Look at the http://bugzilla.mozilla.org/attachment.cgi?id=140802&action=view in the diff viewer: http://bugzilla.mozilla.org/attachment.cgi?id=140802&action=diff If you scroll down to just after the routine "isPantherOrLater" you will see a function body without a function header. This is because the function header begins with a "+" symbol (in obj-c, this denotes a static method). You can see what it should be by looking at the patch itself, which is correct.

This is interesting... John, do you think that this is a bug in PatchReader or in our Bugzilla code?

*** Bug 338486 has been marked as a duplicate of this bug. ***

[Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.9a3pre) Gecko/20070306 SeaMonkey/1.5a] (nightly) (W2Ksp4) Comment 0 stands, with current b.m.o BugZilla version, whatever it is. The missing line from the patch is: {{ ++ (nsMacCursor *) createCursor: (enum nsCursor) aCursor }}

Attachment 190249 [details] [diff] (in bug 288038) fixes this problem. John, is there any plan to land that patch in Patch Reader? cl

[Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.9a5pre) Gecko/20070515 SeaMonkey/1.5a] (nightly) (W2Ksp4) Bug still there, after "2008-03-27 bmo upgrade" (tracked by bug 425607). (That's BugZilla v3.0.3 !?)

Not a blocker. This is a bug in PatchReader itself, not in Bugzilla. And I don't want to implement a workaround in a hurry, especially on a stable branch.

(In reply to comment #5) > Attachment 190249 [details] [diff] (in bug 288038) fixes this problem. Chris, can you confirm that that patch does fix this bug (too), and not (only) bug 303683 ? See bug 288038 comment 9. > John, is there any plan to land that patch in Patch Reader? John, ping ?

FWIW, I filed n°35362 : PatchReader-0.9.5 bug report: ++/-- line issues <http://rt.cpan.org/Public/Bug/Display.html?id=35362>

(In reply to comment #9) > (In reply to comment #5) > > Attachment 190249 [details] [diff] [details] (in bug 288038) fixes this problem. > > Chris, > can you confirm that that patch does fix this bug (too), > and not (only) bug 303683 ? > > See bug 288038 comment 9. I can't recall if I actually tested it or if I just read it and understood it to be true. Someone with a Bugzilla test install should probably verify that it does indeed fix things as it's supposed to. cl

This is particularly annoying now that the default view for patches is diff mode; I thought attachment 458039 [details] [diff] [review] was corrupt, but in fact it's fine.

bmo has taken local patches to PatchViewer before to fix pain points (e.g. bug 362767); if attachment 190249 [details] [diff] [review] in bug 288038 really does fix this problem, can we take it on bmo, too? This can affect not just Obj-C code, but also jar manifests in certain cases (e.g. attachment 474602 [details] [diff] [review]). Particularly as Diff view is now the default view in Bugzilla for most references to patches, this bug is becoming more and more annoying.

This also affects HTML close-comment markers, such as in attachment 493143 [details] [diff] [review].

Fixed in PatchReader 0.9.6.

Great. We should require that version, no?

(In reply to comment #16) > Great. We should require that version, no? Sure, I already planned to file a bug about this, for 4.2. :)