!standard 0.3(60) 05-09-16 AC95-00111/01 !class Amendment 05-09-16 !status received no action 05-09-16 !status received 05-09-16 !subject Editorial comments on Draft 13 !summary [This isn't really No Action, but these items will be processed as corrections, not as separate AIs. So the mail is stored in an AC. - ED] !appendix From: Dan Eilers Date: Monday, August 22, 2005 5:42 AM > I'll now set my stopwatch; I expect Dan to send a list of two dozen spelling > errors about 1.8 seconds after I push send. :-) OK, here they are :) availablity AA-13-13-2.html circularlity AA-10-2-1.html explictly AA-3-8.html, AA-12-3.html Furthmore RM-13-13-2.html intergrity RM-13-12.html namable AA-12-7.html, AA-12-7.html, AA-3-2-2.html, AA-3-6.html, AA-5-4.html progentitor AA-13-14.html a a AA-A-18-2.html, AA-A-18-3.html, AA-A-18-4.html, AA-A-18-7.html be be AA-A-18-10.html from from AA-3-9.html the the RM-3-9-2.html, AA-6-5.html **************************************************************** From: Dan Eilers Date: Wednesday, September 14, 2005 11:18 AM There are unmatched right parens in RM 9.11(3/2), RM A.18.2(153/2), RM A.5.2(53), and AARM G.3.2(161.b/2). RM 2.2(13): there are spurious line breaks before "becomes" and "not equal". RM C.3(22): there is a spurious line break after "10.". RM 12.4(8.4/2) should be updated to match the wording of the parallel paragraph RM 8.5.1(4.4/2). **************************************************************** From: Christoph Grein Date: Friday, September 16, 2005 10:28 AM Introduction 77.4/1 Paragraph numbers {of} unchanged paragraphs should be the same as in the original International Standard. --- Paragraph number 1.1.2(39.f) should be 1.1.2(39.f/2). --- 1.1.4(21.c/2) The example for square brackets has been changed [w]as there is no longer a return_statement syntax rule. --- 13.11.2(17.a) This is not a testable property, since we do not {know} how much --- 13.12.1(4.b/2) It seems to me that the whole section 13.12.1 is new, not just a paragraph thereof. Thus also the paragraph numbering starting with 0 seems incorrect. --- A.3.2 (2/2) [with Ada.Characters.Conversions;] package Ada.Characters.Handling is I do not see the need for the with-clause here. --- A.5.2(32.a.1/2) There is a superfluous ]} at the end of the sentence. --- A.5.2(42) The exponent 15 is on the next line. This should be avoided. --- A.16(129/2) 40 If the target system does not support directories inside of directories, [Is_Directory will always return False, and] Containing_Directory will always raise Use_Error. Is_Directory does not exist (remnant from a previous version). I searched the whole document, and this is the only occurrence. --- A.18(4.v/2) hash functions, et[.] al. The latin "et" is no abbreviation. (IIRC it means "et alii" resp. "et alia" - and others). --- B.1(39.a.1/2) Subprograms named "adainit{"} and "adafinal" should --- B.1(41.c) This paragraph is no longer correct since C++ has been an ISO standard for some time now. --- B.1(49/2) Add the sentence "This paragraph was deleted." --- B.3.3(5.2) first_subtype_local_name is incorrectly italicized: only the part first_subtype must be italicized. **************************************************************** From: Dan Eilers Date: Monday, September 19, 2005 2:31 PM What should the title of the draft RM be? Currently, the title page of draft 13 says: ISO/IEC 8652:1995(E) with COR.1:2001 and AMD.1:200y but the Forward explains that this document is what the project editor would supply to ISO in the event that ISO elects to publish a merged document. The latest WG9 minutes say: > From Maho Takahashi (Ms.), ISO Central Secretariat, 4 May 2005: > > If the base standard, corrigendum and amendment were to be merged into > a single document, it would be published as ISO/IEC 8652:2006 Ed. 3 > (if published in 2006). So I think the title page should be changed to say: ISO/IEC 8652:2006 Ed. 3 possibly with "(draft)" appended, since this draft is intended to be as close as possible to what ISO might publish. **************************************************************** From: Robert Dewar Date: Monday, September 19, 2005 3:14 PM Dan Eilers wrote: > What should the title of the draft RM be? It would be nice if the title or at least an explanatory subtitle said something like "Proposed Ada 2005 draft document". That's something people could understand, unlike the cryptic stuff below :-) **************************************************************** From: Randy Brukardt Date: Monday, September 19, 2005 4:56 PM Dan writes: > What should the title of the draft RM be? > > Currently, the title page of draft 13 says: > ISO/IEC 8652:1995(E) with COR.1:2001 and AMD.1:200y > > but the Forward explains that this document is what the project editor > would supply to ISO in the event that ISO elects to publish a merged > document. That's not what the title page says; that's what's in the page header. The title page writes this out more explicitly: Ada Reference Manual ISO/IEC 8652:1995(E) with Technical Corrigendum 1 and Amendment 1 (Draft 13) > The latest WG9 minutes say: > > > From Maho Takahashi (Ms.), ISO Central Secretariat, 4 May 2005: > > > > If the base standard, corrigendum and amendment were to be merged into > > a single document, it would be published as ISO/IEC 8652:2006 Ed. 3 > > (if published in 2006). > > So I think the title page should be changed to say: > ISO/IEC 