BRITISH COMPUTER SOCIETY FORTRAN SPECIALIST GROUP
Minutes of a meeting on Thursday,
21st October 1971 at B.C.S.
Headquarters, 29 Portland Place,
London, W.1. at 10.30 a.m.
PRESENT:
Mr J.S. Gatehouse (Chairman) GEC-EE Ltd.
Dr J.C. Baldwin Atlas Computer Laboratory, Chilton
Mr P.D. Bond Philips Industries
Mr R.L. Butchart D.T.I.
Mr J.C. Cullen B.P.
Mr I.E. Davidson N.C.R.
Mr R.E. Day Edinburgh R.C.C.
Mr P.J. Hammond B.P.
Mr I.D, Hill M.R.C.
Mr I.D.K. Kelly Olympic Computer Services
Dr M. Kennedy Queens University, Belfast
Mr K. St. Pier G.E.C.
Prof P.A. Samet University College, London
Mr B.H. Shearing Alcock Shearing & Partners
Mr R.E. Small C.A.P.
Mr I.A.G. Snowball A.E.R.E., Harwell
Mr D.T. Muxworthy (Secretary) Edinburgh R.C.C.
APOLOGIES
FOR
ABSENCE:
Dr A.C Day University College, London
Mr M.J. Garside University of Kent at Canterbury
Dr I.C. Pyle A.E.R.E., Harwell
1. APPROVAL OF The minutes of the meeting held on Tuesday 12th January
MINUTES 1971 were approved.
2. MATTERS No matters were raised under this item.
ARISING
3. REPORTS FROM a. Diagnostics working Party. Professor Samet ACTION
THE WORKING reported that the working party consisted of
PARTIES Professor D. Barron, Mr I.D. Hill, Miss E. Wilson
and himself, and that one meeting had been held.
The technical work of the group was outlined at
the meeting and a summary contributed by Mr Hill
is appended to these minutes.
b. Free Format Working Party. No information on
this group was available and its Chairman was to be DTM
asked to provide a written report. It later transpired
that the Chairman had sent a report direct to BCS HQ
but this was not delivered to the meeting; it is
appended to these minutes.
c. Small Machine Working Party. Mr Garside had
indicated, when intimating his apologies for missing
the meeting, that there had been little progress in
this group and that he would welcome the opportunity
to relinquish the chairmanship. Mr Shearing expressed
interest in the group and said that he could devote
time to it in the new year. It was decided to take no
action for the present.
d. Extensions working Party. Mr Shearing reported that
this group had concentrated on two major subjects: general
expressions and type CHARACTER; a technical summary is
appended to these minutes. It had been discovered only
recently that the final ANSI committee meeting at which
new ideas would definitely be considered would be on
November 18 and that the group's report may therefore be
in the form of a discussion rather than definite
recommendations. It was decided that the report would
be considered as a submission to ANSI by the individuals
concerned as there was insufficient time to obtain Fortran
Specialist Group or BCS approval. The report would BHS
however be circulated to the Group and submitted to the
Computer Bulletin. The ANSI committee were to be asked DTM
to supply a copy of the new draft Standard so that it
could be checked for inconsistencies.
4. CURRENT ANSI The Secretary reported on meetings he had had with
ACTIVITY Mr J.C. Noll, a member of the ANSI X3J3 Fortran
Committee. Mr No11 had supplied a number of
documents indicating the lines of thought of
X3J3. The revision of the ANSI Standard was
much wider than had been realised by the BCS Group
and in some cases went beyond what had been implemented
on available compilers. A list of ideas for consideration
as extensions, appendix E of X3J3 meeting 29 in August
1971, is attached. (It will be noted that items from
the BCS Extension Working Party's first report appear
between points 90 to 105; these were usually voted on
and accepted or rejected, verbatim).
Mr Noll a1so explained the reasons for the name changes
of ANSI. The original, American Standards Association,
was objected to by Canadians on the grounds that they too
were American but were not represented. The name of the
country was therefore incorporated, to make United States
of America Standards Institute, but this drew objections
from Ralph Nader who pointed out that the name was
indicative of a government agency, not a voluntary
organization. Hence the name was changed to American
National Standards institute.
