December 2008 - Wine-Bugs - list.winehq.org

[Bug 13374] New: ComboBoxEx list get selected on mouse button down and release
by wine-bugs＠winehq.org Jan. 13, 2009

Jan. 13, 2009

http://bugs.winehq.org/show_bug.cgi?id=13374 Summary: ComboBoxEx list get selected on mouse button down and release Product: Wine Version: CVS/GIT Platform: PC OS/Version: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: comctl32 AssignedTo: wine-bugs(a)winehq.org ReportedBy: dylan.ah.smith(a)gmail.com Created an attachment (id=13272) --> (http://bugs.winehq.org/attachment.cgi?id=13272) comctl32: Added tests to show a ComboBoxEx bug caused by incorrect focus change The quickest way to see the error is to use Wine's clone of wordpad. There are two dropdown lists (ComboBoxEx controls) that are on the toolbar. Click the button to drop down the list, then left click on an item in the list. You will probably notice that the wrong item is selected, and if you do it slowly (i.e. hold down the mouse button for a second then release it), then you will notice that the selection is made and the list disappears when the mouse button is pressed, and when it is released another selection is made even though the list has disappeared. I ran wordpad on Windows XP SP2, and noticed that the correct behavior is the item is only selected when the mouse button is released. Note that there are two other separate bugs that you might notice: - The font size is 0 to start with (richedit bug) - Selecting a new font size will give you an "Invalid number format" error when the control loses control (wordpad bug) - The selected items from the list are not applied until the combobox loses focus (wordpad bug) I have submitted patches that fix these other bugs to the wine-patches mailing list. I have done some created tests that show the bug is in ComboBoxEx from the comctl32 and not in ComboBox from the user32 dll. I have attached the patch that adds these, and will submit it to the wine-patches mailing list after I get the bug number to mention in the commit message. I did some debugging, and to my suprise it was a call to SetFocus was rolling up the ComboBoxEx's list. Digging deeper I found it was actually the WM_KILLFOCUS call to the window that previously had focus, which was the ComboBox (a child of ComboBoxEx). The ComboBox saw that it was losing focus, and what was losing focus it didn't recognize as itself, it's edit control, or its list control, so it incorrectly decided to roll the dropdown up. It turns out that what was losing focus was an edit control that ComboBoxEx and child ComboBox control didn't know about created since, oddly enough, native Windows does not use the child ComboBox's edit control. My tests for ComboBox and ComboBoxEx show what had focus before and after pressing the dropdown button, and selecting an item in the dropdown list. In both causes the top level (main) window had intial focus. With ComboBox the focus was correctly on the edit control after each of the mouse button up/down messages are sent. For ComboBoxEx the focus was supposed to be on ComboBox when the list is dropped down then on the edit control after it's rolled up, according to my crosstests, but instead it was on the edit control after the list is dropped down. Some debugging found that the focus was being changed to the ComboBox, but each time it did that, the ComboBoxEx was notified with a CBN_SETFOCUS command message, which had no case for it in the COMBOEX_Command function, so the default case was used. The default case contains two lines that sets the focus back to the edit control, thus causing the bug, however taking out those lines caused missing notifications of the change. Hopefully this is enough information for someone else more familiar with the code to properly fix the problem. -- Configure bugmail: http://bugs.winehq.org/userprefs.cgi?tab=email Do not reply to this email, post in Bugzilla using the above URL to reply. ------- You are receiving this mail because: ------- You are watching all bug changes.

1 9

[Bug 13361] New: firefox 3 can't handle local paths in location window [ dogfood]
by wine-bugs＠winehq.org Jan. 13, 2009

