 DATA IA L ABC 3 warn854 for N854 The u
86. of the ASSIGN statement prior to the execution of the GOTO statement will cause the program unit to branch to the statement labeled 10 The following assigns the label of a FORMAT statement to a variable for later use in a FORMAT ted input or output statement 15 FORMAT 1X 2F10 5 ASSIGN 15 TO ifmt WRITE ifmt b Execution of the ASSIGN statement prior to the execution of the WRITE statement will write the values of a and in accordance with FORMAT 15 5 2 BACKSPACE Statement The BACKSPACE statement positions a file to the start of the preceding record 48 PromulaFortran Compiler User s Manual Syntax BACKSPACE unit BACKSPACE UNIT unit IOSTAT status ERR err Where unit is an integer expression status isan integer lvalue err is the label of an executable statement Notes The simplest form of the BACKSPACE statement consists of a single unit specification not enclosed in parentheses With this form no additional parameters can be specified If the parenthetical form of BACKSPACE is used then the UNIT specification is optional however when omitted the unit specification must be the first specification Other than the above the order of the parameters within the parenthetical version of the BACKSPACE statement is free Description The BACKSPACE statement backspaces the file currently under the specified unit number one record Any file regardless of its type or open status
87. of the form dlist i init term incr dlist isa list of array element names and implied DO lists i is an integer variable called the implied DO variable init is an integer constant symbolic constant or expression specifying the initial value as for DO loops term isan integer constant symbolic constant or expression specifying the terminal value as for DO loops incr is an integer constant symbolic constant or expression specifying the increment as for DO loops is a list of constants or symbolic constants specifying the initial values Each item in the list can take the form is a constant or symbolic constant r is a repeat count that is an unsigned nonzero integer constant or the symbolic name of such a constant The DATA statement is used to provide initial values for variables arrays array elements and substrings The DATA statement is nonexecutable and can appear anywhere in a program unit after any statements explicitly typing the elements or dimensioning them Usually DATA statements are placed after the specification statements but before the statement function definitions and executable statements Example The following s hows a simple example of a DATA statement INTEG ER K 6 DATA JR 4 DATA AT 5 0 AQ 7 5 DATA NRX SRX 17 0 5 2 57 PromulaFortran Compiler User s Manual DATA K 1 2 3 3 2 1 The variables JR ao and SRX are in
88. of the main entry point of a subroutine subprogram Syntax SUBROUTINE sub d d Where sub is a symbolic name identifying the subroutine subprogram d is a dummy argument that can be a variable name array name dummy procedure name or Notes If there are no dummy arguments either sub or sub can be used Description A subroutine subprogram is a procedure that communicates with the calling program unit either through a list of arguments passed with the CALL statement or through common blocks A subroutine subprogram is executed when a CALL statement naming the subroutine is encountered in a program unit Subroutines begin with a SUBROUTINE statement and end with an END statement Control is returned to the calling program unit when a RETURN statement is encountered If control flows into the END statement then a RETURN is implied Subroutines differ from functions in that the subprogram name is not used to return results to the calling program Subroutines can contain any statement except a PROGRAM BLOCK DATA FUNCTION or another SUBROUTINE statement A subroutine subprogram must not directly or indirectly call itself In a subroutine subprogram the symbolic name of a dummy argument is unique to the program unit and must not appear in an EQUIVALENCE PARAMETER SAVE INTRINSIC DATA or COMMON statement except as a common block name The dummy arguments are replaced with the actual arguments during a subroutine call
89. on any C based platform it will compile and run on almost PromulaFortran Compiler User s Manual any other C base platform Not only are your FORTRAN source codes processed via C but our entire FORTRAN runtime library is also written in C Since C is used as an intermediate language it is trivial to integrate FORTRAN programs with other C based software such as GUIs operating systems data management systems etc We know of no major software library available today which is not usable via C External C functions can be referenced very naturally using FORTRAN CALL statements and function references In addition prototype files can be supplied to ensure that the compiler properly handles call by value versus call by name Runtime efficiency is always a concern At one time the C language was intended for small integer based applications only Early C compilers did not even support floating point arithmetic This is no longer true Contemporary C compilers are fully featured run very quickly and produce efficient code Benchmarks between PROMULA FORTRAN produced executables and those produced by contemporary native FORTRAN compilers when available typically favor PROMULA In fact it is typical for contemporary FORTRAN compilers to share the optimizer back end of the native C compiler PROMULA FORTRAN covers the complete current FORTRAN language as currently used and not simply as defined by the various standards PF passes the GSA FORT
90. operands or operates on a single operand and is written preceding that operand The interpretation of a division can depend on the data types of the operands A set of rules establishes the interpretation of an arithmetic expression that contains two or more operators A precedence among the arithmetic operators determines the order in which the operands are to be combined KE Highest and Intermediate and Lowest For example in the expression A 2 the exponentiating operator has precedence over the negation operator The operands of the exponentiation operator are combined to form an expression used as the operand of the negation operator The expression is the same as the expression A 2 Successive exponentiations are combined from right to left For example A 3 2 is interpreted as A 3 2 Two or more multiplication or division operators are combined from left to right Two or more addition or subtraction operators are combined from left to right Note that arithmetic expressions containing two consecutive arithmetic operators such as A B or A B are not permitted However expressions such as A B and A B are permitted 40 PromulaFortran Compiler User s Manual Subexpressions containing operators of equal precedence are evaluated from left to right The compiler may reorder individual operations that are mathematically associative and or communative to perform optimizations su
91. optional decimal point and with an exponent Commas are not allowed The plus sign can be omitted from either the coefficient or the exponent if they are positive but the minus sign must be present with either or both if either or both are negative Its syntax is as follows coeff exp nD E exp Where n is an unsigned integer constant coeff is a coefficient in one of the following forms n n n n exp is an unsigned integer constant The range of the exponent is 307 to 308 The precision for a real constant is approximately 13 significant digits Examples of valid double precision constants 5 83402 Value 5 834 x 102 583 4 14 5 Value 14 x 1075 00014 9 2 03 Value 9 2 x 103 9200 0 312004 Value 3120 x 104 31200000 0 Examples of invalid double precision constants 7 2D Exponent missing D5 Exponent alone not allowed 24 PromulaFortran Compiler User s Manual 2 001 3D2 Comma illegal 3 14159265 and exponent missing 3 3 4 Complex Constant Complex constants are written as a pair of real or integer constants separated by a comma and enclosed in parentheses The first constant represents the real part of the complex number and the second constant represents the imaginary part The parentheses are part of the constant and must always appear Either constant can be preceded by a plus or minus sign The syntax of a complex constant is as follows real imag Wher
92. procedure receives values from the actual arguments and returns values to the referencing program unit Actual arguments can be constants symbolic names of constants variables array names array elements function references and expressions An actual argument cannot be the name of a statement function within the referencing program unit Dummy arguments appear in the argument list of the referenced procedure Within the referenced procedure the dummy arguments are associated with the actual arguments passed Procedures use dummy arguments to indicate the types of actual arguments the number of arguments and whether each argument is a variable array procedure or statement label Dummy arguments for statement functions can only be variables Since all names are local to the program unit the same dummy argument name can be used in more than one procedure A dummy argument appearing in a SUBROUTINE FUNCTION or ENTRY statement must not appear in EQUIVALENCE DATA PARAMETER SAVE INTRINSIC or COMMON statements except as a common block name Dummy arguments used in array declarations for adjustable dimensions must be type integer Dummy arguments representing array names must be dimensioned When a procedure is executed the actual arguments and dummy arguments are matched up and each actual argument replaces each dummy argument The type of the actual argument and the dummy argument must be the same The actual arguments must be in the same ord
93. rather than a slash Example of message 210 If Line Nl Source all PROGRAM ERR210 2 COMMON CB123 A 10 B 113 2 err210 for E210 The common block identifier is followed by the symbol rather than a slash 3 STOP 4 END 226 PromulaFortran Compiler User s Manual Error 211 Message The equivalence specification is preceded by the symbol token rather than a left parenthesis Example of message 211 If Line Nl Source I PROGRAM ERR211 2 DIMENSION 10 IMAT 5 9 EQUIVALENCE IMAT 2 AMAT 4 3 err211 for E211 The equivalence specification is preceded by the symbol IMAT rather than a left parenthesis 4 STOP 5 END Error 212 Message The equivalence element identifier token is not a valid identifier symbol Example of message 212 If Line Nl Source 1 PROGRAM ERR212 2 DIMENSION AMAT 10 IMAT 5 3 EQUIVALENCE 10 IMAT 2 AMAT 4 3 err212 for E212 The equivalence element identifier 10 is not a valid identifier symbol 4 STOP 5 END Error 213 Message An equivalence relation is being assigned the object ident which is not a variable Example of message 213 If Line Nl Source 1 PROGRAM ERR213 2 PARAMETER IVAL 99 3 DIMENSION AMAT 10 IMAT 5 4 EQUIVALENCE IMAT 2 IVAL 4 err213 for E213 An equivalence relation is being assigned the parameter IVAL which is not a variable 5 STOP 6 END Error 214 Message 227 PromulaFortran
94. specified unit Inquire by file is used to obtain information about the current status of a file Either a file name inquire by file or a unit specifier inquire by unit but not both must be specified in an INQUIRE statement The file or unit need not exist when INQUIRE is executed Following execution of an INQUIRE statement the specified parameters contain values that are current at the time the statement is executed The actual information returned in the parameters is as follows 1 5 51 num nmd fn acc seq dir is the unit number of the file is a logical variable which returns TRUE if the file unit exists and FALSE if the file unit does not exist is the status code for the file is the statement label where execution branches if an error occurs is a logical variable which returns TRUE if the file unit is connected to a unit file and FALSE if the file unit is not connected to a unit file is an integer variable which returns the external unit number of the unit currently associated with the file undefined if the file is not associated with a unit is a logical variable which returns TRUE if the file has a name and FALSE if the file does not have a name is a character variable which returns the name of the file associated with unit u is a character variable which returns the access method of the file SEQUENTIAL if the file is opened for sequential acc
95. subprogram beginning with a FUNCTION statement terminating with an END statement and containing other statements as a single statement written by the user 34 PromulaFortran Compiler User s Manual A main program is a program unit that does not begin with a FUNCTION SUBROUTINE BLOCK DATA or ENTRY statement The main program should have a PROGRAM statement optional and at least one executable statement followed by an END statement The execution of any program begins with the main program unit No executable program can have more than one main program unit The main program can be compiled independently of any subprograms However when a main program is loaded into memory for execution all the required subprograms must be loaded with it prior to its execution 3 8 1 Program Unit and Procedure Communication Communication between the referencing program unit and the referenced procedure is accomplished by passing actual arguments and by using common blocks Common blocks can be used to pass data to a subprogram but not to an intrinsic function or a statement function Data must be passed to these functions through an argument list When passing arguments actual arguments in the referencing program unit are associated with the referenced procedure through dummy arguments Actual arguments appear in the argument list of the referencing program unit The referencing program unit passes actual arguments to the referenced procedure The
96. that they do not overflow the strings from which they are taken This convention provides for rapid execution of substring statements and conforms to the normal FORTRAN conventions which allow the user to play memory games with substrings i e many real FORTRAN programs contain deliberate substring overflows Alternatively if careful programming conventions are being stressed it may be desirable to check for substring overflows at runtime This flag provides this facility 7 3 Executing an Explicit PAUSE In traditional FORTRAN environments the PAUSE statement executed by jobs running in the background sent a message to the operator console and then suspended execution until a go was received from the operator By default the runtime library now treats these PAUSE statements as merely displaying informational messages to the console execution is not physically suspended The explicit PAUSE flag causes the application to pause until the Enter or Return key is pressed or until a record is read from the standard input file Since the concept of an operator s console is no longer well defined for jobs running in the background care should be used with this flag It may cause unexpected results 7 4 Using VAX FORTRAN Runtime Conventions The VAX runtime conventions though they are technically compliant with the 1977 FORTRAN standard are somewhat different from normal conventions Most of these differences can be dealt with at source
97. the FORMAT statement number in the associated input output statement The name specifies the location of the first word of the format information 91 PromulaFortran Compiler User s Manual 5 28 FUNCTION Statement The FUNCTION statement introduces and specifies the name of the main entry point of a function subprogram Syntax typ len FUNCTION fun d d Where typ is any valid type keyword BYTE INTEGER REAL DOUBLECOMPLEX DOUBLEPRECISION LOGICAL COMPLEX CHARACTER len is an integer constant which specifies additional type information fun is the symbolic name which identifies the function subprogram d is a dummy argument that can be a variable name array name or dummy procedure name Notes If there are no dummy arguments either fun or fun can be used Description A function subprogram is a procedure that communicates with the referencing program unit through a list of arguments or common blocks It is usually referred to as an external function A function subprogram performs a set of calculations when the name appears in an expression in the referencing program unit A function subprogram begins with a FUNCTION statement and ends with an END statement Control is returned to the referencing program unit when a RETURN or END is encountered a RETURN statement of the form RETURN exp is not allowed in a function subprogram A function subprogram can contain any statements except PROGRAM BLOCK DATA SUBROUTIN
98. the same first character storage position Equivalencing can indirectly cause the association of other entities for instance when an EQUIVALENCE statement interacts with a COMMON statement If an array element is included in nlist the number of subscript expressions must match the number of dimensions declared for the array name If an array name appears in the list the effect is as if the first element of the array had been included in the list Any subscript expression must be an integer constant expression For character entities any substring expression must be an integer constant expression Equivalencing must not reference array elements in such a way that the storage sequence of the array would be altered The same storage unit cannot be specified as occurring more than once in the storage sequence For example REAL FA 3 EQUIVALENCE FA 1 B FA 3 B would be illegal 72 PromulaFortran Compiler User s Manual Also the normal storage sequence of array elements cannot be interrupted to make consecutive storage words no longer consecutive For example REAL BZ 7 CZ 5 EQUIVALENCE BZ CZ BZ 3 CZ 4 would also be illegal The interaction of COMMON and EQUIVALENCE statements is restricted in two ways First an EQUIVALENCE statement must not attempt the association of two different common blocks in the same program unit For example COMMON LT A T COMMON LX S
99. the statement labeled by it if an error occurs If a unit number is specified and the unit is opened the NAMED NAME ACCESS SEQUENTIAL DIRECT FORM FORMATTED UNFORMATTED RECL NEXTREC OPENED EXIST NUMBER ACCESS and BLANK variables will contain information about the file associated with the unit If a file name is specified the NAMED NAME SEQUENTIAL DIRECT FORMATTED UNFORMATTED OPENED EXIST NUMBER ACCESS FORM RECL NEXTREC and BLANK variables will contain information about the file and the unit it is associated with EXIST returns a TRUE value only if a non empty local file by this name exists or if an empty local file by this name is currently open When EXIST returns a FALSE value the NUMBER NAMED NAME ACCESS SEQUENTIAL DIRECT FORM FORMATTED UNFORMATTED RECL NEXTREC and BLANK variables will contain undefined values This does not result in an error If a file is specified that is associated with more than one unit the NUMBER variable will contain one of the unit numbers or names Note that if a unit that is not associated with a file is specified only the IOSTAT and EXIST variables contain values See also The discussion of the OPEN statement describes the FORTRAN file system in general including the different file types 104 PromulaFortran Compiler User s Manual 5 36 INTRINSIC Statement The INTRINSIC statement identifies a name as representing an intrinsic function Syntax INTRINS
100. the storage allocated to A in the CHARACTER statement because the fourth element of extends beyond the fifth member of A In such cases the storage of the base variable is expanded to allow for the extra storage needed for the equivalence set Thus though A will be treated as though it were a CHARACTER 5 throughout the subprogram storage is allocated to A as though it were a CHARACTER 6 Different data types are associated with the equivalencing of the first storage word of each entity REAL TR 4 COMPLEX TS 2 EQUIVALENCE TR TS R 1 S 1 real part R 2 S 1 imaginary part R 3 S 2 real part R 4 S 2 imaginary part If the equivalenced entities are of different data types equivalencing does not cause type conversion If a variable and an array are equivalenced the variable does not acquire array properties and the array does not lose the properties of an array There are no restrictions about equivalencing character and noncharacter variables and the lengths of the equivalenced character entities can be different Each nlist specification must contain at least two names of entities to be equivalenced In a subprogram names of dummy arguments cannot appear in the list Function and entry names cannot be included in the list Equivalencing specifies that all entities in the list share the same first storage word For character entities equivalencing specifies that all entities in the list share
101. unpredictable data without any error being reported If the length of the iolist in a direct access formatted WRITE statement is less than the record length of the direct access file the unused portion of the record is blank filled A direct access WRITE statement must not write a record longer than the record length A direct access file can be opened for formatted or unformatted input output However neither list directed nor namelist input output can be used with direct access files 5 40 PARAMETER Statement The PARAMETER statement gives a symbolic name to a constant Syntax PARAMETER p e p e Where p isasymbolic name e isa constant or a constant expression Description 114 PromulaFortran Compiler User s Manual The PARAMETER statement is used to give a symbolic name to a constant PARAMETER statements can be used anywhere among the specification statements but each symbolic constant must be defined in a PARAMETER statement before the first reference to the symbolic constant As an example consider the following PARAMETER ITER 20 START 5 CHARACTER CC PARAMETER 14 10 5 DATA COUNT START DO 10 J 1 ITER READ CC IX RX 10 CONTINUE The symbolic constant START is used to assign an initial value to variable COUNT the symbolic constant ITER is used to control the DO loop and the symbolic constant CC is used to specify a character con
102. utest for Processing subprogram utest Processing subprogram anal Processing subprogram ana2 Processing subprogram ana3 Including source file STRUC6 INC Processing subprogram get_date Including source file STRUC6 INC Processing subprogram struc7 Including source file STRUC7 INC Processing subprogram get_d Including source file STRUC7 INC Processing subprogram struc8 Processing subprogram get_da Processing subprogram struc9 Processing subprogram thermal In addition to source files as each new subprogram is compiled a message is generated 2 4 3 Annotated Listing of Source Code The following listing was produced via the Es option for a simple subprogram referencing a single include file PROMULA FORTRAN Compiler V4 00 Date 08 11 92 Time 09 35 Page 1 File utest for If Line Nl Source SUBROUTINE STRUC6 INCLUDE STRUC6 INC STRUCTURE DATE EGER 4 DAY MONTH EGER 4 YEAR UCTURE H STRUCTURE LOGICAL 1 HOUR MINUT END STRUCTURE EL TE NO P E z n ve PromulaFortran Compiler User s Manual 9 S 10 C This is the same as
103. 0 utest for W818 The argument ia is being defined with type integer 4 when it has been passed an argument of type character The W appended to the error number indicates a warning Messages at this level can be ignored but they typically indicate potentially serious problems At the EL2 level the following additional types of messages are issued 54 utest for C870 The array OBUF is being subscripted with less than 1 expressions 380 utest for C815 A data value of type character is being assigned to the variable IA of type integer 4 520 utest for C816 The binary type character is being used where typ integer 2 is expected 558 utest for C861 Data is being allocated to common storage via the variable SEATRD 820 utest for C866 The real 4 type has previously been assigned to UC At this level in addition to warnings other usages are isolated that should either be checked or that represent potentially serious portation problems Finally at the EL3 level the following additional types of messages are issued 37 utest for 858 The identifier THERMA1 with more than 6 characters is nonstandard 184 utest for N864 Declarative statements following executable statements is nonstandard 223 utest for N851 The use of inline comments is nonstandard 295 utest for N872 Equivalencing CBUF of type character 80 with IBUF of type INTEGER 2 is nonstandard 343 utest for N853 The standard
104. 1 HOW BE WORKS e ta e DENT URP ERR Rr PAN FER IRURE 3 2 2 DURUIEROUB BUR DOR DDR RR DORT T IEEE 3 2 3 SUMMARY OF 5 coves EFE ON EROR 4 2 4 ECHO CONTROL OPTIONS HR UENIT ODD ENNIO ERES DERI PEE RIED covey 4 2 4 1 Warnings Notes and Comments eese 5 2 4 2 Monitoring Internal Operations eese eere nennen teen eee et eene enne ene tnee tree tere tren teen trennen rennen 6 2 4 3 Annotated Listing of Source Code 6 2 4 4 Symbol Listing and Cross Reference Table enne nennen tenete trennen rennen 7 2 43 Controlling Page Size ies age e Ee e Ene eee ere ede ere rie Yee e 10 2 5 FORTRAN DIALECT CONVENTION 5 innen innen enne 10 2 3 1 The FORTRAN Integer Typer iei ede eer er EU HE ert FD oer ETE rere Pe coc 10 2 5 2 The Treatment of Short Integer Arithmetic eese eese entente nennen trennen 10 2 5 3 FORTRAN Source Format Used eiiie enii i 12 2 5 4 Default Local Variable Storage 12 2 5 5 The Doloop trip count 12 2 5 0 Specifying Unit Numbers oed tee ida dee eda
105. 11 C 12 STRUCTURE TIME 13 C LOGICAL 1 HOUR MINUTE 14 C END STRUCTURE 15 C RECORD TIME APP TIME 2 16 G 17 STRUCTURE 18 LOGICAL 1 HOUR MINUTE 19 END STRUCTURE 20 21 STRUCTURE APPOINTMENT 22 RECORD DATE APP_DATE 23 24 RECORD TIME APP_TIME 2 25 26 CHARACTER 20 APP_MEMO 4 23 LOGICAL 1 APP FLAG 28 END STRUCTURE 29 RECORD APPOINTMENT NEXT APP APP LIST 7 30 RECORD DATE TODAY 3 WRITE 6 STRUC6 OUT 4 10 1 7 5s d CALL GET DATE I TODAY 6 1 WRITE 6 TODAY DAY TODAY MONTH TODAY YEAR 7 1 APP LIST I APP DATE TODAY 8 I TODAY DAY TODAY DAY 1 9 1 10 END DO 10 5 FORMAT 315 11 NEXT APP APP LIST 1 12 OPEN 1 FILE STRUC6 BIN FORM UNFORMATTED STATUS UNKNOWN 13 WRITE 1 NEXT APP 14 WRITE 1 APP LIST 15 END The heading which is printed at the top of each page contains the name of this compiler along with its current version number on the left hand side of the page The right hand side normally contains the name of the file being compiled the date the time and the page number relative to the file In this case the page width was set to be narrow so the name of the file is placed on a second line The annotated listing itself contains the include file number If the line number within the source file the nesting level of the statement and the actual source code statement The include file number is left blank for statements in the original source fi
106. 140 If Line Nl Source di PROGRAM ERR140 2 PRINT 1 3 1 FORMAT Hello World 3 errl40 for E140 The format specification ends in the symbol END OF STATEMENT rather than a right parenthesis STOP 5 END Error 141 Message A Hollerith string extends beyond the end of the specification Example of message 141 If Line Nl Source 1 PROGRAM ERR141 2 DATA IDAT AHB 2 errl4l for E141 A Hollerith string extends beyond the end of the Specification 3 WRITE A4 IDAT 4 STOP is END Error 142 Message A delimited string has no closing delimeter Example of message 142 If Line Nl Source i PROGRAM ERR142 2 DATA IDAT B 2 errl42 for E142 A delimited string has no closing delimeter 3 WRITE A4 IDAT 4 STOP 5 END Error 143 Message The if conditional expression is preceded by the symbol token rather than a left parenthesis 204 PromulaFortran Compiler User s Manual Example of message 143 If Line Nl Source al PROGRAM ERR143 2 IF I EQ 33 PRINT Hello World 2 errl43 for E143 The if conditional expression is preceded by the symbol I rather than a left parenthesis 3 STOP 4 END Error 144 Message The if conditional expression is followed by the symbol token rather than a right parenthesis Example of message 144 If Line Nl Source 1 PROGRAM 144 2 IF I EQ 33 PRINT Hello World 2 errl44 for E144 The if conditional expre
107. 15 3 02 TOTAL QUANT COST 1 3 WRITE 6 VALUES STOP END produces the following output SVALUES TOTAL 5889 02 QUANT 15E 02 COST 302E 01 SEND No data appears in column 1 of any record Logical constants appear as or F Elements of an array are output in the order in which they are stored Character constants are written with delimiting single quotes Records output by a namelist WRITE statement can be read later in the same program by a namelist READ statement specifying the same group name 5 38 3 Arrays NAMELIST In input data the number of constants including repetitions given for an array name should not exceed the number of elements in the array As a simple example of array use consider the following code fragment INTEGER BAT 10 NAMELIST HAT BAT DOT READ 5 HAT processing the input record HAT BAT 2 3 8 4 DOT 1 05 END The value of becomes 1 05 the array BAT is as follows BAT 1 BAT 2 BAT 3 BAT 4 110 PromulaFortran Compiler User s Manual BAT 5 BAT 6 BAT 7 BAT 8 BAT 9 BAT 10 BRP RE AD As a second example consider the code fragment DIMENSION GAY 5 NAMELIST DAY GAY BAY RAY READ 5 DAY when it processes the input record SDAY GAY 3 7 2 8 3 GAY 5 3 0 2 3 77 25 When data is input in the form array elem
108. 21 CDEXP Exponential Synopsis COMPLEX 16 FUNCTION CDEXP COMPLEX 16 Description CDEXP a is a specific function that returns a double complex result from a double complex argument It returns the exponential of its argument See also CEXP DEXP and EXP 6 22 CDLOG10 Logarithm Base 10 Synopsis COMPLEX 16 FUNCTION CDLOG10 COMPLEX 16 Description CDLOG10 a is a specific function that returns the logarithm base 10 of the argument The argument is double complex and the result is double complex See also ALOG10 CLOG10 DLOG10 and LOG10 6 23 CDLOG Natural Logarithm Synopsis 143 PromulaFortran Compiler User s Manual COMPLEX 16 FUNCTION CDLOG COMPLEX 16 Description CDLOG a is a specific function that returns the natural logarithm of the argument The argument is double complex and the result is double complex See also ALOG CLOG DLOG and LOG 6 24 CDSIN Sine Synopsis COMPLEX 16 FUNCTION CDSIN COMPLEX 16 Description CDSIN a is a specific function that returns a sine The argument is assumed to be in radians The argument and result are both double complex See also CSIN DSIN and SIN 6 25 CDSQRT Square Root Synopsis COMPLEX 16 FUNCTION CDSQRT COMPLEX 16 Description CDSQRT is a specific function that returns the principal square root of its argument The argument and result are double complex Se
109. 3 WRITE 6 100 I 6 1 10 CONTINUE 7 DO 20 I 1 5 8 1 WRITE 6 101 1 2 1 1 9 20 CONTINUE 10 100 FORMAT lt 5 gt dr 101 FORMAT lt I gt F10 lt I 1 11 err298 for E298 A variable lt format specification has no closing gt 12 END Error 299 Message 255 PromulaFortran Compiler User s Manual The format specification has a bad Hollerith string Example of message 299 If Line Nl Source 1 PROGRAM ERR299 2 WRITE 1 3 1 FORMAT 1X 30HHello World 3 err299 for E299 The format specification has a bad Hollerith string 4 STOP 5 END Error 300 Message The executable statement number label was referenced but not defined Example of message 300 If Line Nl Source 1 SUBROUTINE ERR300 IOP 2 WRITE hello world 3 IF IOP EQ 0 GOTO 20 4 STOP 5 END 5 err300 for E300 The executable statement number 20 was referenced but not defined Error 301 Message The format statement number label was referenced but not defined Example of message 301 If Line Nl Source 1 PROGRAM ERR301 2 WRITE 1 hello world 3 STOP 4 END 4 err301 for E301 The format statement number 1 was referenced but not defined Error 302 Message The namelist identifier ident may be used only as an i o target Example of message 302 If Line Nl Source 256 PromulaFortran Compiler User s Manual all SUBROUTINE ERR302 2 NAMELIST NM
110. 4 2 DIMENSION 100 3 DATA 1 10 1 1 100 100 1 2 3 errl94 for E194 The data array A is being subscripted with more than 1 expression s 4 STOP 5 END Error 195 Message 221 PromulaFortran Compiler User s Manual The data substring range for ident contains the symbol token rather than a colon Example of message 195 If Line Nl Source 1 PROGRAM ERR195 2 CHARACTER 100 MESSAGE 3 DATA MESSAGE 56 63 Hello World 3 errl95 for E195 The data substring range for MESSAGE contains the symbol rather than a colon 4 STOP 5 END Error 196 Message The data substring range for ident is terminated by the symbol token rather than a right parenthesis Example of message 196 If Line Nl Source 1 PROGRAM ERR196 2 CHARACTER 100 MESSAGE 3 DATA MESSAGE 56 63 Hello World 3 errl96 for E196 The data substring range for MESSAGE is terminated by the symbol rather than a right parenthesis 4 STOP 5 END Error 197 Message The data statement specification is closed by the symbol token rather than a slash Example of message 197 If Line Nl Source alr PROGRAM ERR197 2 CHARACTER 100 MESSAGE 3 DATA MESSAGE 56 63 Hello World 3 errl97 for E197 The data statement specification is closed by the symbol rather than a slash STOP 5 END Error 198 Message The data statement value list contains too many values Example of message 198 222 PromulaFortr
111. 6 5 27 4 Inpuv Output Conversions e ettet a er rege e E PEERS NE CEPR EEE SEE aeS 77 5 275 Field Separators ein epe De ER ORE TE E oa Er OE ESSAS aR ESTEOS 78 5 27 6 Repeatable and Nonrepeatable Edit Descriptors eese nennen nnne rennen 78 DDI LADO SCFIDUOT i sie Soaked oi a E ADR REG REP VU ERE FEDERER EE ER 78 5 27 8 Single and Double Quote Descriptors esses enne 80 5 27 9 BN and BZ Blank Interpretation eese 80 5 27 10 Carriage Control e ple eter eet eee RO Ee de gere ree Yee dede 80 D2T LED Descriptors iuc uie eaae Ee err eee Oe e 81 3 27 12 E DeSCHIDIOE 81 5 27 13 End of Recotd Slash ciae e nde e Rektor et reed td eod 64 52744 84 85 VOID dU R M 86 3 27 17 L Descriptor 86 5 2748 L EE 87 3 27 19 O DesCriptOFE sieves te costes exo toin ade cie ee ode hodie feet v rode ee e ER D 67 3 27 20 CPI DesCriptOT o re IRR Oder eror dee eyes Saab ine RERO EHE YER 67 2 24 2T SySB SS Plus Sien Gonttol 3 io ee i eter Oa de axes rev ine Ree eee og 89 2 222 T TE ER
112. 6 01 80 4855 07 01 80 1418 08 01 80 0692 09 01 80 0386 10 01 80 0144 11 01 80 0081 12 01 80 0035 End Exit Fn Shift Fn PgUp PgDn Home Arrows Browse Figure 8 8 Actual Values Stored on Datafile Finally suppose that we wish to plot values say average exposure versus concentration The statement below is a simple plot request to do this The result is shown in Figure 8 9 PROMULA PLOT LINE conc avge ITLE Average Exposure as a Function of Concentration EGEND A Character Resolution Plot XLABEL Concentration Average Exposure as a Function of Concentration T4R 4 4 287 0 KOK KR KK 256 0 FO Ok 225 0 FOR ke ek 194 0 163 0 1320970 101 0 70 04 xxx ae ok KKK 39 0 4 4 4 4p 4 0 0 0 1 0 2 05 3 0 4 0 5 0 6 Concentration Character Resolution Plot Figure 8 9 Average Exposure Plot Versus Concentration 188 PromulaFortran Compiler User s Manual As you can see from this example we have extracted information out of a batch FORTRAN program by using PROMULA The example has only scratched the surface Using the PROMULA system as part of your translation strategy can make all of your programs usable in entirely new ways FORTRAN pr
113. 66 dialect Characters are used but are stored in numeric variables therefore this program would not be usable with a pure FORTRAN 77 compiler Except for this presentation this program has no other purpose PROGRAM EXPO C C THIS PROGRAM COMPUTES EXPOSURE LEVELS GIVEN INPUT TIME SERIES C INTEGER I IDAY 50 IMNTH 50 IMOD IOPT IYR 50 LUN1 LUN2 MASS 2 TIME 2 TITLE 20 XLN 2 REAL ABS AVGE 50 AVGI 50 CONC 50 CUME 50 CUMEV CUMI 50 CUMIV DUR EXPOS 50 FREQ PERD 50 SEVER 50 TAKE 50 VBL XMULT LUN1 5 LUN2 6 LUN1 FILE EXPO DAT STATUS OLD LUN2 FILE EXPO RPT STATUS NEW C INITIALIZE THE CUMULATIVE VALUES G CUMEV 0 0 CUMIV 0 0 READ INPUT CONTROLS AND WRITE OUTPUT HEADERS READ LUN1 1000 TITLE I I 1 20 READ LUN1 1005 MASS I I 1 2 XLN I I 1 2 TIME I I 1 2 READ LUN1 1010 IMOD IOPT READ LUN1 1020 ABS XMULT WRITE LUN2 2007 WRITE LUN2 2000 TITLE I I 1 20 MASS I I 1 2 1 XLN I I 1 2 TIME I I 1 2 DETERMINE WHETHER FREQUENCY AND DURATION WILL BE READ IN TIMESERIES C F
114. 8 REAL 8 REAL 8 INTEGER INTEGER INTEGER INTEGER INTEGER REAL REAL REAL REAL REAL REAL NTEGER 2 INTEGER REAL REAL 8 REAL 8 REAL 8 REAL REAL 8 REAL REAL 8 COMPLEX COMPLEX 16 REAL REAL REAL 8 COMPLEX COMPLEX 16 REAL REAL 8 REAL REAL 8 Arguments INTEGER INTEGER 2 REAL INTEGER 2 INTEGER REAL REAL 8 INTEGER INTEGER 2 REAL INTEGER 2 INTEGER REAL REAL 8 INTEGER 2 INTEGER 2 INTEGER INTEGER REAL REAL REAL 8 REAL 8 REAL REAL 8 REAL 8 REAL INTEGER INTEGER INTEGER 2 INTEGER REAL REAL 8 INTEGER 2 INTEGER REAL REAL 8 COMPLEX COMPLEX 16 INTEGER 2 INTEGER 2 INTEGER INTEGER REAL REAL REAL 8 REAL 8 REAL REAL 8 REAL 8 REAL REAL REAL 8 REAL REAL 8 COMPLEX COMPLEX 16 REAL 8 REAL REAL 8 COMPLEX COMPLEX 16 REAL REAL 8 REAL REAL CHARACTER Description Maximum value Maximum value Maximum value Minimum value Minimum value Minimum value Remainder Nearest integer Real value or part Transfer of sign Hyperbolic sine Sine Real value Square root Hyperbolic tangent Tangent Current time 136 PromulaFortran Compiler User s Manual The description of each intrinsic function is given in the following sections 6 1 ABS Absolute Value Synopsis INTEGER 2 FUNCTION ABS INTEGER 2 INTEGER FUNCTION ABS INTEGER REAL FUNCTION ABS REAL
115. 9 for E289 The type of variable KVAL is undefined 7 STOP 8 END Error 290 Message The subprogram argument ident may not be allocated to a common block Example of message 290 If Line Nl Source 1 SUBROUTINE ERR290 IARGU JARGU 2 COMMON ALPHA A B JARGU 2 err290 for E290 The subprogram argument JARGU may not be allocated to a common block 3 RETURN 4 END Error 291 Message The statement function definitions must precede any executable statements Example of message 291 If Line Nl Source 1 PROGRAM ERR291 2 PRINT Hello 3 SFUNC X Y COS X SIN Y 3 err291 for E291 The statement function definitions must precede any executable statements 4 PRINT SFUNC 3 1 5 2 5 STOP 6 END Error 292 Message The object identifier ident may not be assigned a value Example of message 292 If Line Nl Source J PROGRAM ERR292 2 PARAMETER PI 3 14159 3 PI 3 14159 3 err292 for E292 The parameter identifier PI may not be assigned a value 4 STOP 5 END 253 PromulaFortran Compiler User s Manual Error 293 Message The variable ident has type type and not type integer as required in this context Example of message 293 If Line Nl Source ah PROGRAM ERR293 2 DEFINE FILE VEC 100 512 0 2 err293 for E293 The variable VEC has type real 4 and not type integer as required in this context 3 STOP 4 END Error 294 Message The format contains the
116. AL 8 FUNCTION ATAN REAL 8 Description is a generic function that returns an arctangent The result is expressed in radians The result is real or double precision depending on the argument type See also DATAN 6 17 CABS Absolute Value Synopsis REAL FUNCTION CABS COMPLEX Description CABS a returns the real absolute value of its argument The result is the noncomplex square root of the sum of the squares of the two components of the complex number See Also ABS IABS DABS CDABS and I2ABS 6 18 CCOS Cosine Synopsis COMPLEX FUNCTION CCOS COMPLEX Description CCOS a is a specific function that returns a cosine The argument is assumed to be in radians Both the argument and the result are complex See also CDCOS COS and DCOS 6 19 CDABS Absolute Value Synopsis REAL 8 FUNCTION CDABS COMPLEX 16 Description CDABS a returns the absolute value of its argument The result is the double precision square root of the sum of the squares of the two components of the complex number 142 PromulaFortran Compiler User s Manual See Also ABS 5 DABS CABS and I2ABS 6 20 CDCOS Cosine Synopsis COMPLEX 16 FUNCTION CDCOS COMPLEX 16 Description CDCOS a is a specific function that returns a cosine The argument is assumed to be in radians The result and argument are both double complex See also CCOS COS and DCOS 6
117. ASSIGN STATEMENT iit rte t Ria e de REP 48 2 2 BACKSPACE STATEMENT ett eee htt ie e ne et PEU e t Ye trip ettet eri o 48 2 9 BLOCK DATA STATEMENT o reeet Tea entere ret e e br e ete ER EO ERN tnn 49 2 4 BYTE STATEMENT ote ree ER EI e e deett eite be y US 50 SIS CALIZ STATEMENT reri 51 3 0 CLOSE STATEMENT RR RI RN Hetero aede te recen dient 51 3 7 CHARACTER STATEMENT ient Herten er rre ae ete deett 53 5 8 COMMON STATEMENT mE 54 23 9 COMPLEX STATEMENT orte Hh 55 CONTINUE STATEMENT 56 31 DATA STATEMENT erect eene PR HERI ee aee rete 56 5 11 1 Implied DO List Use in DATA Statement i gerorapen en teen paoa i en eene enne ess 58 21 2 Character Dat Initialization i eti te etie eter tee die fee peteret e ethos duns 59 2 12 DECODE STATEMENT HR RR UE 60 3 13 THE DIMENSION STATEMENT cess edite ere n RH PEE Hr a ren eere edem 60 233 1 Adjust ble Dimensions e oe te tmn EE SN Rete 62 23 14 DO STATEMENT 99 i cnet UO OTRO ER E e HER 63 5414 1 Active and Inactive DO Loops eiue edi err i Ra e EE rei Dri creep 64 The Promula Fortran Compil
118. AT statement is reached the line formed internally is output The format control then scans to the left looking for a right parenthesis within the FORMAT statement If none is found the scan stops when it reaches the beginning of the format specification If a right parenthesis is found however the scan continues to the left until it reaches the field separator which precedes the left parenthesis pairing the right parenthesis Output resumes with the format control moving right until either the output list is exhausted or the final right parenthesis of the FORMAT statement is encountered If n slashes are indicated a repetition factor can be used to indicate multiple slashes n 1 lines are skipped on output 5 27 27 Execution Time FORMAT Specification Variable format specifications can be read in as part of the data at execution time and used wherever a normal format can be used The format can be read in under the A specification and stored in a character array variable or array element or it can be included in a DATA statement Formats can also be generated by the program at execution time If an array or array element is used its type can be other than character although character is the preferred type In either case the format must consist of a list of descriptors and editing characters enclosed in parentheses but without the keyword FORMAT and the statement label The name of the entity containing the specifications is used in place of
119. AVGI I 30 CONTINUE 1000 FORMAT 20A4 1005 FORMAT 3 2A4 2X 1010 FORMAT BN 315 1020 FORMAT 8F10 0 1030 FORMAT 312 4X F6 0 2000 FORMAT T2 20A4 1 T2 MASS M UNITS 2A4 2 2 LENGTH L UNITS 2A4 3 2 TIME T UNITS 2A4 2001 FORMAT T45 EXPOSURE ROUTE 1 T45 2007 FORMAT 2 DKK Ck ck ck Ck ck kk ck Ck ck ck Ck ck kk ck Ck ck kk ck Ck ck ck KKK KKK KK 1 1 2 EXPOSURE ANALYSIS MODEL NT 2 T2 VERSION 1 vof 3 Tay VOR de de ede REOR e ee RIOR EE HORROR OR EIR AES 2010 FORMAT T26 Al 2011 FORMAT T25 M BACON DB BI ENT T54 CU T99 CUMULA IVE 8 MONTH U T116 AV 16 YEAR T23 ULATIVE 70 AV pI C T2 DAY ENTRATION 40 5 EXPOSUR E T 86 INTAK T101 INTAK L 3 T41 P T116 70 T T57 M T71 M T 181 PromulaFortran Compiler User s Manual 1 T86 M T T02 M T117 M T 2020 FORMAT T2 23 1 T40 2 T86 2030 FORMAT 2X I2 T9 I2 T17 I2 125 G9 3 140 G9 3 T55 G9 3 170 G9 3 1 185 G9 3 T100 G9 3 T115 G9 3 STOP
120. COPYRIGHT NOTICE for An Enhanced FORTRAN Compiler Version 9 38 Released May 2007 Published by Great Migrations LLC 7453 Katesbridge Ct Dublin Ohio 43017 614 761 9816 This User s manual for the PROMULA FORTRAN Compiler is the property of Great Migrations LLC It embodies proprietary confidential and trade secret information The User s manual and the files of the PROMULA FORTRAN Compiler machine readable distribution media are protected by trade secret and copyright laws The use of the PROMULA FORTRAN Compiler is restricted as stipulated in the Great Migrations LLC License Agreement which came with the PROMULA FORTRAN Compiler product and which you completed and returned to the Great Migrations LLC The content of the machine readable distribution media and the User s manual may not be copied reproduced disclosed transferred or reduced to any electronic machine readable or other form except as specified in the License Agreement with the express written approval of Great Migrations LLC The unauthorized copying of any of these materials is a violation of copyright and or trade secret law DISCLAIMER OF WARRANTIES AND LIMITATIONS OF LIABILITIES THIS USER S MANUAL IS PROVIDED ON AN AS IS BASIS EXCEPT FOR THE WARRANTY DESCRIBED IN THE GREAT MIGRATIONS LLC LICENSE AGREEMENT THERE ARE NO WARRANTIES EXPRESSED OR IMPLIED INCLUDING BUT NOT LIMITED TO IMPLIED WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE AND ALL
121. DATA A I I 1 100 100 1 2 3 errl90 for E190 The data implied do loop variable is followed by the symbol rather than an equals sign 4 STOP 5 END Error 191 Message The variable ident is used in an assigned goto but is never assigned a statement label Example of message 191 If Line Nl Source 1 SUBROUTINE ERR191 2 GOTO IA 10 20 3 10 A 1 4 WRITE A 5 20 RETURN 6 END 6 errl91 for E191 The variable IA is used in an assigned goto but is never assigned a statement label 220 PromulaFortran Compiler User s Manual Error 192 Message The data implied do loop minimum is followed by the symbol token rather than a comma Example of message 192 If Line Nl Source 1 PROGRAM ERR192 2 DIMENSION A 100 3 DATA 1 100 100 1 2 3 errl192 for E192 The data implied do loop minimum is followed by the symbol rather than a comma 4 STOP 5 END Error 193 Message The data implied do loop range is followed by the symbol token rather than a right parenthesis Example of message 193 If Line Nl Source 1 PROGRAM 193 2 DIMENSION 100 3 DATA 1 100 100 1 2 3 errl93 for E193 The data implied do loop range is followed by the symbol rather than a right parenthesis 4 STOP 5 END Error 194 Message The data array ident is being subscripted with more than number expression s Example of message 194 If Line Nl Source 1 PROGRAM ERR19
122. DATA bdname Where bdname is a symbolic name identifying the block data subprogram Description The block data subprogram is a nonexecutable specification subprogram that can be used to enter initial values for variables and array elements in common blocks Both named and blank common may be initialized A program can have more than one block data subprogram Only one block data subprogram can be unnamed The BLOCK DATA statement must appear as the first statement of the block data subprogram The name used for the block data subprogram must not be the same as any local variables in the subprogram The name must not be the same as any other program unit or entry name in the program Block data subprograms can contain IMPLICIT PARAMETER DIMENSION TYPE COMMON SAVE EQUIVALENCE or DATA statements A block data subprogram ends with an END statement Data can be entered into more than one common block in a block data program All variables having storage in the named common must be specified even if they are not all initially defined 5 4 BYTE Statement The BYTE statement defines some user defined entity to be of type byte Syntax BYTE name name Where name has one of the forms var c array d d clist var is a variable function name symbolic constant or dummy procedure array isan array name d specifies the bounds of a dimension clist isalist of constants or symbolic constants specifying the ini
123. DIC SCHAP IOVS de teo devi n ee YO RR e 89 3 27 23 Termination of Format Controls a veio ire e ORE kd 90 3 27 24 X DesCFIDIOT ic a eo enacted ie e T a e f eae ade o iere 90 2 27 25 Descriptotisoiiad eo edo oe t tede tede ive dvo eir tese A do tee e y vee 90 3 27 26 Repetition Factors ioco nete pe f o o ee qoe fete ei ated epe Ronee eae 90 5 27 27 Execution Time FORMAT Specification nss sivede eier eine a nen nennen rene 91 2 20 FUNCTION STATEMENT p ORE NERO DRE CP b irte ER 92 2 20 GENERIC STATEMENT oie er etae Ite i eo ribi 93 3 90 GOTO STATEMENT eet ES nter rti ree ten ero ERU ep RHET vie eius 94 23 90 Unconditional GO IQ soe Ue e ee EE ta Meta eed ea 94 3 202 Computed Statement seri SOC LS NE 94 3 303 Assigned GOTO Statement us eee a Ob EH epatis es 95 3 2 1 IE STATEMENT t Ue EN RR 95 2 94 M Arithimetic IF State mentee innare Eee qa te even aee eet epe ree ea eee en 96 2131 2 Logical IF Statements cg aU eie e Roda den QE PUN 97 2 94 39 BLOCK LE Statements e ee Ree rr t dese vete teet oa YIN Fre a s REPRE en 97 5 32 IMPLICIT STATEMENT E Dr
124. E or another FUNCTION statement A function must not directly or indirectly reference itself The symbolic name of a function subprogram or an associated entry name of the same type is a variable name in the function The symbolic name specified FUNCTION or ENTRY statement must not appear in any other nonexecutable statement except a type statement If the type of a function is specified ina FUNCTION statement then the function name cannot appear in a type statement In an executable statement the symbolic name can appear only as a variable During execution this variable becomes defined and can be referenced or redefined The value of the function is the value of this variable when control returns to the referencing program unit The type of the function name must be the same in the referencing program unit and the referenced function subprogram When type is omitted the type of the function is determined by the first character of the function name Implicit typing by the IMPLICIT statement takes effect only when the function name is not explicitly typed The name cannot have its type explicitly specified more than once The symbolic name of a function subprogram must not be the same as any other name except a variable name or common block name The function subprogram can have more than one entry point although alternate returns are prohibited Multiple entry points are established through the ENTRY statement In a function subprogra
125. END Error 183 Message The referent of the pointer token is not a valid identifier symbol Example of message 183 If Line Nl Source 1 PROGRAM ERR183 2 PARAMETER IP 4 JP 10 3 POINTER P V 0 IP 0 JP 3 err183 for E183 The referent of the pointer is not a valid identifier symbol 4 REAL A 55 5 DATA A 3 123 456 2 3 45 5 7 89 45 6 78 6 LOC a 7 PRINT computed V 0 0 should be A 1 8 STOP 9 END Error 184 Message The data statement subscript expression is too complex to process Example of message 184 If Line Nl Source 1 PROGRAM ERR184 2 DIMENSION IVAL 100 3 DATA I J IVAL J I 50 100 32 3 errl84 for E184 The data statement subscript expression is too complex to process 4 STOP 5 END Error 185 Message The pointer specification ends in the symbol token rather than a right parenthesis Example of message 185 If Line Nl Source 1 PROGRAM ERR185 2 4 10 3 POINTER P V 0 IP 0 JP 218 PromulaFortran Compiler User s Manual 3 errl85 for E185 The pointer specification ends in the symbol END OF STATEMENT rather than a right parenthesis REAL A 55 DATA A 3 123 456 2 3 45 5 7 89 45 6 78 P LOC a PRINT computed V 0 0 should be 1 STOP END Error 186 Message The value number for subscript number of array ident with m