5. OTHER The chairman raised the question of the future activity
BUSINESS of the Group now that much of the initial work related
to the new ANSI Standard was drawing to a close. It
had been suggested that the Group organize lectures or
seminars for both "external" and "internal" education.
The former would inform the public about Fortran matters,
including publicizing Standards; the latter would
be directed more towards existing members of the group
and would discuss such things as syntax checkers, flowcharting
aids, macroprocessors etc. It was also suggested that
the group should examine the new draft standard and should
act as a watchdog on the manufacturers. No decision as
to whether these were suitable activities was made and
more information was requested for the next meeting. JSG/DTM
It was noted that the Society appeared to ignore the
expertise available within the Group which could be
used, amongst other things, for testing algorithms
and for replying to letters on Fortran matters in the
Society journals, The Chairman was to raise this JSG
matter with the Technical Board. It was decided after JSG
a vote (9 to l with 4 abstentions) to reply to Mr. Sandhir's
letter in the October Computer Bulletin (p.376).
A proposal for a formal resolution that the language
specialist groups should form a closer liaison and that
their chairmen should form a language committee of the
Technical Board was he1d over to the next meeting. It
was hoped that by then an individual approach would have
determined the attitude of the Board to such a resolution.
6. DATE OF NEXT The next meeting will be held on Thursday 16th December
MEETING 1971 at 10.30 a.m. at B.C.S. Headquarters.
The "Diagnostics" Working Party consists of:
Prof. P.A. Samet,
Prof. D.W. Barron,
Miss E. Wilson,
Mr. I.D. Hill.
The Working Party met on 6th October 1971; unfortunately
Professor Barron was unable to attend.
It was agreed to concentrate on run-time diagnostics
in the first instance, and to seek to identify the events
that may call for a diagnostic message. The Working Party
are strongly of the opinion that, while speedy running of
a program is very important, speed must take second place
to correctness of results.
An attempt is being made to devise a method that
would be suitable to suggest as a standard "trace" for
Fortran programs. Desirable features seem to be that
the user Should be able to turn the tracing on or off
as between one run and another, as well as between one
part of the program and another.
To allow subprograms to be separately compiled, the
tracing would have to be independent from one subprogram
to another.
The user should be able to specify a channel number
for the output of tracing, and of other diagnostics.
The tracing information should be in "higher level"
form, directly related to the source language. and should
at least include transfers of control both to labels and to and
from subroutines.
The user should also be able to specify a "trace list"
naming variables whose names and values would be printed
in the trace information whenever the value changes. This,
however, needs careful thought if difficulties with array
variables, with COMMON, with EQUIVALENCE and with dummy
arguments are to be avoided.
British Computer Society FORTRAN Specialist Group
Report of the Working Party on Free Format
I attach a letter soliciting contributions on Free Format in FORTRAN.
It is being sent to Universities, Research Establishments and Computer
companies as well as to individual members of the Specialist Group and
delegates to the WORKSHOP. An amended version of the letter is being
sent to the Computer Bulletin.
I would be grateful for any suggestions to widen the circulation list.
David H. Marwick
(Chairman)
18.10.71
British Computer Society FORTRAN Specialist Group
Working Party on Free Format
At the FORTRAN Workshop held in Edinburgh in April, a working party on
Free Format in FORTRAN was set up by the Specialist Group; the remit
being to produce proposals, on Free Format of both instructions and data,
which could be presented as a possible addition to Standard FORTRAN.
The working party are therefore seeking contributions from as many
Sources as possible in an attempt to obtain an overall picture of (a)
what people want and (b) what already exists. We are not seeking
ways of implementing Free Format by subroutines, or any flukes, freaks
or fiddles. Our aim is a Standard FORTRAN Free Format facility.
Enclosed is a copy of the report produced from the Free Format Group at
the Edinburgh Workshop. This indicates the current lines of thought of
the Working Party but are by no means definitive. If you have any
contribution to make, on current implementations, or future requirements,
or just ideas; please send it to the address given below.
David H. Marwick
(Chairman)
Department of Computer Science,
Heriot-Watt University,
37-39 Grassmarket,
Edinburgh, EHl 2HW.
