http://bugs.winehq.org/show_bug.cgi?id=21376
Summary: The Bat! - Access Violations on attachement view since 1.1.34 Product: Wine Version: 1.1.34 Platform: x86 OS/Version: Mac OS X 10.6 Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown AssignedTo: wine-bugs@winehq.org ReportedBy: domiman@gmail.com
Created an attachment (id=25740) --> (http://bugs.winehq.org/attachment.cgi?id=25740) test message with attachement (just a text file)
The Bat! produces Access Violations when you have attachement view set to small icons. This worked fine up to and including Wine 1.1.33.
Tested on Mac OS X 10.6.2, Wine compiled via MacPorts (so no patches added) with a clean prefix, no Winetricks. Wine versions I tested were 1.1.33, 1.1.34, 1.1.35 and 1.1.36. Wine 1.1.34 - 1.1.36 are producing the Access Violations. I didn't see any suspicious console messages.
Steps to reproduce:
- Download and install The Bat! Home (or Professional, it's the same program, only the packaging varies)
- in The Bat! you have to open a new account, you don't have to fill in real information, for this you don't need internet access.
- download the test.msg from here
- import the test.msg in The Bat! via the menus tools->import messages->from .MSG/.EML-files...
- set the attachement view to small icons via the menus view->attached files->small icons
Now you should have the Access Violations. When set to big large icons, all is fine again.
http://bugs.winehq.org/show_bug.cgi?id=21376
Jeff Zaroyko jeffz@jeffz.name changed:
What |Removed |Added ---------------------------------------------------------------------------- Keywords| |download, regression URL| |http://www.ritlabs.com/en/p | |roducts/thebat/download.php Summary|The Bat! - Access |The Bat! - Access |Violations on attachement |Violations on attachement |view since 1.1.34 |view
--- Comment #1 from Jeff Zaroyko jeffz@jeffz.name 2010-01-15 03:10:05 --- Please post the results of your regression test http://wiki.winehq.org/RegressionTesting
http://bugs.winehq.org/show_bug.cgi?id=21376
Jeff Zaroyko jeffz@jeffz.name changed:
What |Removed |Added ---------------------------------------------------------------------------- Attachment #25740|application/octet-stream |text/plain mime type| |
http://bugs.winehq.org/show_bug.cgi?id=21376
domiman@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |bunglehead@gmail.com
--- Comment #2 from domiman@gmail.com 2010-01-15 10:48:19 --- I had hoped I could get away without regression testing :)
bisect identified it as:
fc43895f49a423bdc6d826be7ad16b00c960612c is the first bad commit commit fc43895f49a423bdc6d826be7ad16b00c960612c Author: Nikolay Sivov bunglehead@gmail.com Date: Thu Nov 26 22:42:33 2009 +0300
comctl32/listview: Default column width in LVS_LIST to better value.
:040000 040000 8b5a3f52cc3075466d1a71e9a9e9d1498d9b70a4 8c2aa0f75a319cf7a09127f72d4d18fd3a6c50ba M dlls
I added Nikolay to the CC as suggested in the regression testing page. I hope this helps.
http://bugs.winehq.org/show_bug.cgi?id=21376
--- Comment #3 from Nikolay Sivov bunglehead@gmail.com 2010-01-15 10:51:03 --- Did you try to revert patch?
http://bugs.winehq.org/show_bug.cgi?id=21376
--- Comment #4 from domiman@gmail.com 2010-01-15 11:13:38 --- When I revert wine won't compile anymore.