126. END Note 833 Message 272 PromulaFortran Compiler User s Manual The nonparenthetical form of the parameter statement is nonstandard Example of message 833 If Line Nl Source 1 SUBROUTINE W833 2 PARAMETER 3 4 4 2 warn833 for N833 The nonparenthetical form of the parameter statement is nonstandard 3 WRITE I J 4 END Note 834 Message Using a repeat count with the token format specification is nonstandard Example of message 834 If Line Nl Source 1 SUBROUTINE W834 2 READ 5 10 A 3 10 FORMAT 3 F10 0 3 warn834 for N834 Using a repeat count with the format specification is nonstandard 4 1 0 5 END Note 835 Message Omitting the field after the period in a format specification is nonstandard Example of message 835 If Line Nl Source 1 SUBROUTINE W835 2 WRITE 5 10 1 3 10 FORMAT 1X I3 3 warn835 for N835 Omitting the field after the period in a format Specification is nonstandard 4 WRITE 5 20 2 5 20 FORMAT 1X F6 5 warn835 for N835 Omitting the field after the period in a format Specification is nonstandard 6 WRITE 5 30 3 D0 E 30 FORMAT 1X 1P E20 7 warn835 for N835 Omitting the field after the period in a format Specification is nonstandard 8 END Note 836 273 PromulaFortran Compiler User s Manual Message Omitting the repeat count before the token format specificatio
127. ERR205 2 INTEGER I 4 J 2 K 4 2 err205 for E205 The element length specification for J is closed by the symbol rather than a parenthesis 3 STOP 4 END Error 206 Message The character size specification token for ident is not a constant Example of message 206 If Line Nl Source 1 PROGRAM ERR206 2 INTEGER 4 0 2 2 err206 for E206 The character size specification for is not a constant 3 STOP 4 END Error 207 Message The declaration statement value list contains too many values Example of message 207 If Line Nl Source 1 PROGRAM ERR207 2 CHARACTER 100 MESSAGE Hello World Goodbye 225 PromulaFortran Compiler User s Manual 2 err207 for E207 The declaration statement value list contains too many values 3 STOP 4 END Error 208 Message The statement statement contains the unrecognized symbol token Example of message 208 If Line Nl Source al PROGRAM ERR208 2 DIMENSION A 10 B 10 2 err208 for E208 The dimension statement contains the unrecognized symbol 3 STOP 4 END Error 209 Message The common block identifier token is not a valid identifier symbol Example of message 209 If Line Nl Source T PROGRAM ERR209 2 COMMON 123 A 10 B 113 2 err209 for E209 The common block identifier 123 is not a valid identifier symbol 3 STOP 4 END Error 210 Message The common block identifier is followed by the symbol token
128. FUNC VAL is equivalent to HIGH VAL The EXTERNAL statement also specifies that the procedure is a user written rather than an intrinsic function Consider the following example SUBROUTINE ARGR EXTERNAL SQRT Y SQRT X END FUNCTION SQRT XVAL Since the name SORT is declared external the function reference SORT X references the user written function SQRT rather than the intrinsic function SQRT If an external procedure name is an actual argument in a program unit it must appear in an EXTERNAL statement in the program unit A statement function name must not appear in an EXTERNAL statement If an intrinsic function name appears in an EXTERNAL statement in a program unit the name becomes the name of some external procedure The intrinsic function with the same name cannot be referenced in the program unit 5 27 FORMAT Statement The FORMAT statement is a nonexecutable statement which specifies the formatting of data to be read or written with formatted I O Syntax sl FORMAT flist Where sl is a statement label flist isalist of items separated by commas having the following forms r ed ned r flist ed is a repeatable edit descriptor ned is a nonrepeatable edit descriptor r is a nonzero unsigned integer constant repeat specification Format specifications are used in conjunction with formatted input output statements to produce output or read input that consists of strings of dis
129. Fatal preprocessor errors Each error message is actually a template from which the actual message is built Identifier Description of variable part object An object type such as variable parameter statement function etc ident A user defined identifier of a symbol note that these are always shown in upper case to make them stand out number A number or count derived from the user supplied information token An actual token or tokens within the source statement note that these are always shown upper case to make them stand out type A binary type such as integer real integer 2 etc label A statement or format label option A statement option description statement A statement identifier 9 2 1 Syntax Errors Warnings Comments and Notes When a syntax error is encountered a message is issued and the remainder of the statement is skipped In addition the formation of the intermediate C output is blocked With warnings comments and notes processing continues These are for your information only See the chapter on using the compiler for details on how to control these optional messages Each message consists of four parts separated by a colon as is shown in the sample message below 3 errlO0l for E101 The statement function identifier IALPHA may not be assigned a value The first part of the message is the record number within the source file of the start of the statement causing the messag
130. GRAM ERR262 2 STRUCTURE SHIP 3 of CHARACTER 10 OWNER 4 1 CHARACTER 10 DESTINATION 9e 71 END STRUCTURE SHIP 6 err262 for E262 The structure definition identifier SHIP has already been defined as a structure identifier gi CHARACTER 10 OWNER2 8 CHARACTER 10 DESTINATION2 9 END 10 RECORD SHIP NINA PINTA SANTA_MARIA 11 CHARACTER 10 WHERE 12 WHERE NINA 3 T3 STOP 14 END Error 263 Message The actual argument list contains less than the number argument s required by the object ident Example of message 263 If Line Nl Source PROGRAM ERR263 243 PromulaFortran Compiler User s Manual 2 2 err263 for 2 Error 264 Message ALPHA AMOD BETA E263 The actual argument list contains less than the argument s required by the system function AMOD END The structure or record definition identifier is introduced by the symbol token rather than a slash Example of message 264 If Line Nl Source 6 err264 for 7 8 8 err264 for Error 265 Message PROGRAM ERR264 STRUCTURE SHIP CHARACTER 10 OWNER CHARACTER 10 DESTINATION END RECORD SHIP NINA PINTA SANTA MARIA E264 The structure or record definition identifier is introduced by the symbol SHIP rather than a slash CHARACTER 10 WHERE WHERE NINA OWNER E145 The assignment statement contains the unrecognized symbol STOP END The structure or record identifier token is not a valid id
131. GT 0 033 date tp iday tp imnth tp 100 iyr tp 10000 034 END main 035 END PROGRAM DO main 036 STOP Lines and 2 define the program and the file to be used to contain the actual database description Lines 3 5 specify that an array file is to be defined and Line 6 actually creates that file with the name EXPO DBA Lines 7 9 define the time points classification scheme which classifies the values of the variables to be manipulated Lines 10 23 define the actual variables to be contained on the datafile Notice that along with the variable identifiers and structures the binary types and descriptions are also included This information documents the datafile and is also later used by PROMULA to work with these variables Lines 24 29 define a variable date which is not on the database but which will be used as a KEY to the individual entries in the time points classification scheme The significance of this key relation will become clearer later when we actually work with the values Lines 30 34 contain the minimal amount of startup logic that we will need to work with the actual values Line 31 opens the datafile Line 32 looks at the day values and restricts the time points classification scheme to those entries where a day is defined Line 33 computes actual date values from the raw inputs from the FORTRAN program This is all the logic we need to work with the values The actual working with the values will be done in PROMULA
132. HAR CHARACTER Description ICHAR a is a specific function that returns an integer value from a character argument The value returned depends on the collating weight of the character in the collating sequence used See also CHAR 6 72 Positive Difference Synopsis INTEGER FUNCTION IDIM INTEGER INTEGER Description IDIM al a2 is a specific function that returns an integer positive difference between its integer arguments If a1 is greater that a2 then the value of al a2 Is returned else zero is returned See also DDIM DIM and I2DIM 6 73 IDINT Integer Value Synopsis INTEGER FUNCTION IDINT REAL 8 Description 159 PromulaFortran Compiler User s Manual IDINT a is a specific function that performs conversion to integer and returns an integer result from a double precision argument The conversion to integer is performed via simple truncation See also IFIX INT INT2 and INT4 6 74 IDNINT Nearest Integer Synopsis INTEGER FUNCTION IDNINT REAL Description IDNINT a is a specific function that returns an integer result from a real argument It computes the nearest integer to its real argument In particular the result is int 0 5 if a gt 0 0 and int a 0 5 if a lt 0 0 where int is the FORTRAN intrinsic function INT See also ANINT DNINT I2NINT and NINT 6 75 IFIX Integer Va
133. HARACTER 10 NAME HOME SON 3 DATA NAME 0 9 Freddy 3 err225 for E225 The variable NAME substring subscript 0 is not in the valid range of 1 to 10 4 DATA 1 13 Columbus Ohio 4 err225 for E225 The variable HOME substring subscript 13 is not in the valid range of 1 to 10 5 DATA 5 1 Andy 5 err225 for E225 The variable SON substring subscript 1 is not in the valid range of 5 to 10 6 WRITE NAME gi STOP 8 END Error 226 Message The parameter element ident is followed by the symbol token rather than an equals sign Example of message 226 231 PromulaFortran Compiler User s Manual If Line Nl Source 1 PROGRAM ERR226 2 PARAMETER PI 3 14159 2 err226 for E226 The parameter element PI is followed by the symbol rather than an equals sign 3 PRINT PI 4 STOP 5 END Error 227 Message The parameter element ident definition is followed by the symbol token rather than a constant Example of message 227 If Line Nl Source 1 PROGRAM ERR227 2 PARAMETER PI 3 14159 2 err227 for E227 The parameter element PI definition is followed by the symbol END_OF_STATEMENT rather than a constant 3 PRINT PI 4 STOP 5 END Error 228 Message The namelist statement begins with the symbol token rather than a slash Example of message 228 If Line Nl Source 1 PROGRAM ERR228 2 NAMELIST SHIP 11 12 2 err228 for E228 The nameli
134. IC fun fun Where fun is an intrinsic function name Notes The INTRINSIC statement performs the same action as the GENERIC statement Description The INTRINSIC statement is used to identify a name as representing a generic or intrinsic function This statement enables use of a generic function name as an actual argument As an example SUBROUTINE DC INTRINSIC SORT CALL SUBA X Y SQRT END SUBROUTINE SUBA A B FNC B FNC A END The name SORT is declared generic in subroutine Dc and passed as an argument to subroutine SUBA Within SUBA the reference FNC A references the generic function SQRT Note that if SORT were declared EXTERNAL then a user function would be assumed The appearance of a name in an INTRINSIC statement declares the name as a generic intrinsic function name If an intrinsic function name is used as an actual argument in a program unit it must appear in a GENERIC or INTRINSIC statement in the program unit The following intrinsic function names must not be used as actual arguments Type conversion functions BOOL CHAR CMPLX DBLE FLOAT ICHAR IDINT IFIX INT REAL and SINGL Lexical relationship functions LGE LGT LLE and LLT Largest smallest value functions AMAX0 AMAXI AMINO AMINI DMAXI DMINI MAX MAXO MAXI MIN MINO MINI Logical and masking functions AND OR XOR NEQV EQV The appearance of a generic intrinsic function name in an INTRINSIC statement does not remov
135. IM 2 COMMON ALPHA A IDIM 3 WRITE 4 END 4 err250 for E250 The common variable A has the adjustable dimension IDIM 239 PromulaFortran Compiler User s Manual Error 251 Message The value argument token has been entered where a subprogram argument is required Example of message 251 If Line Nl Source I PROGRAM ERR251 2 EXTERNAL FUNC IFUNC 3 CALL SUB251 FUNC 4 CALL SUB251 IFUNC 5 CALL SUB251 AVAL BVAL 5 err251 for E251 The value argument AVAL BVAL has been entered where a subprogram argument is required 6 STOP u END Error 252 Message The symbol token is not valid in the formal argument list Example of message 252 If Line Nl Source 1 SUBROUTINE ERR252 ALPHA BETA 32 1 err252 for E252 The symbol 32 is not valid in the formal argument list 2 RETURN 3 END Error 253 Message The local variable ident has an adjustable dimension ident Example of message 253 If Line Nl Source ju SUBROUTINE 253 IDIM 2 REAL A IDIM 3 WRITE 4 END 4 err253 for E253 The local variable A has an adjustable dimension IDIM Error 254 Message The formal argument list contains the symbol token rather than a comma or right parenthesis 240 PromulaFortran Compiler User s Manual Example of message 254 If Line Nl Source 1 SUBROUTINE ERR254 A B 1 err254 for E254 The formal argument list contains the symbol ra
136. IP H O01G1 3 err309 for E309 The illegal character G is in the hexidecimal constant H O1G1 4 WRITE IT IP 5 END Error 310 Message The object ident may not be redefined as a statement Example of message 310 If Line Nl Source 1 FUNCTION E310 2 EXTERNAL SINE 3 PARAMETER SINE 3 3 3 err310 for E310 The function SINE may not be redefined as a parameter 4 E310 SINE 5 END Error 311 Message The common variable ident may not be redefined as a statement Example of message 311 If Line Nl Source t FUNCTION ERR311 K 2 COMMON J 259 PromulaFortran Compiler User s Manual 3 PARAMETER J 2 3 err3ll for E311 The common variable J may not be redefined as a parameter 4 E595 J 5 END Error 312 Message The subprogram argument ident may not be redefined as a statement Example of message 312 If Line Nl Source 1 FUNCTION ERR312 2 PARAMETER K 3 2 err312 for E312 The subprogram argument K may not be redefined as a parameter 3 E595 K 4 END Error 313 Message The dimensioned variable ident may not be redefined as a statement Example of message 313 If Line Nl Source T FUNCTION ERR313 K 2 REAL A 10 3 PARAMETER A 3 45 3 err313 for E313 The dimensioned variable may not be redefined as a parameter 4 E595 K A 5 END Error 314 Message The equivalenced variable ident may not be redefined as a statement
137. If Line Nl Source 1 PROGRAM ERR123 2 E005A A A A A 2 errl23 for E123 The statement function argument name A may not be repeated 3 PRINT ERR123 4 END Error 124 Message The implied do loop counter ident is not an integer variable Example of message 124 If Line Nl Source 1 PROGRAM ERR124 2 DIMENSION A 10 3 10 FORMAT 10A10 4 READ IFILE 10 A I R 1 10 4 errl24 for E124 The implied do loop counter R is not an integer variable 5 STOP 6 END Error 125 Message The implied do loop minimum is followed by the symbol token rather than a comma Example of message 125 198 PromulaFortran Compiler User s Manual If Line Nl Source 4 errl25 for 5 6 Error 126 Message PROGRAM ERR125 DIMENSION A 10 10 FORMAT 10A10 READ IFILE 10 A I I 1 10 E125 The implied do loop minimum is followed by the symbol rather than a comma STOP END The implied do loop specification is followed by the symbol token rather than a right parenthesis Example of message 126 If Line Nl Source 4 errl26 for 5 6 Error 127 Message PROGRAM ERR126 DIMENSION A 10 10 FORMAT 10A10 READ IFILE 10 A I I 1 10 E126 The implied do loop specification is followed by the symbol rather than a right parenthesis STOP END The option parameter was not supplied Example of message 127 If Line Nl Source 2 errl27 for 3 4 Error 128 Message PROGRAM E
138. Inclusive disjunction Sum of x1 and x2 EQV Equivalence Is x1 equivalent to x2 NEQV Nonequivalence Is x1 not equivalent to x2 XOR Exclusive disjunction Difference of x1 and x2 A set of rules establishes the interpretation of a logical expression that contains two or more logical operators A precedence among the logical operators determines the order in which the operands are to be combined unless the order is changed by the use of parentheses The precedence of the logical operators is NOT Highest AND OR 42 PromulaFortran Compiler User s Manual EQV Or NEQV or XOR Lowest For example in the expression A OR B AND the AND operator has higher precedence than the OR operator therefore the interpretation is the same as A OR B AND C Logical quantities are combined from left to right when a logical expression contains two or more AND operators two or more OR operators or two or more EQV Or XOR operators The value of a logical factor involving any logical operator is shown below 1 2 2 x1l AND x2 x1 OR x2 2 2 x1 NEQV x2 x1 XOR x2 T T F T T T F F T T F T F T T F ik F F T T F T A logical constant expression contains only logical constants symbolic names of logical constants relational expressions which contain only constant expressions or logical constant expressions enclosed in parentheses 4 1 4 Rela
139. L A B 3 PARAMETER NML 3 3 err302 for E302 The namelist identifier NML may be used only as an i o target 4 REAL NML 4 err302 for E302 The namelist identifier NML may be used only as an i o target 5 EXTERNAL NML 5 err302 for E302 The namelist identifier NML may be used only as an i o target 6 NML 1 1 2 6 err302 for E302 The namelist identifier NML may be used only as an i o target 7 ASSIGN 10 TO NML 7 err302 for E302 The namelist identifier NML may be used only as an i o target 8 GOTO NML 8 err302 for E302 The namelist identifier NML may be used only as an i o target 9 10 READ 5 NML 10 END Error 303 Message The specifier token is not valid for the statement statement Example of message 303 If Line Nl Source 1 SUBROUTINE ERR303 2 CLOSE UNIT 1 END 10 2 err303 for E303 The specifier END is not valid for the close statement 3 INQUIRE UNIT 1 FMT 5 3 err303 for E303 The specifier FMT is not valid for the inquire statement 4 END Error 304 Message Both unit number and unit name may not be specified on the statement statement Example of message 304 If Line Nl Source 1 SUBROUTINE ERR304 2 INQUIRE UNIT 5 FILE DUMMY IOSTAT IOS 2 err304 for E304 Both unit number and unit name may not be specified on the inquire statement 3 WRITE IOS 4 END Error 305 257 PromulaFortran Compiler User s Manual Message The entry parameter identifier ident
140. N runtime library about 300 functions in object form The runtime library may be linked into your executables for royalty free distribution 1 1 User Support If you are a licensed and registered user of the PROMULA FORTRAN compiler then you are entitled to user support from the Great Migrations LLC If you encounter a problem that you cannot resolve on your own by referring to this User s Manual you may call or write us Great Migrations LLC PROMULA FORTRAN Support 7453 Katesbridge Ct Dublin Ohio 43017 614 761 9816 Your comments and suggestions about this product are always welcome If possible we will provide support over the telephone However if the problem involves an apparent compilation or runtime library problem we will probably need a copy of your source FORTRAN code We will protect the full confidentiality of any sample codes that you send to us 1 2 The Design of the Compiler The PROMULA FORTRAN Compiler PF is a comprehensive FORTRAN compiler for C based platforms It works by converting FORTRAN source codes into intermediate C source codes These are then processed by the C compiler of the host platform to produce object code There are four fundamental reasons for this design portability integration efficiency and completeness By using C as an intermediate step applications using this compiler achieve a high degree of portability Our compiler is highly portable Once your code compiles and runs under PF
141. ND Note 871 Message The statement function argument ident is never used Example of message 871 If Line Nl Source 1 SUBROUTINE W871 A B C 2 ASF G B A C 285 PromulaFortran Compiler User s Manual 2 warn871 for 3 4 5 Note 872 Message N871 The statement function argument G is never used A SIN ASF 1 0 RETURN END Equivalencing ident of type type with ident of type type is nonstandard Example of message 872 If Line Nl Source lt lt 10 11 12 13 14 Comment 873 Message 4 warn872 4 warn872 4 warn872 4 warn872 4 warn872 LOr LOr LOr for for SUBROUTINE W872 DOUBLE PRECISION D COMPLEX C CHARACTER 10 51 52 53 54 55 LOGICAL L INTEGER I REAL R EQUIVALENCE EQUIVALENCE EQUIVALENCE EQUIVALENCE EQUIVALENCE 1 51 R 2 D 3 C 54 1 55 WRITE END 872 Equival 872 Equival real 4 872 Equival real 8 872 Equival 872 Equival logical lencing 51 of type c integer 4 is nonstandard lencing S2 of type c is nonstandard lencing S3 of type c is nonstandard lencing S4 of type c complex 8 is nonstandard lencing S5 of type c 4 is nonstandard narac narac narac narac narac ter 10 ter 10 ter 10 ter 10 ter 10 wi wi wi wi wi of of of of of The
142. OTES 190 9 2 2 Fatal Preprocessor i eie eee hee PR e p 299 9 3 RUNTIME ERROR 55 5 299 PromulaFortran Compiler User s Manual 1 INTRODUCTION The PROMULA FORTRAN Compiler is a general purpose multi dialect and portable FORTRAN compiler It runs on multiple platforms and supports both the ANSI FORTRAN 66 and ANSI FORTRAN 77 standard dialects as well as a large number of common extensions such as those found in the following commercial compilers VAX FORTRAN PDP FORTRAN PRIME FORTRAN Data General FORTRAN and Sun FORTRAN Some Fortran 90 extensions are also supported In cases where different versions of FORTRAN have conflicting features or conventions a dialect selection command switch can be used to select the desired set Extended FORTRAN applications will compile with this product on open platforms as is i e without making changes to their FORTRAN source code Validated by the GSA FORTRAN Compiler Validation Test Suite on every platform this compiler works by producing intermediate C source code It requires a C compiler to produce executable code and is the ideal processor for hybrid FORTRAN C applications FORTRAN debugging is supported by the debugger of the host C platform e g the dbx tool for UNIX platforms The product includes both the compiler and an extensive FORTRA
143. OUBLE 7 8 t 1 0 1E 27 QUAD 7 err278 for E278 The kind value 20 associated with ERR278 is not available for the real type 9 END Error 279 Message A statement function argument identifier token is not a valid identifier symbol Example of message 279 If Line Nl Source 1 PROGRAM ERR279 2 SFUNC X 6 SIN X COS Y 2 err279 for E279 A statement function argument identifier 6 is not a valid identifier symbol 3 STOP 4 END 249 PromulaFortran Compiler User s Manual Error 280 Message A statement function argument list is terminated by the symbol token rather than a right parenthesis Example of message 280 If Line Nl Source ah PROGRAM ERR280 2 SFUNC X Y SIN X COS Y 2 err280 for E280 A statement function argument list is terminated by the symbol rather than a right parenthesis 3 STOP 4 END Error 281 Message A statement function parameter list is followed by the symbol token rather than an equals sign Example of message 281 If Line Nl Source 1 PROGRAM ERR281 2 SFUNC X Y SIN X COS Y 2 err281 for E281 A statement function parameter list is followed by the symbol rather than an equals sign 3 STOP 4 END Error 282 Message The left hand side of the assignment statement begins with the symbol token which is not a valid identifier symbol Example of message 282 If Line Nl Source 1 PROGRAM ERR282 2 DIMENSION AVAL 10 10
144. Procedure Communication 35 3 6 2 Adjustable Dimensions ee Ee ee be Hee ede eet etd eee eee ee dee eade 37 3 6 3 Using COMMON BIOCKS 1 tie iE et dee e EE ed ig enger e Henkel echt dede eie 38 4 EXPRESSIONS LVALUES ASSIGNMENTS AND STATEMENT 8 0204 0 eene 39 EXPRESSIONS On Here trt te 39 4 1 Arithmetic EXDF SSIOR S d ham hne 39 4 12 Character Expressions 2 ones ete amete e ette dee 41 4 1 3 LOGICAL EXpressi W eed nne t nde nensi 41 LIA Relational et ertet tase e e Here eei ette ee cach eed ee etes ects d 43 4 1 5 General Rules for Expressions i tnn dnte eee RE etin testet Re 44 42 hhencheRUed d eombhebtnetie ee ote HEIDE 44 4 3 ASSIGNMENT STATEMENTS n e REY D SERERE THREE Pu IRR S Lee 45 4 321 ArithmeticzAsSignmenti i eR RERO RENE RU 45 4 3 2 Character Assignment x ode deni ert e PEERS ERREUR RUE RR FER ERU DR HERE 45 43 3 FogicabAssienmeht od ri E E ERROR e E EORR NE UE stand PN SE tr E totes e e HE er ee 46 44 STATEMENT A LON SA H O I EGER OO UE HT RR 46 5 STATEMENTS SUPPORTED eeeeeeeee eese ense tn atta tuse sees suse tasa an 48 2 1
145. R 287 PromulaFortran Compiler User s Manual 12 CHARACTER 21B C E 3 F 3 13 COMMON E156Y B C 14 COMMON E1562 E F 15 16 FFF T3 WRITE E F 18 END 18 warn876 for C876 Common block E156X is 4 bytes longer than when previously defined 18 warn876 for C876 Common block E156Y is 21 bytes longer than when previously defined 18 warn876 for C876 Common block E1562 is 3 bytes longer than when previously defined Comment 877 Message Common block ident is number bytes shorter than when previously defined Example of message 877 If Line Nl Source WRITE E F END COMMON W877Y B COMMON W8772 E E EEE FFF WRITE E F 8 END SUBROUTINE W877 COMMON W877X A 100 CHARACTER 21 B C E 3 F 3 COMMON W877Y B C COMMON W877Z E F SUBROUTINE W877A COMMON W877X A 99 CHARACTER 21 B C E 3 F 3 18 warn877 for C877 Common block W877X is 4 bytes shorter than when previously defined 18 warn877 for C877 Common block W877Y is 21 bytes shorter than when previously defined 18 warn877 for C877 Common block W877Z is 3 bytes shorter than when previously defined Comment 878 Message The statement statement is ignored in block data subprograms Example of message 878 If Line Nl Source 1 BLOCK DATA W878 2 COMMON PI PI 288 PromulaFortran C
146. RAN Compiler Validation test on every platform that it operates on Other contemporary FORTRAN compilers cover only the FORTRAN standard along with selected extensions from other dialects At PROMULA our primary concern is the dialect coverage of our compiler and its runtime library We not only support the VAX PDP PRIME DG and SUN extensions but we also provide a dialect selection switch to allow for conflicting features The PROMULA FORTRAN Compiler is completely compatible with and shares its front end with PROMULA FORTRAN our FORTRAN to C translator and portation support tool In addition the compiler is compatible with our Application Management system PROMULA Via a simple configuration file PF will add PROMULA data base access instructions to your program during compilation thus unlocking the information within those programs for external management It is the dream of Great Migrations LLC that ultimately there will be only one compiled language be it ANSI C or some other language yet to be designed or discovered All compiler technology will be focused on making this language as efficient as possible on every platform supporting it Operating systems and language processors will be written in this language and will work through this single language thus maximizing portability and minimizing startup time on new platforms Variation and experimentation in hardware design and in programming languages tools and techniques sh
147. RE DEREN ud EE Red 19 REP WIHIEXIMUUP 20 SU FORTRAN STATEMENTS etu iat tpe tene en ertet 20 3 1 1 Standard Fixed Format Lines eee 21 3 1 2 Tab Format Lines tee eee tense ee eben teneo eerie eee eere tede aseo dea vt 21 3 1 3 Normal Free Format Lines a i TAE e e tne 22 3 14 Continuation FreezsFormat LANES uei ode e ente ta get Vane Pret e eee Ree ete ele e Ete y fuse go 22 32 SYMBOLIC NAMES tacet dei I EE ERREUR Ert da solu UTER veda ge Ere Ce je e Rer dine 22 3 99 CONSTANTS run ettet cee ice sack PR UE rn te ERR Gi 22 3 2 4 Integer Constant ede ae ER ee oe Pie ee D ode Ae ei Re e ade ee eade i ete Dried 22 3 3 2 CUNG II EE 23 3 9 9 Double Precision Constant P ee Ev eee ee Recte dee s vade et Pe Eee vede 24 3 3 4 Complex Constant eit dio fete te rode i e et e e ede e e eii eut 25 3 3 3 Double Complex Constant ci cite iet tede Ode e Rr eie deu dea e ivre ee A ante Re Hae 25 3 3 6 Logical temi eie dt i e et ite a ee fede o ebd a 26 3 37 Character Constant st edet edite e ive SA eh ted eg 26 The Promula Fortran Compiler User s Manual 3 3 0 Hollerith Constan i eet tr te T eet t Eee entes edere ries teet 26 3 3 9 Exact Representation
148. RITE statement is free Description The WRITE statement transfers output data from internal storage to records contained in external logical units or internal files There are four basic types of WRITE statements 1 formatted write statements those which contain an n parameter 2 list directed write statements those which contain an asterisk where an n parameter would otherwise appear 3 namelist write statements those which contain a NAMELIST group identifier where an n parameter would otherwise appear 4 unformatted write statements those which contain no n parameter or asterisk or NAMELIST identifier in the n position Formatted list directed and namelist write statements encode information into character records which may either be contained in external logical units or internal files Unformatted write statements perform no encoding operations and may transfer information to external logical units only Formatted and unformatted write statements may operate with either sequential or direct access files while list directed and namelist write statements may operate only with sequential files 129 PromulaFortran Compiler User s Manual In general the parameters on the various WRITE statements are as follows UNIT u is an integer expression a character variable identifier or an asterisk If it is an integer expression then it refers to a specific external logical unit If it is a chara
149. RR127 OPEN FILE ERR127 OUT STATUS NEW E127 The logical unit number parameter was not supplied STOP END Dimension number for variable ident has an invalid extent of number Example of message 128 If Line Nl Source 199 PromulaFortran Compiler User s Manual SUBROUTINE ERR128 2 PARAMETER ITEM 0 JJ 5 3 REAL B 3 2 C 0 4 COMMON 4 ITEM 5 DIMENSION F JJ 0 G JJ 6 INTEGER II 4 Jd 6 7 COMMON H ITEM 13 8 DOUBLE PRECISION DD ITEM JJ DF JJ ITEM 9 R 4 523 10 WRITE 1 B C A E F G II H DD DF R 1 END has an invalid extent of 0 has an invalid extent of O0 has an invalid extent of 4 for variable has an invalid extent of O0 for variable has an invalid extent of 5 for variable II has an invalid extent of 4 for variable II has an invalid extent of O0 for variable DD has an invalid extent of 4 for variable for variable for variable 11 err128 for E128 Dimension 11 err128 for E128 Dimension 11 err128 for E128 Dimension 11 err128 for E128 Dimension 11 err128 for E128 Dimension 11 err128 for E128 Dimension 11 err128 for E128 Dimension 11 err128 for E128 Dimension Q p D OU Error 129 Message The information for the option was supplied redundantly Example of message 129 If Line Nl Source 1 PROGRAM ERR129 2 OPEN 6 FILE ERR127 0OUT UNIT 7 STATUS NEW 2 errl29 for E129 The information
150. SUCH WARRANTIES ARE EXPRESSLY AND SPECIFICALLY DISCLAIMED IN NO EVENT SHALL GREAT MIGRATIONS LLC BE RESPONSIBLE FOR ANY INDIRECT OR CONSEQUENTIAL DAMAGES OR LOST PROFITS EVEN IF GREAT MIGRATIONS LLC HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGE Some states do not allow the limitation or exclusion of liability for incidental or consequential damages so the above limitation or exclusion may not apply to you TRADEMARK PROMULA is a registered trademark of Great Migrations LLC DEFINITION OF PURCHASE The definition of your particular purchase is specified in the Great Migrations LLC License Agreement which came with the PROMULA FORTRAN Compiler product and which you completed and returned to the Great Migrations LLC If you have any questions about your rights or obligations as a PROMULA FORTRAN Compiler user or believe that you have not received the complete PROMULA FORTRAN Compiler package that you purchased please contact Great Migrations LLC 7453 Katesbridge Ct Dublin Ohio 43017 614 761 9816 Table Of Contents 1 INTRODUCTION 1 1 1 USER SUPPORT c oen e ma IR EO REN OR OI 1 1 2 THE DESIGN OF THE COMPILERS e Ie E RO ER ERES ee t ERR ERE 1 1 3 NOTATIONI USED E b DI e ER REOR ee rU 2 PMMUJNICMUIINNUTP 3 2
151. TO 25 5 25 GO TO 30 5 warn843 for N843 It is inefficient to have a statement label on an unconditional goto 6 30 RETURN 7 END Note 844 Message The executable statement number label was defined but not referenced Example of message 844 If Line Nl Source 1 SUBROUTINE W844 2 10 1 0 3 20 FORMAT 1X 3F10 5 4 WRITE 20 A 5 END 5 warn844 for N844 The executable statement number 10 was defined but not referenced 276 PromulaFortran Compiler User s Manual Note 845 Message The format number label was defined but not referenced Example of message 845 If Line Nl Source I SUBROUTINE W845 2 1 0 3 20 FORMAT 1X 3F10 5 4 WRITE A 5 END 5 warn845 for N845 The format number 20 was defined but not referenced Comment 846 Message The actual argument list contains more than the number argument s used when the object ident was previously referenced or defined Example of message 846 If Line Nl Source 1 SUBROUTINE W846 2 CALL W846A 3 CALL W846A A B 3 warn846 for C846 The actual argument list contains more than the 1 argument s used when the subroutine W846A was previously referenced or defined 4 END Note 847 Message More than 19 continuation lines in one statement is nonstandard Example of message 847 If Line Nl Source 1 SUBROUTINE W847 2 DIMENSION 25 3 A B 1 4 5 B 2 5 5 B 3 6 B 4 7
152. Transfer of Sign Synopsis 164 PromulaFortran Compiler User s Manual INTEGER 2 FUNCTION I2SIGN INTEGER 2 INTEGER 2 Description I2SIGN al a2 is a specific function that returns a short integer result from two short integer arguments The result reflects the magnitude of the first value combined with the sign of the second value In particular ifal gt 0 0and a2 lt 0 0 return a1 or ifal lt 0 0 anda2 gt 0 0 return a1 else return al See also DSIGN ISIGN and SIGN 6 88 LEN Number of Characters Synopsis INTEGER FUNCTION LEN CHARACTER Description LEN a is a specific function that returns the length of a character string The argument is a character string and the result is an integer indicating the length of the string See also None 6 89 LGE Lexically Greater or Equal Synopsis LOGICAL 2 FUNCTION LGE CHARACTER CHARACTER Description LGE al a2 is a specific function that returns a result indicating lexically greater than or equal to The arguments are character strings The result is true only if a1 follows a2 or 1 is equal to a2 in the collating sequence See also CHAR LGT LLE and LLT 6 90 LGT Lexically Greater Synopsis LOGICAL 2 FUNCTION CHARACTER CHARACTER Description 165 PromulaFortran Compiler User s Manual LGT al a2 is
153. UMULATIVE EXPOSURE AVERAGE EXPOSURE EXPOSURE M T M T 460 520 380 01 170 01 110 01 680 02 480 02 100E 01 240E 01 500E 01 650E 01 540E 01 120E 03 670 04 250 04 140 04 610 05 265 300 2149 9 80 6 34 3 92 2 71 5 77 13 8 28 8 37 5 31 1 692 01 386 01 144 01 807 02 352 02 CUMULATIVE INTAKE M T 265 300 21 9 9 80 6 34 392 2 77 5 77 13 8 28 8 31 1 692 01 386E 01 144 01 807E 02 352E 02 000p see fe AVERAGE INTAKE M T Figure 8 4 Output Report from EXPO At this point we have a normal FORTRAN program however we have added no value to that program It has the same capabilities strengths and weaknesses as before 8 4 A PROMULA Datafile Description Examining the FORTRAN program we notice that all of the variables in the output report along with two intermediate calculations are stored in arrays It is these values which we would like to examine more closely We now wish to access the information in these variables by means other than this FORTRAN program To do this we need a description of the data on the file so that the individual values can be accessed and manipulated The datafile needs to be described The simple PROMULA program below contains such a description For the sake of readers not familiar with PROMULA notation we have sequenced the lines and wi
154. User s Manual If Line Nl Source 1 SUBROUTINE W859 2 DATA A 2 0 3 WRITE A 4 E 5 1ND 4 warn859 for N859 The continuation of the end statement is nonstandard Note 860 Message The use of return in the main program is nonstandard Example of message 860 If Line Nl Source 1 PROGRAM W860 2 I 1 3 WRITE I 4 RETURN 4 warn860 for N860 The use of return in the main program is nonstandard 5 END Comment 861 Message Data is being allocated to common storage via the variable ident Example of message 861 If Line Nl Source 1 SUBROUTINE W861 2 COMMON A D 3 EQUIVALENCE C A 4 DATA D B C 1 0 2 0 3 0 5 WRITE A B D END 6 warn861 for C861 Data is being allocated to common storage via the variable D 6 warn861 for C861 Data is being allocated to common storage via the variable C Note 862 Message Initializing the function value via a data statement is nonstandard Example of message 862 If Line Nl Source 1 FUNCTION W862 2 DATA W862 4 2 warn862 for N862 Initializing the function value via a data statement 282 PromulaFortran Compiler User s Manual is nonstandard 3 WRITE W862 4 END Comment 863 Message The data variable list is longer than the value list extra ignored Example of message 863 If Line Nl Source 1 SUBROUTINE W863 2 INTEGER 6 2 5 4 3 DATA 2 1 3 warn863 for C863 H he data v
155. When the standard format is selected there is a fixed statement ending column specified This is usually 72 though 132 is not unusual Any characters beyond this column are always ignored In addition there is an inline comment character If this character appears anywhere on the line not in a quoted string or a Hollerith constant then endcol is one position to the left of this character and all characters to the right are part of a comment field The continuation character is nonblank character in column 6 other than the character 0 When the default format is in effect with a maximum ending column of 72 then the maximum number of continuation lines is 61 This assumes that all lines are completely filled each with 66 characters 3 1 2 Tab Format Lines 21 PromulaFortran Compiler User s Manual With tab format the statement label field consists of the characters which precede the first tab character within position 1 8 After the first tab character or beyond position 8 there is either a continuation indicator field or a content field The continuation indicator field is either a numeric character or an ampersand amp The content field continues until an end of line is encountered or until the inline comment character is found not within a quoted string or a Hollerith constant Comment lines are any lines with a nonnumeric character in the first column 3 1 3 Normal Free Format Lines A free form
156. a constant or symbolic constant isa repeat count that is an unsigned nonzero integer constant or the symbolic name of such a constant Description The REAL statement is used to define a variable array symbolic constant function name or dummy procedure name as type real or double precision A length specification immediately following the word REAL applies to each entity not having its own length specification A length specification immediately following an entity is the length specification only for that entity If the length specification for a given entity is 4 or if it is omitted then the type defined is real If the value is 8 then the type is double precision 122 PromulaFortran Compiler User s Manual See also See the discussion of the DIMENSION statement for a description of how dimension bounds are defined See the discussion of the DATA statement for a description of how initial values are defined 5 47 RETURN Statement The RETURN statement terminates a subroutine or function subprogram Syntax RETURN exp Where exp is an arithmetic expression Description The RETURN statement terminates the currently executing subroutine or function subprogram and returns control to the calling program unit If the RETURN is used without the optional expression then it is a simple return If it is used with an expression then it is an alternate return Both subroutine subprograms and function subprograms can
157. a specific function that returns a result indicating lexically greater than The arguments are character strings The result is true only if a1 follows a2 in the collating sequence See also CHAR LGE LLE and LLT 6 91 LLE Lexically Less or Equal Synopsis LOGICAL 2 FUNCTION LLE CHARACTER CHARACTER Description LLE al a2 is a specific function that returns a result indicating lexically less than or equal to The arguments are character strings The result is true only if a1 precedes a2 or al is equal to a2 in the collating sequence See also CHAR LGE LGT and LLT 6 92 LLT Lexically Less Synopsis LOGICAL 2 FUNCTION LLT CHARACTER CHARACTER Description LLT al a2 is a specific function that returns a result indicating lexically less than The arguments are character strings The result is true only if a1 precedes a2 in the collating sequence See also CHAR LGE LGT and LLE 6 93 LOG10 Logarithm Base 10 Synopsis REAL FUNCTION LOG10 REAL REAL 8 FUNCTION LOG10 REAL 8 COMPLEX FUNCTION LOG10 COMPLEX COMPLEX 16 FUNCTION LOG10 COMPLEX 16 Description 166 PromulaFortran Compiler User s Manual LOG10 a is a specific function that returns the logarithm base 10 of the argument The result is always the same type as the argument See also ALOG10 CDLOG10 CLOG10 and DLOG10 6 94 LOG Natura
158. able length string is closed by the symbol token rather than a right parenthesis Example of message 243 If Line Nl Source 1 PROGRAM ERR243 2 CHARACTER 10 ALPHA BETA 2 err243 for E243 The character statement variable length string is closed by the symbol rather than a right parenthesis 8 STOP 4 END Error 244 Message The character statement variable length string token is not a constant Example of message 244 If Line Nl Source al PROGRAM ERR244 2 CHARACTER N ALPHA BETA 2 err244 for E244 The character statement variable length string N is not 237 PromulaFortran Compiler User s Manual a constant 3 STOP 4 END Error 245 Message The argument token of type type has been entered where a call by value argument of type type is required Example of message 245 If Line Nl Source 1 PROGRAM ERR245 2 DOUBLE PRECISION DVAL 3 DVAL DFLOAT 3 45 3 err245 for E245 The argument 3 45 of type real 4 has been entered where a call by value argument of type integer 4 is required STOP 5 END Error 246 Message The argument ident has the local variable ident as an adjustable dimension Example of message 246 If Line Nl Source 1 SUBROUTINE ERR246 I 2 INTEGER I J 3 COMMON JK 4 CALL ETEST I J 5 END 5 err246 for E246 The argument I has the local variable J as an adjustable dimension Error 247 Message The argument token of type typ
159. act term fact term fact fact is an arithmetic factor in one of the forms prim prim fact prim is an arithmetic primary which can be an arithmetic expression enclosed in parentheses or any of the following Unsigned arithmetic constant Arithmetic symbolic constant 39 PromulaFortran Compiler User s Manual Arithmetic variable Arithmetic array element reference Arithmetic function reference An arithmetic expression can be an unsigned arithmetic constant symbolic name of an arithmetic constant arithmetic variable reference arithmetic array element reference or arithmetic function reference More complicated arithmetic expressions can be formed by using one or more arithmetic operands together with arithmetic operators and parentheses Arithmetic operands identify values of type integer real complex short integer byte double precision or double complex The arithmetic operators are shown below Operator Representing Use Meaning Exponentiation xl x2 Exponentiate x1 to the power x2 Multiplication 1 x2 Multiply x1 and x2 Division xl x2 Divide x1 by x2 Addition xl 2 Add x1 and 2 Identity x2 Same as x2 Subtraction xl x2 Subtract x2 from x1 Negation x2 Negate x2 Each of the operators and operates on a pair of operands and is written between the two operands Each of the operators and either operates on a pair of operands and is written between the two
160. age 115 If Line Nl Source 1 SUBROUTINE ERR115 2 INTEGER 1 I1 3 COMPLEX BBB 4 1 BBB 4 errll5 for E115 An expression of type complex 8 is being used in the integer 1 context 5 BBB I1 5 err115 for E115 An expression of type integer 1 is being used in the complex 8 context 6 END 195 PromulaFortran Compiler User s Manual Error 116 Message The code location line number label is being referenced as a format number Example of message 116 If Line Nl Source Jd PROGRAM ERR116 2 100 WRITE Hello World 3 WRITE 100 Goodbye World 3 errll6 for E116 The code location line number 100 is being referenced as a format number 4 STOP 5 END Error 117 Message The code location line number label is being defined twice Example of message 117 If Line Nl Source 1 PROGRAM ERR117 2 100 WRITE Hello World 3 100 WRITE Goodbye World 3 errll7 for E117 The code location line number 100 is being defined twice 4 STOP 5 END Error 118 Message The format line number label is being referenced as a code location Example of message 118 If Line Nl Source 1 PROGRAM ERR118 2 100 FORMAT 1X A20 3 WRITE 100 Hello World 4 GOTO 100 4 errll8 for E118 The format line number 100 is being referenced as a code location b END Error 119 196 PromulaFortran Compiler User s Manual Message The format line number label is being define
161. ains the entry name The entry name can appear in an EXTERNAL statement and for a function entry name in a type statement Each reference to a procedure must use an actual argument list that corresponds in number of arguments and type of arguments with the dummy argument list in the corresponding SUBROUTINE FUNCTION or ENTRY statement Type agreement is not required for actual arguments that have no type such as a dummy subroutine name The dummy arguments for an entry point can therefore be different from the dummy arguments for the primary entry point or another entry point No dummy argument can be used in an executable statement of a procedure unless it has already appeared in a FUNCTION SUBROUTINE or ENTRY statement 5 25 The EQUIVALENCE Statement The EQUIVALENCE statement specifies the sharing of storage by two or more entities in a program unit Syntax EQUIVALENCE nlist nlist Where nlist isa list of variable names array names array element names or character substring names separated by commas Description The EQUIVALENCE statement is used to specify the sharing of storage by two or more entities in a program unit Equivalencing causes association of the entities that share the storage Equivalencing associates entities within a program unit and common blocks associate entities across program units Equivalencing and common can interact The behavior of the EQUIVALENCE statement can best be describ
162. al J 3 1 1 4 1 1 1 J 5 1 1 As second example consider the following PARAMETER 3 14159 REAL Y 5 5 DATA 1 1 1 4 1 1 3 6 The elements of array y initialized to 3 14159 3 1 4 1 4 2 5 1 5 2 and 5 3 as shown below I 3 4 I 5 3 14159 3 14159 3 14159 An iteration count and the values of the implied DO variable are established from init term and the optional incr just as for DO loops except that the iteration count must be positive When the implied DO list appears in a DATA statement the list items in dlist are specified once for each iteration of the implied DO list with appropriate substitution of values for each occurrence of the implied DO variable i S m m m tc The appearance of name implied DO variable DATA statement does not affect the value or definition status of a variable with the same name in the program unit An implied DO variable has the scope of the implied DO list only Each subscript expression used in dlist must be an integer constant expression except that any expression can contain an implied DO variable if the subscript expression is within the corresponding implied DO list 5 11 2 Character Data Initialization For initialization by DATA statement a character item in nlist must correspond to a character constant in clist The initial value is assigned according to the following ru
163. al values separated by a comma and enclosed by parentheses The parentheses are not considered to be a separator The decimal point can be omitted from either of the real constants Each of the real values can be preceded or followed by blanks Character values consist of a string of characters which can include blanks enclosed by single or double quotes A delimiting quote can be represented within a string by two successive occurrences Character values can only be read into variables of any type and character substrings If the string length exceeds the length of the list item the string is truncated If the string is shorter than the list item the string is left justified and remaining character positions are blank filled Logical values consist of an optional period followed by a T or F followed by optional characters which do not include separators slashes or commas Exact representation constants may be read into variables of any type To repeat a value an integer repeat constant is followed by an asterisk and the constant to be repeated Blanks cannot be embedded in the repeat part of the specification A null can be input in place of a constant when the value of the corresponding list entity is not to be changed A null is indicated by the first character in the input string being a comma or by two commas separated by an arbitrary number of blanks Nulls can be repeated by specifying an integer repeat count followed by an asterisk a
164. ameter Error 259 Message The do while conditional is followed by the symbol token rather than the end of statement Example of message 259 If Line Nl Source 1 PROGRAM ERR259 2 DO WHILE A LT B BEGIN 2 err259 for E259 The do while conditional is followed by the symbol BEGIN rather than the end of statement 3 A A B 4 END DO 4 err259 for E166 The statement introduced by the keyword ENDDO is not a valid statement type 5 STOP 6 END Error 260 Message The reference to the object ident is followed by the symbol token rather than a left parenthesis Example of message 260 If Line Nl Source 1 PROGRAM ERR260 2 SFUNC A B SIN A COS 3 ALPHA BETA SFUNC 3 5 4 2 3 err260 for E260 The reference to the statement function SFUNC is 242 PromulaFortran Compiler User s Manual followed by the symbol rather than a left parenthesis STOP 5 END Error 261 Message The argument list for object ident contains the symbol token rather a comma Example of message 261 If Line Nl Source 1 PROGRAM ERR261 2 SFUNC A B SIN A COS 3 ALPHA BETA SFUNC 3 5 3 err261 for E261 The argument list for statement function SFUNC contains the symbol rather a comma 4 STOP fe END Error 262 Message The structure definition identifier token has already been defined as a structure identifier Example of message 262 If Line Nl Source 1 PRO