The "Extensions" Working Party consists of
Mr B.H. Shearing (AS & P)
Mr K.E. Bicknell (Rothamsted)
Mr E. Bodger (IBM)
Dr A.C. Day (UCL)
Mr D. Maisey (ICL)
Mr D.T. Muxworthy (ERCC)
Mr A. Odell (RAE)
Mr C.F. Schofield (ULCC)
Mr R.E. Sma1l (CAP)
It has held four meetings since the Edinburgh Workshop. All the Workshop
Chairmen's notes have been examined and attention has been concentrated on
two items: evaluating expressions and type CHARACTER.
It is suggested- that the procedure for evaluation of expressions be determined
by the hierarchy of operators method commonly used in Fortran textbooks, rather
than the parsing technique of the current ANSI standard. Within the level
**, unary +, unary -, evaluation should be right to left and in other levels left
to right; this imposes a restriction not present in the 1966 Standard but which
is felt to be desirable. The mode of the result of any binary arithmetic
operation between operands of different types is determined according to a table
which includes the new type DOUBLE PRECISION COMPLEX; the numeric value of the
result is the same as that obtained by converting both operands to the same type
as the result before carrying out the operation.
The Working party is concerned that the following CHARACTER facilities are in
the next Standard.
1. The ability to move character strings as single entities.
2, The ability to extract and to replace substrings.
3. The ability to convert single characters to small integers and vice versa.
4. The CHARACTER constant must be the existing Hollerith constant.
The syntax of CHARACTER statements is relatively unimportant.
In this session the Working Party has abandoned its rule that decisions must
be unanimously approved and any recommendations made do not necessarily
reflect the views of individual members.
SUGGESTED ADDITIONAL ACTIVITIES FOR THE FORTRAN SPECIALIST GROUP
A series of public lectures, perhaps once a month, in a decent lecture room,
possibly with a charge for attendance. The topics could include:
What is Standard Fortran.
A code of good. practice for programming in Fortran.
Extensions to Standard Fortran that commonly exist
and can be expected in next revision.
Compiler techniques for Fortran
Desirable applications of Fortran.
Non-technical applications of Fortran.
The interaction of Fortran with the operating system.
The use of Fortran in a time-sharing environment.
J.S.G.
Occasional seminars to discuss and exchange experience of using programming
aids which are of particular relevance to the Fortran programmer, such as:
Syntax checkers.
Automatic Flowcharting programs.
Program profilers.
Macro processors.
D.T.M.
APPENDIX E [of X3J3 minutes of August 1971]
ASSIGNMENTS
GROUP AREA MEMBERS
1. PROGRAM FORM AND PROGRAMS, LOVELL*, EATON, MULLINAX,
FORMAT STATEMENT FREEMAN, (JONES)
2. DATA TYPES, IDENTIFIERS LAIRD*, GREENFIELD, ENGEL,
NOLL, YOUNG, (GIBSON, INFANTE)
3. NONEXECUTABLE STATEMENTS, HOLBERTON*, MIMMACK, BOSWELL,
PROCEDURES AND SUBPROGRAMS, (KLEIN, FLETCHER, YOUNG, BARTH}
INTER-INTRA PROGRAM
RELATIONSHIPS
4. EXPRESSIONS BAILEY*, JONES,
INFANTE, (COLEMAN)
5. EXECUTABLE STATEMENTS KARP*, COLEMAN,
(EATON, INFANTE)
6. INPUT/OUTPUT EXCEPT FORMAT BARTH*, CAMPBELL, GIBSON,
KLEIN, FLETCHER,
(NOLL, FREEMAN)
* DENOTES CHAIRMAN
NAMES IN PARENTHESES INDICATE A SECONDARY RESPONSIBILITY,
GROUP CORRESPONDENCE IS TO BE DISTRIBUTED TO THE COMMITTEE.
IDEAS FOR CONSIDERATION AS EXTENSIONS
STATUS CODE A=APPROVED N=NOT RECOMMENDED
1=MUST BE IN STD 2=SHOULD BE IN STD
3=RECOMMENDED 4=OF SOME VALUE
5=OF TRIVIAL VALUE
* MANUF. CODE C=CDC G=GE H=HONEYWELL I=IBM R=RCA U=UNIVAC X=XDS
B=BURROUGHS D=DEC
NO.-GRP-STATUS EXTENSION OR REVISION IDEA
1-4-1 MIXED REAL-INTEGER ARITHMETIC.