Jan. 13, 2009

http://bugs.winehq.org/show_bug.cgi?id=13361 Summary: firefox 3 can't handle local paths in location window [dogfood] Product: Wine Version: CVS/GIT Platform: Other OS/Version: other Status: NEW Keywords: download, Installer Severity: normal Priority: P2 Component: -unknown AssignedTo: wine-bugs(a)winehq.org ReportedBy: dank(a)kegel.com Today I tried to install Acrobat Reader, and for some reason it decided to use the download manager rather than a direct download. The download was interrupted (that's another story), and the download manager left a handy link on the desktop to resume the download. The launcher contained: env WINEPREFIX="/home/dank/.wine-firefox" wine "C:\Program Files\Mozilla Firefox\firefox.exe" C:\windows\temp\nosget_start_manager.html That's not going to fly for several reasons: 1) the backslashes aren't quoted or escaped (but that's a different bug) 2) Firefox 3 on Wine doesn't seem to recognize local paths as valid URLs; typing c:\foo.txt into the location window yields errors like Firefox can't find the file at /c:/foo.txt. Firefox 2 doesn't seem to have this trouble. (I have not verified that Firefox 3 can handle c:\foo.txt on Windows, but I'll eat a box of doughnuts if it can't.) -- Configure bugmail: http://bugs.winehq.org/userprefs.cgi?tab=email Do not reply to this email, post in Bugzilla using the above URL to reply. ------- You are receiving this mail because: ------- You are watching all bug changes.

1 6

[Bug 13343] New: Regression error in today's git, Microsoft Office 2003 won 't install
by wine-bugs＠winehq.org Jan. 13, 2009

Jan. 13, 2009

http://bugs.winehq.org/show_bug.cgi?id=13343 Summary: Regression error in today's git, Microsoft Office 2003 won't install Product: Wine Version: CVS/GIT Platform: PC-x86-64 OS/Version: Linux Status: UNCONFIRMED Severity: critical Priority: P2 Component: msi AssignedTo: wine-bugs(a)winehq.org ReportedBy: rmuncrief(a)comcast.net Between yesterday and today the installation of Microsoft Office 2003 was broken in git. If you try to run the installation from the CD it will complain that "D:\FILES\SETUP\setup.INI" doesn't exist and exits. I confirmed that this is a filename case sensitivity problem (everything is lower case on the CD - it's D:\files\setup\setup.ini) by copying the CD to disk, changing the case of the directories and filename to those Wine expected, and then the installation worked normally. -- Configure bugmail: http://bugs.winehq.org/userprefs.cgi?tab=email Do not reply to this email, post in Bugzilla using the above URL to reply. ------- You are receiving this mail because: ------- You are watching all bug changes.

1 14

[Bug 13342] New: Winedbg doesn't handle longlong's well
by wine-bugs＠winehq.org Jan. 13, 2009

Jan. 13, 2009

http://bugs.winehq.org/show_bug.cgi?id=13342 Summary: Winedbg doesn't handle longlong's well Product: Wine Version: CVS/GIT Platform: Other OS/Version: other Status: NEW Severity: enhancement Priority: P2 Component: dbghelp AssignedTo: wine-bugs(a)winehq.org ReportedBy: m.b.lankhorst(a)gmail.com Created an attachment (id=13228) --> (http://bugs.winehq.org/attachment.cgi?id=13228) Test program A simple test function that passes LONGLONG's as arguments and is compiled with mingw gets parsed incorrectly by the winedbg. i586-mingw32msvc-gcc test.c -g -O0 -Wall -Werror -gstabs -W will result in having 0xef found as argument i586-mingw32msvc-gcc test.c -g -O0 -Wall -Werror -gstabs+ -W will result in thinking 0xbeefbeef is passed '0xbeefbeef' as argument -- Configure bugmail: http://bugs.winehq.org/userprefs.cgi?tab=email Do not reply to this email, post in Bugzilla using the above URL to reply. ------- You are receiving this mail because: ------- You are watching all bug changes.

1 4

[Bug 13338] New: cygwin installer crashes in /etc/postinstall/00bash.sh
by wine-bugs＠winehq.org Jan. 13, 2009

