July 2009 - Wine-Devel - list.winehq.org

Re: advapi32: test a combination of CreateServiceA and GetSecurityInfo [2nd]
by Paul Vriens July 21, 2009

July 21, 2009

Stefan Leichter wrote: > Am Monday 20 July 2009 14:13:10 schrieb Alexandre Julliard: >> You could insert that into one of the other tests that already create a >> service, so that we don't need to have too many Sleeps in there. > > This testcase demonstrates the current problem in bug 1886 > > > ------------------------------------------------------------------------ > > Hi Stefan, Please stick to the style of the file, curly braces on a separate line. (I see there is only one such style difference in the current file, but that was also introduced by you ;) ) -- Cheers, Paul.

1 0

Bug day a success!
by Scott Ritchie July 21, 2009

July 21, 2009

I'd like to thank everyone that participated in our first Bug Day. I'm definitely going to repeat it again next release, although I may move it back a few days. Some statistics: Our bug target search (bugs unchanged since January 1st with the download tag) started at 574 bugs. Now it's down to 508. That's 66 bugs that were either resolved, confirmed, or at least checked in on. In the past two days there have been 30 bugs newly resolved. By comparison, over the previous 2 weeks before then there were only 16 bugs resolved. So, we're definitely making progress. Quite a few of us had some fun, and I personally get a real satisfaction out of transforming our bug database from a scary mess to a place we actually look forward to visiting. I'd like to give a personal thanks to Luke Benstead, who set a great example by gathering some of his real life friends to participate. Luke has shown himself to be very capable, and he deserves bugzilla permissions so he can help facilitate bug hunts in the future. Again, thank you everyone! Thanks, Scott Ritchie

1 0

Re: ws2/tests: Test AcceptEx with a deferred socket
by Paul Vriens July 21, 2009

July 21, 2009

Mike Kaplinskiy wrote: > This tests the following AcceptEx scenario: > WSAAccept->CF_DEFER->AcceptEx > Windows seems to return the deferred socket with AcceptEx (verified on > XP & 2K3). > > Mike. > > > ------------------------------------------------------------------------ > > Hi Mike, These new tests fail on all NT4 boxes (some 6 and 1 has 4000+ failures) and what looks like to be all Windows x64_64 systems. The strange thing (or maybe not) is that the errors are the same on all of these (one box has 4000+ failures but still has those same 6). See: http://test.winehq.org/data/tests/ws2_32:sock.html Could you have a look? I can do some tests if necessary on NT4, I don't have a Windows x86_64 system however. -- Cheers, Paul.

2 6

FW: Re: please reduce the four registrations on the single winehq.org site.
by Scott Ritchie July 20, 2009

July 20, 2009

S Page wrote: > Hey, I really want to help the Wine project out by sharing my problems > and workarounds. > > But having to register FOUR TIMES on the same winehq.com site (bugs, > AppDB, forums, wiki) is excruciatingly painful and hostile to your > would-be contributors. I realize you're just using open source > tooling, but maybe you can press those four products to > a) implement a single sign-on > b) support OpenID > > Thanks, > -- > =S Page Strangely enough, I was just talking about this today at the Community Leadership Summit. Unfortunately it seems like it's a pretty difficult problem to solve as the tools to get all these things talking to eachother haven't been written. There are packages that integrate a forum and a wiki and such together, but they don't provide an easy way to get the data out of them - switching to them also presents huge data migration issues. Mozilla are farthest on the game here, as they were working on writing some of their own custom stuff based on OpenID, however that then deals with the need of having each app support an OpenID login. I'm not sure if it's a good idea, but there may be some merit in having one of the accounts function as an OpenID Provider and then have the others be able to link to it (but still be "separate accounts"). After that it's just about being able to login by OpenID rather than have a single account moving across each place. Thanks, Scott Ritchie

3 3

D3D shader assembler
by Matteo Bruni July 20, 2009

July 20, 2009

Hi to everybody, I'm sending here the main assembler patch for reviews and suggestions. It is bzipped as it is quite a big patch, but I couldn't find a meaningful way to split it. A quick mile-high overview: the assembler is divided in a parser and a bytecode writer. The parser is made by the flex (asmshader.l) and bison (asmshader.y) source files and the asmparser.c file. The bytecode writer is in bytecodewriter.c, btw, and asmutils.c contains some utility functions used in the assembler. With the asmshader.l SlAssembleShader function the parser processes the shader and returns an intermediate representation of it into the wine_shader struct. Then this struct is passed to the bytecode writer with the SlWriteBytecode function, which produces the assembled shader bytecode. A "point of interest": in the bottom part of asmshader.h there are some definitions borrowed from wined3d. They are used only internally, so they should not be a problem, but I'd like to hear if you think differently (ex. it is better to rename them?). Thank you, Matteo Bruni