(C,G,H,I,R,U,X,B,D)*
2-2-A APOSTROPHE AS HOLLERITH DELIMITER.
(C,G,H,I,R,U,X,B,D)
3-5-A MULTIPLE REPLACEMENT ASSIGNMENT STATEMENT.
(C,G, , , , ,X, , )
4-3-1 MULTIPLE ENTRY SUBPROGRAMS.
(C, , ,I,R,U,X,B, )
5-1-3 SPECIFY A STATEMENT,DELIMITER CHARACTER. ALLOW MORE THAN
ONE STATEMENT PER LINE.
(C,G, , , , ,X, , )
6-2-1 HAVE HOLLERITH OR CHARACTER DATA TYPE. ALLOW HOLLERITH
CONSTANTS IN ASSIGNMENT AND IF STATEMENTS.
(C,G, , , , , ,B, )
7-2-2 MORE THAN THREE DIMENSIONS
(C,G, ,I,R,U,X,B,D)
8-3-2 NONSTANDARD RETURN STATEMENT.
(C,G, ,I,R,U,X,B,D)
9-6-2 REREAD OR DECODE.
(C,G,H, , ,U,X,B,D)
10-6-2 ENCODE OR EQUIVALENT.
(C, , , , ,U,X, ,D)
11-2-2 AUTOMATIC TYPING OF FUNCTION NAMES. I.E. TYPE OF ARGUMENT
DETERMINES WHICH FUNCTION IS ACTUALLY USED.
( , , ,I, ,U, , , )
12-3-5 DATA IN TYPE-STATEMENTS.
( , , ,I,R,U,X,B, )
13-3-1 IMPLICIT STATEMENT.
( , ,H,I,R,U,X,B,D)
14-4-N IMPLIED MULTIPLICATION AFTER A RIGHT PARENTHESIS.
( , , , , , , , , )
15-5 LOGICAL MASKING STATEMENTS.
(C, , , , ,U,X,B,D)
16-1-2 PROGRAM STATEMENT TO IDENTIFY MAIN PROGRAMS.
(C, ,H, ,R, , , , )
17-1-3 SPECIAL FORMAT DESCRIPTOR FOR VERTICAL SPACE CONTROL.
( , , , , , , , , )
18-6-2 PARITY ERROR DETECTION ON READING.
(C, , ,I,R,U,X,B,D)
19-6-1 END FILE DETECTION ON READING.
(C,G, ,I,R,U,X,B,D)
20-2 GENERALIZATION OF SUBSCRIPT EXPRESSIONS.
(C,G, ,I, ,U,X,B,D)
21-6-3 NAMELIST STATEMENT.
(C, , ,I,R,U,X,B,D)
22-3-3 INTERNAL SUBPROGRAMS.
( , , , , ,U,X, , )
23-6-2 FREE FORMAT INPUT DATA, E.G., USE A COMMA AS FIELD DELIMITER.
( ,G,H,I, ,U,X,B,D)
24-2-N NAMES LONGER THAN SIX CHARACTERS.
(C,G, , , , ,X,B, )
25-1 IGNORE TRAILING BLANKS IN INPUT DATA.
( , , , , , , , , )
26-6-N ALLOW HOLLERITH INFORMATION IN OUTPUT LISTS
WITHOUT ASSOCIATED FORMAT FIELD DESCRIPTOR.
( ,G, , , , , , , )
27-3-2 IMPLIED DO LOOP IN DATA STATEMENT.
(C, ,H, , ,U,X,B,D)
28-3-2 ARRAY NAME WITHOUT SUBSCRIPT IN DATA STATEMENT.
(C,G, ,I,R,U,X,B,D)
29-4-2 GROUP SUCCESSIVE EXPONENTIATIONS RIGHT TO LEFT.
( ,G, ,I,R, ,X,B, )
30-3-1 EXPAND BASIC EXTERNAL FUNCTION LIST.
(C,G,H,I,R,U,X,B,D)
31-3 ADD SOME BASIC SUBROUTINE NAMES.
(C,G,H,I,R,U,X, ,D)
32-5 GENERALIZE DO STATEMENT, I.E., ALLOW EXPRESSIONS
WITH ZERO AND NEGATIVE VALUES.