165. an Compiler User s Manual If Line Nl Source L PROGRAM ERR198 2 CHARACTER 100 MESSAGE 3 DATA MESSAGE 56 63 Hello World Goodbye 3 errl198 for E198 The data statement value list contains too many values 4 STOP 9 END Error 199 Message The indexing constant token is of type type rather than of type integer Example of message 199 If Line Nl Source 1 PROGRAM ERR199 2 PARAMETER IPI 100 PI 3 14159 3 DIMENSION A IPI PI 3 errl99 for E199 The indexing constant PI is of type real 4 rather than of type integer STOP 5 END Error 200 Message An unnamed field is being declared with the symbol token rather than with fill Example of message 200 If Line Nl Source 4 PROGRAM ERR200 2 STRUCTURE STRA 3 CHARACTER 1 CHR 4 CHARACTER 3 FULL 00 for E200 An unnamed field is being declared with the symbol SFULL rather than with fill 5 INTEGER 4 ICHR 6 END STRUCTURE 7 8 STOP END Error 201 Message The argument list for object ident ends with the symbol token rather than a closing right parenthesis Example of message 201 223 PromulaFortran Compiler User s Manual If Line Nl Source 1 PROGRAM ERR201 2 SFUNC X Y SIN X COS Y 3 VAL SFUNC 2 3 5 4 3 8 3 err201 for E201 The argument list for statement function SFUNC ends with the symbol rather than a closing right parenthesis 4 STOP 5 END Error 202 Message The di
166. an appear anywhere within the program unit Note that any comment following the END statement is considered part of the next program unit 3 8 User Written Program Units An executable program consists of one main program and optional subprograms Both main programs and subprograms are known as program units A program unit contains a group of FORTRAN statements including optional comments it is terminated by an END statement Program units can be compiled independently of each other but a subprogram cannot be executed except through a main program There are two types of subprograms a specification subprogram and a procedure subprogram A subprogram that begins with a BLOCK DATA statement is a specification subprogram It is used to enter initial values for variables and array elements in named common blocks A subprogram that begins with a SUBROUTINE statement or a FUNCTION statement is a procedure subprogram known as a subroutine subprogram or a function subprogram respectively It can accept one or more values through a list of arguments common blocks or both A procedure is a function or a subroutine subprogram that can be executed many times A subroutine subprogram begins with a SUBROUTINE statement and terminates with an END statement it can return one or more values to the referencing program unit A function is used only in expressions to supply a value to the expression Functions can occur in two forms as a user written function
167. an array element reference an intrinsic function reference reference to a statement function which appears in the same program unit an external function reference a substring reference The symbolic name of the function is a variable and contains the value of the expression after execution During execution the actual argument expressions are evaluated converted if necessary to the types of the corresponding dummy arguments according to the rules for assignment and passed to the function Thus an actual argument cannot be an array name or a function name In addition if a character variable or array element is used as an actual argument a substring reference to the corresponding dummy argument must not be specified in the statement function expression The expression of the function is evaluated and the resulting value is converted as necessary to the data type of the function The symbolic name of a statement function is local and must not be the same as any other local name in the program unit except a common block name The name of a statement function cannot be an actual argument and must not appear in an INTRINSIC or EXTERNAL statement If the statement function is used in a function subprogram then the statement 46 PromulaFortran Compiler User s Manual function can contain a reference to the name of the function subprogram or any of its entry names as a variable but not as a function Each variable reference in the expr
168. aracter Action Blank Space vertically one line then print 0 Space vertically two lines then print 1 Eject to the first line of the next page before printing No advance before printing allows overprinting Any other character Refer to the operating system reference manual Carriage control characters are required at the beginning of every record to be printed including new records introduced by means of a slash Carriage control characters can be generated by any means 80 PromulaFortran Compiler User s Manual 5 27 11 D Descriptor The D descriptor specifies conversion between an internal double precision real number and an external floating point number written with an exponent This descriptor has the form Dw d On input D editing corresponds to E editing and can be used to input all the same forms as E The diagram below illustrates the structure of the input field It shows the characters allowed to start a subfield digit DorE integer fraction exponent subfield subfield On output type D conversion is used to output double precision values D conversion corresponds to E conversion except that D replaces E at the beginning of the exponent subfield For example DOUBLE PRECISION A B C A 111111 11111D0 B 222222 22222D0 C A B WRITE 2 10 A B C 10 FORMAT 3D23 11 produces output of 11111111111 06 22222222222 06 33333333333D406 In general the specification Dw a pr
169. ariable list is longer than the value list extra ignored 4 DATA 5 0 4 warn863 for C863 The data variable list is longer than the value list extra ignored 5 DATA J 2 3 5 warn863 for C863 The data variable list is longer than the value list extra ignored H 6 DATA K 1 2 3 4 7 WRITE I J Y Z 8 END Note 864 Message Declarative statements following executable statements is nonstandard Example of message 864 If Line Nl Source 1 SUBROUTINE W864 2 1 3 3 REAL 1 10 3 warn864 for N864 Declarative statements following executable statements is nonstandard 4 WRITE X I 5 END Note 865 Message Specification statements following data statements is nonstandard Example of message 865 If Line Nl Source t SUBROUTINE W865 2 DATA A 5 6 3 REAL B 283 PromulaFortran Compiler User s Manual 3 warn865 for N865 Specification statements following data statements is nonstandard 4 WRITE 5 END Comment 866 Message The type type has previously been assigned to ident Example of message 866 If Line Nl Source 1 SUBROUTINE W866 AB 2 INTEGER A 2Z 2 3 CHARACTER 5 A 3 warn866 for C866 The integer 4 type has previously been assigned to A 4 COMPLEX Z 4 warn866 for C866 integer 4 type has previously been assigned to 2 D AB 1 6 END Note 867 Message The argument ident has an expression as an adjustable dimension E
170. arn900 for 4 warn900 for 5 Warning 901 Message SUBROUTINE W900 C D REAL C D 10 CALL E569A C D WRITE 2 C I I 1 10 D W900 An implied do must be used for the assumed size array D C870 The array D is being subscripted with less than 2 expressions END Only the real part of a complex value is used in an arithmetic if Example of message 901 If Line Nl Source 3 warn901 for SUBROUTINE W901 C COMPLEX C IF C 2 10 20 20 W901 Only the real part of a complex value is used in an arithmetic if 20 RETURN 10 END 297 PromulaFortran Compiler User s Manual Warning 902 Message The global symbol ident is being used to represent both a subprogram and a common area Example of message 902 If Line Nl Source SUBROUTINE W902 2 COMMON W902 A B C 3 A B C 4 RETURN 5 END 5 warn902 for W902 The global symbol W902 is being used to represent both a subprogram and a common area Warning 903 Message The equivalence element subscript value number for subscript number of array ident with minimum of number and maximum of number is out of range Example of message 903 If Line Nl Source 1 PROGRAM W903 2 PARAMETER NA 10 IA 11 3 DIMENSION AA NA 6 BB 4 4 EQUIVALENCE BB AA IA 2 4 warn903 for W903 The equivalence element subscript value 11 for subscript 1 of array AA with minimum of 1 and maximum of 10 is out of range 5 BB 3 99
171. at line is a sequence of variable length lines with no fixed field boundaries If the first nonblank character on a line is a numeric digit then it is considered an initial line with a statement label If the first nonblank character is an ampersand amp then the line is a continuation line Otherwise the line is initial with no statement label The statement fields continue until the end of line is encountered or until the inline comment character is encountered 3 1 4 Continuation Free Format Lines The continuation free format is unique in that the continuation indication field occurs on the line being continued and not on the continuation line As a result inline comments and comments within continuation sequences are not allowed In this format a comment line begins with a quotation mark in position 1 A statement begins if there is no comment If the last character on the line is a minus sign then the following line continues the present one The continuation minus sign is not part of the statement 3 2 Symbolic Names Symbolic names are assigned by the user They consist of from 1 to 255 letters digits the underscore _ and the dollar sign The first character may not be a digit Letters may be uppercase or lowercase however the case is not significant in establishing the uniqueness of a symbolic name Names that are FORTRAN keywords can be used as user assigned symbolic names without conflict In general however it is
172. be active inactive or nested Each is discussed below 5 14 1 Active and Inactive DO Loops Initially a DO loop is inactive A DO loop becomes active only when its DO statement is executed Once active a loop becomes inactive when any of the following occurs 1 The iteration count is determined to be 0 2 ARETURN STOP or END statement is executed within the program unit containing the loop 3 control variable becomes undefined or is redefined by a process other than loop incrementation 4 115 in the range of another loop that becomes inactive 5 It is in the range of another loop whose DO statement is executed Transfer of control out of the range of a DO loop does not deactivate the loop When such a transfer occurs the control variable retains its most recent value in the loop Control can be returned to the range of the loop provided that the control variable is not redefined outside the range or the program unit containing the loop has not been exited by a RETURN STOP or END statement The loop becomes inactive once the control variable is redefined and should not be reentered except through its DO statement If a DO loop executes zero times the control variable value equals m1 Otherwise if the index variable iterates through the terminal parameter value the control value is the most recent value of the control variable plus the increment parameter value If a DO loop becomes inactive but has not executed to completion
173. ble Dimensions Adjustable dimensions enable creation of a more general subprogram that can accept varying sizes of array arguments For example a subroutine with a fixed array can be declared as SUBROUTINE SUM DIMENSION A 10 The maximum array size subroutine SUM can accept is 10 elements If the same subroutine is to accept an array of any size it can be written as SUBROUTINE SUM A N DIMENSION Value N is passed as an actual argument Adjustable dimensions can also be passed through common variables For example SUBROUTINE SUB A COMMON B M N DIMENSION 07 Dimension of array in subroutine SUB is specified by the values and passed through the common block Character strings and arrays can also be adjustable For example SUBROUTINE MESSAG X CHARACTER X PRINT X The subroutine declares x with a length of to accept strings of varying size Note that the length of the string is not passed explicitly as an actual argument Another form of adjustable dimension is the assumed size array In this case the upper bound of the last dimension of the array is specified by an asterisk The value of the dimension is not passed as an argument but is determined by the number of elements stored in the array If an array is dimensioned the array in the calling program must be large enough to contain all the elements stored in it in the subprogram
174. cation When a slash is used to separate edit descriptors a comma is allowed but not required Consecutive slashes can be used and need not be separated from other elements by commas When a slash is the last format specification to be processed it causes a blank record to be written on output or an input record to be skipped Normally the slash indicates the end of a record during output and specifies that further data comes from the next record during input 5 27 14 F Descriptor The F descriptor specifies conversion between an internal real or double precision number and an external floating point number without an exponent This descriptor has the form Fw d On input the F specification is treated the same as the E specification Examples of F input are shown below Input Field Specification Value Remarks 367 2593 F8 4 367 2593 Integer and fraction field 62543 5 62543 No integer subfield 62543 F6 2 62543 Decimal point overrides d of specification 144 15 03 F112 14415 Exponents are allowed in F in put 50000 F5 2 500 00 No fraction subfield input number converted as 50000 10 2 ANNA F5 2 0 Blanks in input field interpreted as 0 On output the F descriptor outputs a real number without a decimal exponent The plus sign is suppressed for positive numbers If the field is too short all asterisks appear in the output field If the field is longer than required the number is right justified with blanks on
175. ce 1 PROGRAM ERR167 2 CALL 999 A B C 212 PromulaFortran Compiler User s Manual 2 errl67 for E167 The call statement subroutine name is 999 and not an identifier 3 STOP 4 END Error 168 Message The variable ident is used as an assigned format but is never assigned a format label Example of message 168 If Line Nl Source 1 SUBROUTINE ERR168 2 ASSIGN 10 TO IA 3 GOTO IA 10 20 4 10 A 1 5 WRITE IA 6 1 FORMAT 1X F10 0 7 20 RETURN END 8 errl68 for E168 The variable IA is used as an assigned format but is never assigned a format label 169 Message An element in an implicit list is the symbol token rather than a single letter Example of message 169 If Line Nl Source 1 PROGRAM ERR169 2 IMPLICIT INTEGER A BC 2 errl69 for E169 An element in an implicit list is the symbol BC rather than a single letter 3 PRINT A B C 4 STOP 5 END Error 170 Message An upper bound in an implicit list range is the symbol token rather than a single letter Example of message 170 If Line Nl Source T PROGRAM ERR170 2 IMPLICIT INTEGER A DE 2 errl70 for E170 An upper bound in an implicit list range is the symbol DE rather than a single letter 3 PRINT A B C D E 4 STOP 213 PromulaFortran Compiler User s Manual 5 END Error 171 Message The implicit statement is closed by the symbol token rather than a right parenthesis Example o
176. ce of octal or hexidecimal digits depending upon whether an o for octal or an x for hexidecimal is specified The length of these constants depends upon the context in which they are used They are considered to be the typeless numeric constants They assume a type based upon the way in which they are used as follows 1 When used in a binary operation including assignment the constant takes on the type of the other operand 2 When used in some context where a specific data type is required such as in a subscript context or as an argument to an intrinsic function then the constant takes on that data type 3 When used in a context for which no type can be determined then the exact representation constant is assumed to be an integer constant Examples of valid octal constants 02247 0 Value 1191 11070 Value 8 27 PromulaFortran Compiler User s Manual Examples of invalid octal constants 7782 Invalid character 8 0747 No O after second quote Examples of valid hexadecimal constants BF342 X Value 783170 FFB X Value 4091 Examples of invalid hexadecimal constants 82 8 X Invalid Character ROX Missing second quote 3 4 Variables A variable represents a quantity with a value that can be changed repeatedly during program execution Variables are identified by a symbolic name of to 255 letters digits underscore _ or dollar sign beginning with a nondigit A variable is associated with a
177. cecseseseaesececeeseeaaaeeecesesensaaees 176 THE PROMULA INTEREA CEipscicsssccvcissssnscecdsssnstcdassesnssseonesssaceetetesseashetesssdacesosvenseshendesescsvetesessacscessevetssscssssstsonsess 177 8 1 TRANSFORMING A FORTRAN PROGRAM WITH PROMULAA 2 2 00002 0 0 0000000000000000 179 8 2 AN EXPOSURE ANALYSIS MODEL ccsesessccecececsessceseecececsessauececececeessaaeceeseccseneaaaeseeeceesessaeeececeesessaaeeeeess 180 8 3 beceedeeeveserttedecvsosvecueet 182 8 4 A PROMULA DATAFILE DESCRIPTION ccsssesscsnccsececesnsnnncesccsceensnnnesscescessensceusessceceenscaucessccesnsenuccecesceesenneesecesse 183 8 5 THE VIRTUAL COMPILATION PPP RETE ETE Pe PP Ee EE Eee Ev RR Ede 184 8 6 SYNTAX OF THE GI OBALS BILE tertio ee neon eee evo us Dre dee ovo Pen be ebay voee dooce seve 185 9 7 USING EXPO WITH PROMU DA aeee ae eee eie etr veau bene doe vd 185 ERROR MESSAGES c D M 189 9 1 CONTROL PROGRAM 189 9 2 FORTRAN PREPROCESSOR 8 01 1 27 12 2 4222 2 2 0600000000000000000000000000000000000000000000 0 190 9 2 1 SYNTAX ERRORS WARNINGS COMMENTS AND N
178. ch as removal of repeated subexpressions The mathematical results of the reordering are correct but the specific order of evaluation is indeterminate For example the expression A B c is guaranteed to equal algebraically ac B not A but the specific order of evaluation by the compiler is indeterminate An arithmetic constant expression contains only arithmetic constants symbolic names of arithmetic constants or arithmetic constant expressions enclosed in parentheses The exponentiation operator is not permitted unless the exponent is of type integer An integer constant expression is an arithmetic constant expression in which each constant or symbolic name of a constant is of type integer short integer or byte The data type of an arithmetic expression containing one or more arithmetic operators is determined from the data types of the operands When an arithmetic operator combines operands of the same type then the result is of that type When the operands are of different types the lower type is promoted to the higher type prior to the evaluation of the operator Using this terminology the operand types from highest to lowest are double complex complex double precision real integer short integer and byte 4 1 2 Character Expression A character expression is used to express a character string Evaluation of a character expression produces a result of type character The simplest form of a character expression is a character
179. code Codes may be summed to achieve composite effects Code Meaning 0 Standard conventions 2 Interprets carriage control to output 4 Checks substring lengths for overflow 8 Execute an explicit pause for the PAUSE statement 16 Use VAX FORTRAN runtime conventions Assigns the unit whose number is indicated to standard input Assigns the unit whose number is indicated to standard output Assigns the unit whose number is indicated to be a terminal i e reads from standard input writes to standard output Specifies the name of a file which is to be used as the virtual disk file Specifies that the file specified by the V parameter is to be created during the execution of this program Specifies the maximum number of sheets to be allocated for virtual memory 7 1 Interpreting Carriage Control to Output In traditional FORTRAN environments the first character of each character record written contains a format control blank means single space zero means double space one means start a new page By default the runtime library simply writes these characters to coded records without checking or interpreting them If the carriage control convention flag is set however the runtime library strips the first character and replaces it with the appropriate control characters 174 PromulaFortran Compiler User s Manual 7 2 Checking Substring Lengths for Overflow Normally substrings are not checked to make certain
180. comes associated with unit u If FILE is omitted then a file name is created See your system manager for the conventions used for creating this name at your site STATUS sta is a character expression specifying file status Valid values OLD File in currently exists NEW File in does not currently exist SCRATCH Delete the file associated with unit u on program termination or execution of CLOSE that specifies unit 112 PromulaFortran Compiler User s Manual UNKNOWN File status is unknown If STATUS is omitted then an unknown status is assumed ACCESS acc is a character expression specifying the access method of the file Valid values are SEQUENTIAL File is to be opened for sequential access DIRECT File is to be opened for direct access If ACCESS is omitted then sequential access is assumed If the file exists the access method must be valid for the existing file Direct access files are discussed further in a following subsection FORM 2f m is a character expression having one of the following values FORMATTED File is being opened for formatted input output UNFORMATTED File is being opened for unformatted input output If FORM is omitted FORMATTED is assumed for sequential access files and UNFORMATTED is assumed for direct access files For an existing file the specified form must be valid for that file RECL r1 is an integer expression specifying the record length for a direct or se
181. command mode to simplify this discussion The final two lines end the processing of the program and end the session with PROMULA The final step is to process this description using PROMULA itself This is done by compiling the above source file EXPO PRM to produce the following two files EXPO DBA which is the actual datafile EXPO XEQ which contains the additional logic and structure discussed above We are now ready to add value to the FORTRAN program 8 5 The Virtual Compilation Having defined the EXPO DBA datafile we can now re compile the EXPO FOR program by making explicit reference to this datafile This is done as follows pf expo for expo gbl o expo The file expo gbl is used during the compilation It contains the cross reference map linking the variables in the FORTRAN program EXPO FOR with the variables in the PROMULA database EXPO DBA and is listed below expo dba expo iday iday virtual expo imnth imnth virtual expo iyr iyr virtual expo avge avge virtual expo avgi avgi virtual expo conc conc virtual expo cume cume virtual expo cumi cumi virtual 184 PromulaFortran Compiler User s Manual expo expos expos virtual expo perd perd virtual expo sever sever virtual expo take take virtual The new C code produced by this translation EXPO C is different than the one we produced before without making explicit reference to the EXPO DBA datafile The new translation makes virtual reference to the EXPO
182. constant symbolic name of a character constant character variable reference character array element reference character substring reference or character function reference More complicated character expressions can be formed by using one or more character operands together with character operators and parentheses The only character operator available is which performs a concatenation The result of a concatenation operation is a character string concatenated on the right with another string and whose length is the sum of the lengths of the strings For example the value of CDE is the string ABCDE A character expression and the operands of a character expression must identify values of type character Two or more concatenation operators are combined from left to right to interpret the expression For example the interpretation of the character expression CD EE is the same as the interpretation of the character expression AB CD EF The value of the preceding expression is the same as that of the constant ABCDEF Note that parentheses have no effect on the value of a character expression Thus the expression AB CD EF has the same value as the preceding expressions A character constant expression is a character expression in which each operand is a character constant the symbolic name of a character constant or a character constant expression enclosed in parent
183. cter variable then that variable is treated as an internal file of character records If it is an then the write is performed to the standard OUTPUT device If one of the simple forms of the WRITE is used which contain no parentheses and thus no unit number then these also use the standard OUTPUT device FMT fn is integer variable the label of a FORMAT statement a character expression or an array identifier It specifies a format to be used for formatted output If fn is a statement label then it identifies a FORMAT statement in the program unit containing the output statement which specifies how the output records are to be formatted If it is a character expression then that expression specifies how the output records are to be formatted If it is an array identifier then that array is assumed to contain the format information Finally if it is an integer variable then that variable is assumed to have been assigned the statement number of a FORMAT statement by an ASSIGN statement REC rn is an integer expression which specifies the number of the record to be written to the file It must be greater than zero and is valid for files opened for direct access only FMT n1 is the identifier of a NAMELIST group The use of this parameter indicates a namelist write See the NAMELIST statement for a description of the conventions used for the namelist write ERR s1 specifies the label of an executable statement to which control transfers if an
184. d If you specify IMPLICIT NONE you must not include any other IMPLICIT statement in the program unit The IMPLICIT statement has no effect on the default types of intrinsic functions 100 PromulaFortran Compiler User s Manual See also The explicit TYPE statements BYTE INTEGER DOUBLECOMPLEX DOUBLEPRECISION REAL LOGICAL and CHARACTER which describe the actual meanings of the various types 5 33 INCLUDE Statement The INCLUDE statement specifies that a given source file be included as part of a source file Syntax INCLUDE file Where file isthe name of a file Description When the compiler encounters an INCLUDE statement it stops reading from the current file and reads the statements in the included file When it reaches the end of the included file the compiler resumes reading the statement immediately after the INCLUDE statement An INCLUDE statement may appear within an included file however nesting may only be five deep The INCLUDE statement can appear anywhere in a program unit however it may not begin with a continuation line or with a comment within a continued statement Statements within an included file are treated in exactly the same manner as if they had been physically contained within the file containing the INCLUDE 5 34 INTEGER Statement The INTEGER statement defines some user defined entity to be of type integer short integer or byte Syntax INTEGER len name name
185. d twice Example of message 119 If Line Nl Source al PROGRAM ERR119 2 100 FORMAT 1X A20 3 WRITE 100 Hello World 4 WRITE 100 Goodbye World 5 100 FORMAT 1X A30 5 errll9 for E119 The format line number 100 is being defined twice 6 STOP 7 END Error 120 Message The logical unit specification token is of the wrong type Example of message 120 If Line Nl Source 1 PROGRAM ERR120 2 LOGICAL TFILE 3 TFILE TRUE 4 WRITE TFILE Hello World 4 errl20 for E120 The logical unit specification TFILE is of the wrong type 5 STOP 6 END Error 121 Message The internal file specification begins with the symbol token rather than an identifier Example of message 121 If Line Nl Source al PROGRAM ERR121 2 ENCODE 8 100 10 5 Hello World 2 errl21 for E121 The internal file specification begins with the symbol 10 5 rather than an identifier 3 100 FORMAT 8 4 STOP 5 END Error 122 Message The read element is the symbol token rather than an identifier 197 PromulaFortran Compiler User s Manual Example of message 122 If Line Nl Source 1 PROGRAM ERR122 2 10 FORMAT a10 3 READ IFILE 10 I J K 4 READ IFILE Hello World 4 errl22 for E122 The read element is the symbol Hello World rather than an identifier 5 5 6 END Error 123 Message The statement function argument name ident may not be repeated Example of message 123
186. database is also accessible to PROMULA for independent data management manipulation and ad hoc analysis The same code that runs on the mainframe now runs on the desktop as well and reproduces results without making any code changes or sacrificing any of the program s capabilities There is no need to downsize the code to run it on the PC As a result there is no need to maintain two different codes one on the mainframe and one on the PC 3 Write a PROMULA user interface program to manage the program database and to perform the following functions 179 PromulaFortran Compiler User s Manual Browse edit program inputs Edit the program source code Translate compile and link the program source code Run the executable model code Browse or print program outputs Plot program inputs and outputs PROMULA s built in windowing menu management data management and graphics features are used to produce a contemporary user interface The programming requirements for this kind of program are very modest and once you ve done one the code is transferrable to other applications as well 8 2 an Exposure Analysis Model The program listing on the following two pages shows a relatively simple FORTRAN program which calculates the cumulative effects of exposure to various pollutant types It is basically a FORTRAN 66 program though it has OPEN statements and uses T FORMAT specifications Both of these are extensions to the
187. default storage type even if the Sa option is selected 2 5 5 The Doloop trip count assumption Below is the syntax for the FORTRAN DO statement It looks very simple however what you see is unfortunately not what you get FORTRAN compilers vary widely on how the DO statement is executed Syntax DO slab v el e2 e3 12 PromulaFortran Compiler User s Manual Where slab is the label of an executable statement called the terminal statement of the DO loop If slab is omitted the do loop is terminated via an END DO statement v is an integer real or double precision control variable el isaninitial parameter e2 isa terminal parameter e3 isan optional increment parameter default is 1 1 e2 and e3 are called indexing parameters they can be integer real double precision or symbolic constants variables or expressions The DO statement differs widely between FORTRAN 66 and FORTRAN 77 In FORTRAN 66 the value of v is not compared with that of e2 until the bottom of the loop therefore the loop is always executed once In FORTRAN 77 the comparison of v is performed at the top of the loop therefore if 1 exceeds e2 initially the loop is never incremented Officially do loops are to be executed as follows 1 The expressions e1 e2 and e3 are evaluated and then converted to the type of the control variable v if necessary yielding the values m1 m2 and m3 2 The control variable v is assigned th
188. delimeter for a character constant is the single quot 431 utest for N858 The identifier PRANDOM INDX with more than 6 characters is nonstandard 474 utest for N806 0mitting the comma after the I FORMAT specification is nonstandard 545 utest for N823 The COMMON block DIR1 has character 1 variable VFORMS and an unspecified variable OPTEV 722 utest for N801 The INCLUDE statement is nonstandard 1 STRUC6 INC N801 he STRUCTURE statement is nonstandard 800 utest for N833 The nonparenthetical form of the PARAMETER statement is nonstandard 5 PromulaFortran Compiler User s Manual As can be seen the EL3 level generates many messages and is primarily intended for those who are trying to write pure standard conforming code 2 4 2 Monitoring Internal Operations The V1 and V2 flags are used to monitor the internal operations of the FORTRAN preprocessor Only simple messages are displayed These options are generally not used in conjunction with the ES and EX flags which produce structured reports The following is typical output produced at the V1 level Compiling source file utest for Including source file STRUC6 INC Including source file STRUC6 INC Including source file STRUC7 INC Including source file STRUC7 INC In essence a message is generated each time a source file is opened At the V2 level the following output might be produced Compiling source file
189. dom access database for later use by the program and for independent use by other applications Write a database description file in PROMULA and create a PROMULA database that will contain the selected variables Each variable in the database is simply defined by an identifier a structure its dimensions if any a type and a descriptor The dimensionality of variables in the database must be identical to that in the program For example the definitions DEFINE SET week 52 day 7 hour 24 year 10 DEFINE VARIABLE hdata hour day week year TYPE REAL 10 2 Hourly Data describe a four dimensional variable hdat a containing 87 360 values 52 x 7 x 24 x 10 the hourly values for 10 years classified by hour day week and year On PC DOS this variable alone is bigger that 64 Kilobytes and would make it difficult for the program to compile using a typical compiler available on the PC 2 Compile and link the FORTRAN program During the compilaton introduce references to a PROMULA database containing all the variables in the program that cause it to be big larger than the typical memory resources available on say the PC The augmented compilation effectively separates the program s calculations from its database The database can now reside on disk can be as large as we wish and still be accessible to the program via the PROMULA virtual memory manager The program
190. e An association exists for array elements in a character array Note that unless the lengths of the elements in the dummy and actual argument agree the dummy and actual argument array elements might consist of different characters For example if a program unit has the following statements DIMENSION A 2 CHARACTER 2 CALL SUB and the subroutine has the following statements SUBROUTINE SUB DIMENSION B 2 CHARACTER B 1 then the first character of A 1 corresponds to B 1 and the second character of 1 corresponds to B 2 If the actual argument is a noncharacter array element name the size of the dummy argument cannot exceed 1 where as is the size of the actual argument array and av is the subscript value of the array element For example if the program unit has the following statements DIMENSION ARRAY 20 CALL CHECK ARRAY 3 then the value of as is 20 and av is 3 The maximum dummy array size is 18 for the subroutine SUBROUTINE CHECK DUMMY DIMENSION DUMMY 18 SWAP DUMMY 2 Actual argument array elements are associated with dummy argument array elements starting with the first element passed In the example DUMMY 2 is associated with ARRAY 4 and DUMMY 18 is associated with ARRAY 20 The association for characters is basically the same as for noncharacter array elements The actual argument for characters can be an array
191. e The second part is the name of the source file itself The third part is the message identifier which consists of a letter followed by a number This identifier letter indicates the type of the message Letter Type of message E Syntax error W Warning C Comment N Note The identifier number can be used to look up the message in the following listing Under the theory that a picture is worth a thousand words rather than trying to describe the circumstances surrounding each error a typical code fragment generating each error is presented It is the intent of the message system that each message be as clear self contained and specific as possible 190 PromulaFortran Compiler User s Manual Error 101 Message The object identifier ident may not be assigned a value Example of message 101 If Line Nl Source 1 PROGRAM 101 2 IALPHA X IFIX X 2 3 DO 10 IALPHA 1 5 3 errlOl for E101 The statement function identifier IALPHA may not be assigned a value 4 WRITE I 5 10 CONTINUE 6 END Error 102 Message The array ident is being subscripted with more than number expressions Example of message 102 If Line Nl Source 1 PROGRAM ERR102 2 DIMENSION A 10 15 3 A I J A I J K B 3 err102 for E102 The array A is being subscripted with more than 2 expressions 4 STOP 5 END Error 103 Message The statement function either calls itself or there is a missing dimension for
192. e real is areal or integer constant for the real part imag 165 areal or integer constant for the imaginary part Examples of valid complex constants 1 7 54 Value 1 7 541 where i 1 2 1 1 3 24 Value 21 3 241 4 5 Value 4 0 5 01 0 0 1 01 Value 0 0 1 01 Examples of invalid complex constants 12 7D 4 16 1 Comma missing and double precision not allowed 2 1 942 Parentheses missing 3 3 5 Double Complex Constant Double complex constants are written as a real integer or double precision constant paired with a double precision constant separated by a comma and enclosed in parentheses The first constant represents the real part of the double complex number and the second constant represents the imaginary part The parentheses are part of the constant and must always appear Either constant can be preceded by a plus or minus sign The syntax of a double complex constant is as follows real imag Where real is areal integer or double precision constant for the real part imag is areal integer or double precision constant for the imaginary part Either real or imag must be double precision Examples of valid double complex constants 1 7 54D0 Value 1 7 541 where i 1 2 101 3 24 Value 21 3 241 4 5DO Value 4 0 5 01 ODO 1 Value 0 0 1 01 25 PromulaFortran Compiler User s Manual Examples of invalid double complex constants 12 7E 4 16 1 Comma missin
193. e Dynamic and virtual variables can be created via the PROMULA interface described in a later chapter The type of a variable is simply its type specification For records it is the structure type of the record The comment associated with the symbol is a function of its object type For constant integer parameters i e those that may be used in other declaration statements the comment specifies the value of the parameter as specified or computed For variables or records the dimensionality is given for arrays the total size in bytes is also given For subprograms the number of arguments is given along with the assumed type of each argument Note that in the C tradition the PROMULA FORTRAN compiler makes extensive use of subprogram prototypes and always makes certain that argument types are consistent If they are not it issues a warning The symbol reference by line number table is simply that Along with the line number a use type code is also specified means defined M means modified U means used and P means passed to a subprogram If a symbol has multiple references in a single statement then only one reference is reported Note that if include files are involved the line number is followed by the include file sequence number If include files are not involved a simple sequence number is used The statement label types and references table is a numerical listing of the statement labels along with their ty
194. e has been entered where a call by reference argument of type type is required Example of message 247 If Line Nl Source al PROGRAM ERR247 2 COMPLEX IC LC 3 CALL SUB247 AVAL 4 CALL SUB247 IC 5 CALL SUB247 IC LC 5 err247 for E247 The argument IC LC of type complex 8 has been entered where a call by reference argument of type real 4 is required 6 STOP 238 PromulaFortran Compiler User s Manual 7 END Error 248 Message The object argument ident has been entered where a subprogram argument is required Example of message 248 If Line Nl Source 1 PROGRAM ERR248 2 DIMENSION AVAL 10 3 EXTERNAL FUNC IFUNC 4 CALL SUB248 FUNC CALL SUB248 IFUNC 6 CALL SUB248 AVAL 6 err248 for E248 The array variable argument AVAL has been entered where a subprogram argument is required 7 STOP 8 END Error 249 Message The argument token of type type has been entered and cannot be converted into a call by value argument of type type Example of message 249 If Line Nl Source 1 PROGRAM ERR249 2 COMPLEX ALPHA 3 LOGICAL 1 IV 4 ALPHA IV BETA 4 err249 for E249 The argument IV of type logical 1 has been entered cannot be converted into a call by value argument of type real 8 5 STOP 6 END Error 250 Message The common variable ident has the adjustable dimension ident Example of message 250 If Line Nl Source 1 SUBROUTINE ERR250 ID
195. e also CSQRT DSQRT and SORT 6 26 CEXP Exponential Synopsis COMPLEX FUNCTION CEXP COMPLEX Description CEXP a is a specific function that returns a complex result from a complex argument It returns the exponential of its argument See also 144 PromulaFortran Compiler User s Manual CDEXP DEXP and EXP 6 27 CHAR Character Value Synopsis CHARACTER 1 FUNCTION CHAR INTEGER 2 CHARACTER 1 FUNCTION CHAR INTEGER Description CHAR a is a generic function that converts a numeric display code or lexical value or collating weight into its character code The point of this function is that character values on the host processor are not necessarily the same as those on the machine for which a given FORTRAN program was written All numeric display code references in a source FORTRAN program are passed through this function either by the system directly or by the other runtime functions included in this library Note that if you wish this function to return some value other than the host processor values then you must modify it Typically this modification would take the form of a lookup table reference See also ICHAR 6 28 CLOG10 Logarithm Base 10 Synopsis COMPLEX FUNCTION CLOG10 COMPLEX Description CLOG10 a is a specific function that returns the logarithm base 10 of the argument The argument is complex and the result is co
196. e argument type See also CCOS CDCOS and DCOS 6 34 CSIN Sine Synopsis COMPLEX FUNCTION CSIN COMPLEX Description CSIN a isa specific function that returns a sine The argument is assumed to be in radians The argument and result are both complex See also CDSIN DSIN and SIN 6 35 CSQRT Square Root 147 PromulaFortran Compiler User s Manual Synopsis COMPLEX FUNCTION CSORT COMPLEX Description a is a specific function that returns the principal square root of its argument The argument and result are complex See also CDSQRT DSQRT and SQRT 6 36 DABS Absolute Value Synopsis REAL 8 FUNCTION DABS REAL 8 Description DABS a returns the absolute value of its double precision argument The result is double precision and equals a if a is greater than or equal to zero else it is a See Also ABS IABS CABS CDABS and I2ABS 6 37 DACOS Arccosine Synopsis REAL 8 FUNCTION DACOS REAL 8 Description DACOS a returns the double precision arccosine of its double precision argument The result is expressed in radians See also ACOS 6 38 DASIN Arcsine Synopsis REAL 8 FUNCTION DASIN REAL 8 Description DASIN a is a specific function that returns an arcsine The result is expressed in radians The result and argument are double precision See also 148 PromulaFortran Compiler User s Manual ASIN
197. e array not just the first word of the array is written Assumed size array names are legal in output lists On formatted output the olist is scanned and each item in the list is paired with the field specification provided by the FORMAT statement After one item has been output the next format specification is taken together with the next element of the list and so on until the end of the list An implied DO specification has the following form dlist i el e2 3 130 PromulaFortran Compiler User s Manual The elements i ei 2 and e3 have the same meaning as in the DO statement and dlist is an output item list Redundant parentheses are allowed and are ignored The range of an implied DO specification is that of dlist An implied DO can be used to output a single value more than one time For example the list A K B K 1 5 causes the value of variable B to be written five times Output of array elements can be accomplished by using an implied DO The list of variables followed by the DO index is enclosed in parentheses to form a single element of the output item list For example WRITE 5 100 1 1 1 3 has the same effect as the statement WRITE 5 100 A 1 A 2 A 3 A variable cannot be used as a control variable more than once in the same implied DO nest but olist items can appear more than once The value of a control variable within an implied DO specification is defined within that speci
198. e compiler to use unit number num with WRITE or PRINT statements for which no unit number is explicitly shown The default setting UW specifies that WRITE or PRINT statements for which no unit number is explicitly shown should be directed to standard output 13 PromulaFortran Compiler User s Manual The UPnum or UP option tells the compiler to use unit number num with PUNCH statements for which no unit number is explicitly shown The default setting UP specifies that PUNCH statements for which no unit number is explicitly shown should be directed to standard output Note See the chapter on controlling runtime behavior for more discussion or the relationship between particular FORTRAN units and the standard input and output streams 2 5 7 Selecting Dialect Conventions The PROMULA FORTRAN Compiler is a general purpose multi dialect and portable FORTRAN compiler It runs on multiple platforms and supports both the ANSI FORTRAN 66 and ANSI FORTRAN 77 standard dialects as well as a large number of common extensions such as those found in the following commercial compilers VAX FORTRAN PDP FORTRAN PRIME FORTRAN Data General FORTRAN and Sun FORTRAN Some FORTRAN 90 extensions are also supported In cases where different versions of FORTRAN have conflicting features or conventions a dialect selection option switch can be used to select the desired set The particular dialect option which the compiler supports are as follows Option Dial
199. e in the subroutine definition An actual argument in a CALL statement can be a dummy argument name that appears in the dummy argument list of the subprogram containing the CALL statement An asterisk dummy argument cannot be used as an actual argument 5 6 CLOSE Statement 51 PromulaFortran Compiler User s Manual The CLOSE statement disconnects a file from a specified unit Syntax CLOSE unit or CLOSI UNIT unit STATUS clssta IOSTAT status ERR err Where unit isan integer expression clssta is a character expression status is an integer lvalue err the label of an executable statement Notes The simplest form of the CLOSE statement consists of a single unit specification not enclosed in parentheses With this form no additional parameters can be specified If the parenthetical form of CLOSE is used then the UNIT specification is optional however when omitted the unit specification must be the first specification Other than the above the order of the parameters within the parenthetical version of the CLOSE statement is free Description The CLOSE statement disconnects a file from a specified unit and specifies whether the file connected to that unit is to be kept or released A CLOSE statement can appear in any program unit in the program it need not appear in the same program unit as the OPEN statement specifying the same unit A CLOSE statement that references a unit that does
200. e input output list This descriptor is useful in forms where nonlist item edit descriptors follow list item edit descriptors when the iolist is exhausted the subsequent edit descriptors are not processed 5 27 24 X Descriptor The X descriptor is used to skip character positions in an input line or output line X is not associated with a variable in the input output list The X descriptor has the form nX Where n isthe number of character positions to be skipped from the current character position n is a nonzero unsigned integer The specification nX indicates that transmission of the next character to or from a record is to occur at the position n characters forward from the current position When an X specification causes control to pass over character positions on output positions not previously filled during record generation are set to blanks however positions already filled are left unchanged 5 27 25 Z Descriptor The Z descriptor is used for hexadecimal conversion This descriptor has the forms Zw Zw m The form Zw m is meaningful for output only Hexadecimal digits include the digits 0 through 9 and the letters A through F A hexadecimal digit is represented by 4 bits On input embedded blanks are interpreted as zero and an all blank field is equivalent to zero The string is stored right justified with zeros on the left On output the rightmost w 4 bits are converted to hexadecimal and written If m is specifi
201. e input records are formatted If it is an array identifier then that array is assumed to contain the format information Finally if it is an integer variable the variable is assumed to have been assigned the statement number of a FORMAT statement by an ASSIGN statement is an integer expression which specifies the number of the record to be read or written in the file It must be greater than zero and is valid for files opened for direct access only is the identifier of a NAMELIST group The use of this parameter indicates a namelist read See the NAMELIST statement for a description of the conventions used for the namelist read specifies the label of an executable statement to which control transfers when an end of file is encountered during an input operation END s1 is ignored for direct access input operations specifies the label of an executable statement to which control transfers if an error condition is encountered during input processing specifies an integer variable into which an error code is stored as follows means that an end of file was encountered 0 means that the input operation completed normally and a value greater than zero indicates some other error condition is an input item list described below 119 PromulaFortran Compiler User s Manual Regardless of the type of the read if an error occurs as a result and if neither ERR s1 nor IOSTAT ios is supplied then execution will terminate with an