3 8

How to expanding environmental variables?
by Austin English July 20, 2009

July 20, 2009

Howdy all, I'm working on a automated test for Photoshop CS 2. As before, I'm trying to do so in a portable way, so it works on various locales/windows versions (or as much as possible). Because of wine/windows differences, I need a portable way to get to the shared documents folder, e.g.,: Windows XP: C:\Documents and Settings\All Users\Shared Documents WIndows Vista: C:\users\Public\Public Documents Wine: C:\users\Public\Documents Windows has several variables for this sort of thing (http://msdn.microsoft.com/en-us/library/ms933062(WinEmbedded.5).aspx) In particular, I need a way to return what %16430% is on the target system. I've tried using cmd.exe to echo it, but it's not expanding it. Does anyone know how I can expand this type of variable? I figured there's probably something in shell32 that can accomplish this, but nothing popped up at me immediately (and I'm about to run, so I wanted to e-mail before I leave). Surely there's an easy way and I'm just overlooking it... Thanks! -- -Austin

6 7

Re: netapi32: Fix a pointer conversion warning on 64-bit
by Juan Lang July 20, 2009

July 20, 2009

Hi André, @@ -577,7 +577,7 @@ static UCHAR NetBTinetResolve(const UCHAR name[NCBNAMSZ], (*cacheEntry)->numAddresses = i; for (i = 0; i < (*cacheEntry)->numAddresses; i++) (*cacheEntry)->addresses[i] = - (DWORD)host->h_addr_list[i]; + (DWORD_PTR)host->h_addr_list[i]; This isn't correct, as the lvalue is still a DWORD, and needs to remain a DWORD. If you like you can check that the length, host->h_length, is the same as sizeof(DWORD), but this is pretty well implied by the use of gethostbyname(): it only returns addresses of type AF_INET. Basically, the warning is a false positive. --Juan

1 0

Re: XkbSetDetectableAutoRepeat not working on some platforms
by F Capela July 20, 2009

July 20, 2009

From: Dmitry Timoshkov <dmitry(a)codeweavers.com> > "F Capela" <fabio.capela(a)yahoo.com> wrote: > > > XkbSetDetectableAutoRepeat is not working correctly on some > > distributions, including Ubuntu 9.04. In practical terms, this > > means programs running under Wine on such distros are unable to > > ignore autorepeat, doing strange things to games that use keyboard > > controls, for example. One way to work around the problem is to > > manually filter out the key release messages, which can be done > > with a few extra lines in dlls/winex11.drv/event.c . My question is, > > should I send a patch enabling this filtering? From what I have seen, > > patches to work around bugs that are clearly outside Wine are usually > > not accepted. > > Have you reported a bug to XOrg and those distros about that? > > -- Dmitry. I have not yet filled a bug with Ubuntu (which is the distro I'm using), I'll do it as soon as I find how, but while searching for information about this issue I saw a bug filled for Gentoo (which I'm not sure if it's still present), and there are two open bugs under bugs.freedesktop.org (http://bugs.freedesktop.org/show_bug.cgi?id=22194 and http://bugs.freedesktop.org/show_bug.cgi?id=22515) about XkbSetDetectableAutoRepeat not working, both filled in june, one reported under Ubuntu and the other for the GIT version. A bit of extra information, the only software I use that seems to be affected is World of Warcraft; the walk animation stutters when using the keyboard to move, and some things that should not auto repeat (jumps, spell casting, etc) do repeat. BTW, the same bug will happen (and be wine's fault) if the X server does not support DetectableAutorepeat, since when it calls XkbSetDetectableAutoRepeat in dlls/winex11.drv/x11drv_main.c, wine does not care about the function's return value or third argument (although I don't think there is any widely used version of the X server that don't support DetectableAutorepeat). Fábio Oliveira Schmidt Capela

1 0

Re: suggestions about MacOS DYLD_FALLBACK_LIBRARY_PATH patch
by James Mckenzie July 20, 2009

July 20, 2009

>On Sun, Jul 19, 2009 at 12:23 PM, Emmanuel Maillard<mahanuu(a)free.fr> wrote: >> But it's an external appplication, your desktop environment, that handle >> *.desktop file; wine just generate them... > >As per the private email I sent. I could see that working if we can >convince Alexandre of the need of an external helper application for >OS X. Keeping the existing winemenubuilder code mostly as is and >having it write xdg menus and associations would be pretty simple. We >could just change a few lines to have it write xdg in >~/Library/Application Support/Wine/WineMenuBuilder > >> So don't choose ~/Applications but : /Applications/Wine or >> let's the user to choose if want an Applications folder in is Homedir. > >The reason for using ~/Applications as opposed to /Applications is >that Wine has no knowledge or real support for multi-user prefixes. >Each .wine being local to the user that creates it will have their own >applications, etc. If we write the menus to /Applications and another >user logs in then they won't be able to run those applications. Another good reason is that /Applications can only be written to by an administrative user using sudo. James McKenzie