( , , , , ,U,X,B,D)
33-6-A ALLOW VARIABLE FORMAT STATEMENT LABELS IN I/O STATEMENTS,
E.G., ALLOW INTEGER VARIABLE NAME THAT HAS BEEN ASSIGNED
A STATEMENT LABEL 'VALUE' IN AN ASSIGN STATEMENT.
( , , , , ,U,X, , )
34-2 ALLOW MORE PRECISE PRECISION REQUIREMENTS IN TYPE-STATEMENTS.
( , , ,I,R, , ,B, )
35-1 ALLOW ADDITIONAL FORMAT FIELD DESCRIPTORS, E.G., R AND T.
(C, ,H,I,R,U,X,B,D)
36-6-3 DEFINE STANDARD ,INPUT DATA SENTINEL AND MAKE AVAILABLE
NUMBER OF NUMBERS AND/OR NUMBER OF LINES READ.
( , , , , , , , , )
37-0-N SPECIFY SOME LIMITS FOR PROGRAMS, E.G., NUMBER OF LABELS
ALLOWED IN COMPUTED GOTO STATEMENT.
( , , , , , , , , )
38-3-3 PARAMETER STATEMENT OR EQUIVALENT.
( , , , , ,U, , , )
39-2 SOME STANDARD WAY OF PACKING AND UNPACKING MORE THAN ONE
INTEGER VALUE IN ONE STORAGE UNIT.
( , , ,I,R,U, , , )
40-6-1 DIRECT ACCESS L/O STATEMENTS.
( ,G,H,I,R, ,X,B,D)
4L-3-4 STANDARD OVERLAY OR SEGMENTATION SCHEME.
(C,G,H,I, , , ,B,D)
42-6-4 ABILITY TO READ TAPE RECORDS OF UNKNOWN MODE AND LENGTH.
( , , , , , ,X, , )
43-1-3 EXPLICIT FORMAT DESCRIPTION TO SPECIFY RESCAN POINT.
( , , , , , , , , )
44-6-2 OPEN AND CLOSE FILES.
(C,G, ,I, ,U, ,B,D)
45-2 STATEMENT LABEL DATA TYPE.
( , , , , , , , , )
46-5-N REPEAT STATEMENT.
( , , , , , ,X, , )
47-1-3 CONDITIONAL COMPILATION.
( ,G, , , ,U,X, ,D)
48-3-N DEBUG FACILITIES.
(C,G,H,I,R,U,X,B,D)
49-4-3 ERROR CONDITIONS IN COMPUTATIONS.
(C,G,H,I,R,U,X, ,D)
50-5-N ERROR HANDLING MECHANISMS.
( , , ,I, , ,X, , )
51-5-N MULTI-TASKING FACILITIES.
( , , , , ,U, , , )
52-2 CLUSTER DATA.
( , , , , , , , , )
53-1-N MACRO FACILITY
( , , , , ,U, , , )
54-2 STORAGE CLASSES.
55-6-2 BLOCK I/O TRANSFER.
(C, , , , , ,X,B, )
56-6-1 INCLUSION OF READ, PRINT, AND PUNCH STATEMENTS.
(C,G,H,I,R,U,X,B,D)
57-1-A ADDITIONAL SPECIAL CHARACTERS.
(C,G,H,I,R,U,X,B,D)
58-1-2 LOWER CASE ALPHABETICS.
( ,G, , , , , , , )
59-1-3 EMBEDDED COMMENTS.
( , , , , ,U, ,B, )
60-5-A NOT RESTRICT STRINGS IN STOP AND PAUSE TO OCTAL DIGITS.
(C, ,H,I, ,U,X,B,D)
61-2 STRUCTURE DATA TYPE.
( , , , , , , , , )
62-5-N NULL STATEMENT LABELS, E.G. IN IF STATEMENTS.
( , , , , ,U, , , )
63-2-A DELETE SECOND LEVEL DEFINITION.
64-6-2 ALLOW EXPRESSIONS IN OUTPUT LISTS.
( , , , , , ,X,B, )
65-1-N ALLOW COMMENTS BETWEEN CONTINUATION LINES.
