September 2010 - Wine-devel

Re: [PATCH 2/2] shlwapi: Improved UrlEscapeW implementation
by testbot＠testbot.winehq.org 05 Sep '10

05 Sep '10

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=5038 Your paranoid android. === W98SE (32 bit url) === url.c:742: Test failed: UrlCanonicalizeA dwFlags 0x00000000 url 'c:dir\file' Expected "file:///c:dir/file", but got "file:///c:/dir/file", index 42 url.c:742: Test failed: UrlCanonicalizeA dwFlags 0x00000000 url ' c :dir\file' Expected "file:///c:dir/file", but got "file:///c:/dir/file", index 43

1 0

Re: [PATCH 1/2] shlwapi: Improved UrlCanonicalize implementation
by testbot＠testbot.winehq.org 05 Sep '10

05 Sep '10

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=5037 Your paranoid android. === W98SE (32 bit url) === url.c:742: Test failed: UrlCanonicalizeA dwFlags 0x00000000 url 'c:dir\file' Expected "file:///c:dir/file", but got "file:///c:/dir/file", index 42 url.c:742: Test failed: UrlCanonicalizeA dwFlags 0x00000000 url ' c :dir\file' Expected "file:///c:dir/file", but got "file:///c:/dir/file", index 43

1 0

Re: [1/5] urlmon: Implemented IUriBuilder_{Get/Set}Fragment (try 2)
by Jacek Caban 05 Sep '10

05 Sep '10

Hi Thomas, On 9/5/10 8:10 PM, Thomas Mullaly wrote: > Changes since previous attempt: > > 1. Perform lazy allocation/copy of a URI property when the > IUriBuilder_Get* function is called, instead of always allocating when > the IUriBuilder is created (addresses Jacek's comments). > > 2. Created helper functions for Get/Set calls to the IUriBuilder to cut > back on code duplication. > > 3. Added a bit flag member to the UriBuilder object, so that it can store > which properties have been changed. It looks much better now. There is one problem remaining (that I think can be fixed in a follow-up patch): + This->modified_props |= Uri_HAS_FRAGMENT; + return set_builder_component(&This->fragment,&This->fragment_len, pwzNewValue, '#'); The modifier_props should be set only on success. To achieve so, I think it would be cleaner to pass a flag to be set to set_buildier_component. Jacek

1 0

Re: gdi32/tests: Fix a typ description in test_logpen.
by James McKenzie 05 Sep '10

05 Sep '10

Gerald Pfeifer wrote: > winebot claims this is causing a tons of test failures on W98SE, > but that doesn't make sense at all. > > Gerald: What is the job id on the testbot? I don't see a failure message in Wine-Devel. James McKenzie

1 0

Re: [PATCH 1/3] Validate buffers passed to schan_DecryptMessage
by Mikko Rasa 05 Sep '10

05 Sep '10

The patch list requests me to write tests for this patch. I would do that, but there is a problem: I need a fully initialized schannel security context to use DecryptMessage. To acquire one I need to communicate with a server component. Wine is lacking an implementation of at least the AcceptSecurityContext function, so at the moment I cannot do a self-contained test. Communicating with an external server is an option, but I don't know which would be a good server for testing. Implementing the server-side stuff is another, but might take quite a while and I'd like to see this patch applied soon as it's required to play several recent Ubisoft games. Mikko On 30.08.2010 19:37, Mikko Rasa wrote: > --- > dlls/secur32/schannel.c | 50 ++++++++++++++++++++++++++++++++++++++++++---- > 1 files changed, 45 insertions(+), 5 deletions(-)

1 0

Re: msxml3: IXMLDOMText_get_text return trimmed text when XML_TEXT_NODE
by Alexandre Julliard 05 Sep '10

05 Sep '10

Alexandre Goujon <ale.goujon(a)gmail.com> writes: > @@ -181,6 +181,30 @@ static inline BSTR bstr_from_xmlChar(const xmlChar *str) > return ret; > } > > +static inline BSTR bstr_trim(const BSTR str) > +{ > + UINT i, j, len = SysStringLen(str); > + BSTR new_str = NULL; > + > + /* str empty or NULL */ > + if(!len) > + goto err; > + > + for(i=0; isspaceW(str[i]) && i<len; i++); > + for(j=len-1; isspaceW(str[j]) && j>=i; j--); > + > + new_str = SysAllocStringLen(str+i, j-i+1); > + if(!new_str) > + goto err; > + > + SysFreeString(str); > + return new_str; > + > + err: > + SysFreeString(str); > + return SysAllocStringLen(NULL, 0); > +} That's ugly and inefficient, especially since the common case is to not modify it. -- Alexandre Julliard julliard(a)winehq.org

6 7

Re: msvcp90: Fix 64-bit compilation warnings
by Alexandre Julliard 05 Sep '10

05 Sep '10

Piotr Caban <piotr(a)codeweavers.com> writes: > @@ -596,7 +596,7 @@ DEFINE_THISCALL_WRAPPER(MSVCP_basic_string_char_erase, 12) > basic_string_char* __thiscall MSVCP_basic_string_char_erase( > basic_string_char *this, size_t pos, size_t len) > { > - TRACE("%p %d %d\n", this, pos, len); > + TRACE("%p %d %d\n", this, (int)pos, (int)len); You should be casting to long, otherwise you'll potentially truncate the value. -- Alexandre Julliard julliard(a)winehq.org

3 3

Re: winealsa.drv: remove unused variable
by Chris Robinson 05 Sep '10

05 Sep '10

On Saturday, September 04, 2010 11:09:03 pm Austin English wrote: > - frames_per_period = snd_pcm_bytes_to_frames(wwi->pcm, > wwi->dwPeriodSize); > + snd_pcm_bytes_to_frames(wwi->pcm, wwi->dwPeriodSize); The whole call can probably be removed if the result isn't used anywhere. The function just returns the number of frames that would fit in the specified byte count for the specified device's format.

1 0

Re: [setupapi] added tests for SetupDiGetINFClassA (try2)
by testbot＠testbot.winehq.org 04 Sep '10

04 Sep '10

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=5026 Your paranoid android. === WNT4WSSP6 (32 bit devinst) === No test summary line found

1 0

Re: attrib.exe: extendended from stub
by Austin English 04 Sep '10

04 Sep '10

On Sat, Sep 4, 2010 at 1:53 PM, Konrad Wartke <kwartke(a)googlemail.com> wrote: > attrib.exe: extendended from stub > is now parsing arguments and implements them partly + // FIXME: only display attributes Please use C89 /* comments */. A testsuite for attrib.exe (look at cmd.exe and regedit.exe) would also be helpful. -- -Austin

1 0