git show fc43895f49a423bdc6d826be7ad16b00c960612c | patch -p1 -R patching file dlls/comctl32/listview.c Hunk #1 succeeded at 401 (offset -2 lines). Hunk #2 FAILED at 2784. Hunk #3 FAILED at 7452. 2 out of 3 hunks FAILED -- saving rejects to file dlls/comctl32/listview.c.rej patching file dlls/comctl32/tests/listview.c Hunk #1 FAILED at 3124. Hunk #2 succeeded at 3929 with fuzz 1 (offset 8 lines). Hunk #3 FAILED at 3946. 2 out of 3 hunks FAILED -- saving rejects to file dlls/comctl32/tests/listview.c.rej
then when compiling it fails with listview.c listview.c: In function ‘test_getcolumnwidth’: listview.c:3966: error: ‘itema’ undeclared (first use in this function) listview.c:3966: error: (Each undeclared identifier is reported only once listview.c:3966: error: for each function it appears in.) make[2]: *** [listview.o] Error 1 make[1]: *** [comctl32/tests] Error 2
http://bugs.winehq.org/show_bug.cgi?id=21376
--- Comment #5 from Nikolay Sivov bunglehead@gmail.com 2010-01-15 11:26:42 --- Try:
git revert c2a7535e9d112989abe687255b4f0da9a597678e git revert fc43895f49a423bdc6d826be7ad16b00c960612c
on rebased tree of course, to clean merge conflicts.
After that if Bat fails, try to rebase to c2a7535e9d112989abe687255b4f0da9a597678e and check again.
http://bugs.winehq.org/show_bug.cgi?id=21376
--- Comment #6 from domiman@gmail.com 2010-01-15 11:48:14 --- sorry to be bothering, but how does git revert work, I have no clue about git and so far only followed the orders on the regression page. When I do git revert ... I get the revert message, with please enter comitt message and I have no idea to how to proceed from that.
http://bugs.winehq.org/show_bug.cgi?id=21376
--- Comment #7 from Nikolay Sivov bunglehead@gmail.com 2010-01-15 11:55:04 --- (In reply to comment #6)
sorry to be bothering, but how does git revert work, I have no clue about git and so far only followed the orders on the regression page. When I do git revert ... I get the revert message, with please enter comitt message and I have no idea to how to proceed from that.
It shows text edit with default commit comment. To commit you should just exit edit saving a file.
Anyway 'git show <id> | patch p1 -R' will produce same tree state, so use it if you like. The point is that to revert 2 patches this no matter how.
http://bugs.winehq.org/show_bug.cgi?id=21376
--- Comment #8 from domiman@gmail.com 2010-01-15 13:01:52 --- Now I got it to work, thanks. Reverting the first two patches made it work with the attachement view in TB! again!
http://bugs.winehq.org/show_bug.cgi?id=21376
--- Comment #9 from Nikolay Sivov bunglehead@gmail.com 2010-01-15 13:05:58 --- (In reply to comment #8)
Now I got it to work, thanks. Reverting the first two patches made it work with the attachement view in TB! again!
Ok, now try to rebase to commit c2a7535e9d112989abe687255b4f0da9a597678e and build comctl32 again. If it fails please attach get plain output and output with +listview,+seh channels (two files in total).
http://bugs.winehq.org/show_bug.cgi?id=21376
--- Comment #10 from domiman@gmail.com 2010-01-17 06:11:14 --- can you hold my hand on this again? How to rebase? When I tried "git rebase c2a7535e9d112989abe687255b4f0da9a597678e" I got the error on cannot rebase: you have unstaged changes M dlls/comctl32/listview.c M dlls/comctl32/tests/listview.c
On a reset Wine git, I only got "Current branch master is up to date."
And I don't understand what you mean with get plain output.
http://bugs.winehq.org/show_bug.cgi?id=21376
--- Comment #11 from Nikolay Sivov bunglehead@gmail.com 2010-01-17 07:17:27 --- (In reply to comment #10)
can you hold my hand on this again? How to rebase? When I tried "git rebase c2a7535e9d112989abe687255b4f0da9a597678e" I got the error on cannot rebase: you have unstaged changes M dlls/comctl32/listview.c M dlls/comctl32/tests/listview.c
Of course, cause you reverted patches, making these changes not indexed. First of all reset to origin with: 'git reset --hard origin' After that use 'git rebase <...>' again.