( , , , , , , , ,D)
66-3-5 ALLOW 'VAR.=CONSTANT' IN DATA STAT.
(C, , , , , , , , )
67-3-4 ALLOW ARRAY NAME WITHOUT SUBSCRIPT IN EQUIVALENCE.
( ,G,H, , ,U,X,B,D)
68-5-N ALLOW ASSIGNED GOTO IN DIFF. PROG. UNIT THAN ASSIGN.
( , , , , , , , , )
69-2 ALLOW TWO STORAGE UNIT INTEGERS, ESP. IN COMMON.
( , , , , ,U, , , )
70-2 SPECIFY SOME PROCESSOR-DEFINED CONSTANTS, E.G. PI.
( , , , , , , , , )
71-1 ALLOW PARENTHESES AS FIELD SEPARATORS IN FORMATS.
( , , , , , , , , )
72-6-3 ALLOW READING MORE CHARS. THAN WRITTEN, GET BLANKS.
(C, , , , , ,X, , )
73-6-A ALLOW CONSECUTIVE SLASHES TO CAUSE BLANK PRINT LINES.
(C,G,H,I,R,U,X,B,D)
74-1-5 BLANK INPUT FIELD DETECTION.
(C, , , , , , , , )
75-6-N NEG, I/O UNIT IMPLIES BACKWARD OPERATION.
( , , , , , , , , )
76-3-N ALLOW INITIALIZATION OF BLANK COMMON.
( , , , , , , , B, )
77-3-1 SPECIFY UNITS OF STD. FUNCTION ARGS. AND RESULTS.
(C,G,H,I,R,U,X,B,D)
78-2 ALLOW SUBSCRIPT EXPRESSIONS LESS THAN ONE,
( , , , , , ,X, ,D)
79-5-N ALLOW TWO STAT. NO. IF STATEMENT.
(C, , , , , ,X, ,D)
80-5 REDUCE RESTRICTIONS ON EXTENDED RANGE.
( , , , , , ,X,B, )
81-2 ALLOW STATEMENT NOS. AS DATA, E.G. WITH TRAILING S.
82-3-N DYNAMIC STORAGE ALLOCATION OF ARRAYS.
( , , , , , , ,B, )
83-1 COMPLEX FORMAT FIELD DESCRIPTOR.
( , , , , , , , , )
84-1 IGNORE EMBEDDED BLANKS IN NUMERIC INPUT DATA.
( , , , , , , , , )
85-5-A MAKE CHARS. IN STOP ACCESSIBLE.
(C, ,H,I, ,U,X, ,D)
86-1-2 ALLOW MORE PARENTHESE LEVELS IN FORMATS.
( , , , , , ,X,B,D)
87-4-2 ALLOW ARRAY ELEMENTS IN STATEMENT FUNCTION DEFINITION.
(C, , , , , ,X,B,D)
88-5 DO NOT ALLOW EXTENDED RANGE.
( , , , , , , , , )
89-2 BINARY OR BOOLEAN DATA TYPE.
( , , , , ,U, , , )
90-4-1 MIXED REAL, INTEGER, AND DOUBLE PREC. ARITHMETIC.
91-5-A ALLOW ALL ARITH. TYPES OF ASSIGNMENT IN ARITHMETIC
ASSIGNMENT STATEMENT.
92-5-A MAKE COMP. GO TO ACT LIKE CONTINUE WHEN VAR. OUT OF RANGE.
93-5 EXECUTE DO LOOP ONCE WHEN M1>M2.
94-4-N ALLOW MATRIX ARITHMETIC.
95-1 ALLOW G FORMAT DESCRIPTOR WITH ANY ARITH. VARIABLE.
96-1 X FORMAT DESCRIPTOR SKIPS CHARS. ON OUTPUT RATHER THAN
INSERT BLANKS (AT LEAST AFTER T IS USED TO REPEAT COLS.)
97-3-2 ALLOW FUNCTION NAME IN TYPE-STAT. WITHIN THAT FUN. SUBP.
98-1-3 R FORMAT FIELD DESCRIPTOR.
99-6-2 SPECIFY ACTION FOR OMITTED FORMAT STATEMENT.
100-1 SPECIFY ACTION IF OUTPUT NUMBER EXCEEDS FIELD WIDTH.
