October 2010 - Wine-Devel - list.winehq.org

Re: [PATCH 2/2 (try2)] shlwapi: Handle URL_WININET_COMPATIBLITY flag in UrlCanonicalize
by Alexandre Julliard Oct. 8, 2010

Oct. 8, 2010

Andrew Eikum <aeikum(a)codeweavers.com> writes: > try2: Rebase on top of [1/2]. > > dlls/shlwapi/tests/url.c | 4 ++++ > dlls/shlwapi/url.c | 6 ++++-- > 2 files changed, 8 insertions(+), 2 deletions(-) It breaks the tests: ../../../tools/runtest -q -P wine -M cryptnet.dll -T ../../.. -p cryptnet_test.exe.so cryptnet.c && touch cryptnet.ok cryptnet.c:353: Test failed: File URLs not supported make: *** [cryptnet.ok] Error 1 -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: crypt32: Avoid relying on PATH_MAX in import_certs_from_dir helper.
by Austin English Oct. 8, 2010

Oct. 8, 2010

On Thu, Oct 7, 2010 at 6:56 PM, Andrew Nguyen <anguyen(a)codeweavers.com> wrote: > > PATH_MAX is not guaranteed to be available on all platforms, and it is > inadequate as a hardcoded buffer size anyway. Also fixes the first compiler error on GNU Hurd. Thanks Andrew! -- -Austin

1 0

New winetricks 20101008: new verbs abiword, dsound, nvidiasdk9, vd, xinput
by Dan Kegel Oct. 8, 2010

Oct. 8, 2010

Another fortnight-and-a-half, another Winetricks. Online as always at http://kegel.com/wine/winetricks or http://winezeug.googlecode.com (Bug reports to the issue tracker at the above URL, please.) Changes since 20100917: Austin English: - add dsound verb. (Issue 175) - add xinput verb - dxsdk_nov2006, add support for quiet install. - fix dotnet installs for wine 1.2 (broken in recent winetricks release) - introduce $appdata_win/$appdata_unix variables, use it for Safari - refuse to run in 64-bit WINEPREFIX's - remove deprecated pbuffer option - update eadm sha1sum - update flash sha1sums - update kde to 0.9.7-0 - update utorrent sha1sum/url Dan Kegel: - add nvidiasdk9 - remove bogus msvcr80 installed by dotnet20; fixes issue 164. - use symbolic variable uniformly in set_videomemorysize - add abiword, ie6_full - fix two spots WINETRICKS_TMP wasn't quoted; this broke WINEPREFIXes containing spaces. Reported by Roberto Neri. - add vd=1024x768, vd=off options. Fixes bug 172. Scott Ritchie: - new verb cjkfonts (individual verb names may still change) - Add Quicktime 7.6, update quicktimes to depend on vcrun2005 Erik Inge Bolsø - reenable mscoree after directx9 install; fixes issue 25.

3 2

Re: [4/4] msxml3/httprequest: Implement ::get_responseBody()
by testbot＠testbot.winehq.org Oct. 8, 2010

Oct. 8, 2010

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=5972 Your paranoid android. === W98SE (32 bit domdoc) === domdoc.c:3312: Test failed: got 1, expected 1 === W2KPROSP4 (32 bit domdoc) === domdoc.c:3312: Test failed: got 1, expected 1

1 0

Re: [2/4] msxml3/httprequest: Implement ::abort()
by testbot＠testbot.winehq.org Oct. 8, 2010

Oct. 8, 2010

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=5971 Your paranoid android. === W98SE (32 bit domdoc) === domdoc.c:3311: Test failed: got 1, expected 1 === W2KPROSP4 (32 bit domdoc) === domdoc.c:3311: Test failed: got 1, expected 1

1 0

Re: [PATCH 1/3] jscript: test that ActiveScriptSite_OnScriptError is called on script failure. (resend)
by Reece Dunn Oct. 7, 2010