202. e is a variable that is typed explicitly as logical byte It may contain only a value of TRUE or FALSE The storage allocated to a logical byte variable is the same as that allocated to an integer byte variable 3 4 11 Character Variable A character variable is a variable that is typed explicitly as character The length of the character variable is specified when the variable is typed as character There is no storage distinction between a character variable of length n and an n element array of single characters The maximum length of a character variable is 32767 29 PromulaFortran Compiler User s Manual 3 5 Arrays A FORTRAN array is a set of elements identified by a single name The name is composed of 1 to 255 letters digits underscore _ or dollar sign beginning with a nondigit Each array element is referenced by the array name and a subscript The type of the array elements is determined by the array name in the same manner as the type of a variable is determined by the variable name The array name can be typed explicitly with a type statement implicitly with an IMPLICIT statement or by default typing The array name and its dimensions must be declared a DIMENSION COMMON or type statement When an array is declared the declaration of array dimensions takes the form shown below array d d Where array isthe symbolic name of the array d specifies the bounds of an array dimension and takes the for
203. e namelist read statement 131 PromulaFortran Compiler User s Manual 6 FORTRAN INTRINSIC FUNCTIONS Certain procedures that are of general utility or that are difficult to express in FORTRAN statements are contained within the FORTRAN library In general these include the intrinsic functions and various subprograms that interface with different aspects of the operating system An intrinsic function is a procedure that performs a set of calculations when its name appears in an expression in the referencing program unit Intrinsic functions communicate with the referencing program unit through a single value associated with the function symbolic name When the name of an intrinsic function duplicates another element in a program the result depends on the element and the references If a variable array or statement function is defined with the same name as an intrinsic function the name is a local name that no longer refers to the intrinsic function If an external function subprogram is written with the same name as an intrinsic function use of the name references the intrinsic function unless the name is declared as the name of an external function with the EXTERNAL statement Intrinsic functions are typed by default and need not appear in any explicit type statement in the program Explicitly typing a generic intrinsic function name does not remove the generic properties of the name Certain intrinsic functions are generic but ha
204. e same action as the REAL 8 statement Description The DOUBLE PRECISION statement is used to define a variable array symbolic constant function name or dummy procedure name as type double precision The symbol may already have been defined in another declaration statement See also See the discussion of the DIMENSION statement for a description of how dimension bounds are defined See the discussion of the DATA statement for a description of how initial values are defined 5 17 DO WHILE Statement The DO WHILE statement repeats a group of statements while a condition is true Syntax 66 PromulaFortran Compiler User s Manual DO sl WHILE exp block sl exec or DO WHILE exp block END DO Where sl is the label of an executable statement exp is a logical expression exec isanexecutable statement block isa block of executable statements Description The DO WHILE statement is similar to the DO statement The difference is that DO WHILE executes for as long as a logical expression is true while the simple DO executes for a fixed number of iterations The DO WHILE statement tests the logical expression at the beginning of each execution of the loop including the first If the value of the expression is true the statements in the body of the loop are executed if the expression is false control transfers to the statement following the loop If no label appears in a DO WHILE statement the DO WHILE
205. e the generic properties of the name See Also The EXTERNAL statement which allows user subprograms to be used as parameters The GENERIC statement which is equivalent to this statement 105 PromulaFortran Compiler User s Manual 5 37 LOGICAL Statement The LOGICAL statement defines some user defined entity to be of type logical short logical or logical byte Syntax LOGICAL 1en name name Where name has one of the forms var len c array d d len clist len specifies the integer subtype and can be an unsigned nonzero integer constant whose value is 1 2 or 4 var is a variable function name symbolic constant or dummy procedure array isan array name d specifies the bounds of a dimension clist isa list of constants or symbolic constants specifying the initial values Each item in the list take the form is a constant or symbolic constant r isa repeat count that is an unsigned nonzero integer constant or the symbolic name of such a constant Description The LOGICAL statement is used to define a variable array symbolic constant function name or dummy procedure name as type logical short logical or logical byte A length specification immediately following the word LOGICAL applies to each entity not having its own length specification A length specification immediately following an entity is the length specification only for that entity If the
206. e used as the delimiter 5 27 9 BN and BZ Blank Interpretation The nonrepeatable BN and BZ edit descriptors can be used with the repeatable D E F G and I edit descriptors on input to specify the interpretation of blanks other than leading blanks In the absence of a BN or BZ descriptor blanks in input fields are interpreted as zeros or are ignored Their interpretation depends on the value of the BLANK parameter in the OPEN statement that is currently in effect for the input output unit BLANK NULL blanks ignored is the default for input If a BN descriptor is encountered in a format specification all blank characters in succeeding numeric input fields are ignored that is the field is treated as if blanks had been removed the remaining portion of the field right justified and the field padded with leading blanks A field of all blanks has a value of zero If a BZ descriptor is encountered in a format specification all blank characters in succeeding numeric input fields are interpreted as zeros For example assuming if the statement READ 6 I13 BZ BN 13 I J K reads the input record I J and K are assigned the following values I 21 J 200 K 3 5 27 10 Carriage Control Character The carriage control character is the first character of a printer output record and is not printed It appears in other forms of output as data The carriage control characters are shown below Ch
207. e value of m1 3 The iteration count is established as follows ic MAX INT ms m1 m3 m3 mtc where mtc is the minimum iteration count and equals 0 if FORTRAN 66 conventions are desired and 1 if FORTRAN 77 conventions are desired 4 If ic is not zero the loop is executed else execution continues beyond the end of the loop 5 The control variable is incremented by the value m3 ic is decremented by 1 and execution loops back to step 4 The TO and 1 options set the value of mtc in the above description Selecting TO which is the default selects the FORTRAN 77 convention Selecting which forces DO loops to be executed at least once selects the FORTRAN 66 convention 2 5 6 Specifying Unit Numbers FORTRAN has a variety of contexts with I O statements in which no explicit unit number is provided PRINT READ WRITE PUNCH There are a variety of conventions as to what actual units to associate with these statement forms The U command line switch allows you to control this number Note that the PROMULA FORTRAN runtime library associates no significance to any particular unit number value A unit number may be any integer value The URnum or Ur option tells the compiler to use unit number num with READ statements for which no unit number is explicitly shown The default setting of UR specifies that READ statements for which no unit is explicitly shown should be directed to standard input The UWnum or UW option tells th
208. each array element If a length is not specified for an entity the length is 1 If a dummy argument has the length specified the dummy argument assumes the length of the associated actual argument for each reference to the subroutine or function If the associated actual argument is an array name the length assumed by the dummy argument is the length of each array element in the associated actual argument 53 PromulaFortran Compiler User s Manual If a symbolic constant of type character has the length specified the constant has the length of its corresponding constant expression in a PARAMETER statement See also See the discussion of the DIMENSION statement for a description of how dimension bounds are defined See the discussion of the DATA statement for a description of how initial values are defined 5 8 COMMON Statement The COMMON statement provides a means whereby different program units can share the same information Syntax COMMON cb nlist cb nlist Where cb is a symbol name nlist isa list of entities separated by commas which can take the following forms var c array d d clist var is a variable array isan array name d specifies the bounds of a dimension clist isa list of constants or symbolic constants specifying the initial values Each item in the list take the form r cG c is a constant or symbolic constant r isa repeat count that is an unsigned non
209. ecececececececececececececececececececececececececececececececececececececsss 167 6 96 MAXIMUM 2 20 167 The Promula Fortran Compiler User s Manual 6 975 MAX MAXIMUM VALUE eov ert tue A de dee eee e yere eesi 168 6 98 MINO MINIMUM V ABUE eter trie rte ere erede out ea y e reU oe tes 168 6 99 MINA MINIMUM VALUE iore ette rere cete eese oreet dese vec deca tpi og deret 168 6 100 MIN MINIMUM V ABUE 5 reef et eter redet eet eee et ee ei eee c e tithes edet tee es bee re ee eset 169 6 1017 MOD REMAINDER a nidore eter e eres ey eene red esee tee ru oe dere 169 6 102 NINT NEAREST INTEGER 7 5 5 tede rere ree 170 6 103 REALE REAL VAEUE OR PART tret eti eerte voe tree tele eese tee edere lea ee pete e 170 6 104 SIGN CTRANSEER OESIGN eed re eae re ere eru ee epe n 170 6 105 SINLIEZHYPERBOLIG SINE e ut e rel eoe rege ve ee eee loeo e bte e e ire 171 6 106 SIN SINE outs eite e A A 171 6 107 SNGE REAT VALUE ttr dani ete Phe idi tamtn itu 172 6 108 SORT SQUARE het ere mr Dn eta timetis 172 6 109 TANH HYPERBOLIC TANGENT ccccssccccccccsesessececceeceessnscaecececseseseaeeeeeccesessaaececceseseeaaeceeececeeensaeeeeeceeseseaaeeeeess 172
210. ect DVAX Vax FORTRAN DPDP PDP FORTRAN DP77 Prime FORTRAN 77 DPIV Prime FORTRAN IV This particular option should only be used if code is being moved directly from one of these compilers to the PROMULA compiler 2 6 Storage Quantity Values The PROMULA compiler is very conservative in its memory use Extremely large programs even those that exceed the size constraints of some mainframe compilers can be compiled even on the traditional MS DOS platforms There are however a few miscellaneous storage areas which need to be preallocated for efficiency reasons 1 The table of statement line numbers controlled by the QEn option 2 The current input statement controlled by the QIn option 3 The control table of all include files processed controlled by the QHn option and 4 The unresolved externals tables controlled by the QXn option The details of these tables will not be discussed here If one of these areas is exceeded you will get a message telling you explicitly which has been exceeded and which option to use 2 7 Miscellaneous Options The miscellaneous options are used to control the behavior of the compiler and the various other tools used by it 2 7 1 The Compile Only Option 14 PromulaFortran Compiler User s Manual The option tells the compiler that FORTRAN source codes are to be compiled to object form but that no link step should be attempted 2 7 2 The Debugging Flag The flag tell
211. ed The philosophy behind the design of the symbols reference table is that the user will use this table when he wants information about a particular symbol whose status he may not be familiar with but whose identifier he knows or has seen PromulaFortran Compiler User s Manual somewhere in the listing The report consists of a single alphabetical list of each symbol referenced in the subprogram An object type storage status binary type and comment are provided for each symbol The object type is straightforward There are thirteen possible entries constant parameter variable subroutine function intrinsic namelist entry statefunc structure record pointer and common These names correspond directly to the possible FORTRAN object types It should be mentioned that members within structure definitions are treated simply as variables or records This convention is compatible with the approach of using a single alphabetized list of all symbols The storage status of a symbol can be one of four different things For subprogram arguments it is specified as argument For variables in common blocks it is the name of the common block containing the variable For members of structures it is the name of structure containing the member on For simple variables it is one of the following static auto dynamic or virtual See the Sa and Ss switches for a description of static versus auto storag
212. ed the number is printed with leading zeros so that at least m digits are output If the number of hexadecimal digits exceeds w a field of asterisks is written 5 27 26 Repetition Factors 90 PromulaFortran Compiler User s Manual The repeatable edit descriptors can be repeated by prefixing the descriptor with a nonzero unsigned integer constant specifying the number of repetitions required For example 10 FORMAT 314 2E7 3 is equivalent to 10 FORMAT 14 14 14 E7 3 E7 3 Also 50 FORMAT 4612 6 is equivalent to 50 FORMAT G12 6 G12 6 G12 6 G12 6 A group of descriptors can be repeated by enclosing the group in parentheses and prefixing it with the repetition factor If no integer precedes the left parenthesis the repetition factor is 1 For example 1 FORMAT 13 2 15 3 6 1 214 is equivalent to the following specification if the number of items in the input output list does not exceed the number of format conversion codes 1 FORMAT IE E15 3 F6 1 14 14 E15 3 F6 1 14 14 A maximum of five levels of parentheses is allowed in addition to the parentheses required by the FORMAT statement If there are fewer items in the input output list than indicated by the format conversions in the FORMAT specification the excess conversions are ignored If the number of items in the input output list exceeds the number of format conversions when the final right parenthesis in the FORM
213. ed via an example DIMENSION 4 3 2 EQUIVALENCE Y 1 B 3 1 EQUIVALENCE X Y 2 In this example storage is shared so that 6 storage words are needed for v and x The associations are B 1 1 B 2 1 B 3 1 B 1 2 B 2 2 B 3 2 In any equivalence set i e set of variables which occupy the same storage because of EQUIVALENCE statements one variable is always the base variable while other variables are the derived variables The base variable is the 71 PromulaFortran Compiler User s Manual variable which actually allocates storage while the storage locations of the derived variables are computed from the storage location of the base variable In the above example B is the base variable because it is the largest i e occupies the most memory locations variable in the equivalence set B Y X The following example shows an equivalence set of character variables CHARACTER A 5 C 4 D 2 2 EQUIVALENCE A D 1 C D 2 In this example storage is shared so that 5 character storage positions are needed for a c and The associations are A 1 1 D 1 1 1 A 2 2 D 1 2 2 A 3 3 D 2 1 1 C1 A 4 4 D 2 2 2 C 2 2 A 5 5 3 3 C 4 4 Again A is the base variable in the equivalence set A D C because A occupies the most memory Notice however that the total memory allocated by the equivalence set is larger than
214. efault typing The statement specifies the type of variables arrays symbolic constants and functions beginning with the letters ac The IMPLICIT statements in a program unit must precede all other specification statements except PARAMETER statements An IMPLICIT statement in a function or subroutine subprogram affects the type associated with dummy arguments and the function name as well as other variables in the subprogram Explicit typing of a variable name or array element in a type statement or FUNCTION statement overrides an IMPLICIT specification The specified single letters or ranges of letters specify the entities to be typed A range of letters has the same effect as writing a list of the single letters within the range The same letter can appear as a single letter or be within a range of letters only once in all IMPLICIT statements in a program unit The length can be specified implicitly for entities of type character If length is not specified the length is one The length can be specified as an unsigned nonzero integer constant or an integer constant expression enclosed in parentheses with a positive value The specified length applies to all entities implicitly typed as character Note that any explicit typing with a type statement is effective in overriding both the default typing and any implicit typing The IMPLICIT NONE statement overrides all implicit defaults All data types in the program unit must be explicitly declare
215. el with the ASSIGN statement The first three types of assignment are discussed below The ASSIGN statement is discussed in a later chapter As is the normal convention the action of the ASSIGN statement is not considered to be assignment as that term will be used in this discussion 4 3 1 Arithmetic Assignment The arithmetic assignment statement is shown below v g Where v isanlvalue of type integer short integer byte real double precision complex or double complex isan arithmetic expression After evaluation of arithmetic expression e the result is converted to the type of v using the standard intrinsic conversion functions The result is then assigned to v and v is defined or redefined with that value 4 3 2 Character Assignment The character assignment statement is shown below Where v isthe name of a character variable character array element character substring e is a character expression The character expression e is evaluated and the result is then assigned to v The variable v and expression e can have different lengths If the length of v is greater than the length of e e is extended to the right with blank characters until it is the same length as v If the length of v is less than the length of e e is truncated from the right until it is the same length as V Only as much of the value of e must be defined as is needed to define v In the example CHARACTER A 2 B 4 A B the assi
216. ement After one item has been input the next format specification is taken together with the next element of the list and so on until the end of the list An implied DO specification has the following form dlist i el e2 e3 The elements i 1 e2 and e3 have the same meaning as in the DO statement and dlist is an input item list Redundant parentheses are allowed and are ignored The range of an implied DO specification is that of dlist The value of i must not be changed within the range of the implied DO list by the READ statement Changes to the values of e1 2 and e3 have no effect upon the execution of the implied DO However their values can be changed in the READ statement if they are outside the range of the implied DO and the change does have effect For example READ 100 A I I 1 K reads a value into and uses that value as the terminal parameter of the implied DO The statements K 2 READ 10 A I I 1 K 10 FORMAT F10 3 read two records each containing a value for A An implied DO can be used to obtain a single value more than one time For example the list K B K 1 5 causes the value of variable B to be read five times Obviously the last value read will be the one stored in B when the READ is completed Input of array elements can be accomplished by using an implied DO The list of variables followed by the DO index is enclosed in parentheses to form a single element of the i
217. ength of the string defining it in the PARAMETER statement The length must not be changed by another IMPLICIT statement or by subsequent statements Once defined a symbolic constant can appear in the program unit in the following ways In an expression in any subsequent statement In a DATA statement as an initial value or a repeat count In a complex constant as the real or imaginary part A symbolic constant cannot appear in a FORMAT statement 5 41 PAUSE Statement The PAUSE statement temporarily suspends execution of a program Syntax PAUSE n Where 115 PromulaFortran Compiler User s Manual n is an integer or string expression Description The PAUSE statement causes the program to temporarily suspend execution At the same time the message PAUSE n is written to the CONSOLE file The manner in which the program is restarted varies from system to system Consult your system manager 5 42 PRINT Statement The PRINT statement transfers formatted information from the values named in the input output list to the OUTPUT file Syntax PRINT fn iolist Where fn is a format specification iolist is an output list Notes The PRINT statement is a special case of the formatted WRITE statement Description The PRINT statement transfers formatted information from the values specified in the input output list to the OUTPUT file according to the specified format This statement is identical to the statement
218. ent constant constant the constants are stored consecutively beginning with the location given by the array element The number of constants need not be equal but must not exceed the remaining number of elements in the array As a final example DIMENSION 3 5 LOGICAL 1 7 NAMELIST HURRY I1 12 13 K M Y Z L READ 5 HURRY can process the following input record SHURRY I1 1 L TRUE 12 2 13 3 5 Y 3 5 26 Y 1 1 11 12 0 1 13 4 14 2 1 2 16 175 The final values stored are as follows 1 1 1 2 14 0 2 2 2 2 14 0 3 3 2 14 0 3 5 26 0 1 3 14 0 1 1 11 0 16 2 1 120 0 M 17 Y 3 1 13 0 Z 1 2 L TRUE The rest of y is unchanged 5 39 OPEN Statement The OPEN statement can be used to associate an existing file with a unit number to create a new file and associate it with a unit number or to change certain attributes of an existing file Syntax OPEN unit OPEN UNIT u IOSTAT ios ERR sl FILE fin STATUS sta ACCESS acc FORM f m RECL rl1 BLANK b1nk 111 PromulaFortran Compiler User s Manual Where u is an integer expression unit ios is an integer lvalue sl is the label of an executable statement fin is a character expression sta acc fm blink r is an integer expression Notes The simplest form of the OPEN statement consists of a single unit specification not
219. entifier symbol Example of message 265 If Line Nl Source 6 err265 for 7 8 8 err265 for Error 266 Message PROGRAM ERR265 STRUCTURE SHIP CHARACTER 10 OWNER CHARACTER 10 DESTINATION END RECORD SHIP NINA PINTA SANTA_MARIA E265 The structure or record identifier is not a valid identifier symbol CHARACTER 10 WHERE WHERE NINA OWNER E145 The assignment statement contains the unrecognized symbol STOP END 244 PromulaFortran Compiler User s Manual The structure identifier token classifying the record has not been defined Example of message 266 If Line Nl Source 1 PROGRAM ERR266 2 STRUCTURE SHIP 3 CHARACTER 10 OWNER 4 CHARACTER 10 DESTINATION 5 END 6 RECORD SHUP NINA PINTA 6 err266 for E266 has not been defined 7 CHARACTER 10 WHERE 8 WHERE NINA OWNER 8 err266 for E145 symbol 9 5 10 END Error 267 Message SANTA MARIA The structure identifier SHUP classifying the record The assignment statement contains the unrecognized The structure or record definition identifier is followed by the symbol token rather than a slash Example of message 267 If Line Nl Source 1 PROGRAM ERR267 2 STRUCTURE SHIP 3 i CHARACTER 10 OWNER 4 1 CHARACTER 10 DESTINATION S d END 6 RECORD SHIP NINA PINTA 6 err267 for E267 followed by the symbol 7 CHARACTER 10 WHERE 8 WHERE NINA OWNER 8 err267 for E145 symbol 9 STOP 10 END Error 268 Me
220. equivalence with ident is expanding the size of variable ident by number bytes Example of message 873 If Line Nl Source 5 warn873 for SUBROUTINE W873 REAL X 10 Y 13 EQUIVALENCE X 5 Y 1 WRITE X Y HI END C873 The equivalence with Y is expanding the size of variable X by 28 bytes type type type type type 286 PromulaFortran Compiler User s Manual Comment 874 Message The equivalence with ident is expanding the size of common ident by number bytes Example of message 874 If Line Nl Source 1 SUBROUTINE W874 2 REAL X 10 3 COMMON XX Y 4 EQUIVALENCE X 9 Y 5 WRITE HI 6 END 6 warn874 for C874 The equivalence with X is expanding the size of common XX by 4 bytes Note 875 Message Omitting commas from the statement statement is nonstandard Example of message 875 If Line Nl Source 1 SUBROUTINE W875 2 IMPLICIT INTEGER A REAL I 2 warn875 for N875 Omitting commas from the implicit statement is nonstandard 3 AA 2 4 4 5 AA II 6 END Comment 876 Message Common block ident is number bytes longer than when previously defined Example of message 876 If Line Nl Source SUBROUTINE W876 COMMON E156X A 99 CHARACTER 21 B C E 3 F 3 COMMON E156Y B COMMON E1562 E EEE FFF WRITE E F END SUBROUTINE W876A COMMON E156X A 100 MANA R
221. er User s Manual 344 2 Nested DO LOOPS 4 et tee ette ee ener e e Pd rrt eerie eed 65 3 15 DOUBLE COMPLEX STATEMENT dee een unire ere rens 65 5316 DOUBLE PRECISION STATEMENT e sis eee e tee gu oet itu redet een ne ete ee b eee ues 66 23 17 DO WHILE STATEMENT ertet t IR rer eH te en e tetro REF rmt 66 SHE ELSE STATEMENT 67 N19 ENCODE STATEMENT HE er tent rp Idee t re anne 68 3 20 END STATEMENT a Here ite eH Heu eem 69 21 5 9 e rer eite am ET erue ethic mers 69 3 22 THEENDPIEE S TATEMENT i ertet rtr tre ete rr t e ee Eden cet 69 5 28 THE END IE STATEMENT Eaa KSE EEES Eaa Na 70 5 24 ENTRY STATEMENT EEE En EE ES EEEE A Ea 70 25 lt THE EQUIVALENCE STATEMENT iiiter tenente tae 71 5 20 EXTERNAT STATEMENT irs iet th RE DER RIO TERRI UII PIER RES 73 5 24 FORMAT STATEMENT 74 5 27 1 Character FORMAT Specification ccccccccceeccesscesscesscesecesecusecsecaaeeseeeseeeseesceeseesseeesecsecaecuaecsaecsuecsaeeseseneees 75 5 27 2 Noncharacter FORMAT Specifications eese eee 76 D273 Edit DeSCHIDIOTS sxc ssi cass teo rpg ne bum ier e Ebrei 7
222. er and there must be the same number as the dummy arguments in the referenced procedure The actual arguments that are evaluated before the association of arguments include expressions substring expressions and array subscripts If the actual argument is a procedure name the procedure must be available for execution at the time of the reference to the procedure A dummy argument is undefined unless it is associated with an actual argument Argument association can exist at more than one level of procedure reference and terminates within a program unit at the execution of a RETURN or END statement A subprogram reference can cause a dummy argument to be associated with another dummy argument in the referenced procedure For type character both the dummy and actual arguments must be of type character and the length of the actual argument must be greater than or equal to the length of the dummy argument If the length of the actual argument of type character is greater than the length of the dummy argument only the leftmost characters of the actual argument up to the length of the dummy argument are used as the dummy argument If a dummy argument is an array name length applies to the entire array and not to each array element Length of array elements in the dummy argument can be different from length of array elements in the actual argument The total length of the actual argument array must be greater than or equal to the total length of the dum
223. eric function SORT Note that if SORT were declared EXTERNAL then a user function would be assumed The appearance of a name in a GENERIC statement declares the name as a generic intrinsic function name If an intrinsic function name is used as an actual argument in a program unit it must appear in a GENERIC or INTRINSIC statement in the program unit The following intrinsic function names must not be used as actual arguments Type conversion functions BOOL CHAR CMPLX DBLE FLOAT ICHAR IDINT IFIX INT REAL and SINGL Lexical relationship functions LGE LGT LLE and LLT Largest smallest value functions AMAX0 AMAXI AMINO AMINI DMAX1 MAX MAX0 MAXI MIN MINO MINO Logical and masking functions AND OR XOR NEQV EQV The appearance of a generic intrinsic function name in a GENERIC statement does not remove the generic properties of the name See Also 93 PromulaFortran Compiler User s Manual The EXTERNAL statement which allows user subprograms to be used as parameters The INTRINSIC statement which is equivalent to this statement 5 30 GOTO Statement The GOTO statement transfers control to a labeled statement in a program unit Description The three types of GOTO statements are the unconditional GOTO the computed GOTO and the assigned GOTO Each is discussed separately below See Also The ASSIGN statement which assigns labels to variables for use in the assigned GOTO statement 5
224. error code set If either or both are supplied the execution continues despite any errors If supplied the IOSTAT i lvalue receives the runtime error code for the error condition encountered or a zero if no error occurred If ERR s1 is supplied then execution will branch to the statement s1 if an error occurs If END s1 is not supplied then encountering an end of file is treated as any other runtime error except that the error condition code is negative If END s1 is supplied then execution continues at the statement labeled with IOSTAT ios receiving a negative error condition code The input item list portion of an input statement specifies the items to be read and the order of transmission The input item list can contain any number of items List items are read sequentially from left to right If no list appears one or more records are skipped A list item consists of a variable name an array name an array element name a character substring name or an implied DO list List items are separated by commas Subscripts in an input item list can be written as any valid subscript form An array name without subscripts in an input list specifies the entire array in the order in which it is stored The entire array not just the first word of the array is read Assumed size array names are legal in input lists On formatted input the iolist is scanned and each item in the list is paired with the field specification provided by the FORMAT stat
225. eseeeeeeeececeeececececess 163 6 83 2 MAXIMUM VALUE u cccccccccescscscscececeeceseseseeseeseseseseeeseecececeeceseseeeseseseseseececececececeeceecececeeecececececececececess 163 6 84 IZ2MINO MINIMUM VALUE j ccccccccccscecscscseecsceesesesceescseseseseeeeesceceeseseecececececececececececececececececseeceeececsceeseecececececs 164 6 85 PMOD REMAINDER eese hs esee Rs aos inte eese ot eve estes erue 164 6 86 PNIN F E 164 6 87 SIGN TRANSFER OE SIGN xe eoe ue over D into eene e prev 164 6 88 LEN NUMBER OF CHARACTERS ccccccceceesescecececseecececseseseeeseececeeseseseeceeeeeeecesecececscseseececececececeeseseeeeeececececececeeececs 165 6 89 LGE LEXICALLY GREATER OR enne tenen ennt ee th nnns sess ense tensa EN D tena asser ener RESY 165 6 90 TOT EEXICALLY GREATER iier toti eot bee gue ie prove live aeui eoe ee 165 6 91 EEE EEXICALLY LESS OROEQUAET E 166 6 92 ET EEXICAELY EESS iuste rette ote ete E tisse oce eet eNOS 166 6 93 LOGI10 LOGARITHM BASE 10 166 6 94 LOG NATURAL LOGARITHM ccccccccecececececececccseecececseececececececececececcescececececececececececececececececececececececececececececsss 167 6 95 MAX0 MAXIMUM VALUE ccccccccesececscscecececececececececececececec
226. ess input output and DIRECT if the file is opened for direct access input output If the file is not opened acc is undefined is a character variable which returns whether or not the file can be opened for sequential access input output It returns YES if the file can be opened for sequential access input output NO if the file cannot be opened for sequential access input output and UNKNOWN if this cannot be determined is a character variable which returns whether or not the file can be opened for direct access input output It returns YES if the file can be opened for direct access input output NO if the file cannot be opened for direct access input output and UNKNOWN if this cannot be determined 103 PromulaFortran Compiler User s Manual fm is a character variable which returns whether or not the file is formatted It returns FORMATTED if the file is opened for formatted input output and UNFORMATTED if the file is opened for unformatted input output If the file has not been opened fm is undefined fmt is a character variable specifying whether the file can be opened for formatted input output It returns YES if the file can be opened for formatted input output NO if the file cannot be opened for formatted input output and UNKNOWN if it cannot be determined if the file can be opened for formatted input output unf is a character variable specifying whether the file can be opened for unformatted input outp
227. ession e2 is an arithmetic expression e3 is an arithmetic expression exec is an executable statement block is a block of executable statements Description The DO statement is used to specify a loop called a DO loop that repeats a group of statements In the first form s1 is the label of an executable statement called the terminal statement of the DO loop In the second form the ENDDO behaves like a CONTINUE statement and as such may be treated as the terminal statement The parameter v is the control variable for the loop It may be any noncomplex arithmetic type The parameters e1 e2 and e3 are called indexing parameters they can be integer real or double precision The types of these expressions are always converted to the type of the control variable prior to their use The parameter e1 is the initial parameter e2 is the terminal parameter and e3 is an optional increment parameter if omitted its default is 1 The terminal statement of a DO loop is an executable statement that must physically follow and reside in the same program unit as its associated DO statement The terminal statement must not be an unconditional GO TO assigned GO TO arithmetic IF or block IF ELSE IF ELSE END IF RETURN STOP END or DO statement If the terminal statement is a logical IF statement it can contain any statement except a DO block IF ELSE IF ELSE END IF END or another logical IF The range of a DO loop consists of all the executable
228. ession can be either a reference to a variable within the same program unit or to a dummy argument of the statement function Statement functions can reference dummy variables that appear in a SUBROUTINE FUNCTION or ENTRY statement but that statement must precede the statement function Statement function dummy arguments can have the same names as variables defined elsewhere in the same program unit without conflict Any reference to the name inside the function refers to the dummy argument and any reference to the name outside the function definition refers to the variable A statement function is referenced through its statement function name When the statement function name is referenced in an expression the statement function is evaluated The actual arguments are evaluated and converted to the type of the corresponding dummy argument the resulting values are used in place of the corresponding dummy arguments in evaluation of the statement function expression The definition of a statement function must not directly or indirectly reference itself The statement function name can appear anywhere in an expression where an operand of the same type can be used The type of the statement function result is the type of the statement function name The arguments must agree in order and number with the corresponding dummy arguments A statement function can be referenced only in the program unit where the statement function appears 47 PromulaF
229. ex with the argument used as the real part and the imaginary part zero For a single double complex argument the result is the same as the argument For two arguments a1 and a2 the result is double complex with argument a1 used as the real part and argument a2 used as the imaginary part DCMPLX does not have specific names See also CMPLX 6 44 DCONJG Conjugate 150 PromulaFortran Compiler User s Manual Synopsis COMPLEX 16 FUNCTION DCONJG COMPLEX 16 Description DCONJG a is a specific function that returns a conjugate of a double complex argument The result is double complex For a double complex argument ar ai theresultis ar ai with the imaginary part negated See also CONGJ 6 45 DCOSH Hyperbolic Cosine Synopsis REAL 8 FUNCTION DCOSH REAL 8 Description DCOSH a is a specific function that returns a hyperbolic cosine The result and argument are double precision See also COSH 6 46 DCOS Cosine Synopsis REAL 8 FUNCTION DCOS REAL 8 Description DCOS a is a specific function that returns a cosine The argument is assumed to be in radians and the result is in radians The result and argument are both double precision See also CCOS CDCOS and DCOS 6 47 DDIM Positive Difference Synopsis REAL 8 FUNCTION DDIM REAL 8 REAL 8 Description DDIM al a2 is a specific function that returns a double precision positive difference bet
230. exceeded during namelist processing 525 The DEMO license does not allow use of the include statement 526 The PILOT license can only include files less than 1000 bytes long 527 Insufficient value storage for character variable length specification 528 Insufficient system memory to continue processing The source code being processed must be shortened 529 Insufficient value storage for alternate symbol display name 530 Insufficient value storage for symbol reference information control table 531 The include file storage of M bytes is insufficient use the QHn flag to increase it 9 3 Runtime Error Messages 299 PromulaFortran Compiler User s Manual Runtime error messages occur when an error occurs during the execution of programs formed using the PROMULA FORTRAN compiler These are contained within the runtime library and are fatal Each message has a code and a short description as follows Code Text of message Code Text of message ELUN_EOF end of file encountered EWRS_IFS bad format specification EOPN_EOF end of file encountered EWRT_IFS bad format specification ERBV_EOF end of file encountered EWVL_IFS bad format specification ERTX_EOF end of file ENXF EFS bad of format control character EWTX EOF write beyond end of file ENXF BTF bad T format ELUN_RDO write to readonly file ENXF_BUS bad B business format string EINT_NAF no active file structure ENXF_BBF bad BN Z format ELUN_NAF no active file structure EOPN_LNR uni
231. f Line Nl Source 1 SUBROUTINE W806 2 1 FORMAT 1 1 11 World 2 warn806 for N806 Omitting the comma after the H format specification is nonstandard 3 2 FORMAT 1X Good 3 warn806 for N806 Omitting the comma after the X format specification is nonstandard 4 WRITE 1 5 WRITE 2 6 STOP 7 END Comment 807 Message The input format contains a superfluous S specification Example of message 807 If Line Nl Source al PROGRAM W807 2 1 FORMAT SP I6 3 READ 5 SP I5 IV 3 warn807 for C807 The input format contains a superfluous S specification 4 WRITE 6 1 IV 5 READ 5 1 IV 5 warn807 for C807 The input format contains a superfluous 5 specification 6 WRITE 6 1 IV 263 PromulaFortran Compiler User s Manual 7 READ 5 2 IV 8 2 FORMAT 55 18 8 warn807 for C807 The input format contains a superfluous S Specification 9 STOP 10 END Note 808 Message A comma in a format following a left parenthesis is nonstandard Example of message 808 If Line Nl Source al PROGRAM W808 2 1 FORMAT 1H1 Hello World 2 warn808 for N808 A comma in a format following a left parenthesis is nonstandard 3 WRITE 6 1 4 WRITE 6 9H Good Bye 4 warn808 for N808 A comma in a format following a left parenthesis is nonstandard 5 STOP 6 END Note 809 Message comma in a format preceding a right parenthesis is nonstandard Example of message 809 If Line Nl
232. f message 171 If Line Nl Source d PROGRAM ERR171 2 IMPLICIT INTEGER A B C 2 errl71 for E171 The implicit statement is closed by the symbol END OF STATEMENT rather than a right parenthesis 3 PRINT A B C 4 STOP 5 END Error 172 Message An element in the statement list is token rather than an identifier Example of message 172 If Line Nl Source 1 PROGRAM ERR172 2 SAVE A B 99 2 errl72 for E172 An element in the statement list is 99 rather than identifier 3 PRINT A B 4 STOP 5 END Error 173 Message The function keyword token is not well formed in the function declaration Example of message 173 If Line Nl Source 1 PROGRAM ERR173 2 INTEGER TEST 3 PRINT TEST 1 0 2 0 4 STOP 5 END 6 INTEGER 999 FUNCTION TEST 6 errl73 for E173 The function keyword 999 is not well formed in the function declaration 4 INTEGER A B 8 TEST B 9 RETURN 214 PromulaFortran Compiler User s Manual 10 END Error 174 Message The function keyword has been replaced by token in the function declaration Example of message 174 If Line Nl Source 1 PROGRAM ERR174 2 NTEGER TEST 3 PRINT TEST 1 0 2 0 4 STOP is END 6 NTEGER FONCTION TEST 6 errl74 for E174 The function keyword has been replaced by FONCTIONTEST in the function declaration 7 NTEGER A B 8 TEST A B 9 RETURN 10 END Error 175 Message A subroutine or program may not have a ty
233. f the input quantity 314 1592 is read under the specification 2PF8 4 the internal number is 314 1592 x 10 2 3 141592 However if an exponent is read the scale factor is ignored On output for editing the number in the output field is the internal number multiplied by 10 In the output representation the decimal point is fixed the number is adjusted to the left or right depending on whether the scale factor is plus or minus For example the internal number 3 1415926536 can be represented on output under scaled F specifications as shown below 1PF13 6 314159 F13 6 3 141593 1PF13 6 31 415927 3PEI3 96 3141 592654 For E and D editing the effect of the scale factor xP is to shift the output coefficient left x places and reduce the exponent by In addition the scale factor controls the decimal normalization between the coefficient and the exponent such that if is less than or equal to 0 there will be exactly k leading zeros and significant digits after the decimal point if x is greater than 0 there will be exactly x significant digits to the left of the decimal point and a k 1 significant digits to the right of the decimal point For example the number 3 1415926536 is represented on output under the indicated Ew a scaling as shown below 3PE20 4 0003 04 1 20 4 0314 02 20 4 3142E 01 1PE20 4 3 1416 00 3PE20 4 314 16E 02 88 PromulaFortran Compiler
234. fication On exit from the implied DO specification the control variable retains the first value to exceed the upper limit 2 The implied DO can be nested that is the ilist in an implied DO can itself contain an implied DO The first innermost control variable varies most rapidly and the last outermost control variable varies least rapidly For example a nested implied DO with two levels has the form list vl el e2 e3 v2 eel ee2 ee3 Nested implied DO loops are executed in the same manner as nested DO statements The nested form can be used to write from arrays Each execution of a WRITE statement writes at least one record The unformatted WRITE statement transmits one record to the specified unit u from the storage locations or values named in olist Records are not converted no FORMAT statement is used The information is transmitted to the designated file in the form in which it exists in the memory without any conversion The list directed WRITE statement writes data using a free form notation with items separated by commas The actual format used varies from platform to platform In all cases list directed output records can be read by the list directed READ statement See also The discussion of the OPEN statement describes the FORTRAN file system in general including the different file types The discussion of the FORMAT statement describes how format information is interpreted The NAMELIST statement describes th
235. for E221 The formal argument ALPHA has already been defined earlier in the argument list RETURN 3 END Error 222 Message The object ident is not an undimensioned variable as is required in this context Example of message 222 If Line Nl Source all PROGRAM ERR222 2 INTEGER IVEC 10 3 DEFINE FILE IVEC 100 512 U IREC 3 err222 for E222 The array variable IVEC is not an undimensioned variable as is required in this context STOP is END Error 223 Message The equivalence group with the last element ident is closed by the symbol token rather than a right parenthesis Example of message 223 If Line Nl Source 1 PROGRAM ERR 223 230 PromulaFortran Compiler User s Manual 2 DIMENSION AMAT 10 IMAT 5 3 EQUIVALENCE IMAT 2 AMAT 4 3 err223 for E223 The equivalence group with the last element AMAT is closed by the symbol END OF STATEMENT rather than a right parenthesis 4 STOP 5 END Error 224 Message An unexpected end of file was encountered while reading the source file Example of message 224 If Line Nl Source ui PROGRAM ERR22 2 DIMENSION 10 3 READ 5 AMAT 3 err224 for E224 An unexpected end of file was encountered while reading the source file Error 225 Message The variable ident substring subscript number is not in the valid range of number to number Example of message 225 If Line Nl Source 1 PROGRAM ERR225 2 C
236. for the logical unit number was supplied redundantly 3 STOP 4 END Error 130 Message The statement option option is followed by the symbol token rather than by a line number Example of message 130 If Line Nl Source 1 PROGRAM ERR130 2 OPEN 6 FILE ERR127 OUT ERR YES STATUS NEW 2 errl30 for E130 The statement option ERR is followed by the symbol YES rather than by a line number 3 TE STOR 4 END Error 131 Message 200 PromulaFortran Compiler User s Manual The end of statement was reached before the parameter option was supplied Example of message 131 If Line Nl Source al PROGRAM ERR131 2 1 FORMAT Hello World 3 ASSIGN 1 3 Errl3l for E131 The end of statement was reached before the parameter to was supplied 4 WRITE LABEL 5 STOP 6 END Error 132 Message An undefined keyword parameter token has been entered when option was expected Example of message 132 If Line Nl Source 1 PROGRAM ERR132 2 1 FORMAT Hello World 3 ASSIGN 1 FOR LABEL 3 errl132 for E132 An undefined keyword parameter FORLABEL has been entered when to was expected 4 WRITE LABEL 5 STOP 6 END Error 133 Message The variable ident in the statement statement is not an undimensioned integer Example of message 133 If Line Nl Source 1 SUBROUTINE ERR133 2 INTEGER IJ 10 3 ASSIGN 10 TO IJ 33 for E133 The variable IJ in the assign statement is
237. format field width may be too small in this format Example of message 812 If Line Nl Source 1 PROGRAM W812 2 1 1 20 17 2 warn812 for C812 A floating point format field width may be too small in this format 3 2 FORMAT 1X E20 12 4 WRITE 6 1 VAL 5 WRITE 6 2 VAL 6 STOP 7 END 265 PromulaFortran Compiler User s Manual Note 813 Message The option option on the statement statement is nonstandard Example of message 813 If Line Nl Source 1 PROGRAM W813 2 OPEN 1 FILE TEST1 DAT 3 OPEN 2 NAME TEST2 DAT 3 warn813 for N813 The NAME option on the open statement is nonstandard 4 STOP 5 END Warning 814 Message The argument token of type type has been entered where an argument of type type has been used Example of message 814 If Line Nl Source 1 PROGRAM W814 2 CALL 50 1 1 3 CALL SUB1 I X Y 2 3 warn814 for W814 The argument I of type integer 4 has been entered where an argument of type real 4 has been used STOP 5 END Comment 815 Message A data value of type type is being assigned to the variable ident of type type Example of message 815 If Line Nl Source 1 PROGRAM W815 2 INTEGER IVAL 3 DATA IVAL 4HFRED 3 warn815 for C815 A data value of type character is being assigned to the variable IVAL of type integer 4 4 WRITE 1X A4 IVAL 5 STOP 6 END Comment 816 Message 266
238. g ENDIF A block IF structure can contain one ELSE statement to provide an alternative path of execution within the structure This structure is shown below 98 PromulaFortran Compiler User s Manual IF exp THEN if block 1 ELSE if block 2 END IF In the structure with an ELSE statement execution continues with the first statement in if block 1 if exp is true If the last statement of 1 block 1 does not transfer control control transfers to the statement following ENDIF However if exp is false control transfers to the first statement in if block 2 If the last statement in if block 2 does not transfer control execution continues with the statement following END IF A block IF statement can have at most one associated ELSE statement An actual example is as follows READ 2 12 A B IF XSUM LT XLIM THEN 1 4 2 0 2 0 XSUM XSUM X I WRITE 3 14 X I XSUM ELSE Y I A B YSUM Y I WRITE 3 16 YSUM Y I ENDIF An IF structure can contain one or more ELSEIF statements to provide for alternative execution of additional block IF statements An example of this structure is as follows IF expl THEN if block 1 EIF exp2 HEN if block 2 EIF exp3 HEN if block 3 ELSE if block 4 NDIF n 5 This capability allows formation of IF structures containing a number of possible execution paths depending on outco
239. g and single precision not allowed 4 7D0 2 1 942 Parentheses missing 3 3 6 Logical Constant A logical constant takes the form of TRUE or FALSE The periods are part of the constant and must appear Its syntax is as follows TRUE FALSE Where TRUE Represents the logical value true FALSE Represents the logical value false 3 3 7 Character Constant A character constant is a string of characters enclosed in single or double quotes Within a character string the delimiter is represented by two consecutive occurrences of that delimiter Its syntax is as follows Vg Won Where s is a string of characters The minimum number of characters in a character constant is one and the maximum number of characters in a character constant is 32755 The length is the number of characters in the string Blanks are significant in a character constant Any characters in the platform system character set can be used Character positions in a character constant are numbered consecutively as 1 2 3 and so forth up to the length of the constant The length of the character constant is significant in all operations in which the constant is used The length must be greater than zero Examples of valid character constants ABC 123 5 Examples of invalid character constants ABC Terminating quote is missing YEAR S Invalid number of quotes Zero length character constant 3 3 8 Hollerith Consta
