July 2008 - Wine-Devel - list.winehq.org

[crypt32/tests] Don't crash on win9x

by Juan Lang

Hi Paul, thanks very much for looking into the crypt32 tests. - /* Weird--bad add disposition leads to an access violation in Windows. - */ (snip) + if (!on_win9x) + { + /* Weird--bad add disposition leads to an access violation in Windows. + * Both tests crash on win9x + */ This suggests to me that the test should just be removed, as it's very unlikely that an application will depend on such behavior. Leaving it around with an if (0) around it seems reasonable too. Any of these tests whose behavior is inconsistent across Windows versions, especially where some Windows versions crash, are good candidates for removal, IMO. --Juan

July 30, 2008

Re: [PATCH 1/5] quartz: Add tests for Video Mixing Renderer 9

by Alexandre Julliard

"Maarten Lankhorst" <m.b.lankhorst(a)gmail.com> writes: > I want to re-use the code in the videorenderer test. I have some > patches for that but they should be part of a different patchset. If I > didn't want to do that I would have put it in vmr9.c, if you really > want to I can duplicate the files. No, just put it in either vmr9.c or videorenderer.c and reference it from the other file. It means you have to link both of them together to build a stand-alone test, but that's ok. -- Alexandre Julliard julliard(a)winehq.org

July 30, 2008

Re: [PATCH 4/5] quartz: Make the vmr-9 implementation work

by Maarten Lankhorst

Hoi Henri, 2008/7/30 H. Verbeet <hverbeet(a)gmail.com>: > I just looked the patch over quickly since you mentioned d3d9, but > here are a few things I noticed: > >> + hr = IDirect3DSurface9_LockRect(info->lpSurf, &lock, NULL, D3DLOCK_DISCARD); >> + assert(hr == S_OK); > ... >> + hr = IDirect3DDevice9_Clear(This->d3d9_dev, 0, NULL, D3DCLEAR_TARGET, D3DCOLOR_XRGB(0, 0, 0), 1.0f, 0); >> + if (hr != S_OK) > > I realize D3D_OK and S_OK have the same value, but using > SUCCEEDED()/FAILED() would probably be more appropriate. Happens in a > few other places as well. I was just trying to be pedanctic with the return values, but SUCCEEDED is probably ok too. >> +// info.dwFlags = VMR9AllocFlag_TextureSurface; > ... >> + FIXME("SendMessageW: %08x\n", hr); > You'll probably want to remove these. There is a reason I use C++ style comments, mainly because they make sure those lines never make it in the wine tree, it should just be removed, when I grepped vmr9.c after patch 5/5 I couldn't find that line any more, so I probably immediately removed it there, thanks for catching it. Cheers, Maarten

July 30, 2008

Re: quartz: Add tests for avi splitter

by Maarten Lankhorst

Hi Paul, 2008/7/30 Paul Vriens <paul.vriens.wine(a)gmail.com>: > Hi Maarten, > > NT4 doesn't have CreateToolhelp32Snapshot so none of the quartz tests will > run on NT4 currently. > > Should I just create a patch that checks for the existence of this function > and skip some tests? > > Or do we have another way to get the number of threads? We could use > NtQuerySystemInformation but that's not available on Win9x. We could also > have a mix of these two depending on the availability of > CreateToolhelp32Snapshot. Just check for the presence of that function, and skip the tests that count the threads if that function isn't available, that it's skipped on 1 platform is really not that important. Cheers, Maarten.

July 30, 2008

Re: [PATCH 1/5] quartz: Add tests for Video Mixing Renderer 9

by Maarten Lankhorst

Hi Alexandre, 2008/7/30 Alexandre Julliard <julliard(a)winehq.org>: > "Maarten Lankhorst" <m.b.lankhorst(a)gmail.com> writes: > >> From f3544f718dbebd3dca29d94dbbcbdd071884e5a6 Mon Sep 17 00:00:00 2001 >> From: Maarten Lankhorst <m.b.lankhorst(a)gmail.com> >> Date: Mon, 28 Jul 2008 18:18:06 -0700 >> Subject: [PATCH] quartz: Add tests for Video Mixing Renderer 9 >> >> --- >> dlls/quartz/tests/Makefile.in | 5 +- >> dlls/quartz/tests/null.h | 667 +++++++++++++++++++++++++++++ > > You should put that code in the C file, it doesn't belong in a header, > especially since you are not using anywhere else. Building tests > stand-alone is a nice to have, but it's not worth the cost of putting > half the code in a header. I want to re-use the code in the videorenderer test. I have some patches for that but they should be part of a different patchset. If I didn't want to do that I would have put it in vmr9.c, if you really want to I can duplicate the files. >> +#define trace_save trace >> + >> +/* Don't debug null by default */ >> +#define debug(...) do { } while (0) > > Vararg macros are not portable. Ok, I'll change it to (a, b) since there should not be any other arguments anyway. Cheers, Maarten.

July 30, 2008

Re: [PATCH] wine.inf: Change Msi.Package actions to have the full msiexec path.

by Lei Zhang

On Tue, Jul 29, 2008 at 8:19 PM, Vitaliy Margolen <wine-devel(a)kievinfo.com> wrote: > Lei Zhang wrote: >> Hi, >> >> On Windows, the Msi.Package action registry keys have the full path to >> msiexec.exe. >> ------------------------------------------------------------------------ >> >> +HKCR,Msi.Package\shell\Open\command,,0x20002,"""%SystemRoot%\System32\msiexec.exe"" /i ""%1"" %*" >> +HKCR,Msi.Package\shell\Repair\command,,0x20002,"""%SystemRoot%\System32\msiexec.exe"" /f ""%1"" %*" >> +HKCR,Msi.Package\shell\Uninstall\command,,0x20002,"""%SystemRoot%\System32\msiexec.exe"" /x ""%1"" %*" > > Should those 0x20002 be 0x20000 instead? Why do we need to preserve > overrides of something that can be opened with msi only? Sure, we would want users who upgraded Wine to get this in their existing .wine directories. Unlike the registry key for $PATH, I cannot imagine too many people would change this registry key. Will resend patch in a bit.

