CVS difference for ai12s/ai12-0045-1.txt

Differences between 1.2 and version 1.3
Log of other versions for file ai12s/ai12-0045-1.txt

--- ai12s/ai12-0045-1.txt	2012/12/02 01:58:27	1.2
+++ ai12s/ai12-0045-1.txt	2012/12/28 00:23:06	1.3
@@ -1,5 +1,7 @@
-!standard 6.1.1(1/3)                                 12-12-01    AI12-0045-1/01
+!standard 6.1.1(1/3)                                 12-12-27    AI12-0045-1/02
 !class binding interpretation 12-12-01
+!status Amendment 202x 12-12-27
+!status ARG Approved 11-0-0  12-12-06
 !status work item 12-12-01
 !status received 12-09-19
 !priority Medium
@@ -30,11 +32,10 @@
 
 Replace 6.1.1(1/3) with:
 
-   For a subprogram other than the instance of a generic subprogram,
-   a generic subprogram, or entry, the following language-defined aspects
-   may be specified with an aspect_specification:
+   For a noninstance subprogram, a generic subprogram, or entry, the
+   following language-defined aspects may be specified with an
+   aspect_specification (see 13.1.1):
 
-
 !discussion
 
 If we allowed Pre and Post on instances that are subprograms, we
@@ -54,8 +55,18 @@
 should be continued.
 
 We considered allowing the aspects on an instance if none are given on
-the generic unit in order to avoid this problem. But we decided against
-this because of the maintenance problem mentioned above.
+the generic unit in order to avoid this latter problem. But we decided
+against this because of the maintenance problem mentioned above.
+
+!corrigendum 6.1.1(1/3)
+
+@drepl
+For a subprogram or entry, the following language-defined aspects may
+be specified with an @fa<aspect_specification> (see 13.1.1):
+@dby
+For a noninstance subprogram, a generic subprogram, or entry, the
+following language-defined aspects may be specified with
+an @fa<aspect_specification> (see 13.1.1):
 
 !ACATS test
 
@@ -985,7 +996,10 @@
 Heh?  There are lots of compelling examples showing that "not in"
 is useful.  Obviously it's not essential, but that's not what we're talking about.
 
-On the other hand, I haven't seen any compelling examples showing that postconditions on instances are useful.  In fact, it seems to me they're asking for trouble.  When you write a postcondition, you're supposed to write a body that obeys the postconditi
on, but you can't do that for an instance.
+On the other hand, I haven't seen any compelling examples showing that
+postconditions on instances are useful.  In fact, it seems to me they're asking
+for trouble.  When you write a postcondition, you're supposed to write a body that
+obeys the postcondition, but you can't do that for an instance.
 
 > > When a postcondition depends on the  properties of a formal subprogram?
 > > Something along the lines (OK, not realistic, just to show the  idea):

Questions? Ask the ACAA Technical Agent