8652:2006 Ed. 3 > possibly with "(draft)" appended, since this draft is intended to be > as close as possible to what ISO might publish. I agree with this vis-a-vis the page headers, other than that the year should be written :200y for now, as I don't want to predict how long it will take ISO to actually publish the Amendment. The title page itself is rather irrelevant, as it would be completely different in an ISO version. The title was chosen to be as close as reasonable to the Ada 95 "Ada Reference Manual". Robert wrote: > It would be nice if the title or at least an explanatory subtitle > said something like "Proposed Ada 2005 draft document". That's > something people could understand, unlike the cryptic stuff below :-) Dan was incorrect, the title is "Ada Reference Manual", as it has been since the days of Jean Ichbiah. By the time this gets into general circulation (especially by the time Ada Europe produces a printed version), it won't be "proposed" anymore. And the name of the language is "Ada", not "Ada 2005". The subtitle is just to differentiate it from earlier versions of the Ada Reference Manual (and to be consistent with the previous versions of the ARM). But it will be the definition of Ada (and the only definition of Ada) once it is approved. So further discrimination doesn't seem necessary or advisable. **************************************************************** From: Robert Dewar Date: Monday, September 19, 2005 5:04 PM Randy Brukardt wrote: > Dan was incorrect, the title is "Ada Reference Manual", as it has been since > the days of Jean Ichbiah. By the time this gets into general circulation > (especially by the time Ada Europe produces a printed version), it won't be > "proposed" anymore. And the name of the language is "Ada", not "Ada 2005". Yes, of course that's true once it's approved, but not before. Before that, as approved by WG9, the language is to be referred to as Ada 2005, and it would be useful if the title page of this document uses this term (in quotes if you like) so that people know what's going on (the big majority of the Ada community knows this language as Ada 2005 at this stage). > The subtitle is just to differentiate it from earlier versions of the Ada > Reference Manual (and to be consistent with the previous versions of the > ARM). But it will be the definition of Ada (and the only definition of Ada) > once it is approved. So further discrimination doesn't seem necessary or > advisable. I disagree, it is useful at this stage when it is a draft with no official status to use the phrase Ada 2005 to make clear what is being discussed. I believe this is entirely in accordance with the intention of the WG9 vote settling this as the interim name to be used to distinguish this "version" of Ada from other versions of the language. **************************************************************** From: Dan Eilers Date: Monday, September 19, 2005 5:54 PM > That's not what the title page says; that's what's in the page header. OK. I failed to make the distinction between what the title page says and what looks like a title at the top of the title page. > I agree with this vis-a-vis the page headers, other than that the year > should be written :200y for now, as I don't want to predict how long it will > take ISO to actually publish the Amendment. OK. **************************************************************** From: Pascal Leroy Date: Tuesday, September 20, 2005 8:50 AM > A.3.2 (2/2) > [with Ada.Characters.Conversions;] > package Ada.Characters.Handling is > > I do not see the need for the with-clause here. The with_clause is needed because Handling still has a bunch of obsolescent functions (see J.14) and those functions are declared by renaming the functions in Conversions. > A.16(129/2) > 40 If the target system does not support directories inside > of directories, [Is_Directory will always return False, and] > Containing_Directory will always raise Use_Error. > > Is_Directory does not exist (remnant from a previous > version). I searched the whole document, and this is the > only occurrence. I think it should be talking of the File_Kind named Directory here. **************************************************************** From: Christoph Grein Date: Wednesday, September 21, 2005 1:47 AM I see, but then shouldn't the with-clause be moved to J.14? **************************************************************** From: Pascal Leroy Date: Wednesday, September 21, 2005 2:22 AM I suppose it could, but it would be a bit awkward "and by the way note that there is a with_clause, and it's obsolescent even though it wasn't there in Ada 95". I don't think strongly either way, but for clients with_clauses are semantically neutral, so it's no big deal. It would be useful to have a note in the AARM explaining why we need it: you are not the first person to ask that question, and I have a hunch that you won't be the last. **************************************************************** From: Christoph Grein Date: Tuesday, October 11, 2005 12:01 AM 13.11.2(17.a) since we do not {know} how much storage is used 13.14(20.q/2) Added wording to specify that [that] both operational 13.14(20.r/2) of an interface type [is] freezes it. A.10.7(17.3/2) if the length of {the} line exceeds Positive'Last; in [the] {this} case, the line number ****************************************************************