GNU bug report logs -
#25524
26.0.50; diff-mode is broken
Previous Next
Reported by: Dima Kogan <dima <at> secretsauce.net>
Date: Wed, 25 Jan 2017 06:43:01 UTC
Severity: normal
Found in version 26.0.50
Done: Tino Calancha <tino.calancha <at> gmail.com>
Bug is archived. No further changes may be made.
Full log
View this message in rfc822 format
[Message part 1 (text/plain, inline)]
Your bug report
#25524: 26.0.50; diff-mode is broken
which was filed against the emacs package, has been closed.
The explanation is attached below, along with your original report.
If you require more details, please reply to 25524 <at> debbugs.gnu.org.
--
25524: http://debbugs.gnu.org/cgi/bugreport.cgi?bug=25524
GNU Bug Tracking System
Contact help-debbugs <at> gnu.org with problems
[Message part 2 (message/rfc822, inline)]
Tino Calancha <tino.calancha <at> gmail.com> writes:
> Noam Postavsky <npostavs <at> users.sourceforge.net> writes:
>
>> On Wed, Jan 25, 2017 at 11:32 AM, Tino Calancha <tino.calancha <at> gmail.com> wrote:
>>> ;; "index ", "old mode", "new mode", "new file mode" and
>>> -;; "deleted file mode" are output by git-diff.
>>> +;; "deleted file mode" are output by git-diff; "Index: " and
>>> +;; "========..." by SVN.
>>> (defconst diff-file-junk-re
>>> - "diff \\|index \\|\\(?:deleted file\\|new\\(?: file\\)?\\|old\\) mode\\|=== modified file")
>>> + (concat "Index: \\|" (make-string 67 ?=) "\\|"
>>> + "diff \\|index \\|\\(?:deleted file\\|new\\(?: file\\)?\\|old\\) mode\\|=== modified file"))
>>
>> Is it safe to assume exactly 67 "="? Maybe it would be better to use
>> something like "=\\{20,\\}", i.e. a sequence of 20 or more "=".
> Humm probably 20 is OK, but i am not using SVN in a while: i ignore if
> such '====...' might be coustomizable.
> Updated the patch to use >=20 ?=.
Pushed fix to master branch as commit 0073223c23749ffd6bd3f882bc30a82cc37efd2a
[Message part 3 (message/rfc822, inline)]
[Message part 4 (text/plain, inline)]
I'm using a very recent build from git: 0a49f158f15. I see diff-mode
being broken in 2 ways.
I'm attaching a diff file, produced by C-x v D in a project using
subversion (then cut down and de-contented).
Breakage 1:
1. emacs -Q /tmp/tst.patch
2. M-k
I would expect this to kill the first hunk. Instead emacs barfs:
Args out of range: something something
Breakage 2:
1. emacs -Q /tmp/tst.patch
2. M-g g 13 RET move point to the start of the 2nd hunk
3. M-k I would expect this to kill the hunk at point (2nd
hunk). I would then expect the point to remain at
the 2nd hunk
4. M-k Same as before. Should kill 2nd hunk
5. M-k Same as before. Should kill 2nd hunk
Instead, M-k #2 kills the 2nd hunk and then moves the point to the 1st
hunk. So that subsequent M-k #3 kills the 1st hunk.
[tst.patch (text/x-diff, inline)]
Index: Makefile.1
===================================================================
--- Makefile.1 (revision 81382)
+++ Makefile.1 (working copy)
@@ -220,2 +220,3 @@
# 1111
+# 2222
# 3333
Index: Makefile.2
===================================================================
--- Makefile.2 (revision 81382)
+++ Makefile.2 (working copy)
@@ -220,2 +220,3 @@
# 2222
+# 3333
# 4444
@@ -330,2 +330,3 @@
# 3333
+# 4444
# 5555
Index: Makefile.3
===================================================================
--- Makefile.3 (revision 81382)
+++ Makefile.3 (working copy)
@@ -220,2 +220,3 @@
# 4444
+# 5555
# 6666
This bug report was last modified 8 years and 176 days ago.
Previous Next
GNU bug tracking system
Copyright (C) 1999 Darren O. Benham,
1997,2003 nCipher Corporation Ltd,
1994-97 Ian Jackson.