07-184r2 To: J3 From: Van Snyder/Dick Hendrickson Subject: Comments on Clause 2 Date: 8 May 2007 1 Edits Edits refer to07-007r1. Page and line numbers are displayed in the margin. Absent other instructions, a page and line number or line number range implies all of the indicated text is to be replaced by associated text, while a page and line number followed by + (-) indicates that associated text is to be inserted after (before) the indicated line. Remarks are noted in the margin, or appear between [ and ] in the text. [9:4-5] ------------------------------------------------------------------ [The immediately preceding subclause defines the notation used in the standard. We don't find it necessary to specify where to find the notation in the introductory waffle of any other clause. Editor: Delete "The notation . . . 1.7."] [9:8-9] ------------------------------------------------------------------ [It's not necessary to say this here and at [12:43]; it's more proper to say it at [12:43]. Editor: Delete "A . . . both."] [12:29] ------------------------------------------------------------------ [Editor: Replace "a extension" by "an extension".] [13:1] ------------------------------------------------------------------ [Editor: At end of paragraph add " The main program shall be defined by a Fortran or by means other than Fortran, but not both".] [13:9] ------------------------------------------------------------------- [Editor: Replace "in" with "by" for consistency with the remainder of the sentence.] [13:10-12] --------------------------------------------------------------- [Too complicated. Editor: Replace "Unless . . . function value" by "A subroutine or function that is not pure might change the program state by changing the values of data objects accessible to it".] [15:6] ------------------------------------------------------------------- [Editor: Delete "executable" after BLOCK.] [16: Table 2.2]----------------------------------------------------------- [What is the point of footnote 1? Scoping units are described only four pages previously. If we must mutter about the relationship between modules, submodules and scoping units, why not mutter about internal subprograms, type definitions and interface bodies? Editor: Delete footnote 1 and the reference to it.] [17:17] ------------------------------------------------------------------ [Editor: Replace "¡end-program-stmt¿" by "".] [18:9] ------------------------------------------------------------------- [Editor: Delete "contained".] [18:22] ------------------------------------------------------------------ [Editor: Replace "the language" by "\thisstandard".] [18:30] ------------------------------------------------------------------ [Editor: Replace "provide" by "create". Delete "data" since we didn't bother with it at [18:27].] [19:2] ------------------------------------------------------------------- [Editor: Replace "supplied by procedure definitions" by "defined by procedures (7.1.6)".] [22:15-20] --------------------------------------------------------------- [There's no excuse to take two sentences for each of these concepts. Besides, the stuff about disassociation isn't quite correct, and the perfectly good descriptions in Clause 16 don't need to be repeated. Editor: Replace the paragraph by the following:] A pointer is <> with a <> by pointer initialization (4.5.4.5, 5.2.3) or the events described in 16.5.2.2.1. A pointer is <>by explicit or default initialization to NULL(), or by the events described in 16.5.2.2.2. A disassociated pointer is not associated with a target. -------------------------------------------------------------------------- No action on the following item [24:37] ------------------------------------------------------------------ [The term "global procedure" is not defined or used anywhere else. Editor: Insert "external" before "procedures".] [16:11-13] --------------------------------------------------------------- Shouldn't the sentence have or be followed by an "otherwise" part? Without it, I don't see how the images get created. There also seems to be an implicit connection between "instance" (which I thought was related to recursive procedures) and "image".