32234 - prcpucfg.h breaks compiles, needs comment delimeters

Status: RESOLVED FIXED

(NSPR :: NSPR, defect, P3)

Description

[Scott Collins]

This issue always crops up in building automated tests that try to do minimal 
includes.  May I just check in the following patch?

Index: prcpucfg.h
===================================================================
RCS file: /cvsroot/mozilla/nsprpub/pr/src/md/mac/prcpucfg.h,v
retrieving revision 3.3
diff -r3.3 prcpucfg.h
115c115
< #endif NO_NSPR_10_SUPPORT
---
> #endif /* NO_NSPR_10_SUPPORT */

Comment 1

[srinivas]

I have checked in the patch on the tip of NSPR source.
Are you using the tip or the static tag for NSPR 4.0 release (which the Mozilla 
builds use)?

Comment 2

[Scott Collins]

I am using the static tag (and thanks, by the way :-)

Comment 3

[srinivas]

Reassigned.

Comment 4

[Scott Collins]

OK, so if you checked in a fix, then I can just mark this bug fixed, right?  I 
imagine I'll automatically start getting the fixed file the next time the NSPR 
static tag gets updated.  Please correct me if I am mistaken.

Comment 5

[Wan-Teh Chang]

Yes, this fix will eventually get into some static tag.
I need to know the severity of this bug to decide whether
it should go into NSPR 4.0.1 or 4.1.  (Current release is
4.0.)

You said it breaks compiles, which would be a critical bug.
But why does the compile work right now?

Comment 6

[Scott Collins]

It does not break mainline compiles ... I don't know why not.  I always run into 
it when I am trying to compile individual tests.  In particular, in TestCOMPtr 
and in the StringFactoringTests.  The difference is, they don't include as much 
of the world as a normal source file does ... so perhaps something in the rest of 
the include world is counteracting the problem in prcpucfg.h for normal clients.

I've been hitting this problem for about 10 months and living with it ... so I 
don't mind waiting for 4.1 if you have some significant reason not to put it in 
4.0.1.  I'm happy you guys had the time to get it fixed at all :-)

Comment 7

[Wan-Teh Chang]

It could be that you are compiling the individual
tests at a higher warning level.  I decided to make
this part of NSPR 4.0.1.

Checked in the fix on the NSPRPUB_RELEASE_4_0_BRANCH:
/cvsroot/mozilla/nsprpub/pr/src/md/mac/prcpucfg.h, revision 3.2.58.2

Since many files include prcpucfg.h, I will wait until
a good time to check this in on the NSPRPUB_CLIENT_BRANCH,
which the mozilla client pulls.

Comment 8

[Wan-Teh Chang]

Checked in the patch on the NSPRPUB_CLIENT_BRANCH.
/cvsroot/mozilla/nsprpub/pr/src/md/mac/prcpucfg.h, revision 3.3.2.1