240. gnment requires that the substring 1 2 be defined It does not require that the substring 3 4 be defined If vis a substring e is assigned only to the substring The definition status of substrings not specified by v is unchanged 45 PromulaFortran Compiler User s Manual 4 3 3 Logical Assignment The logical assignment statement is shown below VW Where v isthe name of a logical short logical or logical byte variable or logical array element is a logical expression The logical expression is evaluated and the result is then assigned to v Note that e must have a value of either TRUE or FALSE 4 4 Statement Functions A statement function is a user defined procedure which has the same basic syntax as an assignment statement It is a nonexecutable single statement computation that applies only to the program unit containing the definition Within a program unit a statement function must appear after the specification statements and before the first executable statement in the unit A statement function must not directly or indirectly reference itself Syntax fun d d expr Where fun is a symbolic name which identifies the statement function d is a statement function dummy argument There must be at least one dummy argument expr is an expression in which each primary is one of the following an expr enclosed in parentheses a constant a symbolic constant a variable reference
241. has been passed an argument of type real 4 11 L TRUE 12 WRITE S 13 END Warning 819 Message The TL field specification will shift position before column 1 Example of message 819 If Line Nl Source 1 SUBROUTINE W819 2 WRITE 6 20 3 20 FORMAT 1X BIRD TL6 2HXX 3 warn819 for W819 The TL field specification will shift position before column 1 4 END Warning 820 Message The type function ident was previously defined as having type type Example of message 820 If Line Nl Source Jd FUNCTION W820 J 2 W820 FLOAT I 3 END 4 SUBROUTINE W820A I 5 INTEGER W820 6 W820 I I 6 warn820 for W820 The integer 4 function W820 was previously defined as having type real 4 7 END Warning 821 Message This type function has been previously referenced as having type type Example of message 821 If Line Nl Source 1 PROGRAM W821 2 REAL W821A 3 W821A 2 4 END 5 INTEGER FUNCTION W821A I 268 PromulaFortran Compiler User s Manual 6 W821A I 6 warn821 for W821 This integer 4 function has been previously referenced as having type real 4 7 END Note 822 Message The use of initial or redundant commas in symbol lists is nonstandard Example of message 822 If Line Nl Source i PROGRAM W822 2 PARAMETER PI 3 14159 2 warn822 for N822 The use of initial or redundant commas in symbol lists is nonstandard 3 WRITE PI 4 STOP 5 END Note 823
242. he PUNCH file according to the specified format This statement is identical to the statement WRITE u fn iolist where u is a unit number assigned to the PUNCH file As a simple example consider PROGRAM PUNCH CHARACTER B 3 A 1 2 B YES N 19 PUNCH 4 A B N 4 FORMAT G20 6 A 15 PUNCH 50 50 FORMAT END OF FILE STOP END which will convert the values of A B and N into a coded record using format 4 and will then write that record to the PUNCH file followed by a record containing END OF FILE 117 PromulaFortran Compiler User s Manual See also The WRITE statement describes format specifications and iolists in detail See that statement for details on these 5 45 READ Statement The READ statement transmits data from an external unit or internal file to memory storage locations Syntax READ fn ilist READ ilist READ nl ilist READ UNIT u FMT fn REC rn IOSTAT ios ERR sl END sl ilist READ UNIT u FMT nl REC rn IOSTAT ios ERR sl END sl ilist READ UNIT u FMT IOSTAT ios 51 51 ilist READ UNIT u IOSTAT ios REC rn ERR sl END sl ilist Where u is an integer expression or a character variable identifier or an asterisk fn is integer variable the label of a FORMAT statement a character expression or an array identifier r
243. he constants specified by clist The first item of nlist corresponds to the first constant of clist the second item to the second constant and so forth If an unsubscripted array name appears as an item in nlist a constant in clist must be specified for each element of the array The values of the constants are assigned according to the storage order of the array For arithmetic data types the constant is converted to the type of the associated n1ist item if the types differ For all other types the data type of each constant in clist must be compatible with the data type of the nlist item Each subscript expression used in an array element name in nlist must be an integer constant expression except that implied DO variables can be used if the array element name is in dlist Each substring expression used for an item in nlist must be an integer constant expression 5 11 1 Implied DO List Use in DATA Statement An implied DO list can be used as an item in n1ist as the following example shows REAL X 5 5 DATA X J I I 1 J J7 1 5 15 1 0 The elements of array x are initially defined with the DATA statement Elements in the lower diagonal part of the matrix are set to the value 1 0 The elements initialized are 1 1 2 1 2 2 3 1 3 2 3 3 4 1 4 2 4 3 4 4 5 1 5 2 5 3 5 4 and 5 5 as shown below Array x I 1 I 2 I 3 I 4 I 5 J 1 al J 2 1 al 58 PromulaFortran Compiler User s Manu
244. he record is to be encoded iolist is a list of variables arrays or array elements to be transmitted to the location specified by u Description ENCODE is similar to an internal file formatted WRITE Values are transferred to the receiving storage area from the variables specified in iolist under the specified format The parameter c specifies the number of characters to be transferred per record The record length is calculated from c The parameter fn describes the format to be used to encode the values The parameter u specifies the area to receive the encoded information The first record starts with the leftmost character of the location specified by u The internal file must be large enough to contain the total number of characters transmitted by the ENCODE statement If the list and the format specification transmit more than the number of characters specified per record the excess characters are ignored If the number of characters transmitted is less than the record length remaining characters in the record are blank filled See also See the WRITE statement for a discussion of internal files and of formatted output processing 68 PromulaFortran Compiler User s Manual 5 20 END Statement The END statement indicates the end of the program unit to the compiler Syntax END Description Every program unit must physically terminate with an END statement The END statement can be labeled If control flows into or
245. he summary syntax of the PF command is as follows pf options filenames options The remainder of this chapter discusses in detail the options and filenames that are accepted 2 1 How PF Works PF works as follows First it processes all entries on its command line to determine the type of tasks to be performed and the resources required to perform them If there is any entry on the command line which PF cannot interpret PF exits with an error message The chapter on error messages describes these in detail Next PF processes each FORTRAN source file specified along with any prototype and global files to form intermediate C source files These C source files are then converted to compiled form via the native C compiler During this conversion you may get some messages from the host C compiler These depend upon the particular platform Finally if an executable is to be formed all compiled files formed and any additional compiled and library files included on the command line are passed to the linker along with any specified linker options 2 2 File Names PF classifies files by their extension i e by the final 1 2 or 3 characters of the file name that are preceded by a period Note that the PROMULA compiler is not case sensitive however many operating environments are In the following discussion the letters defining the various file types may be either in upper or lower case insofar as the compiler is concerned If the exten
246. heses 4 1 3 Logical Expression 41 PromulaFortran Compiler User s Manual A logical expression is used to express a logical computation Evaluation of a logical expression logexp produces a result of type logical with a value of TRUE or FALSE It has the following syntax logdis logexp EQV logdis logexp NEQV logdis logexp XOR logdis Where logdis logterm logfact logprim is a logical disjunction in either form logterm logdis OR logterm is a logical term in either form logfact logterm AND logfact is a logical factor in either form logprim logprim is a logical primary A logical primary can be a logical expression enclosed in parentheses a relational expression or any of the following Logical constant Logical symbolic constant Logical short logical or logical byte variable Logical short logical of logical byte array element reference Logical function reference The simplest form of a logical expression is a logical constant symbolic name of a logical constant logical variable reference logical array element reference logical function reference or relational expression More complicated logical expressions can be formed by using one or more logical operands together with logical operators and parentheses The logical operators are shown below Operator Representing Meaning NOT Negation Complement x AND Conjunction Product of x1 and x2 OR
247. ic function that returns the value of the largest argument The result is the same type as its first argument See also AMAXI DMAXI 2 MAXO and MAXI 6 98 MIN0 Minimum Value Synopsis INTEGER FUNCTION MINO INTEGER INTEGER 2 FUNCTION MINO INTEGER 2 Description MINO al a2 an is a generic function that returns the value of the smallest argument All arguments are integer and the result is the same type as its first argument See also AMINO AMINI DMINI I2MINO MINI and MIN 6 99 MINI Minimum Value Synopsis INTEGER FUNCTION 1 Description MIN1 al a2 an is a specific function that returns the value of the smallest argument All arguments are real and the result is integer 168 PromulaFortran Compiler User s Manual See also AMINO AMINI DMINI I2MINO MINO and MIN 6 100 MIN Minimum Value Synopsis INTEGER 2 FUNCTION MIN INTEGER 2 INTEGER FUNCTION MIN INTEGER REAL FUNCTION MIN REAL REAL 8 FUNCTION MIN REAL 8 Description MIN al a2 an is a generic function that returns the value of the smallest argument The type of the result is the same as the type of the first argument See also AMINO AMINI DMINI I2MINO MINO and MINI 6 101 MOD Remainder Synopsis
248. ic sine The result is real or double precision depending on the argument type See also DSINH 6 106 SIN Sine Synopsis REAL FUNCTION SIN REAL REAL 8 FUNCTION SIN REAL 8 COMPLEX FUNCTION SIN COMPLEX 171 PromulaFortran Compiler User s Manual COMPLEX 16 FUNCTION SIN COMPLEX 16 Description SIN a is a generic function that returns a sine The argument is assumed to be in radians The result is real double precision complex or double complex depending on the argument type See also CDSIN CSIN and DSIN 6 107 SNGL Real Value Synopsis REAL FUNCTION SNGL REAL 8 Description SNGL a is a specific function that returns a real result from a double precision argument The result is the double precision value converted to real See also FLOAT and REAL 6 108 SQRT Square Root Synopsis REAL FUNCTION SORT REAL REAL 8 FUNCTION SORT REAL 8 COMPLEX FUNCTION SORT COMPLEX COMPLEX 16 FUNCTION SORT COMPLEX 16 Description SORT a is a generic function that returns the principal square root of its argument The result is real double precision complex or double complex depending on the argument type See also CDSQRT CSQRT and DSQRT 6 109 TANH Hyperbolic Tangent Synopsis REAL FUNCTION TANH REAL RE Di 8 FUNCTION TANH REAL 8 172 Promu
249. iler User s Manual where k is a signed or unsigned integer constant called the scale factor Scale factors can either precede D E F and G format specifications or appear independently Forms are as follows kPDw kPEw kPEw kPFw kPGw A scale factor of zero is established when each FORMAT specification is first referenced it holds for all and field descriptors until another scale factor is encountered Once a scale factor is specified it holds for all D E F and G descriptors in that FORMAT specification until another scale factor is encountered To nullify this effect for subsequent D E F and G descriptors a zero scale factor OP must be specified For example 15 FORMAT 2P E14 3 F10 2 G16 2 0P 4F13 2 The 2P scale factor applies to the 14 53 format specification and also to the F10 2 and G16 2 format specifications The OP scale factor restores normal scaling 10 9 1 for the subsequent specification 4F 13 2 Example 20 FORMAT 3P 5X E12 6 F10 3 0PD18 7 1P F5 2 12 6 and F10 3 specifications are scaled by 10 The D18 7 specification is not scaled and the 5 2 specification is scaled by 1071 The specification 3P 319 F10 2 is the same as the specification 319 3PF10 2 On input for F E D and G editing the number is divided by 10 and stored provided that the number in the input field does not have an exponent For example i
250. in the order in which they are written and they normally can be read back only in the same order This can be slow and inconvenient in applications where the order of writing is not the same as the retrieval order In addition such processing requires a continuous awareness of the current file position and the position of the required record To remove these limitations a direct access file capability is provided by the FORTRAN input output statements 113 PromulaFortran Compiler User s Manual In a direct access file any record can be read written or rewritten directly without concern for the position or structure of the file This is possible because the file resides on a random access mass storage device that can be positioned to any portion of a file Thus the entire concept of file position does not apply to a direct access file The notion of rewinding a direct access file is for instance without meaning To create a direct access file the user must specify an OPEN statement with ACCESS DIRECT and include RECL record length parameter For example OPEN 2 FILE DAFL ACCESS DIRECT RECL 120 opens an unformatted file DAFL for direct access The file is associated with unit 2 and has a record length of 120 words All records in a direct access file must have the same length The record length for a formatted direct access file is specified in characters The record length for an unformatted
251. ing here requires virtually no changes in the FORTRAN code The FORTRAN compiler puts into the executable the logic needed to place values on the disk at the point of their creation or use This logic introduces no errors and requires no changes in the FORTRAN source code Here we show through an example how the PROMULA FORTRAN Compiler can add value to an existing FORTRAN program by making its information directly accessible to the PROMULA system If you do not presently use PROMULA itself this will show you how the use of an integrated system can greatly simplify the use of and even breathe life into old batch oriented FORTRAN programs With PROMULA it is possible to enhance the user interface by creating a database that contains the information content of the FORTRAN program what goes in and what comes out as well as intermediate results User Interface Edit Inputs Text Files with a Text Editor Run Code with an Operating System Command Browse Edit Outputs Text Files with a Text Editor Print Outputs with an Operating System Command plus Use Program Database Independently of Program Code Retrieve and Browse Information from the Database Edit Data via Multidimensional Spreadsheet Data Editor or Edit Menus Use Multiple Windows and Menus to Run Program Use Report Generator to Produce Reports Plot Program Variables Perform Ad Hoc Calculations with Database Variables Develop On line Documentation and Context sensitive Help Impo
252. ing letter is an F The messages themselves are self explanatory and are listed below Number Message 501 Insufficient procedure storage for promotion operator 502 Insufficient definition storage for main control table 503 Insufficient symbol table storage 504 Insufficient value storage for scalar data value 505 Insufficient value storage for array values 506 Insufficient definition storage for dummy table 507 Insufficient definition storage for dimension control table 508 Insufficient definition storage for common table 509 Insufficient definition storage for prototype table 510 Insufficient procedure storage for next opcode 511 Insufficient procedure storage for opcode support word 512 Insufficient procedure storage for opcode support byte 513 Insufficient procedure storage for opcode parameters 514 Insufficient value storage for integer constant 515 Insufficient storage for parameter type list 516 Insufficient user space for common size table 517 Insufficient value storage for runtime data string 518 Insufficient value storage for include following record 519 Insufficient unresolved external symbol storage space 520 Insufficient procedure storage for required quantity storage 521 Include statement files nested more than 5 deep 522 Statement contains more than 1 characters use the flag to increase this value 523 Constant string length exceeds 255 characters 524 Length of scratch storage
253. ingle quote ampersand double quote equals sign left parenthesis exclamation mark A 5 represent the letters uppercase or lowercase symbols may be used There is no distinction made between them Remember that in the language syntax descriptions uppercase is always used for reserved words while lowercase is always used for user supplied identifiers Blanks may be inserted anywhere in a statement except within a Hollerith constant or a quoted string Such blanks are simply ignored Alternatively all blanks except within Hollerith constants and quoted strings may be omitted Blanks within Hollerith constants and quoted strings are simply treated as any other character Characters that are not included in the character set described above can be used in Hollerith constants in quoted strings and in comment lines Users should consult with their local system manager for a list of these other locally available characters and their representation Users are warned however that use of such other characters can reduce the portability of their programs A FORTRAN statement is a sequence of characters as described above Four general statement formats are available fixed tab free and continuation To describe these formats the notions of a source line and a statement field are needed A source line a FORTRAN program can be one of three things 1 A blank line 2 A comment line 3 A part of a statement
254. inimum of number and maximum of number is out of range Example of message 186 If Line Nl Source 1 PROGRAM ERR186 2 DIMENSION AVAL 2 15 3 12 3 DATA I J AVAL 2 2 6 4 34 5 3 errl86 for E186 The value 2 for subscript 2 of array AVAL with minimum of 3 and maximum of 12 is out of range 4 STOP 5 END Error 187 Message Data is being assigned the object ident which is not a variable Example of message 187 If Line Nl Source ah PROGRAM ERR187 2 PARAMETER PI 3 14159 3 DATA PI 3 14159 3 errl187 for E187 Data is being assigned the parameter PI which is not a variable 4 STOP 5 END Error 188 Message Data is being assigned to the subprogram argument ident Example of message 188 If Line Nl Source 1 SUBROUTINE ERR188 IARGU 2 DATA IARGU 56 2 err188 for E188 Data is being assigned to the subprogram argument IARGU 219 PromulaFortran Compiler User s Manual 3 RETURN 4 END Error 189 Message The data statement variable list symbol token is not valid Example of message 189 If Line Nl Source 1 PROGRAM ERR189 2 DIMENSION A 100 3 DATA A I I 1 100 100 1 2 3 189 E189 The data statement variable list symbol 100 is not valid 4 STOP 5 END Error 190 Message The data implied do loop variable is followed by the symbol token rather than an equals sign Example of message 190 If Line Nl Source L PROGRAM ERR190 2 DIMENSION A 100 3
255. ion The DECODE statement is the extended internal file input statement It performs a memory to memory transfer of data similar to an internal file formatted READ Starting at location u display code characters in memory are converted according to the specified format and stored in the variables specified in iolist The parameter c specifies the number of characters per record in the internal input file When DECODE processes an illegal character for a given conversion specification a fatal error results DECODE can be used to pack the partial contents of two words into one See also The READ statement for a discussion of internal files and of formatted input processing 5 13 The DIMENSION Statement The DIMENSION statement defines arrays and their bounds Syntax DIMENSION array d d clist Where array isan array name d specifies the bounds of a dimension in one of the forms upper lower upper upper is an expression in which all constants symbolic constants and variables are type integer or an asterisk lower is an expression in which all constants and symbolic constants and variables are of type integer 60 PromulaFortran Compiler User s Manual clist isa list of constants or symbolic constants specifying the initial values Each item in the list can take the form T c c is a constant or symbolic constant r is a repeat count that is an unsigned nonzero integer constant or the symb
256. ion instructions for details If the first character of name is a then the remainder of name is assumed to be the name of an environment variable which contains the include file search paths 2 8 Prototype Files As has been discussed in several other places FORTRAN passes all function arguments by name i e it passes the address of an argument to a subprogram as opposed to its value C on the other hand allows argument values to be passed directly as well as allowing the passing of argument addresses In instances where the value of the argument is scalar and where its value is not being changed by the subprogram passing the value of that argument is far more efficient than passing its address Whenever possible the FORTRAN compiler should use call by value The problem is that it is not always possible to tell whether call by value is valid simply from the source code Some other device is needed to tell the compiler which arguments can be passed by value Another problem with FORTRAN is that it is weak typing By this is meant that it is sometimes valid to pass data of differing binary types via the same subprogram argument The compiler needs to know when this is valid and if it is not 15 PromulaFortran Compiler User s Manual valid it needs to know which binary type is the expected one Many perfectly valid FORTRAN programs fail either at compilation time or at execution time because of differing typing co
257. iption DBLE a is a generic function that performs type conversion and returns a double precision result The argument can be integer real double precision or complex For an integer or real argument the result has as much precision as the double precision field can contain For a double precision argument the result is the argument For a complex argument the real part is used and the result has as much precision as the double precision field can contain See also REAL 6 43 DCMPLX Double Complex Value Synopsis COMPLEX 16 FUNCTION DCMPLX INTEGER 2 COMPLEX 16 FUNCTION DCMPLX INTEGER COMPLEX 16 FUNCTION DCMPLX REAL COMPLEX 16 FUNCTION DCMPLX REAL 8 COMPLEX 16 FUNCTION DC U x C ES EX 16 COMPLEX 16 FUNCTION DC U x ES I x COMPLEX 16 FUNCTION DCMPLX REAL 8 REAL 8 Description DCMPLX 1 2 is a generic function that performs type conversion and returns a double complex value DCMPLX can have one or two arguments A single argument can be integer real double precision complex or double complex If two arguments are used the arguments must be of the same type and must both be integer real or double precision For a single integer real or double precision argument the result is double compl
258. is chapter is to show you how A particularly difficult problem faces anyone porting an existing program to a PC or some other contemporary platform Traditional programs though perfectly useful are rejected by contemporary users because they are not sufficiently user friendly In designing the FORTRAN compiler we felt very strongly that this particular problem could not be ignored Simple compilation was not enough Our solution to this problem is to automatically link the disk image used by the virtual memory manager directly into our PROMULA system and to allow the compiler to use virtual memory for selected variables At issue is more than just the user interface There are myriad tools available on the PC spreadsheets database managers graphics packages statistical packages etc which can be used to work with the data now locked up in existing FORTRAN 177 PromulaFortran Compiler User s Manual programs Once the information can be gotten to disk in a structured form all of these can be used to enhance the value of that information Traditional techniques allow you to work only with the explicit information either read or written by the program Working with the information in the program simply means working with its input files and its output reports If this is not sufficient you must go into the program and modify it This required modification is often time consuming and can introduce errors The technique we are present
259. is nonstandard Example of message 802 If Line Nl Source 1 SUBROUTINE W802 2 1 FORMAT 1X F6 15 E10 2 warn802 for N802 Omitting the period from the F format specification is nonstandard 2 warn802 for N802 Omitting the period from the E format specification is nonstandard 3 WRITE 1 1 A I B 4 RETURN 5 END Note 803 Message The exponent field appended to the D format specification is nonstandard Example of message 803 If Line Nl Source 1 SUBROUTINE W803 A B 2 REAL 4 A 3 REAL 8 B 4 1 FORMAT 1X E26 8E5 D30 20E5 4 warn803 for N803 The exponent field appended to the D format specification is nonstandard 5 WRITE 1 1 A B 6 RETURN 7 END Note 804 Message The token format specification is nonstandard Example of message 804 If Line Nl Source al SUBROUTINE W804 2 DIMENSION IVAL 6 3 1 FORMAT 1X 6010 3 warn804 for N804 The O format specification is nonstandard 4 WRITE 6 1 IVAL 5 RETURN 6 END 262 PromulaFortran Compiler User s Manual Note 805 Message The b business format specification is nonstandard Example of message 805 If Line Nl Source 1 PROGRAM W805 2 WRITE 6 77999 12345 3 WRITE 6 1X B ZZ999 12345 3 warn805 for N805 The b business format specification is nonstandard 4 STOP 5 END Note 806 Message Omitting the comma after the token format specification is nonstandard Example of message 806 I
260. itially defined with the values 4 5 0 7 5 and 5 2 respectively Variable NRX is initially defined with the value 17 after type conversion of the real 17 0 to the integer 17 Array with 6 elements is initially defined with a value for each array element This second example shows the use of a repeat count gt EAL R 10 10 DATA R 50 5 0 50 75 0 The array is initially defined with the first 50 elements set to the value 5 0 and the remaining 50 elements set to the value 75 0 Entities that are initially defined by the DATA statement are defined when the program begins execution Entities that are not initially defined and not associated with an initially defined entity are undefined at the beginning of execution of the program A variable array element or substring must not be initially defined more than once in the program If two entities are associated only one can be initially defined by a DATA statement Names of dummy arguments and functions cannot be initially defined Entities in a common block can be initially defined within a block data subprogram or within any program unit in which the common block appears however the DATA initialization of COMMON variables with DATA statements in non BLOCK DATA subprograms is discouraged Within the DATA statement each list n1ist must have the same number of items as the corresponding list clist A one to one correspondence exists between the items specified by nlist and t
261. ive 158 6 69 TABS ABSOLUTE V ALUE 158 52590 158 6 71 ICHAR INTEGER VALUE OF 2222 159 6 72 IDIM POSITIVE DIFFERENCE 5 debe 159 159 6 74 IDNINT NEAREST INTEGER 5 5242 160 6 T5 IFLIX INTEGER VALUE 160 6 76 INDEX EOCATION OFSUBSTRING 1 352 161 6277 INT2 INTEGER VALUE EA tout e e e Eee eue Neto a 161 678 NTA INTEGER M AEBUE 5 eere a oda ceca eto neis veu ae e eo eee Cs oe 161 6 79 INT INTEGER V AEUB 20 iiu i ie eei uiii edid 162 6 80 SIGN TRANSFER OE SIGN 162 6 81 T2ABS ABSOLUTE eee ie bees share Si ou se OSES eese bee bise iem ve 163 6 82 I2DIM POSITIVE DIFFERENCE cccccccccecscececececcesceceescsesesesceeseseececececceseeeseseecececeescececeeseececces
262. l 4 EXPRESSIONS LVALUES ASSIGNMENTS AND STATEMENT FUNCTIONS Expressions are formed from a combination of operators operands and parentheses Assignment statements are executable statements that use expressions to define or redefine the values of variables 4 1 Expressions Expressions are classified in two ways by whether or not they are constant and by their type A constant expression is an expression in which only constants or symbolic constants and operators are used If an arithmetic expression is written using only constants and operators the expression is an arithmetic constant expression If a character or logical expression is written using only constants and operators the expression is respectively a character constant expression or logical constant expression The types of expressions are arithmetic character logical and relational The relational expressions are not fully independent and are used as parts of logical expressions Each of these types is discussed 4 1 1 Arithmetic Expression An arithmetic expression is a sequence of unsigned constants symbolic constants variables array elements and function references separated by operators and parentheses For example A B F C D E is a valid arithmetic expression Formally the syntax of an arithmetic expression is as follows term term term arithexp term arithexp term Where term is an arithmetic term in one of the forms f
263. l Logarithm Synopsis REAL FUNCTION LOG REAL REAL 8 FUNCTION LOG REAL 8 COMPLEX FUNCTION LOG COMPLEX COMPLEX 16 FUNCTION LOG COMPLEX 16 Description LOG a is a generic function that returns the natural logarithm of the argument The result is always the same type as the argument See also ALOG CDLOG CLOG and DLOG 6 95 MAX0 Maximum Value Synopsis INTEGER FUNCTION INTEGER 2 FUNCTION INTEGER 2 Description al a2 an is a generic function that returns the value of the largest argument The arguments are integer and the result is the same type as the first arguments See also AMAXI MAX DMAXI 2 and MAXI 6 96 MAXI Maximum Value Synopsis INTEGER FUNCTION MAX1 REAL Description 167 PromulaFortran Compiler User s Manual al a2 an isa specific function that returns the value of the largest argument All arguments are real and the result is integer See also AMAXI MAX DMAX1 2 and MAXO 6 97 MAX Maximum Value Synopsis INTEGER 2 FUNCTION MAX INTEGER 2 INTEGER FUNCTION MAX INTEGER REAL FUNCTION MAX REAL REAL 8 FUNCTION MAX REAL 8 Description al a2 an is a gener
264. l argument Another form of adjustable dimension is the assumed size array In this case the upper bound of the last dimension of the array is specified by an asterisk The value of the dimension is not passed as an argument but is determined by the number of elements stored in the array If an array is dimensioned the array in the calling program must be large enough to contain all the elements stored into it in the subprogram See the following for example SUBROUTINE CAT A M N B C REAL A M B N C DO 10 I 1 M 10 C I A I DO 20 I 1 N 20 C I M B I ETURN ND Subroutine CAT places the contents of array A followed by the contents of array into array The dimension of C in the calling program must be greater than or equal to M N Use of the asterisk form of the adjustable dimension to prevent subscript checking for the array is preferable to the common practice of declaring arrays to have dimension 1 See also The description of arrays is in chapter 2 This description covers the properties of arrays the storage of arrays and array references 62 PromulaFortran Compiler User s Manual 5 14 DO Statement The DO statement repeats a group of statements Syntax DO sl v el e2 e3 block sl exec or DO v el e2 3 block ENDDO Where sl is the label of an executable statement is an integer real or double precision variable el is an arithmetic expr
265. l structure into a lower level structure without the use of explicit GOTO statements 99 PromulaFortran Compiler User s Manual 5 32 IMPLICIT Statement The IMPLICIT statement specifies the default typing that occurs according to the first letters of the names Syntax IMPLICIT type ac ac type ac ac or IMPLICIT NONE Where type isany valid type specification ac is a single letter or range of letters represented by the first and last letter separated by a hyphen Description The IMPLICIT statement changes or confirms the default typing that occurs according to the first letters of the names The valid type specifications are as follows BYTE INTEGER 1 INTEGER 2 INTEGER 4 INTEGER REAL 4 REAL 8 REAL DOUBLECOMPLEX DOUBLEPRECISION LOGICAL 1 LOGICAL 2 LOGICAL 4 LOGICAL COMPLEX 8 COMPLEX 16 COMPLEX CHARACTER len The letters ac indicate which variables are implicitly typed to the specified type As an example IMPLICIT CHARACTER 20 M X Z specifies that variables whose names begin with the letter M are typed as character rather than integer and those beginning with x v or Z are character rather than real Default typing is effective in all other cases In this second example IMPLICIT LOGICAL L INTEGER L LX TT variable L is integer rather than logical because it is explicitly typed Lx is integer The name TT is integer because an explicit type overrides the d
266. laFortran Compiler User s Manual Description TANH a is a generic function that returns a hyperbolic tangent The result is real or double precision depending on the argument type See also DTANH 6 110 TAN Tangent Synopsis REAL FUNCTION TAN REAL REAL 8 FUNCTION TAN REAL 8 Description TAN a is a generic function that returns a tangent The argument is assumed to be in radians The result is real or double precision depending on the argument type See also DTAN 6 111 TIME Current Time Synopsis SUBROUTINE TIME CHARACTER Description TIME a obtains the current time as set within the system The time is returned as an 8 character character string of the form hh mm ss If the argument is too short to receive eight characters the return value is truncated If the argument is longer that eight then trailing positions are padded with blanks See Also DATE 173 PromulaFortran Compiler User s Manual 7 CONTROLLING RUNTIME BEHAVIOR Once an application has been processed by the compiler and converted into an executable form its runtime behavior can be controlled by passing execution control switches to the runtime system via the command line for the application These include the following Switch Ccode Inumber Onumber Tnumber Vname Svalue Zvalue Description of Use Establishes the runtime conventions
267. le The nesting level indicator is used with declaration statements when structures are being defined It indicates the level of nesting within the structure For executable statements the nesting level indicator indicates the degree of nesting within DO and or IF statements If there is no current nesting the nesting level is left blank 2 4 4 Symbol Listing and Cross Reference Table The following symbol listing and cross reference table was produced using the Ex option Not counting the heading which was described above the table consists of 4 sections include files symbols referenced symbol references by line number and statement label types and references In general in these tables all user defined symbols are shown in uppercase while all descriptive symbols are shown in lowercase PROMULA FORTRAN Compiler V4 00 Date 08 11 92 Time 09 35 Page 2 File utest for Include files used in unit Seq Filename PromulaFortran Compiler User s Manual 1 STRUC6 INC Symbols referenced in SUBROUTINE STRUC6 Identifier Object Storage Type Comment APP_DATE record APPOINTMEN DATE scalar APP_FLAG variable APPOINTMEN logical 1 scalar APP_LIST record static APPOINTMENT ld array 679 APP_MEMO variable APPOINTMEN character 20 1 80 APP_TIME record APPOINTMEN IME 1 4 APPOINTMENT structure DATE struc
268. length of the list item is used CHARACTER NAME 30 PHONE 7 READ 5 A A NAME PHONE On output if w is less than the length of the list item the leftmost characters in the item are output For example if a variable A declared CHARACTER 4 9 contains SAMPLE and A is output with the statement WRITE 6 1X A4 A then the characters SAMP are output If w is greater than the length of the list item the characters are output right justified in the field with blanks on the left For example if A in the previous example is output with the statements WRITE 6 40 A 40 FORMAT 1X A12 output is as follows SAMPLE If w is omitted the length of the character list item determines the length of the output field 79 PromulaFortran Compiler User s Manual 5 27 8 Single and Double Quote Descriptors Character strings delimited by a pair of single quote 7 or double quote symbols can be used as alternate forms of the H specification for output The paired symbols delineate the string If the string is empty or invalidly delimited a fatal compilation error occurs and an error message is printed The single and double quote descriptors are ignored on input A single or double quote within a string delimited by the same symbol can be represented by two consecutive occurrences of the symbol Alternatively if a single quote or double quote appears within a string the other symbol can b
269. ler and PROMULA can add significant value to FORTRAN applications A typical program user interface for old batch oriented FORTRAN programs is shown below The user manipulates what goes in and what comes out of the program via a text editor The program itself is a computational black box whose logic transforms the program inputs into the program outputs The instructions controlling the running of the program are usually in a batch procedure of commands to the operating system written in JCL the Job Control Language of the operating system User Interface Edit Inputs Text Files with a Text Editor Run Code with an Operating System Command Browse Edit Outputs Text Files with a Text Editor Print Outputs with an Operating System Command Program Inputs gt Program Code gt Program Outputs Figure 8 1 A Typical FORTRAN Program User Interface In this scheme the program s logic and the program s information are locked in the program code Typically the program s inputs and outputs are separate text files that can be accessed sequentially by the program and can be manipulated by the user via a text editor How do you make this interface more user friendly Can you do it without diminishing the current functionality of the program Can you add value and functionality without modifying the current code The answer is yes The PROMULA tools make it possible to add value to existing FORTRAN programs The purpose of th
270. les 1 If the length of the character item in nlist is greater than the length of the corresponding character constant the additional character positions in the item are initially defined as blanks 2 if the length of the character item in nlist is less than the length of the corresponding character constant the additional characters in the constant are ignored Note that initial definition of a character item causes definition of all character positions Each character constant initially defines exactly one character variable array element or substring The following is a character data initialization example CHARACTER STR1 6 STR2 3 DATA STR1 ABCDE DATA STR2 FKGJK The character variables STR1 and STR2 are initially defined Variable srR1 is set to ABCDE with the sixth character position defined as blank Variable stR2 is set to FKG with the fourth and fifth characters of the constant ignored 59 PromulaFortran Compiler User s Manual 5 12 DECODE Statement The DECODE statement is the extended internal file input statement Syntax DECODE c fn u iolist Where is an unsigned integer constant or variable having a value greater than zero fn is a statement label of a FORMAT statement or a character expression whose value is a format specification u is a variable array element or array name iolist is a list of character variables arrays or array elements Descript
271. ll give a brief discussion of the elements of this description 001 DEFINE PROGRAM Exposure Analysis Model 002 OPEN SEGMENT expo xeq STATUS NEW 003 DEFINE FILE 004 dbase TYPE ARRAY 005 ND FILE 006 OPEN dbase STATUS NEW 007 DEFINE SE 008 tp 50 Time Points 009 END 010 DEFINE VARIABLE dbase 011 iday tp YPE INTEGER 5 Day 012 imnth tp TYPE INTEGER 5 Month 013 iyr tp YPE INTEGER 5 Year 014 avge tp YPE REAL 10 1 Average Exposure M T 015 avgi tp YPE REAL 10 1 Average Intake M T 016 conc tp YPE REAL 10 5 Ambient Concentration M L 3 017 cume tp YPE REAL 10 1 Cumulative Exposure M 018 cumi tp YPE REAL 10 1 Cumulative Intake M 019 expos tp TYPE REAL 10 1 Exposure M T 020 perd tp YPE REAL 10 1 Period 021 sever tp TYPE REAL 10 1 Severity 022 take tp YPE REAL 10 1 Intake M T 023 END 024 DEFINE VARIABLE 025 date tp TYPE DATE 10 Date of exposure 183 PromulaFortran Compiler User s Manual 026 END 027 DEFINE RELATION 028 KEY tp date 029 END 030 DEFINE PROCEDURE main 031 OPEN dbase expo dba STATUS OLD 032 SELECT tp IF iday tp
272. lue Synopsis INTEGER FUNCTION IFIX INTEGER 2 INTEGER FUNCTION IFIX INTEGER INTEGER FUNCTION IFIX REAL INTEGER FUNCTION IFIX REAL 8 INTEGER FUNCTION IFIX COMPLEX INTEGER FUNCTION IFIX COMPLEX 16 Description IFIX a is a generic function that performs conversion to integer and returns an integer result The conversion to integer for noncomplex arguments is performed via simple truncation For complex arguments only the real part is used and truncated See also IDINT INT INT2 and INT4 160 PromulaFortran Compiler User s Manual 6 76 INDEX Location of Substring Synopsis INTEGER FUNCTION INDEX CHARACTER CHARACTER Description INDEX al a2 is a specific function that returns the location of a substring within a string Both arguments must be character string arguments If string a2 occurs as a substring within string a1 the result is an integer indicating the starting position of the substring a2 within a1 If a2 does not occur as a substring within a1 the result is 0 If a2 occurs as a substring more than once within a1 only the starting position of the first occurrence is returned See also None 6 77 INT2 Integer Value Synopsis INTEGER 2 FUNCTION INT2 INTEGER INTEGER 2 FUNCTION INT2 INTEGER 2 INTEGER 2 FUNCTION INT2 REAL
273. ly create a database rather it converts references to variables in the program to references to variables on the database It must have access to the database so that it can locate the values The subsequent lines simply specify the program variables their corresponding database variables and the access method to be used The fields themselves are free format The first field defines a global symbol common block name or subprogram name which contains a desired variable The second field contains the local identifier of the variable itself The third field contains the name of the corresponding variable on the database Note that in the example above the local names and the database names are always the same In general this is not required The final field defines the actual access method to be used At this point the only two methods are virtual and dynamic With virtual access variable values are accessed using a virtual memory scheme Using purely virtual access there is no limit on the amount of information that can be referenced however there may be a performance penalty With dynamic access the memory needed to contain the variable is allocated at runtime and the entire set of values is read in at that time When the use of the variable is completed the values are written back out and the memory released Depending upon the situation dynamic access can be more efficient 8 7 Using EXPO with PROMULA As far as the user is concer
274. m lower upper Where lower specifies the lower bound of the dimension The lower bound is an integer expression with a positive zero or negative value If omitted the lower bound is assumed to be 1 upper Specifies the upper bound of the dimension The upper bound is an integer expression with a positive zero or negative value The upper bound must be greater than or equal to the lower bound In the case of an assumed size array the upper bound of the last dimension can be specified as There is no limit on the number of dimensions that an array can have The dimension bounds can be positive negative or zero If the lower bound is omitted the lower bound is assumed to be one In this case the upper bound must be positive The general rule is that the upper bound must always be greater than or equal to the lower bound The size of each dimension is indicated by the distance between the lower bound and upper bound i e the span of an array dimension is given by upper lower 1 where upper is the upper dimension bound and lower is the lower dimension bound For example DIMENSION RX 0 5 declares a 1 dimensional array of six elements such as shown below The values are the element locations relative to the location of the first element Row 0 Row 1 Row 2 Row 3 Row 4 Row 5 Alternatively 30 PromulaFortran Compiler User s Manual DIMENSION TABLE 4 3 declares a 2 dime
275. m the symbolic name of a dummy argument is unique to the program unit and must not appear in an EQUIVALENCE PARAMETER SAVE INTRINSIC DATA or COMMON statement except as a common block name The dummy arguments are replaced with the actual arguments during a function reference Dummy arguments that represent array names must be dimensioned by a DIMENSION or type statement Adjustable dimensions are permitted in function subprograms 92 PromulaFortran Compiler User s Manual The type of the function result is the type of the function name The arguments must agree in order number and type with the corresponding dummy arguments Function subprograms can be referenced in any procedure subprogram 5 29 GENERIC Statement The GENERIC statement identifies a name as representing a generic function Syntax GENERIC fun fun Where fun isa generic function name Notes The GENERIC statement performs the same action as the INTRINSIC statement Description The GENERIC statement is used to identify a name as representing a generic or intrinsic function This statement enables use of a generic function name as an actual argument As an example UBROUTINE DC ENERIC SQRT L SUBA X Y SQRT ROUTINE SUBA A B FNC AWNAADN D The name SORT is declared generic in subroutine and passed as an argument to subroutine SUBA Within SUBA the reference FNC A references the gen
276. m VS FORTRAN Note that in an integer context universal promotion to integer is followed Also in all cases the intermediate addition result is promoted to long But for some reason the multiplication result is allowed to overflow while the addition result is not 11 PromulaFortran Compiler User s Manual Using the Cl and Cs options you can select whether you want universal promotion to long or no automatic promotion to long There is no provision for selective promotions Note that selective and universal promotion differ only in overflow conditions so users from such environments should use the universal promotion to long convention The default convention is no universal promotion Cs Automatic promotion is selected via the Cl option 2 5 3 FORTRAN Source Format Used In the good old days the one thing that was always the same was the basic line format used to enter FORTRAN programs But all good things must end Now there are at least 5 different major variations of the FORTRAN entry format that we know of The F command line switch allows you to specify the entry format that you are using There is an extensive discussion of the different formats in the chapter on the FORTRAN language elements in this manual That discussion will not be repeated here The individual settings associated with this flag are mutually exclusive and are as follows FSnum Selects the standard fixed format with an ending column of n The default setting is
277. m argument is required CALL PROCESS CFUNC IVAL 234 PromulaFortran Compiler User s Manual 5 err234 for E248 The variable argument CFUNC has been entered where a subprogram argument is required 6 STOP 7 END Error 235 Message The entry identifier token is not a valid identifier symbol Example of message 235 If Line Nl Source 1 SUBROUTINE ERR235 A 2 10 3 ENTRY ERR235A A 3 err235 for E235 The entry identifier is not a valid identifier symbol 4 PRINT A 5 RETURN 6 END Error 236 Message The alternate return indicated by the symbol token may be used only in subroutines Example of message 236 If Line Nl Source 1 FUNCTION ERR236 1 err236 for E236 The alternate return indicated by the symbol may be used only in subroutines 2 ERR236 5 0 3 RETURN 1 4 END Error 237 Message The filename token within the parenthetical form of the include statement is not a valid identifier symbol Example of message 237 If Line Nl Source 1 PROGRAM ERR237 2 INCLUDE DDL237 2 err237 for E237 The filename DDL237 within the parenthetical form of the include statement is not a valid identifier symbol 3 STOP 4 END 235 PromulaFortran Compiler User s Manual Error 238 Message The include statement filename token is not a valid string constant or identifier symbol Example of message 238 If Line Nl Source ah PROGRAM ERR238