101-6-A REWIND HAS NO EFFECT WHEN UNIT IS AT INITIAL POSITION.
102-5-A MAKE COMMA AND LIST OPTIONAL IN ASSIGNED GOTO.
103-5-A ALLOW INTEGER EXPRESSION IN COMPUTED GOTO.
104-3-4 ALLOW NO. OF CONSTS. TO DIFFER FROM NO. OF VARS. IN DATA STATEMENT.
105-3-N MAKE MOD FUNCTIONS ALWAYS GIVE NON-NEGATIVE RESULTS.
106-5-A ALLOW OPTIONAL COMMA AFTER STAT. NO. IN DO STAT.
107-3-1 ALLOW NAMING OF BLOCK DATA SUBPROGRAMS.
108-3-1 ALLOW MORE THAN ONE BLOCK DATA SUBP. IN ONE PROGRAM.
109-1 PROVIDE NUMERIC OUTPUT FOR STD, INFO. INTERCHANGE.
110-5-N ALLOW MORE THAN ONE STAT. AS PART OF LOGICAL IF.
111-3 ALLOW REFERENCING BEYOND DUMMY ARGUMENT ARRAY.
112-3-1 ALLOW EQUIVALENCE TO GIVE TWO NAMES TO ONE VARIABLE.
113-3-1 ALLOW SUBPROGRAMS TO RETAIN LOCAL VARS. BETWEEN REFS.
114-3-N DO NOT REQUIRE LABELED COMMON BLOCKS TO BE SAME LENGTH.
115-6-3 DEFINE RECORD LENGTH.
116-6-5 PROVIDE FOR MULTI-REEL FILES.
117-1-3 ALLOW NON-FORTRAN CHARS ON COMMENT LINES.
118-1-4 MAKE END LINE A STATEMENT,
119-1 SPECIFY NPEW.D OUTPUT WHEN N>0.
120-4-3 ALLOW CONTROL OF ROUNDING IN FL. PT. OPERATIONS.
121-1 PICTORIAL FORMAT SPECIFICATIONS.
122-6-5 SPECIFY NORMAL SYSTEM I/O UNIT NUMBERS.
123-4-2 ADD < AND > TO CHARACTER SET AND ALLOW IN REL. EXPR.
124-2 DO NOT REFER TO A BLOCK NAME AS A DATA NAME IN STD.
125-3-1 COMBINE BASIC EXT. FUNS. AND INTRINSIC FUNS.
126-3-2 SPECIFY STATUS OF USER SUBP. WITH NAME OF BASIC EXT. FUN.
127-6-N GET CONVERSION WHEN VAR. TYPE AND FORMAT DISAGREE.
128-2 CONSTANT AND INPUT VALUE OF IDENTICAL FORM MUST HAVE SAME VALUE.
129-1-N ALLOW ALPHANUMERIC STAT. LABELS.
130-3-4 ADD MORE TRUNCATION AND REMAINDERING FUNCTIONS.
131-3-2 REPEAT MORE THAN ONE CONSTANT IN DATA STATEMENT.
132-2 SPECIFY PARTS CF COMPLEX NO. IN STORAGE.
133-1 SPECIFICATION OF INTEGERS IN FORMATS DURING EXECUTION.
134-3-4 ALLOW SUBP. REFS. TO USE LESS THAN NO. OF DUMMY ARGS.
135-3-5 ALLOW NAMELIST TO BE USED FOR A LIST OF ACTUAL ARGUMENTS.
136-4-3 MORE PERMISSIVE LOGICAL IF, E.G. IF(I.EQ.2.OR.3).
ALSO MORE PERMISSIVE LOGICAL EXPRESSIONS.
137-3-1 ALLOW COMMON TO BE INITIALIZED IN ANY PROGRAM UNIT.
138-5 MAKE END LINE IMPLY A STOP IN MAIN PROGRAM.
139-5 MAKE END LINE IMPLY A RETURN IN SUBPROGRAM.
140-5 ALLOW BRANCHING INTO RANGE OF ANY UNSATISFIED DO.
141-5 ALLOW REAL NOS. AS PARAMETERS IN DO STATEMENT.
142-6-2 ALLOW HOLLERITH CONSTANTS IN OUTPUT LISTS WITH
ASSOCIATED FORMAT FIELD DESCRIPTOR.