Jan. 13, 2009

http://bugs.winehq.org/show_bug.cgi?id=13338 Summary: cygwin installer crashes in /etc/postinstall/00bash.sh Product: Wine Version: CVS/GIT Platform: Other URL: http://cygwin.com OS/Version: other Status: NEW Keywords: download, Installer Severity: normal Priority: P2 Component: -unknown AssignedTo: wine-bugs(a)winehq.org ReportedBy: dank(a)kegel.com Installation was smooth until it tried to run the postinstall scripts; it got to 14%, then complained 393321 [proc_waiter] bash 37! _cygtls::handle_exceptions: Exception: STATUS_ACCESS_VIOLATION fixme:advapi:ImpersonateLoggedOnUser (0xd8) fixme:advapi:ImpersonateLoggedOnUser (0xd8) 393865 [proc_waiter] bash 37! open_stackdumpfile: Dumping stack trace to bash.exe.stackdump but no good stack dump is output. You can reproduce this by running 'wine bash.exe' in cygwin's bin. +relay seems to show it's inside a call to ReadFile(): 003d:Call KERNEL32.ReadFile(000000e4,7e39e790,000000a4,7e39e78c,00000000) ret=61099eaa trace:file:ReadFile 0xe4 0x7e39e790 164 0x7e39e78c (nil) 003c:Call KERNEL32.WaitForSingleObject(00000110,0000ea60) ret=61097f94 003c:Ret KERNEL32.WaitForSingleObject() retval=00000000 ret=61097f94 003c:Call KERNEL32.CloseHandle(00000110) ret=61097fa8 003c:Ret KERNEL32.CloseHandle() retval=00000001 ret=61097fa8 003c:Call KERNEL32.WaitForMultipleObjects(00000002,0067b928,00000000,ffffffff) ret=61048886 003c:Ret KERNEL32.WaitForMultipleObjects() retval=ffffffff ret=61048886 trace:seh:raise_exception code=c0000005 flags=0 addr=0x610476d9 trace:seh:raise_exception info[0]=00000000 trace:seh:raise_exception info[1]=0000019c trace:seh:raise_exception eax=00000000 ebx=611697b8 ecx=ffffffff edx=61124b01 esi=00000000 edi=611030e4 trace:seh:raise_exception ebp=0067c528 esp=0067c4b0 cs=0073 ds=007b es=007b fs=0033 gs=003b flags=00010286 trace:seh:call_stack_handlers calling handler at 0x61018970 code=c0000005 flags=0 -- Configure bugmail: http://bugs.winehq.org/userprefs.cgi?tab=email Do not reply to this email, post in Bugzilla using the above URL to reply. ------- You are receiving this mail because: ------- You are watching all bug changes.

1 5

[Bug 13329] New: World of Warcraft (WoW) trial installer crashes with unimplemented function js3250.dll. JS_SetGCParameter
by wine-bugs＠winehq.org Jan. 13, 2009

Jan. 13, 2009

http://bugs.winehq.org/show_bug.cgi?id=13329 Summary: World of Warcraft (WoW) trial installer crashes with unimplemented function js3250.dll.JS_SetGCParameter Product: Wine Version: CVS/GIT Platform: Other URL: https://signup.worldofwarcraft.com/trial/ OS/Version: other Status: NEW Keywords: download, Installer Severity: normal Priority: P2 Component: shdocvw AssignedTo: wine-bugs(a)winehq.org ReportedBy: dank(a)kegel.com Running TryWoW.exe doesn't get too far. Our mshtml doesn't seem to be good enough for it. Unhandled exception: unimplemented function js3250.dll.JS_SetGCParameter called in 32-bit code (0x7bc42f17). ... Backtrace: =>1 0x7bc42f17 stub_entry_point+0x4c(dll=0x149afb6, name=0x149afb6) [/data/dkegel/wine-git/dlls/ntdll/loader.c:196] in ntdll (0x0032c790) 2 0x0147e87c in xpc3250 (+0xe87c) (0x0032c7e8) ... 11 0x0039575c in xpcom_core (+0x575c) (0x0032ce14) 12 0x7d563ebd NSContainer_Create+0x23(doc=<register EAX not in topmost frame>, parent=0x0) [/data/dkegel/wine-git/dlls/mshtml/nsembed.c:1583] in mshtml (0x0032ce74) -- Configure bugmail: http://bugs.winehq.org/userprefs.cgi?tab=email Do not reply to this email, post in Bugzilla using the above URL to reply. ------- You are receiving this mail because: ------- You are watching all bug changes.

