J3/07-102 To: J3 From: Malcolm Cohen Subject: Editorial Report at m178 Date: 2006/11/30 1. Introduction This paper responds to the editorial comments in the papers at meeting 178 which were passed directly to /EDIT. 2. Responses by paper 06-292 ------ Agreed in principle. Did the first edit. Second edit deferred pending interp resolution (and I would omit the final "types"). Re the PS: The editor does not agree with the assertion that unless qualified, "type" means "dynamic type". Often it means "declared type", and sometimes it means "both declared and dynamic type". The lack of a general rule is mostly not a problem, because it is mostly blindingly obvious which meaning it is. For example, when referring to declarations, it has to mean "declared type" since declarations don't have a dynamic type. 06-299 ------ - As the paper implied (but did not say) deleted the witter at [326:2-3], i.e. in between the 12.6.2 heading and the 12.6.2.1 heading, because it is redundant with 12.6.2.3 (which it references). - Reworded the first paragraph to make it more direct, and remove terms we don't use. - Reworded the second paragraph to correct it, viz "procedure"->"subprogram". - Deleted Note 12.40a instead of moving it, since it is now said in normative text in the first paragraph. 06-300 ------ "this standard" fixed 8:13, 50:1, 52:8, Also fixed several not noted in the paper: [460:30-31], [543:16-17], [550:15-16], [564:2-3] Also fixed [442:19] "the standard"->"this part of ISO/IEC 1539". Ditto [505:4]. Also, after "The wording of a definition here is not necessarily the same as" changed "in the standard" to "its normative definition". (The previous was self-contradictory.) Can we not just delete annex A? Again "the standard" at [507:21,22,23,24,25]. [508:38] "the standard" -> "this edition of this part of ISO/IEC 1539". Again "The standard" at [535:40] -> "This part of ISO/IEC 1539". Similarly [545:9], [545:35], [563:19], [30:7+2], [31:17+2]. And [xv:6] "this part of the standard" -> "this part of ISO/IEC 1539". Also [554:38] "The standard set operations" -> "The usual set operations". Deleted [1:15-16] "Throughout ... 1539-1.". Reworded [3:2-3]. Did 20:12 at 20:29. For 124:13, deleted irrelevant nonconformant witter instead, viz the entirety of [124:11-17+2]. For 68:14-15, instead, I split syntax rule R453 into two productions, one with (), the other with [ => ], and deleted C468. In the () one, made the double colon and the nonoptional (the DEFERRED keyword is required so it cannot validly be omitted). I think this is much easier to read. For 129:4-7, I also - incorporated the text, suitably modified, about not depending on anything allocated/deallocated in the statement (so this now appears once instead of twice). - Note: the text for deallocate didn't mention length type parameters, so this is a bug fix to the standard. - Note: the text for allocate prohibited any dependence on length type parameters, but since only deferred type parameters can vary by ALLOCATE/DEALLOCATE, I have fixed that too. - Also extracted the STAT= text (almost identical for ALLOCATE/DEALLOCATE) into its own subclause, and put it at the end ahead of ERRMSG=. And incorporated the text, suitably modified, about not depending on anything allocated/deallocated in the statement (so this now appears once instead of twice). - Did not swap NULLIFY and DEALLOCATE. For 171:6+2, deleted the note instead. It doesn't say anything useful, and says lots that is incorrect (even after Van's correction, it is still wrong). For 171:9 first edit, instead, reworded to avoid both present and appear. For 171:35, also ".TRUE."->"true" (this wording should be identical to that of 171:9, since it is describing precisely the same semantics). Rejected: 75:11,75:14 ok as is 88:1 correct as is 102:15 seems ok as is 111:21 wrong "fix" for the "problem"; if we want to nail it down, we should say that a type has default init if any direct component has default init. 120:12-13 not an improvement 131:6,132:6 better as is, and not useful for future 142:39 correct as is 146:8,150:8 mooted by 06-312 06-301 ------ Done. 06-312 ------ Unfortunately this edit was not just a simple drop-in, as it had some wording problems, it undid some of the ISO conformance work done in previous drafts, and lost/hid some important semantics. I have accepted this with substantial changes. - deleted the opening witter between the heading "7" and "7.1". - added 7.1.1 General after 7.1 Expressions - added 7.1.2.1 "Expression categories" after 7.1.2 "Form of an expression" - added 7.1.5.1 Definitions after 7.1.5 "Intrinsic operations" - changed all the cross-references for "the permitted types ... are specified in" to point to the new 7.1.5.1 instead of just 7.1.5, since almost all those refs are *in* 7.1.5 already! - moved the opening defintions in 7.1.5.2 "Numeric intrinsic operations" into 7.1.5.1. Ditto 7.1.5.3 "Character intrinsic operation", 7.1.5.4 "Logical intrinsic operations", 7.1.5.5 "Bits intrinsic operations". - Also into 7.1.5.1 goes the first paragraph of 7.1.5.6 "Relational intrinsic operations" except for the second sentence (the one about "<" and ".LT." being the same), which becomes the new second paragraph of 7.1.5.6.1 "Interpretation of relational intrinsic operations"). Finally, the second paragraph of 7.1.5.6 becomes the new third paragraph of 7.1.5.6.1. - Added 7.1.6.1 Definitions after 7.1.6 "Defined operations". - Repeated the paragraph about "<" and ".LT." having the same interpretation in 7.1.6.2 "Interpretation of a defined operation". (These were the lost or hidden semantics.) - deleted opening witter of 7.2 Assignment as it was redundant and contradictory with the opening witter of 7.2.x.1. - moved the opening witter of 7.2.3 "Masked array assignment - WHERE" into 7.2.3.1 "General form of the masked array assignment", where I appended it to the first paragraph, changing "The masked array assignment" to "It". - deleted opening witter of 7.2.4 FORALL as it added little (other than vaguer wording) to the existing opening of 7.2.4.1 and 7.2.4.2). - Changed 7.2.4.1 "The FORALL construct" to "Form of the FORALL Construct". (The other subclauses about the FORALL construct are not subclauses of this one.) - added 7.2.4.2.1 "Execution stages" after 7.2.4.2 "Execution of the FORALL construct". 06-315 ------ Deleted UTI 007. This was a glitch; as noted in both the editor's report and the list of outstanding issues, 007 was accepted as being resolved (and the editor's report said I'd deleted it). If people had read these things they could have just dropped me an email saying "you said you deleted UTI 007 but it is still there", there was no need for a J3 paper with its attendant overhead. 06-316 ------ Deleted UTI 020 and 023 (no edits needed to resolve). 06-324 ------ Rejected. The editor does not agree with most of the assertions in this paper. The logic in the paper (suddenly deciding to lump rank in) is flawed. 06-355 ------ [339:note 13.1] This note is incorrect and not useful. Deleted it instead. Done. 06-364 ------ Done. ===END===