143-6-4 PROGRAM INTERLOCKS.
144-1 DO NOT ALLOW NON-FORTRAN IN PROCEDURES IN STD PROGRAMS.
145-5 DEFINE VALUE OF CONTROL VAR. AFTER SATISFYING DO LOOP.
146-5 PROVIDE FOR INDICATING ABNORMAL TERMINATION OF A PROGRAM.
147-3 ALLOW SUBARRAY SPECIFICATION STATEMENT.
148-4-3 ALLOW ARRAY ARITHMETIC.
149-4-2 ALLOW THREE NEW CHARACTERS FOR .AND., .OR., AND .NOT.
150-6 ALLOW MORE THAN ONE FILE ON ONE I/O UNIT.
151-3 MAKE COMMA AFTER SLASH IN DATA STAT. OPTIONAL.
152-3 MAKE COMMA BETWEEN LISTS IN EQUIVALENCE STAT. OPTIONAL.
153-2 SPECIFY PARTS OF DOUBLE PREC. NO. IN STORAGE.
APPENDIX F [of X3J3 minutes of August 1971]
EXTENSIONS AND REVISIONS APPROVED
1. INCLUDE APOSTROPHE IN CHARACTER SET. (/21,4.1)
2. ALLOW DECIMAL DIGITS IN STOP AND PAUSE. (/21,4.2)
3. MAKE CHARS. IN STOP STAT. ACCESSIBLE. (/21,4.3)
4. REMOVE CONCEPT OF $ECOND LEVEL DEFINITION. (/24,4.1)
5. ALLOW APOSTROPHES TO DELIMIT HOLLERITH STRINGS. (/24,4.2)
6. ALLOW T FORMAT FIELD DESCRIPTOR. (/24,4.3 AND 4.4)
7. ALLOW APOSTROPHE TO DELIMIT CHAR. STRING IN PAUSE. (/24,4.5)
8. SPECIFY THAT CONSECUTIVE SLASHES IN FORMATS CAUSE
BLANK LINES WHEN PRINTED. (/24,4.6)
9. ALLOW INTEGER EXPRESSION IN COMPUTED GOTO. (/21,4.1)
1O. WHEN COMPUTED GOTO EXPR. IS OUT OF RANGE, EXECUTE
FOLLOWING STATEMENT NEXT. (/25,4.1)
11. ALLOW AS TERMINAL STAT. OF DO LOOP ANY EXEC. STAT. THAT
ALLOWS EXECUTION OF THE FOLLOWING STAT. (/25,4.2)
12. REQUIRE ASSIGN STAT. AND ASSOCIATED ASSIGNED GOTO STAT.
TO BE IN SAME PROGRAM UNIT. (/25,4.3)
l3. ALLOW ASSIGN STAT. TO SET VAR. FORMAT. (/25,4.4)
l4. FIX NO. OF RANGE EXECUTIONS AND INCR. AT DO STMT. (/25,4.5)
15. ALLOW NEG. VALUES IN DO PARAMETERS. (/25,4.5)
16. ALLOW REDEF. OF DO PARAMETERS WITH NO EFFECT ON NO.
OF ITERATIONS, (/25,4.5)
17. ALLOW MULTIPLE ASSIGNMENT STATEMENTS. (/27,4.3)
18. OPTIONAL COMMA IN COMPUTED AND ASSIGNED GOTO (/27,4.4)
19. OPTIONAL LIST IN ASSIGNED GOTO. (/21,4.5)
20. OPTIONAL COMMA IN DO AFTER STAT. LABEL. (/27,4.6)
21. ALLOW FUNCTIONS WITH NO ARGUMENTS. (/27,4.7)
22. ALLOW ALL TYPES OF ARITHMETIC ASSIGNMENT STATEMENTS.(/29,4.1)
23. ALLOW INTEGER EXPRESSIONS IN DO STATEMENTS. (/29,4.2)
24. REWIND HAS NO EFFECT AT INiTIAL POSITION. (/29,4.3)
25. ALLOW END= IN READ STATEMENTS. (/29,4.4)
REAL VARIABLE IN DO
ZERO-TRIP COUNT
IMPLICIT STATEMENT