July 30, 2008

Re: WWN announcements on WineHQ

by Jeremy Newman

The problem is that currently our news articles do not have separate pages like Slashdot does. So there is no page to actually link to. Also, linking to the actual WWN issue is done in the body of the news post. In order to have the WWN issue link properly, the news XML would need a <link> tag or something so the RSS generator knows where to set the link. Francois Gouget wrote: > On Wed, 30 Jul 2008, Jeremy Newman wrote: > >> Not to mention, for people who subscribe to the RSS feed, it is a great >> way to let them know an Issue has been released. > > I'm not using RSS for WineHQ so I just discovered that they are broken. > The behavior I'm expecting is the same as for Slashdot's RSS feed: if I > click on 'Nasa Turns 50' I get to the page for that story. > > But if I click on 'Wine Weekly News Issue 348', I get to the top of > Wine's news page (essentially http://www.winehq.org/?news=archive) > There's two issues with that: > * when the page opens on my laptop the news item about WWN 348 is not > even visible. So for general news items we should at least link to > the relevant entry. That means adding anchors as currently there are > none. > * for news about WWN issues the link should point straight to the > issue, rather than to the news about it. The same as clicking on a > Slashodt story does not point you to the main index but to the story > itself. That probably means that news items should have a url field > to specify where the RSS entry should point to. >

July 30, 2008

Re: dinput: the DIK_ keycode is not the same as the scancode.

by Aric Stewart

Doing tests with shift show that the presence of the shift key does not affect MapVirtualKey so SHIFT+2 still returns '2' since it is scancode based i doubt that numlock would either. So I do not think we need to worry about that. -aric Vitaliy Margolen wrote: > Aric Stewart wrote: >> It is mapped with the keyboard mapping to the resulting character. so >> the key 'A' is DIK_A nomatter what its scancode or vkey would be. This >> is relevent to Japanese keymapping where the '@' key is in the '[' >> location the scancode for both is 0x22 but dinput generates DIK_AT in >> japanese and DIK_LBRACKET in us_qwerty >> >> reworked to remove the giant case statement and hopefully be more clear. >> remove japan specific stuff as with the proper keyboard scancodes it >> works out. > > I'm still not clear why do you need to translate the v_key into char? This > involves a round trip to the X server for each key press. > > Besides this will depend on the state of the num-lock, shift, etc. Which > should not be the case for dinput. It should always return the same code for > the key regardless of anything else. > > Vitaliy. > >

July 30, 2008

Re: WWN announcements on WineHQ

by Francois Gouget

On Wed, 30 Jul 2008, Francois Gouget wrote: > On Wed, 30 Jul 2008, Zachary Goldberg wrote: > [...] > > How then would people find the index of WWNs? I think having that box > > on the side is really convenient actually and is the only way I > > usually go to get to issues. > > With the 'More Issues...' link which would be right below the index in > the news item. I guess with all these changes the news file would look something like shown below with some caveats: * likely wrapping by my email tool * Don't know if one can put a link in the <title> field. If not, then add an 'Issue 349' line to the body. * I don't have a WineHQ mirror here so this is untested. <news> <date>July 13, 2008</date> <title><a href="http://www.winehq.org/?issue=349">Wine Weekly News Issue 349</a></title> <body> <p><ul> <li><a href="http://www.winehq.org/?issue=349#News:%20Stable%20and%20Development%20versi…">News: Stable and Development versions of Wine released</a></li> <li><a href="http://www.winehq.org/?issue=349#Wineconf%202008%20Announced">Wineconf 2008 Announced</a></li> <li><a href="http://www.winehq.org/?issue=349#GSoC%20Midterms">GSoC Midterms</a></li> <li><a href="http://www.winehq.org/?issue=349#Wine%20and%20CUDA%20(Nvidia%20GPGPU)">Wine and CUDA (Nvidia GPGPU)</a></li> <li><a href="http://www.winehq.org/?issue=349#Valgrind%20Updates">Valgrind Updates</a></li> <li><a href="http://www.winehq.org/?issue=349#Weekly%20AppDB/BugZilla%20Status%20Changes">Weekly AppDB/BugZilla Status Changes</a></li> </ul></p> <p><a href="http://www.winehq.org/?issue=back">More Issues...</a></p> </body> </news> -- Francois Gouget <fgouget(a)free.fr> http://fgouget.free.fr/ f u kn rd ts, ur wy 2 gky 4 ur wn gd.

July 30, 2008

Re: WWN announcements on WineHQ

by Francois Gouget

On Wed, 30 Jul 2008, Zachary Goldberg wrote: [...] > How then would people find the index of WWNs? I think having that box > on the side is really convenient actually and is the only way I > usually go to get to issues. With the 'More Issues...' link which would be right below the index in the news item. -- Francois Gouget <fgouget(a)free.fr> http://fgouget.free.fr/ Any sufficiently advanced bug is indistinguishable from a feature. -- from some indian guy

July 30, 2008