I saw it twice today, on b.m.o.. Second time is: https://bugzilla.mozilla.org/attachment.cgi?id=178811 {{ ++ content/global/bindings/tree.xml (widgets/tree.xml) }} is missing from https://bugzilla.mozilla.org/attachment.cgi?id=178811&action=diff Then requesting 'Raw Unified' has the same bug.

First time was: https://bugzilla.mozilla.org/attachment.cgi?id=156373&action=diff which misses {{ --- mozilla/mailnews/base/search/resources/locale/en-US/filter.properties 28 Feb 2003 00:21:23 -0000 1.16 +++ mozilla/mailnews/base/search/resources/locale/en-US/filter.properties 17 Aug 2004 21:13:29 -0000 }} from https://bugzilla.mozilla.org/attachment.cgi?id=156373 Then requesting 'Raw Unified' has the same bug.

Wow. This seems to be two different bugs; comment 0 refers to a line that starts with a plus being missing in the diff. Comment 1 refers to a patch missing a whole hunk from it.

[Mozilla/5.0 (Windows; U; Win98; en-US; rv:1.7.8) Gecko/20050511] (release) (W98SE) [Mozilla/5.0 (Windows; U; Win98; en-US; rv:1.8b2) Gecko/20050701] (nightly) (W98SE) Another issue: 'moz' is (now !?) added at the beginning of the filenames... {{ <https://bugzilla.mozilla.org/attachment.cgi?id=178811&action=diff> (+) mozmozilla/toolkit/content/jar.mn (-1 lines) (+) mozmozilla/toolkit/content/widgets/tree.xml (-8 / +3 lines) }} Or could this one be a "new wanted feature" !??

(In reply to comment #3) > (+) mozmozilla/toolkit/content/jar.mn (-1 lines) [Mozilla/5.0 (Windows; U; Win98; en-US; rv:1.8b3) Gecko/20050710 SeaMonkey/1.0a] (nightly) (W98SE) Or on the filename: {{ <https://bugzilla.mozilla.org/attachment.cgi?id=175059&action=diff> (+) mozilla/xpinstall/wizard/windows/setup/diadialogs.c (-16 / +32 lines) }}

(In reply to comment #4) [Mozilla/5.0 (Windows; U; Win98; en-US; rv:1.8b3) Gecko/20050710 SeaMonkey/1.0a] (nightly) (W98SE) Or in-between: {{ <https://bugzilla.mozilla.org/attachment.cgi?id=152585&action=diff> (+) mozilla/xpinstall/packager/winwin_mfcembed/config.it (-1 / +1 lines) +) mozilla/xpinstall/packager/winwindows/config.it (-1 / +1 lines) }} Sigh :-(

This is irritating, certainly ('ve seen it happen), but not a release blocker.

(In reply to comment #6) > This is irritating, certainly ('ve seen it happen), but not a release blocker. Additional bad behaviour example related to the link part: [Mozilla/5.0 (Windows; U; Win98; en-US; rv:1.8b3) Gecko/20050714 SeaMonkey/1.0a] (nightly) (W98SE) https://bugzilla.mozilla.org/attachment.cgi?id=152140&action=diff displays as (-) mozmozilla/mailnews/addrbook/resources/content/addressbook.js (-4 / +8 lines) and links to http://lxr.mozilla.org/mozilla/source/la/mailnews/addrbook/resources/content/addressbook.js which obviously lead to a "This file does not exist." I guess this bug is a regression, but I can't verify it since I use b.m.o. site only. Even if it's not a "core" (database, ...) feature, it's seems useful enough to be a blocker; but that's only my thought...

Looks like this may actually be a flaw in PatchReader/Raw.pm. Gerv provided this patch, and it fixes the problem for several diffs on a test instance of b.m.o. Here are the diffs on the original instance: https://bugzilla.mozilla.org/attachment.cgi?id=187534&action=diff&headers=1&context=file https://bugzilla.mozilla.org/attachment.cgi?id=187534&action=diff&headers=1&context=file https://bugzilla.mozilla.org/attachment.cgi?action=diff&id=187534&collapsed=&headers=1&context=30 And here they are on the test instance: https://bugzilla.mozilla.org/test/attachment.cgi?id=187534&action=diff&headers=1&context=file https://bugzilla.mozilla.org/test/attachment.cgi?id=187534&action=diff&headers=1&context=file https://bugzilla.mozilla.org/test/attachment.cgi?action=diff&id=187534&collapsed=&headers=1&context=30

(In reply to comment #8) > Created an attachment (id=190249) [edit] > patch for PatchReader/Raw.pm that fixes problem > https://bugzilla.mozilla.org/attachment.cgi?id=187534&action=diff&headers=1&context=file https://bugzilla.mozilla.org/attachment.cgi?id=187534&action=diff&headers=1&context=file https://bugzilla.mozilla.org/attachment.cgi?action=diff&id=187534&collapsed=&headers=1&context=30 > https://bugzilla.mozilla.org/test/attachment.cgi?id=187534&action=diff&headers=1&context=file https://bugzilla.mozilla.org/test/attachment.cgi?id=187534&action=diff&headers=1&context=file https://bugzilla.mozilla.org/test/attachment.cgi?action=diff&id=187534&collapsed=&headers=1&context=30 (You might want to check in both list why your 2 first URLs are identical...) Anyway, I tested comment 0 example: Using the '/test/' URL fixes the issue on the filename links :-) (Should review be asked for this patch !?) Then, we need another patch for the content issue.

John Keiser maintains Patch Reader; this bug needs to be assigned to him, really. Gerv

Comment on attachment 190249 [details] [diff] [review] patch for PatchReader/Raw.pm that fixes problem This patch affects PatchReader/Raw.pm which is not part of the Bugzilla code. This has nothing to do with us and requesting review doesn't make sense. See gerv's previous comment.

Is this a dupe of bug 233695?

Let's see: Comment 0: Looks much like bug 233695. Comment 1: Might be bug 233695 too !? Comment 3: WorksForMe now. Comment 4: WorksForMe now. Comment 5: WorksForMe now. Comment 7: WorksForMe now. (Comment 8 'test' links are '404 Not Found' now.) Still very much interested in a solution, whether we patch the tool, or use a version that works ... after (one, no:) two years :-/

(In reply to comment #14) > *** Bug 365383 has been marked as a duplicate of this bug. *** Quote from that bug: { [Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.8.1.2pre) Gecko/20061229 SeaMonkey/1.1] (nightly) (W2Ksp4) "Bugzilla was upgraded on December 26th at 6pm PST (2am Wednesday UTC)." regressed comment 7. } Changing W98 to W2K, as I am currently dropping my old W98SE.

[Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.9a3pre) Gecko/20070306 SeaMonkey/1.5a] (nightly) (W2Ksp4) For the record, my comment 13 stands, with current b.m.o BugZilla version, whatever it is. *** I could have thought there was more data/discussion/work done here than there ... but, all right, let's V.Duplicate.

(In reply to comment #15) > (In reply to comment #14) > > *** Bug 365383 has been marked as a duplicate of this bug. *** > > "Bugzilla was upgraded on December 26th at 6pm PST (2am Wednesday > UTC)." regressed comment 7. To be explicit, I noted there that that (temporary) regression is now WFM: comment 13 stands as it was written.