278. me of the associated IF tests In this structure the initial block IF statement and each ELSEIF or ELSE statement has an associated if block Only one if block in this structure is executed if no nested levels appear Each logical expression is evaluated until one is found that is true Control then transfers to the first statement of the associated if block When execution of the if block has completed and if control has not been transferred control transfers to the statement following ENDIF If none of the logical expressions are true and no ELSE statement appears no if blocks are executed control transfers to the statement following ENDIF In this structure at most one if block is executed If an ELSE statement appears it must follow the last ELSE IF statement If no logical expression is true control transfers to the statement following ELSE There are no restrictions about branching in or out of if blocks If a statement branches into an if block then execution continues from that point in exactly the same manner as if control had normally flowed to that statement via the logic of the block IF Any given if block may be empty i e contain no statements Finally block IF structures can be nested that is any if block within a structure can itself contain block IF structures Within a nesting hierarchy control can transfer from a lower level structure into a higher level structure however control cannot transfer from a higher leve
279. mension specification is terminated by the symbol token rather than a right parenthesis Example of message 202 If Line Nl Source 1 PROGRAM ERR202 2 DIMENSION A 10 10 1 0 2 err202 for E202 The dimension specification is terminated by the symbol rather than a right parenthesis 3 STOP 4 END Error 203 Message The value number is not within the valid statement or format number range Example of message 203 If Line Nl Source d PROGRAM ERR203 2 10 FORMAT 1X 110 3 READ 5 0 IA 3 err203 for E203 The value 0 is not within the valid statement or format number range 4 IF IA EQ 45 GOTO 12345678 4 err203 for E203 The value 12345678 is not within the valid statement or format number range 5 WRITE 6 0 IA 5 err203 for E203 The value 0 is not within the valid statement or format number range 6 12345 STOP 7 END Error 204 Message The object ident is of base type type and cannot be assigned a variable length string specification 224 PromulaFortran Compiler User s Manual Example of message 204 If Line Nl Source 1 PROGRAM ERR204 2 INTEGER I 4 J 2 K 2 err204 for E204 The variable K is of base type integer 4 and cannot be assigned a variable length string specification STOP 4 END Error 205 Message The element length specification for ident is closed by the symbol token rather than a parenthesis Example of message 205 If Line Nl Source 1 PROGRAM
280. ments decode information from character records which may either be contained in external logical units or internal files Unformatted read statements perform no decoding operations and may transfer information from external logical units only Formatted and unformatted read statements may operate with either sequential or direct access files while list directed and namelist read statements may operate only with sequential files In general the parameters on the various READ statements are as follows UNIT u FMT fn REC rn FMT n1 END s1 ERR s1 IOSTAT i0s ilist is an integer expression a character variable identifier or an asterisk If it is an integer expression then it refers to a specific external logical unit If it is a character variable then that variable is treated as an internal file of character records If it is an then the read is performed from the standard INPUT device If one of the simple forms of the READ is used which contain no parentheses and thus no unit number then these also use the standard input device is an integer variable the label of a FORMAT statement a character expression or an array identifier It specifies a format to be used for formatted input If n is a statement label then it identifies a FORMAT statement in the program unit containing the input statement which specifies how the input records are formatted If it is a character expression then that expression specifies how th
281. mplex See also ALOG10 CDLOG10 DLOG10 and LOG10 6 29 CLOG Natural Logarithm Synopsis COMPLEX FUNCTION CLOG COMPLEX Description CLOG a is a specific function that returns the natural logarithm of the argument The argument is complex and the result is complex See also ALOG CDLOG DLOG and LOG 145 PromulaFortran Compiler User s Manual 6 30 CMPLX Complex Value Synopsis COMPLEX FUNCTION CMPLX INTEGER 2 COMPLEX FUNCTION CMPLX INTEGER COMPLEX FUNCTION CMPLX REAL COMPLEX FUNCTION REAL 8 COMPLEX FUNCTION CMPLX COMPLEX COMPLEX FUNCTION COMPLEX 16 COMPLEX FUNCTION CMPLX REAL REAL COMPLEX FUNCTION CMPLX REAL 8 REAL 8 COMPLEX FUNCTION CMPLX REAL 8 REAL COMPLEX FUNCTION CMPLX REAL REAL 8 Description 1 2 is a generic function that performs type conversion and returns a complex value CMPLX can have one or two arguments A single argument can be integer real double precision complex or double complex If two arguments are used the arguments must be of the same type and must both be integer real or double precision For a single integer real or double precision argument the result is complex with the argumen
282. my argument array When an actual argument is a character substring the length of the actual argument is the length of the substring If the actual argument expression involves concatenation the sum of the lengths of the operands is the length of the actual argument 35 PromulaFortran Compiler User s Manual A variable in a dummy argument can be associated with a variable array element substring or expression in the actual argument A procedure can define or redefine the associated dummy argument if the actual argument is a variable name array element name or substring name The procedure cannot redefine the dummy argument if the actual argument is a constant a symbolic constant a function reference an expression using operators or an expression enclosed in parentheses The array declaration in a type COMMON or DIMENSION statement provides the information needed for the array during the execution of the program unit The actual argument array and the dummy argument array can differ in the number of the dimension and size of the array A dummy argument array can be associated with an actual argument that is an array array element or array element substring If the actual argument is a noncharacter array name the size of the actual argument array cannot be less than the size of the dummy argument array Each actual argument array element is associated with the dummy argument array element that has the corresponding subscript valu
283. n global symbols below for more discussion on this topic 2 8 5 Multiple Forms Another problem that comes up has to do with multiple forms Here a single function in FORTRAN might have to be compiled in different ways either because of some weak typing convention or because the function is to be used in both virtual and non virtual mode See the chapter on the PROMULA interface for a discussion of virtual mode As an example consider that you have a statistical analysis function which computes the mean and variance of a vector of values You have compiled it twice once using virtual conventions and once using memory conventions Let us first see how these two versions of the following utility can be produced SUBROUTINE ANADAT VAL N XBAR VAR DIMENSION VAL XBAR 0 0 VAR 0 0 DO 10 J 1 N XBAR XBAR VAL J 10 CONTINUE XBAR XBAR N 15 J 1 N S VAL J XBAR VAR VAR S S 15 CONTINUE VAR VAR N 1 RETURN END Compiling it with the following prototype produces a memory version anadat void manadat float short double double This version is like the FORTRAN original except that its external name is manadat and not Now compiling the identical FORTRAN code with the following prototype produces a virtual version anadat void vanadat float short double double This version is called vanadat and assumes that the vector to be analysed is on disk and is no
284. n be used freely elsewhere Complex constants can be broken across records between the real part and the comma and between the comma and the imaginary part More than one record can be used as input data in a namelist group The first column of each input record is ignored All input records containing data should end with a constant followed by a comma however the last record can be terminated by a without the final comma Constants be preceded by a repetition factor followed by an asterisk Omitting a constant constitutes a fatal error Constants can be integer real double precision complex logical exact representation or character Each constant must agree with the type of the corresponding input list item as follows 1 Character constants and exact representation constants may be associated with any input list item type A character constant is truncated from the right or extended on the right with blanks if necessary to yield a constant of the same length as the variable array element or substring 2 A logical or complex constant must be of the same type as the corresponding input list item 3 An integer real or double precision constant can be used for an integer real or double precision input list item The constant is converted to the type of the list item Logical constants have the following forms TRUE FALSE De Tece Feee If the third form is used any words starting with T o
285. n entity is 8 or if it is omitted then the type defined is complex If the value is 16 then the type is double complex See also See the discussion of the DIMENSION statement for a description of how dimension bounds are defined See the discussion of the DATA statement for a description of how initial values are defined 5 10 CONTINUE Statement The CONTINUE statement performs no operation Syntax CONTINUE Description The CONTINUE statement performs no operation It is an executable statement that can be placed anywhere in the executable statement portion of a source program without affecting the sequence of execution The CONTINUE statement is most frequently used as the last statement of a DO loop It can provide loop termination when a GOTO or IF would normally be the last statement of the loop 5 11 DATA Statement The DATA statement provides initial values for storage elements Syntax DATA nlist clist nlist clist Where nlist isa list of names to be initially defined Each name in the list can take the form Var 56 PromulaFortran Compiler User s Manual clist Description array element substr dolist var is a variable name array is an array name element is a subscripted array element name Subscript expressions must consist of integer constants and active control variables from DO lists substr is a substring of a character variable or array element dolist is animplied DO list
286. n is an integer expression nl is the identifier of a NAMELIST group ios is an integer lvalue sl is the label of an executable statement ilist is an input item list Notes The simplest form of the READ statement consists of a single format specification not enclosed in parentheses possibly followed by an input item list With this form no additional parameters can be specified If one of the parenthetical forms of READ is used then the UNIT and FMT specifications are optional however when omitted the unit specification must be the first specification and the n specification must be second Other than the above the order of the parameters within the parenthetical versions of the READ statement is free Description The READ statement transfers input data to internal storage from records contained in external logical units or internal files There are four basic types of READ statements 1 formatted read statements those which contain an fn parameter 118 PromulaFortran Compiler User s Manual 2 list directed read statements those which contain an asterisk where an fn parameter would otherwise appear 3 namelist read statements those which contain NAMELIST group identifier where an fn parameter would otherwise appear 4 unformatted read statements those which contain no fn parameter or asterisk or NAMELIST identifier in the fn position Formatted list directed and namelist read state
287. n is defined in the PARAMETER statement thus the first bound goes from 1 to 300 while the second goes from 0 to 5 Within the same program unit only one definition of an array is permitted Note that dimension information can be specified in COMMON statements and type statements The dimension information defines the array dimensions and the bounds for each dimension 61 PromulaFortran Compiler User s Manual 5 13 1 Adjustable Dimensions Adjustable dimensions enable creation of a more general subprogram that can accept varying sizes of array arguments For example a subroutine with a fixed array can be declared as SUBROUTINE SUM A DIMENSION A 10 The maximum array size subroutine SUM can accept is 10 elements If the same subroutine is to accept an array of any size it can be written as SUBROUTINE SUM A N DIMENSION A N Value N is passed as an actual argument Adjustable dimensions can be passed through common variables For example Ij SUBROUTINE SUB A COMMON B M N DIMENSION Dimension of array A in subroutine SUB is specified by the values M and N passed through the common block B Character strings and arrays can also be adjustable For example SUBROUTINE MESSAG X CHARACTER X PRINT X The subroutine declares x with a length of to accept strings of varying size Note that the length of the string is not passed explicitly as an actua
288. n is nonstandard Example of message 836 If Line Nl Source al SUBROUTINE W836 2 WRITE X 11HHello World 2 warn836 for N836 Omitting the repeat count before the X format specification is nonstandard 3 END Comment 837 Message The variable ident has already been dimensioned Example of message 837 If Line Nl Source 1 SUBROUTINE 11837 2 2 REAL 10 3 COMMON AB B 10 A 6 3 warn837 for C837 The variable A has already been dimensioned 4 REAL B 10 4 warn837 for C837 The variable B has already been dimensioned 5 Z 1 6 END Warning 838 Message The data initialization for ident will be ignored Example of message 838 If Line Nl Source al BLOCK DATA W838 2 PARAMETER N 3 3 COMMON E607A A 3 4 REAL 2 5 INTEGER IOTA 6 DATA A I I 1 N 1 2 3 7 DATA K 2 8 END 8 warn838 for W838 The data initialization for K will be ignored Note 839 Message The variable ident is assigned a statement label but is never used in an 274 PromulaFortran Compiler User s Manual assigned goto Example of message 839 If Line Nl Source al SUBROUTINE W839 2 ASSIGN 10 TO IA 3 GOTO 20 4 10 1 5 WRITE A 6 20 RETURN 7 END 7 warn839 for N839 The variable IA is assigned a statement label but is never used in an assigned goto Comment 840 Message In this character assignment the variable ident is self referential Example of message 840 If Li
289. n the UNIT specification is optional however when omitted the unit specification must be the first specification Other than the above the order of the parameters within the parenthetical version of the REWIND statement is free Description The REWIND statement positions a file at beginning of information so that the next input output operation references the first record in the file If the file is already at beginning of information no action is taken If an error occurs as a result of the rewind and if neither err nor status are supplied execution will terminate with an error code set If either or both are supplied the execution continues despite any errors If supplied the status lvalue receives the runtime error code for the error condition encountered or a zero if no error occurred If err is supplied then execution will branch to the statement err if an error occurs 5 50 SAVE Statement 124 PromulaFortran Compiler User s Manual The SAVE statement is used to retain the definition status of entities within a subprogram after that subprogram has executed Syntax SAVE a al Where a is a variable name array name or common block name enclosed in slashes Description The SAVE statement is used to retain the definition status of entities after the execution of a RETURN or END statement in a subprogram A SAVE statement in a main program is optional and has no effect Dummy argument names proced
290. name array element name or array element substring name If the actual argument begins at character 36 PromulaFortran Compiler User s Manual storage position acu of an array then the first character storage position of the dummy argument array becomes associated with character storage position acu of the actual argument array and so forth to the end of the dummy argument array A dummy argument that is a dummy procedure can be associated only with an actual argument that is an intrinsic function external function subroutine or another dummy procedure If the dummy argument is used as an external function the actual argument that is passed must be a function or dummy procedure The type of the dummy argument must agree with the type of result of all specific actual arguments that become associated with the dummy argument When a dummy argument is used as an external function and is the name of an intrinsic function the intrinsic function name corresponding to the dummy argument name is not available If the dummy argument is referenced as a subroutine the actual argument must be the name of a subroutine or dummy procedure and the dummy argument must not appear in a type statement or be referenced as a function A dummy argument that is an asterisk can only appear in the argument list of a SUBROUTINE or ENTRY statement in a subroutine subprogram The actual argument is an alternate return specifier in the CALL statement 3 8 2 Adjusta
291. namelist group name is used the list must be omitted from the input output statement A variable or array name can belong to one or more namelist groups Assumed size arrays cannot appear in a namelist group Data read by a single namelist name READ statement must contain only names listed in the referenced namelist group All items in the namelist group or any subset of the group can be input Values are unchanged for items not input Variables need not be in the order in which they appear in the defining NAMELIST statement The program in the following example PROGRAM NMLIST NAMELIST SHIP 11 12 READ SHIP END 10 IF C GT 0 0 THEN A B C 1 11 12 WRITE SHIP ENDIF STOP 10 PRINT NO DATA FOUND STOP END can read the following input record SSHIP A 14 7 B 12 3 C 3 4 11 58 12 8SEND and if the above were read would produce the following output SSHIP A 157E 02 B 123E 02 34E 01 Il 66 I2 8 SEND 107 PromulaFortran Compiler User s Manual The following sections discuss namelist input and output and arrays in namelist The syntax descriptions are simplified to focus the discussion on the namelist See the sections on READ WRITE PRINT and PUNCH for a general discussion of these statements 5 38 1 NAMELIST READ Statement The namelist READ statement reads input data from a designated file Its simplified syntax i
292. nation and scaling for numeric editing The nonrepeatable edit descriptors cannot be repeated The nonrepeatable edit descriptors are single quote double quote BN BZ colon slash nH xP S SP SS Tn TL TR and dollar sign 5 27 7 A Descriptor The A descriptor is used with an input output list item of type character or noncharacter The following paragraphs discuss the A descriptor for input output list items of type character and noncharacter The form of the A descriptor for character list items is A or Aw On input if w is less than the length of the list item the input quantity is stored left justified in the item the remainder of the item is filled with blanks If w is greater than the length of the item the rightmost characters are stored and the remaining characters are ignored If w is omitted the length of the field is equal to the length of the list item This example shows a character list item 78 PromulaFortran Compiler User s Manual CHARACTER A 9 READ 5 10 A 10 FORMAT A7 Input record EXAMPLE In location A EXAMPLE The second example shows truncation when input is wider than the list item CHARACTER B 10 READ T 20 B 20 FORMAT A13 Input record 1 1 3 SPECIFICATION In location B 1 10 CIFICATION The following example shows that if no length is specified for an A edit descriptor then the
293. nce bar and press ENTER Copyright 1988 1991 PROMULA Development Corporation ALL RIGHTS RESERVED 1 Application Management System Figure 8 5 PROMULA Main Menu We wish to work with an existing program so press the F6 key which results in the dialog shown below Enter the filename of the program to be executed expo xeq Upon completion of this request we are returned to the Main Menu again as shown in Figure 8 5 Here we press the F10 key to work with PROMULA directly We will not do extensive work here only enough to demonstrate that the values from the program are indeed now fully exposed in a highly usable form Initially we can use the BROWSE VARIABLE statement shown below to obtain a browsable listing of the variables available This listing is shown in Figure 8 6 Note that the variable descriptions are all obtained from our initial definitions IH PROMULA BROWSE VARIABLI 186 AVGE Average Exposure M T AVGI Average Intake M T CONC Ambient Concentration M L 3 PromulaFortran Compiler User s Manual Exposure M CUME Cumutative E CUMI Cumulative Intake M EXPOS Exposure M T Rn Parind Press any key to continue Figure 8 6 BROWSE VARIABLE Listing Next we can look at a particular variable in the list say expos To do this we enter the command shown below The result is shown in Figure 8 7 Note that the values match those in the o
294. nd any value separator The next value begins immediately after a repeated null A null cannot be used for either the real or imaginary part of a complex constant however a null can represent an entire complex constant 121 PromulaFortran Compiler User s Manual When the value separator is a slash remaining list elements are treated as nulls and the remainder of the current record is discarded Input values must correspond in type to variables in the input output list Note that the form of a real value can be the same as that of an integer value See also The discussion of the OPEN statement describes the FORTRAN file system in general including the different file types The discussion of the FORMAT statement describes how format information is interpreted The NAMELIST statement describes the namelist read statement 5 46 REAL Statement The REAL statement defines some user defined entity to be of type real or double precision Syntax REAL len name name Where name has one of the forms var len c array d d len clist len specifies the real subtype and can be an unsigned nonzero integer constant whose value is 4 or 8 var is a variable function name symbolic constant or dummy procedure array isan array name d specifies the bounds of a dimension clist isa list of constants or symbolic constants specifying the initial values Each item in the list can take the form c is
295. ne Nl Source 1 SUBROUTINE W840 2 CHARACTER 40 A B 3 A ABCDEFGHIJKLMNOPORSTUVWXYZ01234567894 4 A 12 40 A 4 warn840 for C840 In this character assignment the variable A is self referential 5 B NOT USED 6 A 0123456789 A 6 warn840 for C840 In this character assignment the variable A is self referential 7 WRITE A 8 END Comment 841 Message The switch value for this computed goto is a constant Example of message 841 If Line Nl Source 1 SUBROUTINE W841 2 PARAMETER J 2 3 GOTO 1051 2 3 warn841 for C841 The switch value for this computed goto is a constant 4 GOTO 10 1 2 5 GOTO 10 1 2 3 5 warn841 for C841 The switch value for this computed goto is a constant 6 1 CONTINUE 7 2 CONTINUE 8 10 RETURN 9 END 275 PromulaFortran Compiler User s Manual Note 842 Message The variable ident is assigned a format label but is never used as an assigned format Example of message 842 If Line Nl Source T SUBROUTINE W842 2 ASSIGN 10 TO IA 3 GOTO IA 10 20 4 10 1 5 ASSIGN 1 TO 6 WRITE 1 A 7 1 FORMAT 1X F10 0 8 RETURN 9 END 9 warn842 for N842 The variable IA is assigned a format label but is never used as an assigned format Note 843 Message It is inefficient to have a statement label on an unconditional goto Example of message 843 If Line Nl Source 1 SUBROUTINE W843 2 1 3 IF I EQ 0 GO TO 20 4 20 IF I NE 0 GO
296. ned we can now run the program and get the same results As before the program reads the input data from an input text file does some calculations and writes the output report on an output text file EXPO DAT expo EXPO RPT The run time may have gone up a little because the program now reads from and writes to the PROMULA database EXPO DBA To use the EXPO program with PROMULA we execute it in precisely the same manner as before except we must tell the virtual memory system that it is to use the file EXPO DBA as the actual disk file To do this we enter the following at the operating system level 185 PromulaFortran Compiler User s Manual expo Vexpo dba The V option tells the program to use the EXPO DBA datafile After the run is complete a report file EXPO RPT is created and the file EXPO DBA now contains the variable values as they were computed within the program To verify this we will use PROMULA again We will take you through the individual steps First enter PROMULA at the operating system level to get the PROMULA Main Menu as shown in Figure 8 5 below PROMULA V3 00 09 01 91 IBM PC Version Main Menu Function Exit PROMULA Restart PROMULA Run the PROMULA Tutorial Edit a source file Compile a source program Run a program from the console Resume an interrupted program Run a program from a disk file Run a menu of applications Use the PROMULA Language Press desired key or move bou
297. nent if they are positive but the minus sign must be present with either or both if either or both are negative Its syntax is as follows Where n is an unsigned integer constant coeff is a coefficient in one of the following forms Nes n exp is an unsigned integer constant The range of the exponent is 307 to 308 The precision for a real constant is approximately 7 significant digits Optionally a real constant can be followed by a decimal exponent written as the letter E and an integer constant that indicates the power of ten by which the number is to be multiplied If the E is present the integer constant following the letter E must not be omitted The plus sign can be omitted if the exponent is positive but the minus sign must be present if the exponent is negative Examples of valid real constants 23 PromulaFortran Compiler User s Manual 7 5 3 22 4000 Examples of invalid real constants 33 500 Comma not allowed 2 5 Letter not allowed Examples of valid real constants with exponents 42 1 Value 42 10 420 0 00028E 5 Value 00028 x 10 28 0 6 205E6 Value 6 205 x 106 6205000 0 700 E 2 Value 700 x 102 7 0 3 3 3 Double Precision Constant A double precision constant is written in the same way as a real constant with exponent except that the exponent is prefixed by the letter D instead of E It consists of a string of decimal digits coefficient written with an
298. nks Generally w must be at least d 6 or d e 4 for negative numbers and w must be at least d 5 d e 3 for positive numbers Positive numbers need not reserve a space for the sign of the number unless an SP specification is in effect If the field is not wide enough to contain the output value asterisks are inserted throughout the field If the field is longer than the output value the quantity is right justified with blanks on the left If the value being converted is indefinite an I is printed in the field if it is out of range an R is printed The Ew d specification produces output in the following formats S d aE ee for values where the magnitude of the exponent is less than 100 and S a a eee for values where the magnitude of the exponent exceeds 100 Where 83 PromulaFortran Compiler User s Manual 5 is a minus sign if the number is negative and omitted if the number is positive a a the most significant digits of the value correctly rounded ee e digits in the exponent When the specification Ew dEe is used the exponent is preceded by E and the number of digits used for the exponent field not counting the letter and sign is determined by e If e is specified too small for the value being output the entire field width as specified by w will be filled with asterisks 5 27 13 End of Record Slash The slash indicates the end of a record anywhere in the FORMAT specifi
299. not an undimensioned integer 4 ASSIGN 20 TO R 4 errl33 for E133 The variable R in the assign statement is not an undimensioned integer 5 R 10 0 6 GOTO IJ 10 20 30 6 errl33 for E133 The variable IJ in the goto statement is not an u 3 err ndimensioned integer E R R he variable R in the goto statement is not an undimensioned integer 9 20 WRITE IJ 0 GOTO IJ 3 10 WR 8 GOT 8 errl33 tobr EL33 201 PromulaFortran Compiler User s Manual 10 err133 for E133 The variable IJ in the goto statement is not an undimensioned integer 11 30 GOTO R 10 20 11 err133 for E133 The variable in the goto statement is not an undimensioned integer 12 END Error 134 Message The symbol token within the data value list is not a constant Example of message 134 If Line Nl Source 1 PROGRAM ERR134 2 DIMENSION A 5 3 DATA A 1 0 2 0 BETA 4 0 5 0 3 errl34 for E134 The symbol BETA within the data value list is not a constant 4 WRITE A 5 STOP 6 END Error 135 Message The sign character has been applied to a non numeric constant value token Example of message 135 If Line Nl Source 1 PROGRAM ERR135 2 INTEGER A 3 3 DATA A 77 x ff x l x 3 errl35 for E135 The sign character has been applied to a non numeric constant value 1 4 WRITE A 5 STOP 6 END Error 136 Message The first member of the complex val
300. nput item list For example READ 5 100 A I I 1 3 has the same effect as the statement READ 5 100 A 1 A 2 A 3 120 PromulaFortran Compiler User s Manual A variable cannot be used as a control variable more than once in the same implied DO nest but ilist items can appear more than once The value of a control variable within an implied DO specification is defined within that specification On exit from the implied DO specification the control variable retains the first value to exceed the upper limit 2 The implied DO can be nested that is the ilist in an implied DO can itself contain an implied DO The first innermost control variable varies most rapidly and the last outermost control variable varies least rapidly For example a nested implied DO with two levels has the form list vl el e2 e3 v2 eel ee2 ee3 Nested implied DO loops are executed in the same manner as nested DO statements The nested form can be used to read values into arrays Each execution of a READ statement processes at least one record The formatted READ statement transmits data to storage locations named in ilist according to FORMAT specification fn Once a READ is initiated the FORMAT statement determines when a new record will be transmitted The unformatted READ statement transmits one record from the specified unit u to the storage locations named in ilist Records are not converted no FORMAT statement is used The info
301. nsional array of four rows and three columns for a total of twelve elements as shown below with the values being the element locations relative to the location of the first element Column 1 Column 2 Column 3 Row 1 0 4 8 Row 2 1 5 9 Row 3 2 6 10 Row 4 3 7 11 Finally INTEGER STOR 3 4 2 declares a 3 dimensional array of three rows four columns and two pages or planes for a total of 24 elements It can be viewed as follows again with the values being the element locations relative to the location of the first element Plane 1 Column 1 Column 2 Column 3 Column 4 Row 1 0 3 6 9 Row 2 1 4 7 10 Row 3 2 5 8 11 Plane 2 Column 1 Column 2 Column 3 Column 4 Row 1 12 15 18 21 Row 2 13 16 19 22 Row 3 14 17 20 23 3 5 1 Array Storage The elements of an array have a specific storage order with elements of any array stored as a linear sequence of storage locations The first element of the array begins with the first storage location or character storage position and the last element ends with the last storage location or character storage position The number of storage words reserved for an array is determined by the type of the array and its size Integer real and logical arrays all occupy the same amount of storage referred to as a storage word Thus for these types the number of storage words in an array equals the array size For complex and double precision arrays the number of storage words reserved is twice the
302. nt Example of message 159 If Line Nl Source 1 PROGRAM ERR159 2 PARAMETER IVAL N N 3 DIMENSION A 100 4 DATA A I I 1 IVAL 100 1 2 4 errl59 for E159 The parameter IVAL being used to control the data range is not an integer constant 5 STOP 6 END Error 160 Message The do statement minimum value is followed by the symbol token rather than a comma Example of message 160 If Line Nl Source 1 PROGRAM ERR160 2 INTEGER 4 I 3 DO 10 I 1 TOS5 3 errl60 for E160 The do statement minimum value is followed by the symbol TO5 rather than a comma PRINT I 10 CONTINUE STOP END YAU BS Error 161 Message The do statement maximum value is followed by the symbol token rather than a comma 210 PromulaFortran Compiler User s Manual Example of message 161 If Line Nl Source 1 PROGRAM ERR161 2 INTEGER 4 I 3 10 1 5 BY 1 3 errl6l for E161 The do statement maximum value is followed by the symbol BY1 rather than a comma PRINT I 10 CONTINUE STOP END d Error 162 Message The object ident may not be dimensioned Example of message 162 If Line Nl Source 1 SUBROUTINE ERR162 2 PARAMETER ICON 3 CON 3 3 DIMENSION ICON 3 3 errl62 for E162 The parameter ICON may not be dimensioned 4 WRITE A Z 5 END Error 163 Message The computed goto is opened by the symbol token rather than a left parenthesis Example of message
303. nt A Hollerith constant is an unsigned integer constant followed by the letter followed by the specified number of characters Its syntax is as follows 26 PromulaFortran Compiler User s Manual nHs Where n isan unsigned nonzero integer constant s isa string of exactly n characters The minimum number of characters in a Hollerith constant is one and the maximum number of characters in a Hollerith constant is 32755 The length is the number of characters in the constant and must always be equal to the specified value of n Blanks are significant in a Hollerith constant Any characters in the platform system character set can be used Character positions in a character constant are numbered consecutively as 1 2 3 and so forth up to the length of the constant The length of the Hollerith constant is significant in all operations in which the constant is used The length must be greater than zero Examples of valid Hollerith constants 3HABC 3H123 6HYEAR S Examples of invalid Hollerith constants HABC Missing count OH Zero length 3 3 9 Exact Representation Constants Exact representation constants specify an exact sequence of bits using either octal or hexadecimal notation This use of these constants is highly nonportable They are alternative ways to represent numeric values and may not be assigned to character variables Exact representation constants have the following syntax tet OX Where s is a sequen
304. nt references a specific element and takes the form shown below array e e Where array is the symbolic name of the array e is a subscript expression that is an integer real or double precision Each subscript expression has a value that is within the bounds of the corresponding dimension Non integer subscripts are converted to integer by truncation prior to their use An array element reference must specify a value for each dimension in the array Array element references are not legal unless a value is supplied for each dimension Each subscript value after conversion to integer must not be less than the lower bound or greater than the upper bound of the dimension If the array is an assumed size array with the upper bound of the last dimension specified as asterisk the value of the subscript expression must not exceed the actual size of the dimension The results are unpredictable if an array element reference exceeds the size of an array For each array element reference evaluation of the subscript expressions yields a value for each dimension and a position relative to the beginning of the complete array The position of an array element is calculated as shown below for an array with from to 7 dimensions The position indicates the storage location of an array element relative to the first Dimensions Position of Array Element 1 81 11 2 51 31 52 32 1 3 51 31 52 32 1 53 33 2
305. nventions and differing internal representations There is no general solution to the compilation of weak typing FORTRAN programs on multiple platforms Such programs may work on one platform but not on another The compiler must be told what to do Some device is needed to describe subprogram arguments A similar problem has faced C programmers as well Consequently the new ANSI C has introduced the notion of a function prototype which describes the arguments of functions in terms of their binary type and in terms of their pointer status The conventions developed there are exactly those needed by the FORTRAN compiler as well although they need to be extended slightly to deal with virtual variables multiple forms and external name clash The C prototype system and its extensions to FORTRAN are discussed in this section It is strongly recommended that you take the time to develop a set of prototypes for the subprograms within a FORTRAN program once you have operationalized it to obtain optimal runtime efficiency It should be pointed out that the storage of prototype information read from a configuration file has been carefully optimized which means that there are minimal performance penalties for using large prototype files 2 8 1 Formation of External Symbols Before moving into the discussion of the prototype syntax a word about how the compiler forms its own external symbols is needed Each COMMON block name and each subprogram name a
306. nverts the argument to its real floating point equivalent See also REAL and SNGL 6 68 GETCL Get Command Line Synopsis SUBROUTINE GETCL CHARACTER Description GETCL returns any command line string entered along with the request to execute the program If that string is shorter than the character length of a then it is truncated If the command line string is shorter than the character length of a then a is padded with blanks See also None 6 69 IABS Absolute Value Synopsis INTEGER FUNCTION IABS INTEGER INTEGER 2 FUNCTION IABS INTEGER 2 Description is a generic function that returns an absolute value of its argument The result is integer or short integer depending on the argument type It is a if a is greater than or equal to zero else itis See Also ABS DABS CABS CDABS and I2ABS 6 70 IAND Bitwise And Synopsis 158 PromulaFortran Compiler User s Manual INTEGER 2 FUNCTION IAND INTEGER 2 INTEGER FUNCTION IAND INTEGER Description IAND al a2 an is a generic function which accepts short integers or integers The return value is the successive bitwise AND of its arguments and is the same type as the first parameter See also None 6 71 ICHAR Integer Value of Character Synopsis INTEGER 2 FUNCTION IC
307. nvironments There is a real semantics issue here the same program compiled in different environments behaves differently even though these environments have the same word size Consider the following FORTRAN program which does short integer additions multiplications and divisions in a variety of contexts PROGRAM VARI2 INTEGER 2 I1 12 13 14 INTEGER 2 ISUM IPROD IQUOT INTEGER 4 JSUM JPROD JQUOT REAL 4 RSUM RPROD RQUOT REAL 8 DSUM DPROD DQUOT 1 20000 I2 30000 I3 200 10 PromulaFortran Compiler User s Manual ISUM 1 12 lt Note addition overflow IPROD 13 14 lt Note multiplication overflow IQUOT 11 I2 14 lt Note intermediate overflow JSUM 1 12 JPROD 13 14 JQUOT I1 I2 14 RSU 11 I2 RPROD 13 14 RQUOT 1 12 14 DSU 11 I2 DPROD 13 14 DQUOT 1 12 14 WRITE 24H Short Integer Results 3113 2E ISUM IPROD IQUOT WRITE 24H Long Integer Results JSUM JPROD JQUOT WRITE 24H Short Real Results 3 13 5 RSUM RPROD RQUOT H D WRITE 24H Long Real Results 3 13 5 DSUM DPROD DQUOT STOP END In running this example with various FORTRAN compilers always on machines with 16 bit short integers VAX IBM mainframe IBM PC we have obtained the following three results 1 Universal prom
308. o width or free form format value editing specifications is nonstandard 3 30 FORMAT I0 3 warn889 for N889 The use of zero width or free form format value editing Specifications is nonstandard 4 40 FORMAT L0 4 warn889 for N889 The use of zero width or free form format value editing Specifications is nonstandard 60 FORMAT 00 0 5 warn889 for 889 The use of zero width or free form format value editing specifications is nonstandard 6 70 FORMAT E0 0 6 warn889 for N889 The use of zero width or free form format value editing Specifications is nonstandard 7 80 FORMAT 60 0 7 warn889 for N889 The use of zero width or free form format value editing Specifications is nonstandard 8 WRITE 10 9 WRITE 30 10 WRITE 40 11 WRITE 60 12 WRITE 70 13 WRITE 80 14 END Comment 890 Message The zero statement label is ignored Example of message 890 If Line Nl Source 1 SUBROUTINE W890 2 warn890 for C890 The zero statement label ignored 2 0000 REAL A 3 warn890 for C890 The zero statement label ignored 3 0000 1 4 WRITE 5 END Comment 891 Message 293 PromulaFortran Compiler User s Manual The statement function ident has been previously declared as being external Example of message 891 If Line Nl Source 1 SUBROUTINE W891 I 2 EXTERNAL JJ 3 JJ K 3 K I KK 3 warn891 for C891 The statement function JJ has been p
309. oduces output in the following format S ateee for values where the magnitude of the exponent is greater than or equal to 100 and s aDtee for values where the magnitude of the exponent is less than 100 where 5 Minus sign if the number is negative or blank if the number is positive a One or more most significant digits ee e Digits in the exponent 5 27 12 E Descriptor 81 PromulaFortran Compiler User s Manual The E descriptor specifies conversion between an internal real or double precision value and an external number written with an exponent This descriptor has the forms w d w d On input the width w includes plus or minus signs digits decimal point E and exponent If an external decimal point is not provided d acts as a negative power of 10 scaling factor The internal representation of the input quantity is integer subfield X 10 4 x 10 exponent subfield For example if the specification is E10 8 the input quantity 3267E 05 is converted and stored as 3267 X 107 x 10 3 267 If an external decimal point is provided it overrides d e if specified has no effect on input An input field consisting entirely of blanks is interpreted as zero The following diagram illustrates the structure of the E input field It shows the characters allowed to start a subfield digit E or D integer fraction exponent subfield subfield The integer subfield begins with a or sign
310. ograms can share their information with users and other programs 9 ERROR MESSAGES The error messages for the FORTRAN compiler are divided into three groups 1 Control program errors 2 FORTRAN preprocessor errors 3 Runtime errors 9 1 Control Program Errors The control program interprets the command line as entered by the user and then directs the operations of the FORTRAN preprocessor the C compiler and the linker as requested In particular the control program works as follows First it checks to make certain that all of its support files are present in their expected locations If it is unable to locate some file then it issues the following error message and exits PROMULA FORTRAN Compiler installation error If you get this error either something was done incorrectly during the installation of the compiler or the file system on your computer has been damaged Please refer to the platform specific installation instructions which are separate from this manual for further instructions Second the control program processes all entries on its command line to determine the type of tasks to be performed and the resources required to perform them If there is any entry on the command line which it cannot interpret it exits with the following message Bad argument The bad argument message is followed by the particular command line argument which the control program was unable to interpret Refer to the chapter on using the
311. olic name of such a constant Description The DIMENSION statement defines symbolic names as array names and specifies the bounds of each array More than one array can be declared in a single DIMENSION statement Dummy argument arrays specified within a procedure subprogram can have adjustable dimension specifications A further explanation of adjustable dimension specifications appears below The parameter upper is the upper bound of the dimension and 1ower is the lower bound of the dimension If only the upper bound is specified the value of the lower bound is one The earlier discussion of arrays in Chapter 2 presented the exact form of a dimension bound expression The following shows the relation between the DIMENSION statement and a type declaration statement REAL NIL DIMENSION NIL 6 2 2 is equivalent to REAL NIL 6 2 2 and both are equivalent to DIMENSION NIL 6 2 2 REAL NIL In the following example CHARACTER 8 XR DIMENSION XR 0 4 the array XR contains 5 character elements with each element having a length of 8 characters A reference to the third and fourth characters of the second element would be XR 1 3 4 The following example shows an expression in a DIMENSION bound specification PARAMETER N 100 DIMENSION ARR 1 N 3 0 5 Array ARR is a two dimensional array that contains 1800 elements The value of N in the dimension bound expression for the first dimensio
312. ompiler User s Manual 3 DATA PI 3 141592 4 PI 3 14 4 warn878 for C878 The assignment statement is ignored in block data subprograms 5 CALL E159A 5 warn878 for C878 The call statement is ignored in block data subprograms 6 END Note 879 Message The conditional expression was of type type rather than logical Example of message 879 If Line Nl Source 1 SUBROUTINE W879 2 IF A B GO 10 2 warn879 for N879 The conditional expression was of type real 4 rather than logical 3 4 2 4 10 IF A LE 10 B 0 5 END Note 880 Message The use of redundant commas to indicate missing arguments is nonstandard Example of message 880 If Line Nl Source 1 SUBROUTINE W880 A B 2 A W880B A 2 warn880 for N880 The use of redundant commas to indicate missing arguments is nonstandard 3 WRITE 4 END Note 881 Message The type control value for the computed goto has been truncated to integer Example of message 881 If Line Nl Source I SUBROUTINE W881 I J 2 REAL I 3 DOUBLE PRECISION J 4 GOTO 10 20 30 4 warn881 for 881 real 4 control value for the computed goto has been truncated to integer 289 PromulaFortran Compiler User s Manual 5 10 I 7 8 6 RETURN 7 20 GO 0 10 30 2 7 warn881 for 881 The real 8 control value for the computed goto has been truncated to integer 8 RETURN 9 30 I207 8 10 RETURN T3 END Comment 882
313. or a real constant is approximately 7 significant digits 3 4 5 Double Precision Variable A double precision variable is a variable that is typed explicitly as double precision The value of a double precision variable can range from 10 307 to 10308 with approximately 17 significant digits of precision 3 4 6 Complex Variable A complex variable is a variable that is typed explicitly as complex A complex variable may be thought of as a sequence of two real values with the first being the real part of the complex number and the second being the imaginary part of the number 3 4 7 Double Complex Variable A double complex variable is a variable that is typed explicitly as double complex A double complex variable may be thought of as a sequence of two double precision values with the first being the real part of the double complex number and the second being the imaginary part of the number 3 4 8 Logical Variable A logical variable is a variable that is typed explicitly as logical It may contain only a value of TRUE or FALSE The storage allocated to a logical variable is the same as that allocated to an integer variable 3 4 9 Short Logical Variable A short logical variable is a variable that is typed explicitly as short logical It may contain only a value of TRUE or FALSE The storage allocated to a short logical variable is the same as that allocated to a short integer variable 3 4 10 Logical Byte Variable A logical byte variabl
314. ortran Compiler User s Manual 5 STATEMENTS SUPPORTED This chapter describes the statements of FORTRAN other than assignment statements Each statement is identified by the keyword which introduces it The statement descriptions themselves are organized alphabetically to simplify using this manual as a reference guide 5 1 ASSIGN Statement The ASSIGN statement assigns a statement label to an integer variable Syntax ASSIGN label TO identifier Where label is the label of an executable statement or a FORMAT statement identifier is the identifier of a scalar integer variable Description The ASSIGN statement assigns a statement label to an integer variable The value assigned represents the label of an executable statement or a FORMAT statement It is not the value of the label itself The labeled statement must appear in the same program unit as the ASSIGN statement While the variable contains an assigned value it cannot be used in any statement other than an assigned GOTO statement or in the FORMAT position of an input output statement The ASSIGN statement is an executable statement therefore a variable must be ASSIGNed a label to execution of the assigned GOTO statement or the input output statement that references the assigned label Examples The following assigns the label of an executable statement to a variable for later use in an ASSIGNed GOTO statement ASSIGN 10 TO iswit GOTO iswit 5 10 15 20 10 STOP Execution
315. otion to long Short Integer Results 15536 5536 166 Long Integer Results 50000 60000 166 Short Real Results 50000 00000 60000 00000 166 00000 Long Real Results 50000 00000 60000 00000 166 00000 2 No automatic promotion to long Short Integer Results 15536 5536 o Long Integer Results 15536 5536 51 Short Real Results 15536 00000 5536 00000 51 00000 Long Real Results 15536 00000 5536 00000 51 00000 3 Selective promotion to long Short Integer Results 15536 5536 166 Long Integer Results 50000 60000 166 Short Real Results 50000 00000 5536 00000 166 00000 Long Real Results 50000 00000 5536 00000 166 00000 In reviewing these results a negative number means that short integer overflow has occurred The typical FORTRAN result is the first one In this instance the output of all integer calculations is a long That result is then converted to the desired result type Notice that even the intermediate addition in the division example is calculated as a long In the second case the result of any short integer calculation is always also short regardless of the surrounding context This type of result is unusual for mainframe FORTRANs and is common for PC FORTRANS Note that Microsoft FORTRAN allows the user to select which type of convention is to be followed as a side effect of the WORDSIZE metacommand The third case is strange and difficult to deal with The particular result above can be gotten fro