1 5

[Bug 13311] New: winetest always fails on shell32:shelllink, but "make test " doesn't
by wine-bugs＠winehq.org Jan. 13, 2009

Jan. 13, 2009

http://bugs.winehq.org/show_bug.cgi?id=13311 Summary: winetest always fails on shell32:shelllink, but "make test" doesn't Product: Wine Version: CVS/GIT Platform: Other OS/Version: other Status: NEW Severity: normal Priority: P2 Component: testcases AssignedTo: wine-bugs(a)winehq.org ReportedBy: dank(a)kegel.com Every recent report at http://test.winehq.org says shell32:shelllink fails, yet "make test" doesn't show it failing for me. Is there some timing issue in winetest? -- Configure bugmail: http://bugs.winehq.org/userprefs.cgi?tab=email Do not reply to this email, post in Bugzilla using the above URL to reply. ------- You are receiving this mail because: ------- You are watching all bug changes.

1 10

[Bug 13280] New: Klipfolio 4 hangs on startup
by wine-bugs＠winehq.org Jan. 13, 2009

Jan. 13, 2009

http://bugs.winehq.org/show_bug.cgi?id=13280 Summary: Klipfolio 4 hangs on startup Product: Wine Version: CVS/GIT Platform: PC URL: http://www.klipfolio.com/ OS/Version: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown AssignedTo: wine-bugs(a)winehq.org ReportedBy: nodisgod(a)yahoo.com Created an attachment (id=13129) --> (http://bugs.winehq.org/attachment.cgi?id=13129) Klipfolio 4 logs When trying to start Klipfolio 4 with today's Git (wine-1.0-rc1-104-g2470b0b), a nonpainted invisible window appears and does not respond to UI input. In the middle of the logs, a page fault occurs in wininet, so I have attached that information. -- Configure bugmail: http://bugs.winehq.org/userprefs.cgi?tab=email Do not reply to this email, post in Bugzilla using the above URL to reply. ------- You are receiving this mail because: ------- You are watching all bug changes.

1 8

[Bug 13269] New: ntdll/tests/exception crashes
by wine-bugs＠winehq.org Jan. 13, 2009

Jan. 13, 2009

http://bugs.winehq.org/show_bug.cgi?id=13269 Summary: ntdll/tests/exception crashes Product: Wine Version: CVS/GIT Platform: PC-x86-64 OS/Version: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: ntdll AssignedTo: wine-bugs(a)winehq.org ReportedBy: leffeman(a)gmail.com test_exceptions crashes in NtSetContextThread on access to the passed CONTEXT. I have managed to avoid the crash by doing one of the following things: * Put the CONTEXT on the stack. * Skip the memset call. * WINEDEBUG=trace+relay Toolchain bug? Wine magic? -- Configure bugmail: http://bugs.winehq.org/userprefs.cgi?tab=email Do not reply to this email, post in Bugzilla using the above URL to reply. ------- You are receiving this mail because: ------- You are watching all bug changes.

1 9

[Bug 13176] New: Fails to compile
by wine-bugs＠winehq.org Jan. 13, 2009

Jan. 13, 2009

http://bugs.winehq.org/show_bug.cgi?id=13176 Summary: Fails to compile Product: Wine Version: CVS/GIT Platform: PC-x86-64 OS/Version: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown AssignedTo: wine-bugs(a)winehq.org ReportedBy: grahame(a)regress.homelinux.org Seeing as RC1 has been released I decided it was time to try compiling and testing wine again. I checked out the latest version of wine as of 13th May 2008 and tried to compile using the /tools/wineinstall script. The compile faillls with the following error: gcc -m32 -c -I. -I. -I../../include -I../../include -D__WINESRC__ -D_WINX32_ -D_REENTRANT -fPIC -Wall -pipe -fno-strict-aliasing -Wdeclaration-after-statement -Wwrite-strings -Wpointer-arith -g -O2 -o netconnection.o netconnection.c netconnection.c: In function 'NETCON_init': netconnection.c:134: error: 'SONAME_LIBCRYPTO' undeclared (first use in this function) netconnection.c:134: error: (Each undeclared identifier is reported only once netconnection.c:134: error: for each function it appears in.) make[2]: *** [netconnection.o] Error 1 I have searched with google and looked for any obvious missing dev packages. I'm using gcc version Debian 4.2.3-5 and I'm trying to build on an AMD64 platform. If any further information is required please let me know. -- Configure bugmail: http://bugs.winehq.org/userprefs.cgi?tab=email Do not reply to this email, post in Bugzilla using the above URL to reply. ------- You are receiving this mail because: ------- You are watching all bug changes.

1 6