Oct. 7, 2010

On 7 October 2010 14:35, Jacek Caban <jacek(a)codeweavers.com> wrote: > Hi Reece, > > These tests need more thoughts. They should show how jscript exceptions are > translated into IActiveScriptError. Your code doesn't provide mechanism to > do so, so I would suggest to wait with sending patches that don't move us > any forward until you have a clue how it should work. Otherwise you will > have to deal with eventually bad decisions for no good reason. Ok, so if I keep the current tests/ActiveScriptSite as they are and have something like: test_script_error(source, scode, description, lineno, charpos, linesource) { EXPECT CALL ActiveScriptSite_OnScriptError hres = parse_script_with_error(source); ok(hres == 0x8012345, ...) CHECK CALL ActiveScriptSite_OnScriptError ok(script_excep.scode == scode, ...) ok(script_excep.bstrDescription == description, ...) ok(script_lineno == lineno, ...) ok(script_charpos == charpos, ...) ok(script_linesource == linesource, ...) } test_script_errors() { test_script_error("var foo = new Object();\nfoo.bar;", 0x8012345, "Object does not support this property", 20, 11, "foo.bar;") test_script_error("new <>%^^;", 0x8012322, "Syntax error", 0, 0, "new <>%^^;") ... } where parse_script_with_error does the same thing as parse_script, but uses an ActiveScriptSite implementation that checks OnScriptError calls and stores the IActiveScriptError object for testing within test_script_error (like with my 3rd patch, but checking specific values). 1/ Does this look like a sensible approach to you? 2/ Do we care about the tests working on non-English locales (not sure how to get the above working on other languages -- specifically the description text)? 3/ Are we ok with making wine match the error description strings exactly, given the above tests? NOTE: 2 and 3 can be avoided by just checking the scode and checking that the description is set and non-NULL (as per my 3rd patch). > Also merging parsing_htmlscript is not a good idea. Its purpose is different > than parse_script. Ok, I'll keep them separate. - Reece

2 1

Re: [PATCH 1/2] shlwapi/tests: Mark Win98 failures as broken
by Jacek Caban Oct. 7, 2010

Oct. 7, 2010

Hi Andrew, Can we please find a way to detect such old (broken) shlwapi and just skip the whole test (preferable all tests in the file)? Failing UrlGetPart would do the trick, hopefully there is better way. This will result in cleaner code and will save us time in future. Jacek

1 0

Re: ntdll: fix compile on systems without termios.port.c_ispeed/termios.port.c_ospeed structs
by Alexandre Julliard Oct. 7, 2010

Oct. 7, 2010

Austin English <austinenglish(a)gmail.com> writes: > @@ -498,7 +498,7 @@ static NTSTATUS set_baud_rate(int fd, const SERIAL_BAUD_RATE* sbr) > return STATUS_NOT_SUPPORTED; > #endif /* Don't have linux/serial.h or lack TIOCSSERIAL */ > } > -#elif !defined(__EMX__) > +#elif !defined(__EMX__) && defined (_HAVE_STRUCT_TERMIOS_C_ISPEED) && defined (_HAVE_STRUCT_TERMIOS_C_OSPEED) > switch (sbr->BaudRate) > { > case 0: port.c_ospeed = B0; break; You'd want to use the cflag method in this case. Which platform needs this? -- Alexandre Julliard julliard(a)winehq.org

2 1

Mentor summit for Google Summer of Code - time sensitive
by Jeremy White Oct. 7, 2010

Oct. 7, 2010

As a heads up, Bradley reminds me that the Mentor summit for GSOC: http://gsoc-wiki.osuosl.org/index.php/2010 is coming up; the registration deadline is tomorrow. Google may reimburse travel expenses, so if cost is an issue, we may be able to help with that. If you need such help, let me know, and I'll connect you with Bradley. Cheers, Jeremy

1 0

Re: [6/6]strmbase: Implement BaseFilter in strmbase
by Alexandre Julliard Oct. 7, 2010

Oct. 7, 2010

Aric Stewart <aric(a)codeweavers.com> writes: > --- > dlls/qcap/vfwcapture.c | 134 +++++-------------------- > dlls/quartz/acmwrapper.c | 20 ++-- > dlls/quartz/avidec.c | 12 +- > dlls/quartz/avisplit.c | 6 +- > dlls/quartz/dsoundrender.c | 223 > ++++++++++++------------------------------- > dlls/quartz/filesource.c | 129 +++++-------------------- > dlls/quartz/mpegsplit.c | 4 +- > dlls/quartz/nullrenderer.c | 177 ++++++++-------------------------- > dlls/quartz/parser.c | 156 ++++++++++-------------------- > dlls/quartz/parser.h | 9 +-- > dlls/quartz/transform.c | 201 ++++++++------------------------------- > dlls/quartz/transform.h | 10 +-- > dlls/quartz/videorenderer.c | 226 > +++++++++++++------------------------------ > dlls/quartz/waveparser.c | 4 +- > dlls/strmbase/Makefile.in | 1 + > dlls/strmbase/filter.c | 158 ++++++++++++++++++++++++++++++ > include/wine/strmbase.h | 23 +++++ > 17 files changed, 517 insertions(+), 976 deletions(-) > create mode 100644 dlls/strmbase/filter.c I'm seeing occasional crashes in the qedit:mediadet test with this patch. Unfortunately the debugger fails to get a backtrace. -- Alexandre Julliard julliard(a)winehq.org

2 1