To: J3 Members J3/16-130 From: Van Snyder Subject: Editorial remarks concerning Clause 13 Reference: 16-007 Date: 2016 January 25 [327:11 13.3.1p5] In the last line of 13.3.1p5, insert a comma before "and". [328:12+2 NOTE 13.4] In the first line of NOTE 13.4, replace "of these" with "in the descriptions of". [328:12+5 NOTE 13.4] The formula at the end of NOTE 13.4 could be the same as in 13.4p3 because f1 is defined to be nonzero at [328:3]. [331 Table 13.1 EPSILON] Description is not informative. Replace it with "Smallest positive model number such that the result of adding it to 1 is different from 1." [331 Table 13.1 FRACTION] Insert "floating-point" before number. [332 Table 13.1 ISHFTC] Replace "of the" with "of the SIZE". [333 Table 13.1 RANGE] Delete "(13.4)". It doesn't appear in the description of most other functions that inquire properties of the floating-point model, e.g., EPSILON, PRECISION, RADIX, or TINY. [334 Table 13.1 SPACING] Delete "(13.4)". It doesn't appear in the description of most other functions that inquire properties of the floating-point model, e.g., EPSILON, PRECISION, RADIX, or TINY. [343:10 13.8.20p3 VALUE] Replace "representable in ATOMIC_INT_KIND" with "representable by an integer with kind specified by the ATOMIC_INT_KIND named constant from the ISO_FORTRAN_ENV intrinsic module (13.9.2.2)" [343:24-25 13.8.21p3 VALUE] Replace "representable in ATOMIC_INT_KIND" with "representable by an integer with kind specified by the ATOMIC_INT_KIND named constant from the ISO_FORTRAN_ENV intrinsic module (13.9.2.2)" [344:35 13.8.24p3 VALUE] Replace "representable in ATOMIC_INT_KIND" with "representable by an integer with kind specified by the ATOMIC_INT_KIND named constant from the ISO_FORTRAN_ENV intrinsic module (13.9.2.2)" [345:9-10 13.8.25p3 VALUE] Replace "representable in ATOMIC_INT_KIND" with "representable by an integer with kind specified by the ATOMIC_INT_KIND named constant from the ISO_FORTRAN_ENV intrinsic module (13.9.2.2)" [345:27-28 13.8.26p3 VALUE] Replace "representable in ATOMIC_INT_KIND" with "representable by an integer with kind specified by the ATOMIC_INT_KIND named constant from the ISO_FORTRAN_ENV intrinsic module (13.9.2.2)" [346:3-4 13.8.27p3 VALUE] Replace "representable in ATOMIC_INT_KIND" with "representable by an integer with kind specified by the ATOMIC_INT_KIND named constant from the ISO_FORTRAN_ENV intrinsic module (13.9.2.2)" [346:21-22 13.8.28p3 VALUE] Replace "representable in ATOMIC_INT_KIND" with "representable by an integer with kind specified by the ATOMIC_INT_KIND named constant from the ISO_FORTRAN_ENV intrinsic module (13.9.2.2)" [347:9-10 13.8.30p3 VALUE] Replace "representable in ATOMIC_INT_KIND" with "representable by an integer with kind specified by the ATOMIC_INT_KIND named constant from the ISO_FORTRAN_ENV intrinsic module (13.9.2.2)" [351:9 13.8.42p5] Delete "model for the"; insert "described" before "in 13.3". Compare to, e.g., 13.6.40p6 at [350:19-20]. [360:14 13.8.63p1] Description is not inforative. Replace it with "Smallest positive model number such that the result of adding it to 1 is different from 1." [361:29,33 13.8.67p4] Replace "will have" with "has" twice. Using "will have" leaves one to wonder "when will that be?" [366:43-44 13.8.77p3 LENGTH] Replace "It is processor ... blanks" with "The significant length may include trailing blanks if the processor allows command arguments with significant trailing blanks." Compare to [366:5-6 13.8.76p3 LENGTH]. [367:31,33 13.8.78p3 VALUE] Move "VALUE is assigned all blanks", preceded by a comma, to the end of the sentence. Capitalize first "if". [367:39 13.8.78p3 STATUS] Replace "set to" with "assigned the value". [367:40-41 13.8.78p3 STATUS] Delete "STATUS is set to -1". Capitalize "if". Insert ", it is assigned the value -1" after the first "environment variable". [367:41-42 13.8.78p3 STATUS] Delete "It is assigned the value 1". Capitalize "if". Insert "it is assigned the value 1." after "exist,". [367:42 13.8.78p3 STATUS] Replace "or 2 if" with "If". After "variables" insert "it is assigned the value 2". [368:23 13.8.79p5] Replace "and" with "or". [372:5 13.8.88p5] After "collating sequence" insert "associated with the specified kind type parameter". Compare to 13.8.44p5 at [351:31-32]. [373:37 13.8.92p5(iv)] The final sentence is wrong. Replace it with "The interpretation of a bit sequence with $w_{z-1} = 1$, where $z$ is the model parameter in 13.3, is processor dependent." [376:26 13.8.101p3 ARRAY] Before "shall" insert "may be of any type; it" [377:10 13.8.101p8] Add another example. Replace "and" with a comma. Append "and LBOUND (A(2:3,7:10)) is [1,1]". Without careful reading of subclause 6 (quite far distant), some readers assume the result would be [2,7]. [377:37 13.8.103p5] Replace "leftmost" with "most significant". [378:32 13.8.106p3 STRING_A] Insert "of type" before "default". Compare to, e.g., VERIFY, which says "shall be of type character", not "shall be character". [379:9 13.8.107p3 STRING_A] Insert "of type" before "default". Compare to, e.g., VERIFY, which says "shall be of type character", not "shall be character". [379:21 13.8.108p3 STRING_A] Insert "of type" before "default". Compare to, e.g., VERIFY, which says "shall be of type character", not "shall be character". [380:5 13.8.109p3 STRING_A] Insert "of type" before "default". Compare to, e.g., VERIFY, which says "shall be of type character", not "shall be character". [389:16+ 13.8.129p3+] Insert a paragraph (or maybe a note): "If FROM and TO are the same entity, they both become deallocated because TO is INTENT(OUT)." {This is the subject of an interp, which it would be nice to moot.} [390:24 13.8.130p3 S] Append "unless the processor distinguishes positive zero from negative zero". [390:29+3 NOTE 13.20] Replace "On many systems there are" with "A processor might have". "On many..." is questionable. [394:29 13.8.139p4] Insert "a" before "scalar". [397:19 13.8.147] Replace "in" with "on". [398:35 13.8.150p1] Delete "(13.4)". It doesn't appear in the description of most other functions that inquire properties of the floating-point model, e.g., EPSILON, PRECISION, RADIX, or TINY. [400:29 13.8.153p5(ii)] Replace "is" with "are". [402:32 13.8.157p6] Replace "will be" with "is". Using "will be" raises the question "When will it be?" [408:19 13.8.172p1] Delete "(13.4)". It doesn't appear in the description of most other functions that inquire properties of the floating-point model, e.g., EPSILON, PRECISION, RADIX, or TINY. [409:15 138.8.174p3] Replace "Unless X is complex" with "If X is not complex" or "If X is real". We don't usually use "Unless" where there are only two possibilities. [412:30 13.8.182p5] Replace "rightmost" with "least significant". [414:9 13.8.186p3 ARRAY] Before "shall" insert "may be of any type; it" [414:33 13.8.186p6] Replace "and" with a comma and append ", and UBOUND ( ( A ) ) is [2,4]". Without careful reading of subclause 6 (quite far distant), some readers assume the result would be [3,10]. [417:1 13.9.2.1p2] Replace "type" with "types" (there are two now). [420:17 13.9.20p1] Before "(9.6.4.3)" insert "or the unit used for a PRINT statement". Compare to 13.9.2.11 INPUT_UNIT, which specifies that it applies to a READ statement without a control information list. [510:37+] Insert a list item: " o the values assigned to INTENT(OUT) or INTENT(INOUT) arguments to some intrinsic subroutines (13)." Otherwise, we'd need several more list items. [511:4] After "ERRMSG" insert ", EXITSTAT". [511:10+] Insert a list item: " o the order of command arguments (13.8.77)." [511:18] Move after [511:20] [511:20] Replace "(13.6)" with "(13.8.148)". [511:23] Delete list item for 13.8.177 because it's covered by the one added at [510:37+]. [511:26] After "(13.9.2)" insert ", and the lengths of those values". [512:15-16] move to [510:38+].