March 2011 - Wine-devel - list.winehq.org

Re: cryptnet/tests: Partially revert commit de6e33f306a3b1b424ad1a9c41e85d3692ef9e4d
by Joris Huizer 11 Mar '11

11 Mar '11

Hello, I found a small (copy/paste) mistake in this patch: - /* w2k3,XP, newer w2k: CRYPT_E_NO_MATCH */ - ok(!ret && (GetLastError() == CRYPT_E_NO_MATCH), + /* w2k3,XP, newer w2k: CRYPT_E_NO_MATCH, older w2k: CRYPT_E_ASN1_BADTAG */ + ok(!ret && (GetLastError() == CRYPT_E_NO_MATCH || + broken(GetLastError() == CRYPT_E_NO_MATCH)), The comment indicates the alternative error, but the actual code doesn't check that one HTH, Joris

1 0

Re: wininet(4/7): Implement commiting a url cache entry that already existed (try 2)
by Alexandre Julliard 11 Mar '11

11 Mar '11

Juan Lang <juan.lang(a)gmail.com> writes: > With the previous patches applied, this no longer fails for me on > Alexandre's apparently corrupt index.dat. Note that the index.dat I sent you was created from scratch, by running the wininet tests in an empty wineprefix. So if it's really corrupt it means wininet has corrupted it. It would probably be a good idea to fix this first. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: msxml3: Ignore XML Property ResolveExternals
by Nikolay Sivov 11 Mar '11

11 Mar '11

On 3/11/2011 13:46, Alistair Leslie-Hughes wrote: > Hi, > > > Changelog: > msxml3: Ignore XML Property ResolveExternals Please correct naming: static const WCHAR PropertyNewParserW[] = {'N','e','w','P','a','r','s','e','r',0}; static const WCHAR PropValueXPathW[] = {'X','P','a','t','h',0}; static const WCHAR PropValueXSLPatternW[] = {'X','S','L','P','a','t','t','e','r','n',0}; +static const WCHAR PropValueResolveExternalsW[] = {'R','e','s','o','l','v','e','E','x','t','e','r','n','a','l','s',0}; It's not a value, it's a property with boolean value, so it's PropertyResolveExternalsW. > > > Best Regards > Alistair Leslie-Hughes

1 0

Re: wininet(7/7): Delete the cache file when it's no longer referenced (resend)
by Marvin 11 Mar '11

11 Mar '11

Hi, While running your changed tests on Windows, I think I found new failures. Being a bot and all I'm not very good at pattern recognition, so I might be wrong, but could you please double-check? Full results can be found at http://testbot.winehq.org/JobDetails.pl?Key=9868 Your paranoid android. === WINEBUILD (build) === Patch failed to apply

1 0

Re: wininet(6/7): Delete cache entries when their use count reaches 0
by Marvin 11 Mar '11

11 Mar '11

Hi, While running your changed tests on Windows, I think I found new failures. Being a bot and all I'm not very good at pattern recognition, so I might be wrong, but could you please double-check? Full results can be found at http://testbot.winehq.org/JobDetails.pl?Key=9867 Your paranoid android. === WINEBUILD (build) === Patch failed to apply

1 0

Re: [PATCH 6/6] user32: Add support for sequence-based animated cursors.
by Henri Verbeet 11 Mar '11

11 Mar '11

On 10 March 2011 07:42, Erich Hoover <ehoover(a)mines.edu> wrote: > @@ -940,6 +954,7 @@ done: > #define ANI_seq__ID RIFF_FOURCC('s', 'e', 'q', ' ') > #define ANI_fram_ID RIFF_FOURCC('f', 'r', 'a', 'm') > #define ANI_rate_ID RIFF_FOURCC('r', 'a', 't', 'e') > +#define ANI_seq_ID RIFF_FOURCC('s', 'e', 'q', ' ') Just happened to notice this, but I doubt there's much point in having two different defines for sequence chunks.

2 1

Re: [4/4] d3d8: Test depth stencil size mismatches
by Henri Verbeet 10 Mar '11

10 Mar '11

On 10 March 2011 22:58, Stefan Dösinger <stefan(a)codeweavers.com> wrote: > @@ -264,6 +264,8 @@ struct IDirect3DSurface8Impl > > /* If set forward refcounting to this object */ > IUnknown *forwardReference; > + > + UINT width, height; > }; You could just ask wined3d for the surface size.