316. ould be encouraged Our design makes this possible 1 3 Notation Used The notation used to describe FORTRAN statements throughout this manual is as follows UPPERCASE indicates a statement keyword or character that is to be written as shown lowercase indicates a name number symbol or entity that is to be supplied by the programmer indicate an optional item that can be used or omitted indicate that only one of the enclosed items can be used indicates that the preceding optional item in brackets can be repeated as necessary In program examples a vertical ellipsis indicates that other FORTRAN statements or parts of the program have not been shown because they are not relevant to the example Finally the symbol indicates a blank character in cases where such is not obvious PromulaFortran Compiler User s Manual 2 USING THE COMPILER PF is the program that controls the compilation of FORTRAN programs It guides files of source and object code through each phase of compilation and linking PF has many options to assist in the compilation of FORTRAN programs in essence however all you need do to produce an executable file from your FORTRAN program is to type PF followed by the name of the file or files that hold your program PF checks whether the file names you give are reasonable selects the right phase for each file and performs other tasks that ease the compilation debugging and analysis of your programs T
317. pe statement or format and a listing of the lines where they are referenced If the subprogram contains EQUIVALENCE statements then a fifth table type is generated the equivalence pairs table Given the code fragment below 48 SUBROUTINE ANA1 49 INTEGER BUF1 2048 BUF2 2048 BUF3 2048 50 BYTE OBUF 32767 51 EQUIVALENCE BUF2 1 OBUF 1 52 EQUIVALENCE BUF3 OBUF 62 END The following equivalence pairs table is produced Equivalence pairs Dependent Base Offset BUF 2 OBUF 0 BUF3 OBUF 0 PromulaFortran Compiler User s Manual The base variable is the variable whose storage is being used to contain the dependent variable The offset is the byte offset within the base variable of the start of the dependent variable 2 4 5 Controlling Page Size The PH and PW flags are used in conjunction with the ES and EX flags to specify the desired output page height and width The default setting for PH is 80 lines and that for PW is 132 characters The reports produced are quite large and the minimum setting allowed for PW is 80 characters Thus a command like the following pf 56 pw120 S x demo for would compile a FORTRAN program demo for and would produce tables 120 characters wide with 56 lines per page Note that the symbol table widths are adjusted to fit into the specified page size however the source code listing is truncated if the source line
318. pe specification Example of message 175 If Line Nl Source 1 PROGRAM ERR175 2 INTEGER TEST 3 PRINT TEST 1 0 2 0 4 STOP 5 END 6 INTEGER SUBROUTINE TEST A B 6 errl75 for E175 A subroutine or program may not have type Specification INTEGER A B 8 TEST B 9 RETURN 10 END Error 176 Message The program or subprogram identifier token is not well formed Example of message 176 If Line Nl Source 215 PromulaFortran Compiler User s Manual 1 PROGRAM 176 errl76 for E176 The program or subprogram identifier 176 is not well formed 2 PRINT Hello World 3 STOP 4 END Error 177 Message The program statement closing parenthesis is missing Example of message 177 If Line Nl Source 1 PROGRAM ERR177 INPUT OUTPUT 1 errl77 for E177 The program statement closing parenthesis is missing 2 PRINT Hello World 3 STOP 4 END Error 178 Message Entry points are allowed in subprograms only Example of message 178 If Line Nl Source 1 PROGRAM ERR178 2 ENTRY SAM 2 errl78 for E178 Entry points are allowed in subprograms only 3 WRITE HI 4 END Error 179 Message A statement may not goto itself Example of message 179 If Line Nl Source 1 SUBROUTINE ERR179 2 1 3 10 GOTO 20 30 10 1 3 errl79 for E179 statement may not goto itself 4 20 T I 1 5 GOTO 10 6 30 STOP 7 END 216 PromulaFortran Compiler User s Manual Erro
319. play code characters On input data is converted from a specified format to its internal binary representation On output data is converted from its internal binary representation to the specified format before it is transmitted Formats can be specified by 1 the statement label of a FORMAT statement 2 an integer variable which has been assigned the statement label of FORMAT statement see ASSIGN statement 74 PromulaFortran Compiler User s Manual 3 a character array name or any character expression except one involving assumed length character entities 4 anoncharacter array name The FORMAT statement is used in conjunction with formatted input and output statements It can appear anywhere in the program after the PROGRAM FUNCTION or SUBROUTINE statement An example of a FORMAT statement and its associated READ statement is as follows READ 5 10 INK NAME AREA 10 FORMAT 10X 14 12 F7 2 The format specification consists of edit descriptors in parentheses Blanks are not significant except in H quote and apostrophe descriptors Generally each item in an input output list is associated with a corresponding edit descriptor in a FORMAT statement The FORMAT statement specifies the external format of the data and the type of conversion to be used Complex variables always correspond to two edit descriptors Double precision variables correspond to one edit descriptor when using D E F or G The D
320. ptional however when omitted the unit specification must be the first specification Other than the above the order of the parameters within the parenthetical version of the ENDFILE statement is free Description The ENDFILE statement writes end of file to the designated unit Depending upon the file type this end of file may or may not be a physical record ENDFILE is not permitted on units opened for direct access The end of file can be later detected by the END and IOSTAT parameters on other input statements See also The discussion of the OPEN statement describes the FORTRAN file system in general including the different file types The discussion of the READ statement describes the END parameter for detecting the end of file condition 5 23 END IF Statement The END IF statement terminates a block IF construct Syntax END IF Description The END IF statement is always the last statement of a block IF construct Whenever a block of statements below the block IF itself or below one of the ELSE statements has been executed control is transferred to the first statement immediately following the END IF statement Alternatively if none of the conditional expressions within the block IF are true and if there is no ELSE with no conditional expression then control passes directly from the start of the block IF to the statement immediately following the END IF See also See the block IF statement for a complete discu
321. quential access file RECL is required for direct access files If omitted for sequential files variable length records are assumed BLANK bink is a character expression having one of the following values NULL Blank values in numeric formatted input fields are ignored except that a field of all blanks is treated as zeros ZERO Blanks other than leading blanks are treated as zeros If omitted blanks are ignored in numeric input fields Once properties of a file have been established in an OPEN statement only the BLANK parameter can be changed in a subsequent OPEN statement for that file unless the file is first closed in a CLOSE statement Once a file has been associated with a particular unit the file can be associated with another unit in a subsequent OPEN statement The file is then associated with more than one unit In this case the unit numbers refer to the same file Actions taken on one unit also affect the other unit For example closing a unit closes all other units associated with the same file If a file is associated with a unit and a succeeding OPEN statement associates a different file with the same unit the effect is the same as performing a CLOSE without a STATUS parameter on the currently associated file before associating the new file with the unit 5 39 1 Direct Access Files Direct access file manipulations differ from conventional sequential file manipulations In a sequential file records are stored
322. r 180 Message The pointer specification is preceded by the symbol token rather than a left parenthesis Example of message 180 If Line Nl Source I PROGRAM ERR180 2 PARAMETER IP 4 JP 10 3 POINTER P V 0 IP 0 JP 3 err180 for E180 The pointer specification is preceded by the symbol P rather than a left parenthesis REAL A 55 DATA A 3 123 456 2 3 45 5 7 89 45 6 78 P LOC a PRINT computed V 0 0 should be A 1 STOP END Error 181 Message The pointer identifier token is not a valid identifier symbol Example of message 181 If Line Nl Source 1 PROGRAM ERR181 2 PARAMETER IP 4 JP 10 3 POINTER P V 0 IP 0 JP 3 errl81l for E181 The pointer identifier is not a valid identifier symbol 4 REAL A 55 5 DATA A 3 123 456 2 3 45 5 7 89 45 6 78 6 LOC a PRINT computed V 0 0 should be A 1 8 STOP 9 END Error 182 Message The pointer is separated from its referent by the symbol token rather than by a comma Example of message 182 If Line Nl Source 1 PROGRAM ERR182 2 PARAMETER IP 4 JP 10 3 POINTER V 0 IP 0 JP 3 err182 for E182 The pointer is separated from its referent by the symbol rather than by a comma 217 PromulaFortran Compiler User s Manual REAL A 55 DATA A 3 123 456 2 3 45 5 7 89 45 6 78 P LOC a PRINT computed V 0 0 should be 1 STOP
323. r W825 The subscript value 5 for subscript 2 of array A is not in the range of 1 to 4 6 END Comment 826 Message The variable ident substring subscript is of type type rather than integer Example of message 826 If Line Nl Source 1 SUBROUTINE W826 2 CHARACTER 10 ABCD 3 ABCD 1 2 4 5 5 ABCD 3 0 4 CD 5 warn826 for C826 The variable I substring subscript is of type real 4 rather than integer 6 R 5 ABCD 1 7 warn826 for C826 The variable R substring subscript is of type real 4 rather than integer 8 WRITE ABCD 9 END Comment 827 Message 270 PromulaFortran Compiler User s Manual The function ident has been previously referenced or defined as a subroutine Example of message 827 If Line Nl Source 1 SUBROUTINE W827 2 CALL W827A J 1 3 END 4 FUNCTION W827A J I 4 warn827 for C827 The function W827A has been previously referenced or defined as a subroutine 5 CALL W827C J I 6 I W827C J 1 6 warn827 for C827 The function W827C has been previously referenced or defined as a subroutine Y W827A I Jd 8 END Comment 828 Message The subroutine ident has been previously referenced or defined as a function Example of message 828 If Line Nl Source 1 SUBROUTINE W828 2 R W828A J 1 3 END 4 SUBROUTINE W828A J I 4 warn828 for C828 The subroutine W828A has been previously referenced or defined as a function
324. r 2 double real 8 unsigned short logical 2 char integer 1 or byte long integer 4 16 PromulaFortran Compiler User s Manual float real 4 unsigned long logical 4 dcomplex complex 16 complex complex 8 string character name is any valid identifier to be used in the C output is one of following special characters indicates a memory pointer to the indicated type indicates a virtual pointer to the indicated type indicates that a value conversion should be made to the indicated type Only a single prototype definition may occur on each record and the notation below type name means that the function has no arguments It does not mean that the function has some unspecified number of unspecified arguments The type specifiers have their traditional C interpretation The complex and dcomplex types refer to single precision complex and double precision complex respectively The string type refers to the FORTRAN style character string A string argument actually consists of a pointer and a length specification PROTOTYPE definitions affect both the compilation of references to subprograms and the definitions of subprograms The prototype file must be supplied both when subprograms are being compiled and when they are being referenced 2 8 3 Value Parameters If no c specifier follows the type specification a value parameter is being defined When a reference to this argument is processed the value of the a
325. r F may be used Note that either upper or lower case may be used A character constant must have delimiting single or double quotes If a character constant occupies more than one record each continuation of the constant must begin in column two A complex constant has the form real constant real constant Blank characters appearing within noncharacter constants are ignored The BLANK parameter in an OPEN statement has no effect on namelist If a constant other than a character constant contains no characters other than blanks a fatal error results 5 38 2 NAMELIST WRITE PRINT PUNCH Statements The three namelist output statements are WRITE PRINT and PUNCH Their simplified syntax is shown below Syntax WRITE u name PRINT name PUNCH name 109 PromulaFortran Compiler User s Manual Where u is an integer expression name is the identifier of a NAMELIST group Description All variables and arrays and their values in the list associated with the namelist group name are output on the file associated with unit u OUTPUT or PUNCH They are output in the order of specification in the NAMELIST statement Output consists of at least three records The first record is a in column 2 followed by the group name the last record is a in column 2 followed by the characters END Each group begins with a new record As an example the following PROGRAM NAME NAMELIST VALUES TOTAL QUANT COST DATA QUANT COST
326. re for output under conversion w must be greater than or equal to d 6 The 6 extra spaces are required for sign and decimal point plus four spaces for the exponent field If the Gw dEe form is used for a number output without an exponent then e 2 spaces are inserted to the right of the field See the examples below Y 77 132 WRITE 7 20 Y 20 FORMAT G10 3 Output before printing 77 1 EXIT 1214635 1 WRITE 4 10 EXIT 20 FORMAT G10 3 Output before printing 121E 07 85 PromulaFortran Compiler User s Manual READ 5 50 SAMPLE 50 FORMAT 20 5 WRITE 6 60 SAMPLE 60 FORMAT 1X G14 8 Data Read By Output before Format Option READ Statement printing 1415926535 E 10 14159265E 10 E conversion 8079323846 89793238 F conversion 2643383279 26433833 10 693 9937510 603 99375 F conversion 5 27 16 H Descriptor The H descriptor is used to output strings of characters This descriptor is not associated with a variable in the output list The H descriptor has the form nHstring Where n is the number of characters in the string including blanks string is a string of characters The H descriptor cannot be used on input Note that although using quotes to designate a character string precludes the need to count characters the H descriptor may be more convenient if the string contains quotes 5 27 17 I De
327. re passed to the linker in a slightly modified form an underscore character is added to the back or each name This has pretty much become the defacto standard for FORTRAN compilers The additional underscore reduces the likelihood of name clash with standard system resources 2 8 2 Function Prototype Syntax In C a function prototype declaration defines the name return type and storage class of a function In addition it can define the types of some or all of the arguments for that function The prototype declaration has the same format as the function declaration except that it is terminated by a semicolon and the argument identifiers are optional If used argument identifiers have scope only within the prototype declaration and serve only as place holders The syntax used for prototypes by the FORTRAN compiler is similar to that used by C however it is unfortunately not the same The reason is that additional information must be supplied to the compiler since additional problems are introduced by the weak typing conventions of FORTRAN The syntax of the PROMULA prototype definition is as follows fname type name type c typel c type name type c typelc 1 Where fname is the name of the subprogram used in the FORTRAN code without the added underscore type is one of the following C binary type specifiers the corresponding FORTRAN type is also shown void Any type or a record type short intege
328. ression for the position of the first character of the substring If first is omitted the value is one last specifies an integer real or double precision expression for the position of the last character in the substring If 1ast is omitted the value is the length of the string The specification of the first character in the substring is an integer real or double precision expression that is evaluated and converted as necessary to integer via truncation The expression can contain array element references and function references but evaluation of a function reference must not alter the value of the other expression in the substring reference If the specification of first is omitted the value is 1 and all characters from 1 to the value of the specification of last are included in the substring The specification of last in the substring is an expression subject to the same rules as the specification of irst If last is omitted the value is the length of the string and all characters from the specified first position to the end of the string are included in the substring For a string length 1en the value of irst must be at least 1 and must not exceed 1ast the value of 1ast must not exceed the value of 1en The following is an example of a string reference CHARACTER 6 51 52 DATA S1 STRING 52 S1 Reference to 51 is reference to the full string 1 1 3 Value STR 51 3 4 Value 51 4 Value 51 4 Value
329. result is al a2 See also None 6 59 DSIGN Transfer of Sign Synopsis REAL 8 FUNCTION DSIGN REAL 8 REAL 8 Description DSIGN 1 2 is a specific function that returns a double precision result from two double precision arguments The result reflects the magnitude of the first value combined with the sign of the second value In particular if al 0 0 and a2 0 0 return a1 or if al lt 0 0 and a2 gt 0 0 return a1 else return al See also ISIGN I2SIGN and SIGN 155 PromulaFortran Compiler User s Manual 6 60 DSINH Hyperbolic Sine Synopsis REAL 8 FUNCTION DSINH REAL 8 Description DSINH a is a specific function that returns a hyperbolic sine The result and argument are double precision See also SINH 6 61 DSIN Sine Synopsis REAL 8 FUNCTION DSIN REAL 8 Description DSIN a is a specific function that returns a sine The argument is assumed to be in radians The argument and result are both double precision See also CSIN CDSIN and SIN 6 62 DSQRT Square Root Synopsis REAL 8 FUNCTION DSQRT REAL 8 Description DSQRT a is a specific function that returns the principal square root of its argument The argument and result are double precision See also CDSQRT CSQRT and SQRT 6 63 DTANH Hyperbolic Tangent Synopsis REAL 8 FUNCTION DTANH REAL 8 Description DTANH a is a specific function tha
330. reviously declared as being external 4 LL I KK I 5 I LL I JJ I 6 END Note 892 Message List directed I O for internal files is nonstandard Example of message 892 If Line Nl Source 1 SUBROUTINE W892 2 CHARACTER 128 LINE 3 A 3 4 WRITE LINE A A 4 warn892 for N892 List directed I O for internal files is nonstandard 5 WRITE LINE 6 END Warning 893 Message There is an equivalence contradiction for ident Example of message 893 If Line Nl Source 1 SUBROUTINE W893 2 REAL A 10 B 20 3 EQUIVALENCE A 1 B 10 A 10 B 5 4 WRITE A B 5 END 5 warn893 for W893 There is an equivalence contradiction for B Comment 894 Message The do loop ends on a transfer statement Example of message 894 294 PromulaFortran Compiler User s Manual If Line Nl Source SUBROUTINE W894 INTEGER 6 15 war 75 DO 80 I 1 6 A I 894 for C894 The do loop ends on a transfer statement statement 1 2 3 DATA J 1 4 DO 10 I 1 6 5 6 10 IF J 2 15 15 30 6 warn894 for C894 The do loop ends on a transfer statement 7 15 DO 20 I 1 6 8 9 20 ENDDO 0 30 DO 60 I 1 6 1 2 60 RETURN 12 warn894 for C894 The do loop ends on a transfer 3 65 DO 70 I 1 6 4 5 70 STOP 70 n 6 4 8 80 GOTO 65 75 85 105 I 18 warn894 for C894 The do loop ends transfer 9 85 DO 90 I 1 6 20
331. rgument is passed and not its address If the actual argument does not have the proper type then an error occurs The only exception to this occurs when processing numeric constants A constant with a decimal point or exponent specified may be used as either a float or double value parameter and a numeric constant without a decimal indication with a value in the range 32767 to 32767 may be used in either a long or short environment 2 8 4 External Name Clash A problem that pervades C in particular is the external name clash problem One tends to use many different libraries with C It is not at all unusual to have the same names used by different libraries As FORTRAN moves into contemporary environments it must face the same problem With PROMULA a replacement name may be specified in the prototype Thus the following two prototypes void fread short long short void fwrite short long short which specify argument types for two functions could also be written fread void ftread short long short fwrite void ftwrite short long short This second form specifies not only the argument types but also changes the external names of the subprograms The decision to change these names is made entirely in the prototype file no changes are needed in the actual FORTRAN 17 PromulaFortran Compiler User s Manual source Note that this renaming overrides the normal addition of an underscore to the back of the name See the section o
332. rm specified in a previously declared structure Syntax RECORD sname rlist sname rlist Where sname is the name of a previously declared structure 123 PromulaFortran Compiler User s Manual rlist isa list of one or more variable names array names or array declarators separated by commas Description The RECORD statement creates records of the form specified in a previously declared structure All of the records named in any given rlist have the same structure and are allocated separately in memory The RECORD statement is comparable to that of an ordinary type declaration except that aggregate data items are declared instead of scalar data items Record names can be used in COMMON and DIMENSION statements They cannot be used in DATA EQUIVALENCE NAMELIST or SAVE statements Records initially have undefined values unless you have defined their values in their structure declarations See also The STRUCTURE statement describes how record structures are defined 5 49 REWIND Statement Syntax REWIND unit REWIND UNIT unit IOSTAT status ERR err Where unit is an integer expression status is integer lvalue err the label of an executable statement Notes The simplest form of the REWIND statement consists of a single unit specification not enclosed in parentheses With this form no additional parameters can be specified If the parenthetical form of REWIND is used the
333. rmation is transmitted from the designated file in the form in which it exists on the file without any conversion If the number of words in the list exceeds the number of words in the record an execution diagnostic results If the number of locations specified in ilist is less than the number of words in the record the excess data is ignored If ilist is omitted the unformatted READ skips one record The list directed READ statement reads data into the storage locations named in ilist with the input data items being free form with separators rather than in fixed size fields A list directed READ following a list directed READ that terminated in the middle of a record starts with the next data record Input data consists of a string of values separated by one or more blanks or by a comma or slash either of which can be preceded or followed by any number of blanks Also a line boundary such as end of record or end of card serves as a value separator however a separator adjacent to a line boundary does not indicate a null value Embedded blanks are not allowed in input values except character values and complex numbers The format of values in the input record is as follows Integers use the same format as for integer constants For real numbers any valid FORTRAN format for real or double precision numbers may be used In addition the decimal point can be omitted it is assumed to be to the right of the mantissa Complex numbers consist of two re
334. ror messages themselves are discussed in a later chapter in this manual The echo control options may be used to send additional information to the standard output file This information includes the following Option Information sent to standard output Warnings about potentially serious inconsistencies or usages FORTRAN source code EL2 Comments about possibly nonportable code or about code that may be incorrect and the warnings from above PromulaFortran Compiler User s Manual EL3 Notes about standard FORTRAN violations and other miscellaneous observations and comments and warnings from above ES An annotated listing of the source code EX An alphabetical listing of the symbols used in the source along with a summary description of each and a cross reference listing of symbol references by line number VI A message each time the preprocessor opens a new source file or include file V2 A message each time the preprocessor starts a new subprogram as well as the messages from above 2 4 1 Warnings Notes and Comments The form and meaning of the warnings notes and comments are discussed in a later chapter on error messages Suffice to say that the PF does extensive syntax checks while it is processing the source code and extensive consistency checks after it has processed each subprogram The following is a sample listing produced which shows the type of messages that might be produced at EL1 level 39
335. rs There is only one character operator No precedence exists among the relational operators The precedences among the operators are Arithmetic Highest Character Relational Logical Lowest An expression can contain more than one kind of operator For example the logical expression L OR A B EH C where a and c are of type real and is of type logical contains an arithmetic operator a relational operator and a logical operator This expression would be interpreted as L OR A B GE C Any arithmetic operation whose result is not mathematically defined is prohibited for example neither dividing by zero nor raising a zero valued primary to a zero valued or negative valued power is allowed 4 2 LVALUES An lvalue is the name of a variable or array element of type integer short integer byte real double precision complex or double complex The type of the lvalue is the same as the type of the variable or array element which forms it The term value is an abbreviation of the term left hand value The name is derived from the fact that only lvalues may appear on the left hand side of an assignment statement Lvalues are contrasted with expressions in that expressions produce a value while lvalues can receive a value 44 PromulaFortran Compiler User s Manual 4 3 Assignment Statements There are four types of assignment statements Arithmetic Character Logical Statement lab
336. rt Export Data from to Other Programs Program Database Program Inputs Program Outputs Program Code 178 PromulaFortran Compiler User s Manual Figure 8 2 A PROMULA Enhanced FORTRAN Program User Interface 8 1 Transforming a FORTRAN Program with PROMULA PROMULA and the FORTRAN compiler can be used in tandem to upgrade existing FORTRAN codes The interface between these two systems truly adds value to existing FORTRAN programs Traditionally FORTRAN programs are computational boxes they read in data and write out reports By making the internal data structure of a FORTRAN program a PROMULA dataset you are able to unlock the information content of the FORTRAN box and make it immediately accessible to users via a database and a structured user friendly interface Consider the following scenario You have been given the assignment to port a useful but large FORTRAN program from the company mainframe to the PC or some other desktop platform The program runs on a mainframe requires about 8 Megabytes of system memory and has a batch user interface its inputs are fixed format sequential text files and so are its outputs Your mission is to run this program on the desktop without losing any of its functionality reproduce its behavior and its results and make it more user friendly With PROMULA this portation project will go like this 1 Make a list of all the program variables that you wish to deposit in a ran
337. s execution will branch to statement 25 See also The discussion of the OPEN statement describes the FORTRAN file system in general including the different file types 5 7 CHARACTER Statement The CHARACTER statement defines some user defined entity to be of type character Syntax CHARACTER len name name Where name one of the forms var len c array d dl len clist len specifies the length and can be an unsigned nonzero integer constant an integer constant expression enclosed in parentheses with a positive value or an asterisk enclosed in parentheses var is a variable function name symbolic constant or dummy procedure array is an array name d specifies the bounds of a dimension clist isa list of constants or symbolic constants specifying the initial values Each item in the list can take the form EXC is constant or symbolic constant ra is a repeat count that is an unsigned nonzero integer constant or the symbolic name of such a constant The CHARACTER statement is used to define a variable array symbolic constant function name or dummy procedure name as type character A length specification immediately following the word CHARACTER applies to each entity not having its own length specification A length specification immediately following an entity is the length specification only for that entity Note that for an array the length specified is for
338. s a constant or symbolic constant r isa repeat count that is an unsigned nonzero integer constant or the symbolic name of such a constant Notes The DOUBLE COMPLEX statement performs the same action as the COMPLEX 16 statement Description The DOUBLE COMPLEX statement is used to define a variable array symbolic constant function name or dummy procedure name as type double complex The symbol may already have been defined in another declaration statement See also See the discussion of the DIMENSION statement for a description of how dimension bounds are defined 65 PromulaFortran Compiler User s Manual See the discussion of the DATA statement for a description of how initial values are defined 5 16 DOUBLE PRECISION Statement The DOUBLE PRECISION statement defines some user defined entity to be of type double precision Syntax DOUBLE PRECISION name name Where name has one of the forms var c array d d clist var is a variable function name symbolic constant or dummy procedure array isan array name d specifies the bounds of a dimension clist isa list of constants or symbolic constants specifying the initial values Each item in the list can take the form 6 r c c is a constant or symbolic constant isa repeat count that is an unsigned nonzero integer constant or the symbolic name of such a constant Notes The DOUBLE PRECISION statement performs th
339. s not normally a problem since options do not usually end with extensions while all valid filenames must The following is a complete list of the options Note that with one exception the PF command line options are NOT case sensitive i e the characters and A are considered to be the same on the command line Option Information 1 8 D vax pdp p77 piv El 1 2 3 Es Ex F Snum t f v 9 FI sil G I name LJ name name PH numb PW numb QE numb QI numb QH numb numb 5 0 1 UR numb UW numb UP numb V 1 2 Z name 2 4 Echo Control Options Sec Characteristic Affected by Switch Compile only Treatment of short arithmetic FORTRAN dialect conventions Echo errors at specified level Echo source code Echo symbol cross reference information FORTRAN source format used FORTRAN INTEGER type Debugging flag C compiler include file pathname Library file name or pathname for linker Name of executable Output page height Output page width Quantity of line number storage Quantity of input record storage Quantity of include file storage Quantity unresolved external symbol storage Storage auto or static Doloop trip count assumption Input unit number Output unit number Punch unit number Verbose flag Include file pathname for FORTRAN processor PF is normally silent unless a fatal error is encountered Messages are sent to standard output and may be redirected Note that er
340. s preceding exponents since those signs are always provided 5 27 22 TL TR Descriptors The T TL and TR descriptors provide for tabulation control These descriptors have the forms Tn TLn TRn Where n isa nonzero unsigned decimal integer When a Tn descriptor is encountered in a format specification input or output control skips right or left to column n the next edit descriptor is then processed When a TLn descriptor is encountered control skips backward left n columns If n is greater than or equal to the current character position control skips to the first character position When a TRn descriptor is encountered control skips forward right n characters TRn is the same as nX On input control can be positioned beyond the end of record but a succeeding descriptor would read only blanks 89 PromulaFortran Compiler User s Manual With a T TR or TL specification the order of a list need not be the same as that of the input or output record The same information can be read more than once When a T TR TL specification causes control to pass over character positions on output positions not previously filled during record generation are set to blanks those already filled are left unchanged 5 27 23 Termination of Format Control A colon in a format specification terminates format control if there are no more items in the input output list The colon has no effect if there are more items in th
341. s shown below Syntax READ u name Where u is an integer expression name is identifier of a NAMELIST group Description When a READ statement references a namelist group name input data in the format described below is read from the designated file Syntax Snam ntity value entity value SEND Where name is the name of the namelist that contains the entity or entities whose values are being specified entity is a namelist defined entity It can be a variable array name subscripted variable variable with a substring or a subscripted variable with a substring value is a constant a list of constants a repetition of constant in the form r c or a repetition of values in the form r Where c is a constant or symbolic constant isa repeat count that is an unsigned nonzero integer constant or the symbolic name of such a constant As an example consider the following SAGRP Group name XVAL 5 0 ARR 5 1 7 2 4 Five complex numbers CHAR HI THERE SEND Group terminator In each record of a namelist group column one is reserved for carriage control and must be left blank Data items following name are read until another is encountered 108 PromulaFortran Compiler User s Manual Blanks must not appear in the following locations 1 Between and namelist group name 2 Between and END 3 Within array names and variable names Blanks ca
342. s the compiler that line number information and symbol information is to be left in the object forms so that the standard debugger can be used See the information on the debugger for your platform for details on its use 2 7 3 Compiler Include File Pathname The Iname option specifies the path to be used by the C compiler to find its standard include files This option should only be used if your C compiler has been installed in a nonstandard way 2 7 4 Library File Name or Pathname for Linker The Lname uppercase version specifies the pathname name to be used in locating the runtime libraries specified via the option The lowercase version of this option specifies additional libraries to be added into the link step via shorthand name See the installation instructions for the conventions for your platform 2 7 5 Name of executable The Oname option specifies the name of the executable to be produced No extension may be specified for this name If this option is omitted the name of the executable is taken from the first source or object file name encountered on the command line 2 7 6 Location of FORTRAN Files To Be Included It is often desirable to have include files that are referenced from within FORTRAN programs stored in some other directories The Zname option can be used to specify directories to be searched for include files The syntax of name varies for operating system to operating system See the installat
343. scriptor The I descriptor specifies integer conversion This descriptor has the forms Iw Iw m On input the plus sign can be omitted for positive integers When a sign appears it must precede the first digit in the field The Iw and Iw m specifications have the same effect on input An all blank field is considered to be zero Decimal points are not permitted The value is stored in the specified variable Any character other than a decimal digit blank or the leading plus or minus sign in an integer field on input will cause an error See example of I input below OPEN 2 BLANK NULL READ 2 10 I J K L M N 10 FORMAT 13 17 12 13 12 14 Input Record jesse CELA TAG In storage 86 PromulaFortran Compiler User s Manual I contains 139 J contains 15 K contains 18 L contains 7 M contains 0 N contains 14 If BLANK ZERO had been specified on the OPEN statement J would contain 1500 and N would contain 104 Other values would not be affected On output if the integer is positive the plus sign is suppressed unless an SP specification is in effect Leading zeros are suppressed If Iw m is used and the output value occupies fewer than m positions leading zeros are generated to fill up to m digits If m 0 a zero value will produce all blanks If m w no blanks will occur in the field when the value is positive and the field will be too short for any negative value If the field is too short asteri
344. se of r or 1 character constants is nonstandard 4 WRITE 1 1X A3 IA 5 STOP 6 END Note 855 Message The use of radix 50 constants is nonstandard Example of message 855 If Line Nl Source ab PROGRAM W855 2 DATA IA 3RABC 2 warn855 for N855 The use of radix 50 constants is nonstandard 3 WRITE 1 1X 110 IA 4 STOP 5 END Note 856 Message 280 PromulaFortran Compiler User s Manual A logical or arithmetic if with less than 3 branches is nonstandard Example of message 856 If Line Nl Source 1 PROGRAM W856 2 0 3 0 0 5 6 3 warn856 for N856 logical or arithmetic if with less than 3 branches is nonstandard 4 5 WRITE I 5 6 CONTINUE 6 END Note 857 Message The parenthetical form of the program statement is nonstandard Example of message 857 If Line Nl Source 4 PROGRAM 1857 OUTPUT TAPE99 INPUT 1 warn857 for N857 The parenthetical form of the program statement is nonstandard 2 A 1 3 WRITE A 4 END Note 858 Message The identifier token with more than 6 characters is nonstandard Example of message 858 If Line Nl Source al SUBROUTINE W858 2 1 3 JABCDEF I 3 warn858 for N858 The identifier JABCDEF with more than 6 characters is nonstandard 4 WRITE I JABCDEF 5 END Note 859 Message The continuation of the end statement is nonstandard Example of message 859 281 PromulaFortran Compiler
345. sion elements PromulaFortran Compiler User s Manual COMPLEX A B C D WRITE 6 10 A 10 FORMAT F7 2 E8 2 READ 5 11 B C D 11 FORMAT 2E10 3 2 F8 3 F4 1 Different types of data can be read by the same FORMAT specification For example 10 FORMAT I5 F15 2 specifies two values the first of type integer the second of type real Example CHARACTER R 4 READ 5 15 INK A B R 15 FORMAT 315 2F7 2 A4 reads three integer values two real values and one character string 5 27 5 Field Separators Field separators are used to separate descriptors and groups of descriptors The format field separators are the slash the comma the colon and the dollar sign The slash is also used to specify demarcation of formatted records while the dollar sign blocks the formation of a new record when the end of the format specification is encountered 5 27 6 Repeatable and Nonrepeatable Edit Descriptors The repeatable edit descriptors are used to specify numeric logical or character data conversions The repeatable edit descriptors can be repeated by prefixing the descriptor with a nonzero unsigned integer constant specifying the number of repetitions required The repeatable edit descriptors are A D E F G I L O and Z The nonrepeatable edit descriptors are used for numeric input output control tabulation control character output control format control end of record desig
346. sion of the DATA statement for a description of how initial values are defined See the description of the EQUIVALENCE statement for a discussion of how COMMON areas can be extended 5 9 COMPLEX Statement The COMPLEX statement defines some user defined entity to be of type complex or double complex Syntax COMPLEX len name name Where name one of the forms var len c array d d len clist len specifies the complex subtype and can be an unsigned nonzero integer constant whose value is 8 or 16 var is a variable function name symbolic constant or dummy procedure array isan array name d specifies the bounds of a dimension 55 PromulaFortran Compiler User s Manual clist isa list of constants or symbolic constants specifying the initial values Each item in the list can take the form a is a constant or symbolic constant isa repeat count that is an unsigned nonzero integer constant the symbolic name of such constant Description The COMPLEX statement is used to define a variable array symbolic constant function name or dummy procedure name as type complex or double complex A length specification immediately following the word COMPLEX applies to each entity not having its own length specification A length specification immediately following an entity is the length specification only for that entity If the length specification for a give
347. sion of the file name begins with an then PF assumes that it is a FORTRAN source code and passes it to the FORTRAN preprocessor and then to the C compiler for compilation Upon completion a file with the extension or obj is formed This file may then be passed on to the linker If the file name extension is o or obj on some platforms then PF assumes that the file contains compiled code and passes it to the linker unchanged If the file name extension is a or lib on some platforms then PF assumes that the file is a library of compiled codes and passes it to the linker unchanged If the file name extension is pro then PF assumes that the file contains a list of function prototypes to be used by the FORTRAN preprocessor This file is passed to that preprocessor along with the names of the FORTRAN source codes Multiple prototype files may be supplied The content of prototype type files is described in a later section of this chapter If the file name extension is glb then PF assumes that the file contains a list of the program variables that are to be made global for use by the PROMULA Application Development System The content of this file and the general topic of the PROMULA interface are described in a later chapter in this manual PromulaFortran Compiler User s Manual 2 3 Summary of Options Each option on the PF command line may be preceded by so that it will not be interpreted as a file name This i
348. sion result is ai where the complex argument is ar ai See also AIMAG 6 51 DINT Truncation Synopsis REAL 8 FUNCTION DINT REAL 8 Description DINT a returns a double precision integer value after the truncation of its double precision argument The result is O if the absolute value of a is less than 1 If the absolute value of a is greater than or equal to 1 the result is the largest integer with the same sign as argument a that does not exceed the magnitude of a See also AINT 6 52 DLOG10 Logarithm Base 10 Synopsis REAL 8 FUNCTION DLOG10 REAL 8 Description DLOG10 a is a specific function that returns the logarithm base 10 of the argument The argument is double precision and the result is double precision See also ALOG10 CDLOG10 CLOGI0 and LOG10 6 53 DLOG Natural Logarithm Synopsis REAL 8 FUNCTION DLOG REAL 8 Description DLOG a is a specific function that returns the natural logarithm of the argument The argument is double precision and the result is double precision See also 153 PromulaFortran Compiler User s Manual ALOG CDLOG CLOG and LOG 6 54 DMAX1 Maximum Value Synopsis REAL 8 FUNCTION DMAX1 REAL 8 Description DMAX1 al a2 an is a specific function that returns the value of the largest argument All arguments are double precision and the result is double precision See also AMAXI MAX
349. sitive difference REAL 8 Exponential REAL REAL Positive difference REAL 8 REAL 8 REAL REAL 8 REAL 8 REAL INTEGER 2 INTEGER 2 INTEGER INTEGER INTEGER INTEGER COMPLEX 16 Imaginary part REAL 8 Truncation REAL 8 Logarithm base 10 REAL 8 Natural logarithm REAL S Maximum value REAL S Minimum value REAL 8 REAL 8 Remainder REAL 8 Nearest integer REAL REAL Product REAL 8 REAL 8 REAL 8 REAL 8 REAL 8 REAL 8 REAL 8 INTEGER REAL REAL 8 COMPLEX COMPLEX 16 INTEGER 2 INTEGER CHARACTER INTEGER Transfer of sign Hyperbolic sine Sine Square root Hyperbolic tangent Tangent Stop program execution Exponential Real value Get command line Absolute value 134 PromulaFortran Compiler User s Manual Table 6 1 Intrinsic Functions Name Type Arguments Description INTEGER 2 INTEGER 2 IAND INTEGER 2 INTEGER 2 Bitwise and INTEGER INTEGER ICHAR INTEGER 2 CHARACTER Integer value of character IDIM INTEGER INTEGER INTEGER Positive difference IDINT INTEGER REAL 8 Integer value IDNINT INTEGER REAL Nearest integer IFIX INTEGER INTEGER 2 Integer value INTEGER INTEGER INTEGER REAL INTEGER REAL 8 INTEGER COMPLEX INTEGER COMPLEX 16 INDEX INTEGER CHARACTER CHARACTER Location of substring INT2 INTEGER 2 INTEGER Integer value INTEGER 2 INTEGER 2 INTEGER 2 REAL INTEGER 2 REAL 8 INTEGER 2 COMPLEX INTEGER 2 COMPLEX 16 INT4 INTEGER INTEGER 2 Integer value
350. sks occupy the field 5 27 18 L Descriptor The L descriptor is used to input or output logical items This descriptor has the form Lw On input if the first nonblank characters in the field are T or T the logical value TRUE is stored in the corresponding list item which should be of type logical If the first nonblank characters are F or F the value FALSE is stored If the first nonblank characters are not T T F or F a diagnostic is printed An all blank field has the value FALSE On output variables under the L specification should be of type logical A value of TRUE or FALSE in memory is output as a right justified T or F with blanks on the left 5 27 19 O Descriptor The O descriptor letter O is used to input or output items in octal format This descriptor has the forms Ow Ow m The form Ow m means the same as Ow on input The octal digits are the digits O through 7 On input blanks are allowed and a plus or minus sign can precede the first octal digit Blanks are interpreted as zeros and an all blank field is interpreted as zero A decimal point is not allowed On output the rightmost w digits are output If m is specified the number is printed with leading zeros so that at least m digits are printed 5 27 20 P Descriptor The P descriptor is used to change the position of a decimal point of a real number when it is input or output The P descriptor has the form KE 87 PromulaFortran Comp
351. ssage SANTA MARIA The structure or record definition identifier is rather than a slash The assignment statement contains the unrecognized The member identifier token associated with the object ident is not a valid identifier symbol Example of message 268 If Line Nl Source 1 PROGRAM ERR268 2 STRUCTURE SHIP 3 1 CHARACTER 10 OWNER 4 1 CHARACTER 10 DESTINATION 5 1 END 6 RECORD SHIP NINA PINTA SANTA MARIA 245 PromulaFortran Compiler User s Manual 7 CHARACTER 10 WHERE 8 WHERE NINA 3 8 err268 for E268 The member identifier 3 associated with the structure SHIP is not a valid identifier symbol 9 STOP 10 END Error 269 Message The member identifier token is not a member of the object ident Example of message 269 If Line Nl Source ab PROGRAM ERR269 2 STRUCTURE SHIP 3 al CHARACTER 10 OWNER 4 1 CHARACTER 10 DESTINATION 5 d END 6 RECORD SHIP NINA PINTA SANTA MARIA F CHARACTER 10 WHERE 8 WHERE NINA OWNR 8 err269 for E269 The member identifier OWNR is not a member of the structure SHIP 9 STOP 10 END Error 270 Message The dimension specification for ident must be the last dimension Example of message 270 If Line Nl Source all SUBROUTINE ERR270 A B C 2 DIMENSION 3 2 err270 for E270 The dimension specification for A must be the last 3 DIMENSION B 2 2 5 4 DIMENSION 2 5 4 err270 for E270 The dimension specifica
352. ssion is followed by the symbol vather than a right parenthesis 3 STOP 4 END Error 145 Message The statement statement contains the unrecognized symbol token Example of message 145 If Line Nl Source ah PROGRAM ERR145 2 IF I EQ 45 THEN PRINT Hello World 2 errl45 for E145 The if statement contains the unrecognized symbol PRINT 3 X STOP 4 1 END Error 146 Message The object ident may not be allocated to common Example of message 146 If Line Nl Source 1 SUBROUTINE ERR146 2 PARAMETER ICON 3 CON 3 3 REAL A B 4 INTEGER Z CON 205 PromulaFortran Compiler User s Manual D COMMON E CON 5 errl46 for E146 The parameter CON may not be allocated to common 6 WRITE A Z 7 END Error 147 Message The members of a complex constant are separated by the symbol token rather than a comma Example of message 147 If Line Nl Source 1 PROGRAM ERR147 2 COMPLEX A 3 3 DATA A 0 0 1 0 2 0 4 0 0 0 0 0 3 errl47 for E147 The members of a complex constant are separated by the symbol rather than a comma 4 WRITE A 5 STOP 6 END Error 148 Message The second member of a complex constant is the symbol token rather than a numeric constant Example of message 148 If Line Nl Source 1 PROGRAM ERR148 2 COMPLEX A 3 3 DATA 0 0 1 0 2 0 4 0 0 0 0 0 3 errl48 for E148 The second member of a complex constant is the symbol 0 0 ra