On a reset Wine git, I only got "Current branch master is up to date."
It's ok.
And I don't understand what you mean with get plain output.
I mean terminal output - http://wiki.winehq.org/FAQ#get_log.
To activate debug channels use: --- WINEDEBUG=+listview,+seh wine <.exe> &> ~/log.txt.
Finally attach two logs here.
http://bugs.winehq.org/show_bug.cgi?id=21376
--- Comment #12 from domiman@gmail.com 2010-01-17 07:57:27 --- Created an attachment (id=25766) --> (http://bugs.winehq.org/attachment.cgi?id=25766) normal output log
http://bugs.winehq.org/show_bug.cgi?id=21376
--- Comment #13 from domiman@gmail.com 2010-01-17 07:58:14 --- Created an attachment (id=25767) --> (http://bugs.winehq.org/attachment.cgi?id=25767) +seh,+listview log
http://bugs.winehq.org/show_bug.cgi?id=21376
--- Comment #14 from domiman@gmail.com 2010-01-17 08:00:12 --- thanks for holding hand :)
I started The Bat! with Wine and since the message was already selected the access violation appeared right away, clicked ok, and exited the bat! via its menu. Same steps for both logs.
http://bugs.winehq.org/show_bug.cgi?id=21376
--- Comment #15 from Nikolay Sivov bunglehead@gmail.com 2010-01-17 08:21:21 --- Actually I can't reproduce this. I tried with Bat 4.2.18.
Tried importing message and create new one. Switching to small icons is fine for me. So I can't test it.
http://bugs.winehq.org/show_bug.cgi?id=21376
--- Comment #16 from domiman@gmail.com 2010-01-17 08:50:42 --- hmm, that's strange, so it might be a OS X only thing if no one else can reproduce this on Linux. It's really strange since I can always reproduce it. And it's not only with my own built but also with Mike Kronenbergs' OS X builts (1.1.33 works, 1.1.36 doesn't) and also with Crossover Games for Mac (which is at 1.1.28x, if I'm not mistaken). Always with a new prefix, to make sure nothing else interferes.
http://bugs.winehq.org/show_bug.cgi?id=21376
--- Comment #17 from domiman@gmail.com 2010-01-23 09:51:13 --- I got a confirmation on Mike Kronenberg's site. AFAIK Mike's Wine builts are now proper Wine and not infested with all kinds of special patches. His built is up to 1.1.35 (not 1.1.36 as I previously wrote).
One thing I overlooked in the reproduction text, you need to switch to another message first and then back when you have switched from large icons to small icons. Just keeping the view on the mail with the attachement will not show the access violation when you switch icon size.
I'd still like to help figuring this out :)
http://bugs.winehq.org/show_bug.cgi?id=21376
domiman@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|UNCONFIRMED |RESOLVED Resolution| |WORKSFORME
--- Comment #18 from domiman@gmail.com 2010-05-08 10:40:49 --- This works now. Tested again with Wine 1.1.43. I don't know which Wine release fixed it, since I didn't have time in the last few weeks to keep up (and just now compiling 1.1.44).
I set the status to to Resolved and Works for me, don't know if that closes the report or not. If not, feel free to close it. Thanks for the help :)
http://bugs.winehq.org/show_bug.cgi?id=21376
Vitaliy Margolen vitaliy@kievinfo.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Resolution|WORKSFORME |FIXED
http://bugs.winehq.org/show_bug.cgi?id=21376
Alexandre Julliard julliard@winehq.org changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED
--- Comment #19 from Alexandre Julliard julliard@winehq.org 2010-05-21 14:39:36 --- Closing bugs fixed in 1.2-rc1.
http://bugs.winehq.org/show_bug.cgi?id=21376
Austin English austinenglish@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- OS/Version|Mac OS X 10.6 |Mac OS X
-
wine-bugs＠winehq.org