2 1

Re: [PATCH 3/4] mshtml: Added IHTMLPrivateWindow_GetAddressBarUrl implementation
by Marvin 10 Mar '11

10 Mar '11

Hi, While running your changed tests on Windows, I think I found new failures. Being a bot and all I'm not very good at pattern recognition, so I might be wrong, but could you please double-check? Full results can be found at http://testbot.winehq.org/JobDetails.pl?Key=9859 Your paranoid android. === WXPPROSP3 (32 bit htmldoc) === htmldoc.c:3564: Test failed: unexpected address bar url: L"about:blank" htmldoc.c:3526: Test failed: unexpected address bar url: L"about:blank" htmldoc.c:3557: Test failed: unexpected address bar url: L"about:blank" htmldoc.c:3564: Test failed: unexpected address bar url: L"about:replace" === WVISTAADM (32 bit htmldoc) === htmldoc.c:3564: Test failed: unexpected address bar url: L"about:blank" htmldoc.c:3526: Test failed: unexpected address bar url: L"about:blank" htmldoc.c:3557: Test failed: unexpected address bar url: L"about:blank" htmldoc.c:3564: Test failed: unexpected address bar url: L"about:replace" === W7PRO (32 bit htmldoc) === htmldoc.c:3564: Test failed: unexpected address bar url: L"about:blank" htmldoc.c:3526: Test failed: unexpected address bar url: L"about:blank" htmldoc.c:3557: Test failed: unexpected address bar url: L"about:blank" htmldoc.c:3564: Test failed: unexpected address bar url: L"about:replace" === W7PROX64 (32 bit htmldoc) === htmldoc.c:3564: Test failed: unexpected address bar url: L"about:blank" htmldoc.c:3526: Test failed: unexpected address bar url: L"about:blank" htmldoc.c:3557: Test failed: unexpected address bar url: L"about:blank" htmldoc.c:3564: Test failed: unexpected address bar url: L"about:replace" === W7PROX64 (64 bit htmldoc) === htmldoc.c:3564: Test failed: unexpected address bar url: L"about:blank" htmldoc.c:3526: Test failed: unexpected address bar url: L"about:blank" htmldoc.c:3557: Test failed: unexpected address bar url: L"about:blank" htmldoc.c:3564: Test failed: unexpected address bar url: L"about:replace"

1 0

Wine Gecko 1.2.0 RC1
by Jacek Caban 10 Mar '11

10 Mar '11

Hi all, I'm happy to announce that the new update of Wine Gecko packages is almost there. It took long time since last beta to get to RC point from beta, not only because of many changes on our side, but mostly because of delays in Mozilla release process. We've done everything I've hoped to be done (and even more than I expected). I'm also happy that it was much better tested than previous times (mostly because they were available for so long). So far I'm aware of only one regression in Outlook that probably won't block the release because it seems to be a problem on Wine side (I can't be sure as I wasn't able to reproduce it so far). If I've missed something, please let me know. I'd appreciate help with testing. Builds are on Sourceforge [1]. To use it, just grab MSI files and put it into the directory where usually CAB files would go and apply the attached patch. Please let me know about any regression you notice. * CHANGES * - The CAB file is created without external dependencies Thanks to Alexandre for cabinet work. - We use MSI installers Thanks to Hans for winemsibuilder and preparing scripts that made it possible. - winedbg works with Gecko debug builds Thanks to Eric for many fixes to dbghelp/winedbg. Once he will merge his work to Wine, we will be able to use winedbg for comfortable Gecko debugging - win64 build now works - a few bug fixes - I've got rid of some Gecko ugly things like using API hooks * FURTHER PLANS * Firefox 4 RC1 came out yesterday and is expected to be the final build (so far nothing that would require another RC was found). Mozilla estimates release to be out in a week or two. I'm tempted to not wait and release early next week (so that it will live in Git for at least a few days before Wine release), but we will see how it will work. If we'd find any new blocker, it would most likely mean that we will release a week later. * FOR PACKAGERS * If you are interested in building the packager yourself from sources (so far I've heard about Debian and Fedora attempts to do so), everything is ready to support it now. Building instructions changed a bit after switching to use MSI files, but I've made things as easy as possible. Building the package itself is a matter of invoking one script with proper arguments. The tricky part is that Wine is needed for building, so we get circural build-time dependency. The script, however, makes sure that Wine doesn't need Gecko during compiling Gecko, so it shouldn't be a big problem to handle. Also, unless you manually set WINEPREFIX, the script created a temporary wineprefix directory, so there is nothing to worry if your ~/.wine is in broken state. It can also use Wine from its build dir if you set WINE_BUILD_DIR environment variable. Please let me know if there are more problems than need to be addressed and feel free to contact me if you have any questions. Thanks, Jacek [1] http://sourceforge.net/projects/wine/files/Wine%20Gecko/1.2.0-rc1/

3 4

Re: wininet(6/10): Implement commiting a url cache entry that already existed
by Alexandre Julliard 10 Mar '11

10 Mar '11

Juan Lang <juan.lang(a)gmail.com> writes: > As a nice side effect, the aggravating fixme about not knowing what to > do disappears. The tests just hang for me: ../../../tools/runtest -q -P wine -M wininet.dll -T ../../.. -p wininet_test.exe.so http.c && touch http.ok fixme:wininet:set_cookie persistent cookies not handled (L"expires=Sat, 09 Apr 2011 14:14:39 GMT; path=/") fixme:wininet:InternetSetOptionW Option 43 STUB fixme:wininet:InternetSetOptionW Option 44 STUB fixme:wininet:CommitUrlCacheEntryInternal Trying to delete entry of unknown format "\x00\x00\x00\x00" -- Alexandre Julliard julliard(a)winehq.org

2 1