353. ssion of the various component statements within the block IF construct 5 24 ENTRY Statement The ENTRY statement defines additional entry points for a procedure established by the SUBROUTINE or FUNCTION statement Syntax ENTRY ep d d 1 Where ep is a symbolic name which identifies the entry point d is a dummy argument that can be one of the following a variable name an array name a dummy procedure name an asterisk only if in a subroutine subprogram 70 PromulaFortran Compiler User s Manual Description The ENTRY statement can be used to define additional entry points for a procedure established by the SUBROUTINE or FUNCTION statement Each procedure subprogram has a primary entry point established by the statement that begins the program unit Usually a subroutine call or function reference invokes the procedure at the primary entry point and the first statement executed is the first executable statement in the program unit ENTRY statements are used to define other entry points A procedure that contains one or more ENTRY statements is said to have multiple entry points An ENTRY statement can appear anywhere after the SUBROUTINE or FUNCTION statement in the subprogram When an entry name is used to reference a procedure execution begins with the first executable statement that follows the referenced entry point An entry name is available for reference in any program unit except in the procedure that cont
354. st statement begins with the symbol SHIP rather than a slash 3 0 0 4 READ SHIP END 10 4 err228 for E129 The information for the an option was supplied redundantly 5 IF C NE 0 0 STOP 6 10 PRINT NO DATA FOUND B STOP 8 END Error 229 Message The namelist identifier token is not a valid identifier symbol 232 PromulaFortran Compiler User s Manual Example of message 229 If Line Nl Source 2 err229 for 3 4 4 err229 for Error 230 Message PROGRAM ERR 229 namelist SHIP 1 12 E229 The namelist identifier is not a valid identifier symbol 0 0 READ SHIP END 10 E129 The information for the an option was supplied redudantly IF C NE 0 0 STOP 10 PRINT NO DATA FOUND STOP END The namelist identifier ident has already been defined as a object Example of message 230 If Line Nl Source 3 err230 for 1 5 Error 231 Message PROGRAM ERR230 INTEGER SHIP AMELIST SHIP 11 12 E230 The namelist identifier SHIP has already been defined as a variable C 0 0 READ SHIP END 10 IF C NE 0 0 STOP 10 PRINT NO DATA FOUND STOP END The namelist identifier ident is followed by the symbol token rather than a slash Example of message 231 If Line Nl Source 22 rr2981 for O1 BW PROGRAM ERR231 NAMELIST SHIP 11 12 E231 The namelis
355. stant format specification A constant expression can contain a constant a previously defined symbolic constant the arithmetic operators an extended constant expression enclosed in parentheses If a symbolic name is of type integer real double precision complex or double complex the corresponding expression must be an arithmetic constant expression If the symbolic name is of type character or logical the corresponding expression must be a character constant expression or logical constant expression Each symbolic name becomes defined with the value of the expression that appears to the right of the equals according to the rules for assignment Any symbolic constant that appears in an expression must have been previously defined in the same or a different PARAMETER statement in the program unit A symbolic name of a constant can be defined only once in a program unit and can identify only the corresponding constant The type of a symbolic constant can be specified by an IMPLICIT statement or type statement before the first appearance of the symbolic constant in a PARAMETER statement If the length of a symbolic character constant is not the default length of one the length must be specified in an IMPLICIT statement or type statement before the first appearance of the symbolic constant The easiest way to do this is to explicitly type the symbolic constant as character with length The actual length of the constant is determined by the l
356. statements following the DO statement up to and including the terminal statement The DO statement execution is as follows 1 The expressions 1 2 and are evaluated and if necessary converted to the type of the control variable Ms 2 Control variable v is assigned the value of e1 3 The iteration count is established this value is determined by the following expression 63 PromulaFortran Compiler User s Manual INT m2 m1 m3 m3 mtc Where ml m2 m3 are the values of the expressions 1 2 and e3 respectively after conversion to the type of v mtc is the minimum trip count normally has a value of 0 It may be set to 1 if a very old dialect of FORTRAN is being processed 4 the iteration count is not 0 the range of the DO loop is executed If the iteration count is 0 execution continues with the statement following the terminal statement of the DO loop the control variable retains its most recent value which is typically m2 m3 5 Control variable v is incremented by the value of e3 6 The iteration count is decremented by one Steps 4 through 6 are repeated until the iteration count has a value of 0 If a DO loop appears within an IF block the range of the DO loop must be entirely contained within the IF block If a block IF statement appears within the range of a DO loop the corresponding END IF statement must also appear within the range of that DO loop A DO loop can
357. storage location Whenever a variable is used it references the value currently in that location A variable does not have be defined before being referenced for its value The possible types of variables are as follows 1 Integer 2 Short integer 3 Byte 4 Real 5 Double precision 6 Complex 7 Double complex 8 Logical 9 Short logical 10 Logical byte 11 Character When the type of a variable is not explicitly specified its type is determined by its first character It is integer if the first letter is I J K L M or N and is real if the first letter is any other character Note that the IMPLICIT statement is used to override this default convention 3 4 1 Integer Variable An integer variable is a variable that is typed explicitly implicitly or by default as integer The range of an integer variable is from 2147483648 to 2147483647 3 4 2 Short Integer Variable A short integer variable is a variable that is typed explicitly as a short integer The range of a short integer variable is from 32768 to 32767 3 4 3 Byte Variable 28 PromulaFortran Compiler User s Manual A byte variable is a variable that is typed explicitly as a byte It behaves like an integer variable however its range is from 0 to 255 3 4 4 Real Variable A real variable is a variable that is typed explicitly implicitly or by default as real The range of a real variable is from 107 307 to 10308 The precision f
358. string consists of two characters only the function or subroutine prefix and suffix characters As an example consider the following piece of a prototype file for a FORTRAN based system in which the FORTRAN compiler appends a underscore character GLOBALS void actday long long long long void valdt long long long short void fixdt long long long long long long long short void weeknd long long long long long long long short The actual public names are actdat_ valdt fixdt and weeknd The GLOBALS string preceding these tells the compiler to make this change in the external names 2 8 7 Renaming Identifiers Only A final capability of the prototype file is to allow the user to enter simple identifier renaming requests The following notation oldname newname in a prototype input file will rename all occurrences of o1dname with newname 19 PromulaFortran Compiler User s Manual 3 LANGUAGE ELEMENTS This chapter discusses the language elements of FORTRAN FORTRAN statements symbolic names constants variables arrays character substrings statement order and user written program units 3 1 FORTRAN Statements FORTRAN statements are written using the FORTRAN character set which consists of 26 letters 10 digits and 16 special characters as follows blank right parenthesis period or decimal point underline plus sign dollar sign minus sign slash asterisk colon s
359. t assigned to an integer variable by the execution of a prior ASSIGN statement Syntax GOTO iv slab slab Where slab 15 the label of an executable statement iv is an integer variable Description The variable iv must not be defined by any statement other than an ASSIGN statement The list of statement labels is optional The possible branches taken by the GOTO are determined entirely by the assignments made to the variable within the program unit In the following example ASSIGN 50 TO JUMP 10 GOTO JUMP 20 30 40 50 20 CONTINUE 30 CAT ZERO HAT 40 CAT 10 1 3 50 25 2 7 3 statement 50 is executed immediately after statement 10 5 31 IF Statement The IF statement evaluates an expression and conditionally transfers control or executes another statement depending on the outcome of the test Syntax IF aexp slabl slab2 slab3 95 PromulaFortran Compiler User s Manual IF lexp stat IF lexp THEN ELSEIF lexp THEN 221 ENDIF Where aexp is an arithmetic expression lexp is a logical expression slab1 labels of executable statements slab2 slab3 stat is any executable statement except a DO block IF ELSE ELSE IF END END IF or another logical IF statement Description The IF statement evaluates an expression and conditionally transfers control or executes another statement depending on the outcome of the test The kinds of
360. t directly stored in memory To achieve these two different versions two compilations have been made but no changes have been made to the FORTRAN source Now the following prototype and FORTRAN code can be compiled For the compilation we will also read a globals file which will request that the variable resides on disk while the remaining ones reside in memory Again see the chapter on the PROMULA interface for more information on global files anadat void vanadat float short double double void manadat float short double double SUBROUTINE TES DIMENSION A 10 B 20 C 100 REAL 8 ABAR AVAR BBAR BVAR CBAR CVAR CALL ANADAT A 10 ABAR AVAR CALL ANADAT B 20 BBAR BVAR CALL ANADAT C 100 CBAR CVAR RETURN END 18 PromulaFortran Compiler User s Manual In this function ANADAT will be compiled to as manadat when the vector is in memory and vanadat when the vector is virtual 2 8 6 Global Symbols and Prototypes An additional problem that comes up is that some of the runtime libraries have been implemented via FORTRAN and others via C Some FORTRANS and or some Cs append additional characters to each external symbol be it a function or a subroutine This requires that groups of global symbols but not all use a modified naming convention These modifications are achieved via GLOBAL strings which may be entered into prototype files The GLOBALS
361. t identifier SHIP is followed by the symbol rather than a slash 0 0 READ SHIP END 10 IF C NE 0 0 STOP 10 PRINT NO DATA FOUND STOP END 233 PromulaFortran Compiler User s Manual Error 232 Message The namelist ident element identifier token is not a valid identifier symbol Example of message 232 If Line Nl Source 2 err232 for 4 CO Error 233 Message PROGRAM ERR232 NAMELIST 5 11 12 E232 The namelist SHIP element identifier is not a valid identifier symbol 0 0 READ SHIP END 10 IF C NE 0 0 STOP 10 PRINT NO DATA FOUND STOP END The namelist ident element ident is a subprogram argument Example of message 233 If Line Nl Source 2 err233 for 3 234 Message SUBROUTINE ERR233 J1 K1 11 NAMELIST SHIP 11 12 E233 The namelist SHIP element I1 is a subprogram argument 0 0 READ SHIP END 10 IF C NE 0 0 RETURN 10 PRINT NO DATA FOUND STOP END The external element identifier token is not a valid identifier symbol Example of message 234 If Line Nl Source 2 err234 for 3 4 4 err234 for 5 PROGRAM ERR234 EXTERNAL AFUNC BFUNC CFUNC E234 The external element identifier is not a valid identifier symbol CALL PROCESS AFUNC IVAL CALL PROCESS BFUNC IVAL E248 The variable argument BFUNC has been entered where a subprogra
362. t is not executed execution continues with the next statement As an example IF P AND Q RES 7 2 TEMP RES Z if P and Q are both true the value of the variable RES is replaced by 7 2 otherwise the value of RES is unchanged In either case TEMP is set equal to RES times Z As another example IF A LT B CALL SUB1 20 ZETA TEMP RES4 if A is less than B the subroutine SUB1 is called Upon return from this subroutine statement 20 is executed If A is greater than or equal to B statement 20 is executed and SUB1 is not called 5 31 3 Block IF Statement The block IF statement allows conditional execution of a block of executable statements 97 PromulaFortran Compiler User s Manual Syntax IF expl THEN ELSEIF exp2 THEN ELSE ENDIF Where expl logical expressions exp2 Description The block IF statement is used with the END IF and optionally the ELSE and ELSE IF statements to form block IF structures If the logical expression is true execution continues with the next executable statement If exp1 or exp2 is false control transfers to an ELSE or ELSE IF statement or if none are present to an END IF statement If control is transferred to an ELSEIF it behaves exactly as the original IF except that its logical expression is tested If control is transferred to an ELSE then the statements immediately following it are
363. t number out of range EINT_TMF too many files open E XF_DEL mising terminating delimeter ELUN_TMF too many files open ENXF_HOL bad Hollerith string too many files open ERCK_BUF internal buffer exceeded ESIO_TMF too many files open ERDX MLP missing left parenthesis ECLO PCF physical close failure ERDX COM missing comma EOPN POF physical open failed ERDX MRP missing right parenthesis ECLO POF physical open failure ERDZ MLP missing left parenthesis EBCK FRT at front of file ERDZ COM missing comma EBCK DIR direct access file ERDZ MRP missing right parenthesis ELUN PWF physical write failed ERNL MNI missing namelist identifier ERWV PWF physical write failure ERNL MVI missing variable identifier EWBV PWF physical write error ERNL_UVI undefined variable identifier EWEF_PWF physical write failure ERNL_SSV subscripted scalar variable ERDB_IFS invalid format specification ERNL_NNS non numeric subscripts ERDD_IFS invalid format specification ERNL_TMS too many subscripts ERDF_IFS invalid format specification ERNL_EQL missing equals sign ERDI IFS invalid format specification ERNL BSI bad string input ERDL IFS invalid format specification ERNL MLP complex missing left pren ERDS IFS invalid format specification ERNL COM complex missing comma ERDT IFS invalid format specification ERNL MRP complex missing right pren EWRB IFS bad format specification ESTD NNC non numeric character in field EOPN BFZ setting buffer size failed 300
364. t returns a hyperbolic tangent The result and argument are double precision 156 PromulaFortran Compiler User s Manual See Also TANH 6 64 DTAN Tangent Synopsis REAL 8 FUNCTION DTAN REAL 8 Description DTAN a is a specific function that returns a tangent The argument is assumed to be in radians The result and argument are double precision See also TAN 6 65 EXIT Stop Program Execution Synopsis SUBROUTINE EXIT INTEGER Description call to EXIT stops program execution It performs the identical operation as would be performed by executing the FORTRAN statement STOP a See also STOP statement description 6 66 EXP Exponential Synopsis REAL FUNCTION EXP REAL REAL 8 FUNCTION EXP REAL 8 COMPLEX FUNCTION EXP COMPLEX COMPLEX 16 FUNCTION EXP COMPLEX 16 Description EXP a is a generic function that returns an exponential The result is real double precision complex or double complex depending on the argument type See also 157 PromulaFortran Compiler User s Manual CDEXP CEXP and DEXP 6 67 FLOAT Real Value Synopsis REAL FUNCTION FLOAT INTEGER 2 REAL FUNCTION FLOAT INTEGER Description FLOAT a is a specific function that returns a real result from an integer argument It simply co
365. t structure declaration block which is used to define the form and possibly the content of records to be defined later in the program unit Syntax 125 PromulaFortran Compiler User s Manual STRUCTURE sname fdecl END STRUCTURE Where sname is a symbolic name fdecl has one of the following forms data substruc union data substruc union Description is any valid type statement BYTE INTEGER REAL DOUBLEPRECISION LOGICAL COMPLEX CHARACTER has the form STRUCTURE sname flist fdecl END STRUCTURE flist is a list of symbolic names has the form UNION mapdecl END UNION mapdecl has the form MAP fdecl END MAP DOUBLECOMPLEX The form of a record is defined by the multistatement STRUCTURE declaration This declaration is composed of the STRUCTURE statement itself which provides a symbolic name for the STRUCTURE being defined followed by the body of the declaration which is composed of one or more field declarations The order of the declarations determines the order of the fields within a structure The STRUCTURE declaration is terminated by an END STRUCTURE statement Field declarations within structure declarations consist of the following 126 PromulaFortran Compiler User s Manual 1 Typed data declaration statements which are simply standard FORTRAN type declarations Fields can be any valid type and are dimensioned in the normal way
366. t used as the real part and the imaginary part zero For a single complex argument the result is the same as the argument For two arguments 1 and a2 the result is complex with argument a1 used as the real part and argument a2 used as the imaginary part CMPLX does not have specific names See also DCMPLX 6 34 CONJG Conjugate Synopsis COMPLEX FUNCTION CONJG COMPLEX COMPLEX 16 FUNCTION CONJG COMPLEX 16 Description CONJG a is a generic function that returns a conjugate of a complex or double complex argument The result is the same type as the argument For a complex or double complex argument ar ai the result is ar ai with the imaginary part negated See also DCONGJ 146 PromulaFortran Compiler User s Manual 6 32 COSH Hyperbolic Cosine Synopsis REAL FUNCTION COSH REAL REAL 8 FUNCTION COSH REAL 8 Description COSH a is a generic function that returns a hyperbolic cosine The result is real or double precision depending on the argument type See also DCOSH 6 33 COS Cosine Synopsis REAL FUNCTION COS REAL REAL 8 FUNCTION COS REAL 8 COMPLEX FUNCTION COS COMPLEX COMPLEX 16 FUNCTION COS COMPLEX 16 Description COS a is a generic function that returns a cosine The argument is assumed to be in radians The result is real double precision complex or double complex depending on th
367. th character constants nonstandard i WRITE 10 A 8 WRITE 20 A 9 WRITE 30 B 10 END Note 885 Message An implicit high low range specification is nonstandard Example of message 885 If Line Nl Source 1 SUBROUTINE E885 2 IMPLICIT INTEGER D A 2 warn885 for N885 An implicit high low range specification is nonstandard 3 A 1 4 END Note 886 Message Including a repeat count before the token format specification is nonstandard Example of message 886 If Line Nl Source 1 SUBROUTINE W886 2 WRITE 10 3 10 FORMAT 1X 1S 11 6 2SP 12 3SS 13 is is is is 3 warn886 for N886 Including a repeat count before the S format Specification is nonstandard 3 warn886 for N886 Including a repeat count before the format Specification is nonstandard 3 warn886 for N886 Including a repeat count before the S format Specification is nonstandard 3 warn886 for N886 Including a repeat count before the S format Specification is nonstandard 4 WRITE 20 291 PromulaFortran Compiler User s Manual 5 20 FORMAT 10T10 6 A 2TL2 5 B 6TR6 C 7 5 warn886 for N886 Including a repeat count before the T format Specification is nonstandard 5 warn886 for N886 Including a repeat count before the format Specification is nonstandard 5 warn886 for N886 Including a repeat count before the T format Specification is nonstandard 5 warn886
368. the higher type using the same rules as are used for arithmetic expressions prior to the comparison A character relational expression has the logical value TRUE only if the values of the operands satisfy the relation specified by the operator The character expression x1 is considered to be less than x2 if the value of x1 precedes the value of x2 in the collating sequence x1 is greater than x2 if the value of x1 follows the value of x2 in the collating sequence Note that the collating sequence in use determines the result of the comparison If the operands are of unequal length the shorter operand is extended on the right with blanks to the length of the longer operand 4 1 5 General Rules for Expressions The order in which operands are combined using operators is determined by Use of parentheses Precedence of the operators Right to left interpretation of exponentiations Left to right interpretation of multiplications and divisions Left to right interpretation of additions and subtractions in an arithmetic expression Left to right interpretation of concatenations in a character expression Left to right interpretation of operators Left to right interpretation of AND operators Left to right interpretation of OR operators 0 Left to right interpretation of EQV NEQV and XOR operators in a logical expression or Boolean expression Precedences exist among the arithmetic and logical operato
369. the left The specification Fw d outputs a number in the following format sn n where n is a field of decimal digits 84 PromulaFortran Compiler User s Manual 5 is a minus sign if the number is negative or omitted if the number is positive The following shows some examples of F output Value of A FORMAT Output Before Printing 32 694 F6 3 432 694 32 694 F10 3 32 694 32 694 F6 3 32694 4 3 66 3 327 327 32 694 F6 0 ADMINS 5 27 15 G Descriptor The G descriptor specifies conversion between an internal real or double precision number and an external floating point number written either with or without an exponent depending on the magnitude of the number This descriptor has the forms Gw d Gw dEe On input the G specification is treated the same as the E specification The rules which apply to the E specification also apply to the G specification For example READ 5 11 A B C 11 FORMAT G13 6 2G12 4 On output results depend on the size of the floating point number being edited For values in the range greater than or equal to 1 and less than 10d the number is output under F format For values outside this range Gw d output is identical to Ew d and dEe is identical to Ew dEe If a number is output under the Gw d specification without an exponent four spaces are inserted to the right of the field these are reserved for the exponent field Therefo
370. the number of digits in the exponent m Unsigned integer constant specifying the minimum number of digits to be output k Integer constant scale factor n Positive nonzero decimal integer The following paragraphs discuss input output conversions field separators repeatable and nonrepeatable edit descriptors and repetition factors 5 27 4 Input Output Conversions For the D E F and G input conversions a decimal point in the input field overrides the decimal point specification of the field descriptor Leading blanks are not significant in numeric input conversions other blanks in numeric conversions are ignored unless BLANK ZERO is specified for the file on an OPEN statement or a BZ edit descriptor is in effect Plus signs can be omitted An all blank field is considered to be zero except for logical input where an all blank field is considered to be FALSE The output field is right justified for all output conversions If the number of characters produced by the conversion is less than the field width leading blanks are inserted in the output field unless w m is specified in which case leading zeros are produced as necessary The number of characters produced by an output conversion must not be greater than the field width Ifthe field width is exceeded asterisks are inserted throughout the field Complex data items are converted on input output as two independent floating point quantities The format specification uses two conver
371. ther than a comma or right parenthesis STOP 3 END N Error 255 Message The alternate return label token is not an integer constant Example of message 255 If Line Nl Source ah PROGRAM ERR255 2 CALL SUB255 A B STOP 2 err255 for E255 The alternate return label STOP is not an integer constant 3 PRINT Hello 4 100 STOP 5 END Error 256 Message The actual argument list contains the symbol token rather than a comma or right parenthesis Example of message 256 If Line Nl Source 1 PROGRAM ERR256 2 CALL SUB256 A B amp 100 2 err256 for E256 The actual argument list contains the symbol amp rather than a comma or right parenthesis 3 PRINT Hello 4 100 STOP 5 END Error 257 Message The actual argument list contains more than the number argument s required by the object ident Example of message 257 If Line Nl Source 1 PROGRAM ERR257 2 AVAL ALOG10 BVAL 100 0 241 PromulaFortran Compiler User s Manual 2 err257 for E257 The actual argument list contains more than the 1 argument s required by the system function ALOG1O 3 STOP 4 END Error 258 Message The adjustable dimension ident for ident is not a variable or parameter Example of message 258 If Line Nl Source 1 SUBROUTINE ERR258 FUNC N 2 DIMENSION FUNC N 3 CALL N FUNC 4 RETURN 5 END 5 err258 for E258 The adjustable dimension N for FUNC is not a variable or par
372. ther than a numeric constant 4 WRITE A 5 STOP 6 END Error 149 Message The default else is not the last part of the blocked if Example of message 149 If Line Nl Source PROGRAM ERR149 IF I 0 99 THEN PRINT Goodbye ELSE PRINT What now ELSE IF I EQ 98 or E149 The default else is not the last part of the blocked herrer OF 206 PromulaFortran Compiler User s Manual if 72 2 PRINT Hello 8 1 ENDIF 8 errl49 for E166 The statement introduced by the keyword ENDIF is not a valid statement type 9s AL END Error 150 Message Negative and zero if branches are separated by the symbol token rather than a comma Example of message 150 If Line Nl Source i PROGRAM ERR150 2 IF I 6 10 15 20 2 errl50 for E150 Negative and zero if branches are separated by the symbol rather than a comma 3 10 PRINT Hello 4 15 PRINT Goodbye 5 20 STOP 6 END Error 151 Message The zero if branch is the symbol token rather than an integer constant Example of message 151 If Line Nl Source 1 PROGRAM ERR151 2 IF I 6 10 IJK 20 2 errl5l for E151 The zero if branch is the symbol IJK rather than an integer constant 3 10 PRINT Hello 4 15 PRINT Goodbye 5 20 STOP 6 END Error 152 Message The positive if branch is token rather than an integer constant Example of message 152 If Line Nl Source 1 PROGRAM ERR152 2 IF I 6 10 15 IJK 2
373. tial values Each item in the list can take the form is constant or symbolic constant 50 PromulaFortran Compiler User s Manual f is a repeat count that is an unsigned nonzero integer constant or the symbolic name of such a constant Notes The BYTE statement performs the same action as INTEGER 1 statement Description The BYTE statement is used to define a variable array symbolic constant function name or dummy procedure name as type byte The symbol may already have been defined in another declaration statement A byte entity behaves like an integer entity however its range is 0 to 255 See also See the discussion of the DIMENSION statement for a description of how dimension bounds are defined See the discussion of the DATA statement for a description of how initial values are defined 5 5 CALL Statement The CALL statement transfers control to a subroutine subprogram Syntax CALL sub a a Where sub is the name of a subroutine or dummy procedure a is an actual argument that can be one of the following An expression An array name An intrinsic function name An external procedure name An alternate return specifier of the form s s is the statement label of an executable statement that appears in the same program unit as the CALL statement Description The CALL statement can contain actual arguments and statement labels which must correspond in order number and type to thos
374. tial values are defined 5 35 INQUIRE Statement The INQUIRE statement obtains information about the current status of a file or unit Syntax INQUIRE UNIT u IOSTAT ios ERR s1 OPENED od NUMBER num NAMED nmd NAME fn ACCESS acc SEQUENTIAL seq DIRECT dir FORM fm FORMATTED fmt UNFORMATTED unf RECL rcl NEXTREC nr BLANK b1nk EXIST ex INQUIRE FILE fin IOSTAT ios ERR s1 OPENED od NUMBER num NAMED nmd NAME fn ACCESS acc SEQUENTIAL seq DIRECT dir FORM fm FORMATTED fmt UNFORMATTED unf RECL rcl NEXTREC nr BLANK b1nk EXIST ex Where u is an integer expression fin is a character expression ios is an integer lvalue sl is the label of an executable statement ex are logical lvalues od nmd num are integer lvalues rcl nrz 102 PromulaFortran Compiler User s Manual fn acc seq fm fmt unf blnk dir Notes are character lvalues The UNIT specification is optional however when omitted the unit specification must be the first specification Other than the above the order of the parameters within the INQUIRE statement is free Description There are two forms of the INQUIRE statement Inquire by unit is used to obtain information about the current status of a
375. tifier DUBLE has not yet been defined as a constant integer parameter 9 END Error 277 Message The kind identifier ident is a object rather than a constant integer parameter Example of message 277 248 PromulaFortran Compiler User s Manual If Line Nl Source 1 PROGRAM ERR277 2 INTEGER SHORT DUBLE QUAD 3 PARAMETER SHORT 2 DOUBLE 8 QUAD N 4 PRINT 23 0 1234567 42_1 42_SHORT 5 PRINT 13 5 0 1234567 123 45678 00 30 DOUBLE 5 err277 for E277 The kind identifier DOUBLE is a parameter rather than a constant integer parameter 6 PRINT 3 0 3 12345 0 1234567 QUAD 6 err277 for E277 The kind identifier QUAD is a parameter rather than a constant integer parameter T PRINT 1 1 3 14 1 0 3 14_DOUBLE 7 8 1 0 1E 27_QUAD 7 err277 for E277 The kind identifier DOUBLE is a parameter rather than a constant integer parameter 9 END Error 278 Message The kind value number associated with ident is not available for the type type Example of message 278 If Line Nl Source d PROGRAM ERR278 2 INTEGER SHORT DOUBLE QUAD 3 PARAMETER SHORT 2 DOUBLE 8 QUAD 20 4 PRINT 23 0 1234567 42 1 42 SHORT 5 PRINT 13 5 0 1234567 123 45678 00 30 DOUBLE 6 PRINT 2 0 3 12345 0 1234567 QUAD 6 err278 for E278 The kind value 20 associated with ERR278 is not available for the real type 7 PRINT 1 1 3 14 1 0 3 14 D
376. tion for C must be the last 5 WRITE 1 1 2 2 2 2 2 2 2 6 END Error 271 Message The equivalence element substring specification token for ident is not a constant Example of message 271 If Line Nl Source 246 PromulaFortran Compiler User s Manual 1 PROGRAM ERR271 2 CHARACTER AMAT 10 IMAT 5 3 EQUIVALENCE 2 5 3 3 err271 for E271 The equivalence element substring specification N for AMAT is not a constant STOP 5 END B Error 272 Message The symbol token is not a valid subscript expression separator a comma was expected Example of message 272 If Line Nl Source 1 PROGRAM ERR272 2 CHARACTER 10 ERRMES 10 20 3 ERRMES 1 3 5 Bad Version 3 err272 for E272 The symbol is not a valid subscript expression separator a comma was expected 4 STOP 5 END Error 273 Message The symbol token is not a valid substring range separator a colon was expected Example of message 273 If Line Nl Source al PROGRAM ERR273 2 PARAMETER N 5 3 CHARACTER AMAT 10 IMAT 5 4 EQUIVALENCE IMAT 2 5 AMAT N N 3 4 err273 for E273 The symbol is not a valid substring range separator a colon was expected 5 STOP 6 END Error 274 Message The equivalence element substring specification is closed by the symbol token rather than a right parenthesis Example of message 274 If Line Nl Source 1 PROGRAM ERR274 2
377. tional Expression A relational expression can appear only within logical expressions Evaluation of a relational expression produces a logical result with a TRUE or FALSE value A relational expression used as a primary in a logical expression is in one of the forms arithexp rop arithrexp charexp rop charexp Where rop is one of the relational operators LE EQ NE GT GE arithexp is an arithmetic expression charexp is a character expression A relational expression is used to compare the values of two arithmetic or two character expressions A relational expression cannot be used to compare the value of an arithmetic expression with the value of a character expression The relational operators are shown below Operator Representing Meaning Less than Is x1 less than x2 LE Less than or equal to Is x1 less than or equal to x2 EQ Equal to Is x1 equal to x2 NE Not equal to Is x1 not equal to x2 GT Greater than Is x1 greater than x2 GE Greater than or equal to Is x1 greater than or equal to x2 An operand of type complex or double complex is permitted only when the relational operator is EQ or NE 43 PromulaFortran Compiler User s Manual An arithmetic relational expression has the logical value TRUE only if the values of the operands satisfy the relation specified by the operator If the two arithmetic expressions are of different types then the operand of the lower type is promoted to
378. ture DAY variable DATE integer 4 scalar GET_DATE subroutine 2 args integer 4 unspecified HOUR variable IME logical 1 scalar I variable static integer 4 scalar INUTE variable IME logical 1 scalar ONTH variable DATE integer 4 scalar EXT_APP record static APPOINTMENT scalar IME structure TODAY record static DATE scalar YEAR variable DATE integer 4 scalar Symbol references by line number in SUBROUTINE STRUC6 Identifier Line If u D defined M modified U used P passed APP_DATE 22s 0D 7 00 U APP_FLAG 27 01 D APP_LIST 2 97 0150 7 00 M 11 00 U 14 00 U APP_MEMO 26 01 D APP_TIME 24 01 D APPOINTMENT 21013 D DATE 1 01 5D DAY 2 50 89 6 00 U 8 00 U DATE 5 00 U HOUR 18 01 D I 4 00 5 00 P 7 00 U MINUTE MONTH 2 014 6 00 U NEXT_APP 29 OLED 11 00 13 00 0 17 01 D TODAY 30 00 D 5 00 P 6 00 U 7 00 U 8 00 YEAR 3 01 6 00 0 Statement label types and references line number SUBROUTINE STRUC6 Label Line If u D defined U used A assigned 5 format 10 00 D 10 statement 4 00 U 9 00 0 The include files section simply lists all include files encountered to date in the compilation along with the sequence number assigned to them Note that the base source file has a number of zero If there are no include files referenced in the current subprogram then this section is omitt
379. uccessive appearance of the common block name or no name for blank common adds more entities to the common block and is treated as a continuation of the specification Variables and arrays are stored in the order in which they appear in the specification Alignment bytes are often inserted between members of different types within common blocks therefore great care must be taken when the types of variables within COMMON blocks vary from subprogram to subprogram The actual size of any common block is the number of storage words required for the entities in the common block plus any extensions associated with the common block by EQUIVALENCE statements Extensions can only be made by adding storage words at the end of the common block Entities in common blocks can be initially defined by a DATA statement in a block data subprogram or by a DATA statement in any program unit If data is assigned to a common block in a subprogram other than a BLOCK DATA subprogram then that data may not be defined until that subprogram has been executed If data is assigned to the same storage location in a COMMON block in multiple non BLOCK DATA subprograms then the content of that location may vary as the different subprograms are initially executed The initialization of COMMON variables in non BLOCK DATA subprograms is highly discouraged See also See the discussion of the DIMENSION statement for a description of how dimension bounds are defined See the discus
380. ue token is not a numeric constant Example of message 136 If Line Nl Source T PROGRAM ERR136 2 COMPLEX A 3 3 DATA A 0 0 1 0 2 0 4 0 0 0 0 3 errl36 for E136 The first member of the complex value 0 is not a numeric constant 202 PromulaFortran Compiler User s Manual 4 WRITE A 5 STOP 6 END Error 137 Message The symbol token was used for the option option rather than a variable identifier Example of message 137 If Line Nl Source 1 PROGRAM ERR137 2 OPEN 6 FILE ERR127 OUT IOSTAT 32 STATUS NEW 2 errl37 for E137 The symbol 32 was used for the option IOSTAT rather than a variable identifier 3 STOP 4 END Error 138 Message The format does not have a line number Example of message 138 If Line Nl Source ai PROGRAM ERR138 2 PRINT 1 3 FORMAT Hello World 3 err138 for E138 The format does not have a line number 4 STOP 5 END Error 139 Message The format specification is preceded by the symbol token rather than a left parenthesis Example of message 139 If Line Nl Source 1 PROGRAM ERR139 2 PRINT 1 3 1 FORMAT Hello World 3 errl39 for E139 The format specification is preceded by the symbol Hello World rather than a left parenthesis 4 STOP 5 END 203 PromulaFortran Compiler User s Manual Error 140 Message The format specification ends in the symbol token rather than a right parenthesis Example of message
381. uivalence element subscript specification N for AMAT is not a constant 4 STOP 5 END Error 218 Message The symbol token is not a valid format specifier Example of message 218 If Line Nl Source 1 PROGRAM ERR218 2 10 FORMAT 1X I10 3 READ 5 FMT 10 0 IA 3 err218 for E218 The symbol 10 0 is not a valid format specifier 4 STOP 5 END Error 219 Message The equivalence element ident is closed by the symbol token rather than a right parenthesis Example of message 219 If Line Nl Source 1 PROGRAM ERR219 2 DIMENSION 10 5 3 EQUIVALENCE IMAT 2 AMAT 4 3 err219 for E219 The equivalence element IMAT is closed by the symbol rather than a right parenthesis 4 STOP 5 END Error 220 Message The first element ident of the equivalence is followed by the symbol token rather than a comma introducing the next element 229 PromulaFortran Compiler User s Manual Example of message 220 If Line Nl Source 1 PROGRAM ERR220 2 DIMENSION AMAT 10 IMAT 5 3 EQUIVALENCE 2 AMAT 4 3 err220 for E220 The first element IMAT of the equivalence is followed by the symbol AMAT rather than a comma introducing the next element STOP 5 END Error 221 Message The formal argument ident has already been defined earlier in the argument list Example of message 221 If Line Nl Source 1 SUBROUTINE ERR221 ALPHA BETA ALPHA 1 err221
382. unrecognized specification character token Example of message 294 If Line Nl Source 1 PROGRAM ERR294 2 WRITE 1 Hello World 3 1 FORMAT 1 10 5 3 err294 for E294 The format contains the unrecognized specification character Y STOP 5 END Error 295 Message The format specification has a bad t format specification Example of message 295 If Line Nl Source 1 PROGRAM ERR295 2 WRITE 1 Hello World 3 1 FORMAT T A10 3 err295 for E295 The format specification has a bad t format specification 4 STOP 5 END Error 296 Message 254 PromulaFortran Compiler User s Manual The format specification has a bad business format string Example of message 296 If Line Nl Source 1 PROGRAM ERR296 2 1 FORMAT 1X B 2999999999999999999999999999999999999999999999 2 err296 for E296 The format specification has a bad business format string 3 WRITE 6 1 12345 4 STOP 5 END Error 297 Message A b format specification is not followed by an n or a z Example of message 297 If Line Nl Source 1 PROGRAM ERR297 2 1 FORMAT BC 1X I5 2 err297 for E297 A b format specification is not followed by an n or a 2 3 WRITE 6 1 12345 4 STOP 5 END Error 298 Message A variable lt format specification has no closing gt Example of message 298 If Line Nl Source 1 PROGRAM ERR298 2 DIMENSION A 5 3 DATA A 1 2 3 4 5 4 DO 10 I 1 10 5
383. ure names and names of entities in a common block must not appear in the SAVE statement A common block name or indicating blank common has the effect of specifying all of the entities in the common block A SAVE statement with no list is treated as though it contained the names of all allowable items in the program unit If a common block name is specified in a SAVE statement in a subprogram the common block name must be specified by a SAVE statement in every subprogram in which the common block appears Execution of a RETURN statement or an END statement within a subprogram may cause the entities within the subprogram to become undefined except in the following cases 1 entities specified by SAVE statements 2 entities that have been initially defined in a DATA declaration If a local variable or array that is specified in a SAVE statement and is not in a common block is defined in a subprogram at the time a RETURN or END statement is executed that variable or array remains defined with the same value at the next reference to the subprogram 5 51 STOP Statement The STOP statement terminates program execution Syntax STOP n Where n is an integer or string expression The STOP statement writes the message STOP n in the CONSOLE file and then terminates program execution The actual value of n at the time of termination is shown in the STOP message 5 52 STRUCTURE Statement The STRUCTURE statement introduces a multistatemen
384. ut It returns YES if the file can be opened for unformatted input output NO if the file cannot be opened for unformatted input output and UNKNOWN if it cannot be determined if the file can be opened for unformatted input output rcl is an integer variable which returns the record length of a file opened for direct access If the file is FORMATTED xc1 contains the record length in characters if UNFORMATTED the record length is in words It is undefined if the file is not opened for direct access nr is an integer variable which for a direct access file returns the record number of the last record read or written plus one If no records have been read or written nr contains 1 For sequential files it is undefined blink isa character variable which specifies the treatment of blanks It returns NULL if null blank control is in effect for a file opened for formatted input output It returns ZERO if zero blank control is in effect for a file opened for formatted input output It is undefined if the file is not opened for formatted input output If an error occurs as a result of the inquire and if neither s1 nor ios are supplied then execution will terminate with an error code set If either or both are supplied the execution continues despite any errors If supplied the ios lvalue receives the runtime error code for the error condition encountered or a zero if no error occurred If 51 is supplied then execution will branch to
385. utput report shown in Figure 8 4 The date keys are used to label the rows in the report as specified in the KEY RELATION defined in the program Line 28 of the program listing The number of decimal places shown is 1 as corresponds to the type specification for the variable Finally the range of values has been restricted to the 17 actual ones as specified in the SELECT statement Line 32 of the program listing PROMULA BROWSE expos Exposure M T an 08 01 79 09 01 79 10 01 79 11 01 79 12 01 88 01 01 88 02 01 88 03 01 88 04 01 88 13 05 01 88 28 06 01 88 37 07 01 88 31 08 01 88 9284 2 09 01 88 8639 2 10 01 88 4418 2 11 01 88 8738 3 12 01 88 5119 3 End Exit Shift Fn PgUp PgDn Home Arrows Browse Figure 8 7 Values of expos Suppose next that we wish to see these values shown to 4 decimal places so that we can see all of the significant digits The statement shown below does this The result is shown in Figure 8 8 This display emphasizes that it is the actual values within the program that are stored on the datafile and not the reported values as shown in the report PROMULA BROWSE exposl 10 4 187 PromulaFortran Compiler User s Manual Exposure M T 1 08 01 79 2820 09 01 79 8840 10 01 79 9146 11 01 79 0039 12 01 79 23437 01 01 80 9216 02 01 80 7682 03 01 80 7678 04 01 80 8488 05 01 80 8358 0
386. ve related specific functions For example the generic function name LOG computes the natural logarithm of an argument Its argument can be real double precision or complex and the type of the result is the same as the type of the argument Specific function names ALOG DLOG and CLOG also compute the natural logarithm The specific function name ALOG computes the log of a real argument and returns a real result Likewise the specific name DLOG is for double precision arguments and results and the specific name CLOG is for complex arguments and results If a generic name and specific names exist a generic name can be used in place of a specific name and is more flexible than a specific name Except for type conversion generic functions the type of the argument determines the type of the result Only a specific name can be used as an actual argument when passing the function name to a user defined procedure or function The following table lists the intrinsic functions in alphabetical order by name Table 6 1 Intrinsic Functions Name Type Arguments Description ABS INTEGER 2 INTEGER 2 Absolute Value INTEGER INTEGER REAL REAL REAL 8 REAL 8 REAL COMPLEX REAL 8 COMPLEX 16 ACOS REAL 8 REAL 8 Arccosine REAL REAL AIMAG REAL COMPLEX Imaginary part REAL 8 COMPLEX 16 AINT REAL REAL Truncation REAL 8 REAL 8 ALOGIO REAL REAL Logarithm base 10 ALOG REAL REAL Natural logarithm 132 PromulaFortran Compiler User s Manual
387. ween its double precision arguments If a1 is greater that a2 then the value of 151 PromulaFortran Compiler User s Manual al a2 is returned else zero is returned See also DIM and I2DIM 6 48 DEXP Exponential Synopsis REAL 8 FUNCTION DEXP REAL 8 Description DEXP is a specific function that returns a double precision result from a double precision argument It returns the exponential of its argument See also CDEXP CEXP and EXP 6 49 DIM Positive Difference Synopsis REAL FUNCTION DIM REAL REAL REAL 8 FUNCTION DIM REAL 8 REAL 8 REAL 8 FUNCTION DIM REAL REAL 8 REAL 8 FUNCTION DIM REAL 8 REAL INTEGER 2 FUNCTION DIM INTEGER 2 INTEGER 2 INTEGER FUNCTION DIM INTEGER INTEGER Description DIM al a2 is a generic function that returns a positive difference The result is integer real or double precision depending on the type of the first argument If a1 is greater that a2 then the value of al a2 is returned else zero is returned See also DDIM IDIM and I2DIM 6 50 DIMAG Imaginary Part Synopsis REAL 8 FUNCTION DIMAG COMPLEX 16 152 PromulaFortran Compiler User s Manual Description DIMAG a is a specific function that returns the imaginary part of a double complex argument The double preci
388. witch specifies that the file whose name is specified by the Vvalue parameter is to be created during the execution of this program The parameter value specifies the size of that file in bytes If the S switch is used with no corresponding name specified then the name pfc dba is used Note that the value to be used with this switch is displayed by the compiler when an application requiring virtual memory is processed 7 10 Specifying a Virtual Sheet Count The Zvalue command line switch specifies the maximum number of sheets to be allocated for virtual memory This parameter is needed when both dynamic and virtual allocations are being made It prevents the virtual manager from exhausting all memory before the dynamic manager has a chance to satisfy its memory needs The maximum setting for this parameter is 254 176 PromulaFortran Compiler User s Manual 8 THE PROMULA INTERFACE The PROMULA FORTRAN compiler converts FORTRAN source codes into executable code The PROMULA Application Management System henceforth PROMULA is an integrated programming environment which is designed to use a data structure which is completely compatible with the one used by FORTRAN Based on multi dimensional arrays dimensioned variables the PROMULA data structure is a generalization of the FORTRAN data structure The PROMULA language itself is an elegant notation for managing and manipulating such arrays Using the two tools together the FORTRAN compi
389. xample of message 867 If Line Nl Source 1 SUBROUTINE W867 A B C D I 2 COMMON J K 3 REAL A I B J 1 C K 4 WRITE A B C D 5 END 5 warn867 for N867 The argument B has an expression as an adjustable dimension Note 868 Message Implicit statements following specification statements is nonstandard Example of message 868 If Line Nl Source 1 SUBROUTINE W868 SAM A 2 INTEGER SAM REAL 4 IMPLICIT COMPLEX S 4 warn868 for N868 Implicit statements following specification statements 284 PromulaFortran Compiler User s Manual is nonstandard 5 WRITE 6 END Warning 869 Message The do index variable ident is already the index for an outer loop Example of message 869 If Line Nl Source 1 SUBROUTINE W869 2 REAL A 10 3 DATA A 10 4 5 4 DO 10 J 1 10 St a 10 WRITE 6 A J J 1 3 6 DO 20 I 1 10 dd DO 20 1 10 7 warn869 for W869 The do index variable I is already the index for an outer loop 8 2 20 A I I 9 END Comment 870 Message The array ident is being subscripted with less than number expressions Example of message 870 If Line Nl Source 1 SUBROUTINE W870 2 REAL 10 10 7 6 3 3 DATA 2 3 2 3 warn870 for C870 The array Z is being subscripted with less than 2 expressions 4 DO 10 J 1 100 10 Y J 0 5 warn870 for C870 The array is being subscripted with less than 2 expressions 6 WRITE Y Z 7 E
390. zero integer constant or the symbolic name of such a constant Description The COMMON statement provides a means of associating entities in different program units The use of common blocks enables different program units to define and reference the same data without using arguments and to share storage units Within one program unit any entity in a common block is known by a specific name Within another program unit the same data can be known by a different symbolic name that is valid only within the scope of that program unit The cb parameter is a common block name identifying a named common block containing the entities in nlist If the name is omitted the nlist entities are in blank common Other than having no name blank common has no characteristics different from those of labeled common blocks 54 PromulaFortran Compiler User s Manual A single variable name or array name can appear only once in any COMMON statement within the program unit Function or entry names cannot be included in common blocks In a subprogram names of dummy arguments cannot be included in common blocks If the common block name is omitted the common block is blank common When the first specification in the COMMON statement is for blank common the slashes can also be omitted If a common block name is specified the common block is a named common block Within a program unit declarations of common blocks are cumulative nlist following each s

Download Pdf Manuals

Related Search

Related Contents

  XBF-TC04 User Manual  Samsung Refrigerador French Door 589 L manual do usuário    Seguridad Taller Tecnologia - Blog de Ana María Gallardo Suárez  Betriebsanleitung NOVACAT 7800  2.4 GHz - Televes  SMC Barricade SMC7004VBR (SMC7004VBR  Samsung WF-J125N Manual de Usuario  Quick Guide iCom  

Copyright © All rights reserved. Failed to retrieve file