J3/07-008 Date: 2006/01/01 To: J3 From: Malcolm Cohen Subject: Editorial enhancements for future Draft revisions This is the editor's list of things which have been and are being considered for future Draft revisions. Sections A and B have been applied to the LaTeX sources and are included in the 007. These sections are by reference to 04-007, and are now closed. Sections C to E will become progressively smaller; suggestions that are implemented will be removed from this document. The next revision of this document will remove all 04-007 references and become a simple list of changes that are either under consideration or rejected for the time being. A. Post-DIS pre-publication edits --------------------------------- Except for item 0, these are not in 04-007. 0. Revert to working draft form for headings, cover page, and line numbering. 1. Edits for table 15.2 in c15.tex Replace line 419 by \multicolumn{3}{|l|}{} \\[-10pt] (making it the same as line 409). Replace the "|l|" at line 483 by "|p{5.625in}|". 2. Change comma to period at end of [126:4] 3. Delete stray '"' at end of [367:35] 4. [409:27] Change "index-name" to "". B. Agreed post-publication edits -------------------------------- [throughout] Hyperlinks should be black (same as non-hyperlink text) not red. [xiii] Replace with boilerplate ready for addition of F2008 feature descriptions. [3:10-] Insert new subclause 1.6.1 "Fortran 2003 compatibility" with obvious contents waiting for something incompatible to be done. [3:11-12] Delete no longer applicable "the preceding ... Standard,". [36:12] Change "A processor shall" to "The processor shall". [37:27] Ditto. [40:12] Ditto. [44:1] Ditto. [42:11+] Change "instances" to "occurrences". [45:8] Index "statements!TYPE". [73:11] Insert "An" between "(R504)" and "". [86:13] Remove blanks/newlines between \mindex items; this causes excessive space (here, between "module." and "The"). [112:26] Change reference for "allocation transfer procedure" to point to the actual definition of MOVE_ALLOC (13.7.82) instead of to the unhelpful one-line paragraph. [117:27+5] Replace "" by "" [117:27+6] Delete. [118:1-1-] Insert new lines to Note 7.1 "X%KIND "KIND ". [139:3-1] Add whitespace inside the bottom of table 7.8. [158:2] \mindex space removal between "The" and "CASE". [160:7] Index "statements!ASSOCIATE". [162:6] Index "statements!SELECT TYPE". [162:16] Index "statements!TYPE IS". [162:17] Index "statements!CLASS IS". [163:6] Insert "kind" after "and". [165:36] Font change "action-stmt" -> "". [167:25] \mindex space removal between "a" and "CYCLE". [168:4] \mindex space removal between "The" and "EXIT". [169:9] "or" should not be in obsolescent font. [170:18] \mindex space removal between "a" and "CONTINUE". [209:13] Change "flush" to "FLUSH". [211:7+] Insert a constraint parallel to C946 at [209:12-13]. [259:11] Typeset ", alternate return," in obsolescent font. [265:18] Remove extraneous quotation mark at the end of the line. [267:20] Delete "used in the ". [277:18+1] Change "instances of a generic function" to "specific procedures of a generic procedure" [277:18+2] Change "Given" to "For example, given". [292:26] \mindex space removal between "a" and "model". [307:26] Insert "for the default character type" at the end of the sentence. [331:16-17] Change all forward and backward double quotes to forward single quotes. [334:21-22] Change all forwards and backwards double quotes to forwards single quotes. [335:40-41] Change "true... than 1" to "independent of the declared lower bounds for A". [336:3] Change "true ... than 1" to "independent of the declared lower bounds for B". [395:28] Typo "CPTR" -> "CPTR%PX." (Hint: % is the LaTeX comment char.) [409:21] Change "implied-DO" to "implied DO". [419:1] Change "procedure instance" to "subprogram instance". [437:34+1-2] Replace with "ISO R 1539-1972 {\sc Fortran} 66 ISO 1539-1980 {\sc Fortran} 77" C. Edits under consideration ---------------------------- 1. [279:22-25] Constraints C1234 and C1235 refer to RESULT which no longer appears in the referenced syntax rule R1224. Perhaps they should be reworded, moved, or the reference changed. 2. The construction "...; otherwise ..." sometimes has a comma after the otherwise, sometimes not. Even when the elided text is otherwise identical. The editor should pick one and change all the dozens of others to match. 3. "alternatively may be" and "may alternatively be" should be consistent. And grammatical. 2 in c03, 2 in c05 (see [74:25]), 1 in c10. 5. [144:5-6] and [267:15-17]. The texts of C727 and C1229 should be identical. 6. Sometimes we say "neither the XYZ attribute nor the ABC attribute", sometimes "neither the XYZ nor ABC attribute". These should be consistent. One occurrence of each in c04, one of the first in c07, one of the second in c12. 8. [368:22] could be read as defining IEEE_OTHER_VALUE as being in the "normal" class. D. Rationale for accepted edits ------------------------------- This section contains explanations for some (less obvious) editorial changes. 1. Problem: Note 7.1 (Examples of a ) uses syntax terms that do not appear in the definition of a , and does not include examples of new syntax terms. In particular, and have become and we have new and Solution: Change and delete (the former example seems more useful). Insert example of . We could ignore the , but for completeness I've added it. Edits: to [117] and [118]. E. For future consideration --------------------------- 1. [83:24+] Define <>. 2. [280:23-25] Delete "The specifications ... (12.2)" because it duplicates 12.2. If "information in the procedure heading" includes RECURSIVE -- which appears not to be excluded -- it's incorrect as well. 3. [407:24ff] 16.2.3 belongs in 12.3.2 (perhaps with constraints, since it won't be in 16 any more). 4. Fix item 21 in 16.5.5 5. Delete C530 as being redundant with C1238. 6. Delete [411:36] sentence saying that local identifiers of a subprogram are not accessible to its host. Or reword for any scoping unit. And/or make it into a note. F. Rejected suggestions. ------------------------ 1. Q. [71:10] Index "statements!TYPE". A. This is not a TYPE statement. We don't index the other keywords here either (CLASS statement, INTEGER statement, REAL statement, ...). 2. Q. [75:8] Index "TYPE" A. There is already an index entry on this very page for "TYPE type specifier", which happens to be the correct entry. We don't index bare keywords. 3. Q. [162:2] Index "SELECT TYPE" A. We don't index bare keywords. 4. Q. [148:4-5] Reword to make it clearer what happens if the is scalar. A. It is already clear enough, rewording would just make it longer. 5. Q. Introduce a syntax term for so that we know what it means. A. There does not appear to be any problem with the existing text. 6. Q. [26:40] Replace "be given to" by ``identify''. A. This would cause interpretation problems with [27:2-7]. 7. Q. [xiii, item(3)]: "complex" => "complicated". A. This page will be replaced in F2008, so making this change is a waste of time.