1 0

RE: D3D shader assembler (Matteo Bruni)
by Matteo Bruni July 20, 2009

July 20, 2009

2009/7/20 "Luis C. Busquets Pérez" <luis.busquets(a)ilidium.com>: > I read several months ago that the itnetion was to create the assembler > inside the wined3d and then use its functionality inside the diverse d3dx9 > implementations. Has this changed? I think it was Stefan Dösinger sho > proposed it: > > d3dx -> wined3d assembler > the wined3d will be re-usable and the d3dx functions will be extremely easy. > > > Yes, the plan changed in the meantime. Now d3dx8 isn't in Wine anymore (it wasn't used before anyway, the apps using d3dx8 are usually linked against a static version of the library) so the assembler is needed only by d3dx9. The assembler can be put into d3dx9_36 and the other d3dx9_xx dlls can forward to it, as it is for the other d3dx9 functions. DirectX 10 doesn't support anymore assembler shaders, only HLSL one, so no worry from there. That said, maybe a part of the assembler, the bytecode writer, could be useful for the shader compiler. So that part could be moved into wined3d to have it in common between the assembler and the compiler, but this is not something that will happen (if it will happen) soon.

1 0

Re: advapi32: test a combination of CreateServiceA and GetSecurityInfo
by Alexandre Julliard July 20, 2009

July 20, 2009

Stefan Leichter <Stefan.Leichter(a)camline.com> writes: > + /* Wait a while. The following test also does a CreateService for the > + * same servicename and this would result in an ERROR_SERVICE_MARKED_FOR_DELETE > + * error if we do this to quick. Vista seems more picky then the others. > + */ > + Sleep(1000); > + > + svc_handle1 = CreateServiceA(scm_handle, servicename, NULL, SERVICE_ALL_ACCESS, SERVICE_WIN32_SHARE_PROCESS, > + SERVICE_DEMAND_START, SERVICE_ERROR_NORMAL, pathname, NULL, NULL, NULL, NULL, NULL); > + ok(svc_handle1 != NULL, "Could not create the service : %d\n", GetLastError()); > + if (svc_handle1 != NULL) { > + PSID sidOwner, sidGroup; > + PACL dacl, sacl; > + PSECURITY_DESCRIPTOR pSD; > + HRESULT retval = GetSecurityInfo(svc_handle1,SE_SERVICE,DACL_SECURITY_INFORMATION,&sidOwner,&sidGroup,&dacl,&sacl,&pSD); > + todo_wine ok(ERROR_SUCCESS == retval, "Expected GetSecurityInfo to succeed: result %d\n",retval); > + } You could insert that into one of the other tests that already create a service, so that we don't need to have too many Sleeps in there. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: dsound: Fix some pointer conversion warnings on 64-bit
by Alexandre Julliard July 20, 2009

July 20, 2009

André Hentschel <nerv(a)dawncrow.de> writes: > --- a/dlls/dsound/propset.c > +++ b/dlls/dsound/propset.c > @@ -109,7 +109,7 @@ static HRESULT WINAPI IKsBufferPropertySetImpl_Get( > S(prop).Set = *guidPropSet; > S(prop).Id = dwPropID; > S(prop).Flags = 0; /* unused */ > - S(prop).InstanceId = (ULONG)This->dsb->device; > + S(prop).InstanceId = (ULONG_PTR)This->dsb->device; If the destination is a DWORD changing the cast is only hiding the problem. -- Alexandre Julliard julliard(a)winehq.org

1 0

Doing better than barely keeping up with bug reports - Bug Day this Monday (July 20)
by Scott Ritchie July 20, 2009

July 20, 2009

So, briefly: Over the past few months, users have added an average of between 12 and 14 bugs every day. Since June 1st: - 412 total bugs filed - 87 bugs resolved invalid - 227 bugs resolved fixed - 133 bugs confirmed but not resolved (status new) - 292 bugs created but unconfirmed Doing some subtraction that means we have 292 new untriaged bugs, but we triaged or fixed (412-292-87-227-133) = 327 old bugs. So, we're swimming above water, which is good. But at this rate it'll be years before we triage every bug. So, let's do something :) Bug Jam this Monday! One idea that has been tried in the past is to hold regular bug days. We've had a lot of success with them in Ubuntu as a way of organizing non developer volunteers, especially when we focus the event on a particular package. Today's bug day, for instance, tackled over a 100 Synaptic bugs: https://wiki.ubuntu.com/UbuntuBugDay/20090716 I'd like to attempt this in Wine. We tried it once in the past, but no one's organizing it now. Even if it's just me and 2 people testing a handful of bugs in our favorite apps, it'll still sink a good number of bugs and help drain the swamp. I'm picking Monday for bug day for a few reasons - it's both after a new release and after the weekend, so users will have already had time to play their games and see if they're still affected. If there's any sort of success, hopefully this will become a regular event. So, if you're into bug triage, please come and join me in #winehackers this Monday. The purpose of triaging bugs is to ultimately get them fixed, so if you're a developer and would rather work on patches then by all means do that instead. I've created a wiki page for the event here: http://wiki.winehq.org/BugDay When we're done I'll poke bugzilla for some stats and we can see how much of a success the event was. Thanks, Scott Ritchie

4 4

Re: msi: Fix some pointer conversion warnings on 64-bit(try2)
by Mike Kaplinskiy July 20, 2009

July 20, 2009

André Hentschel wrote: > didnt want to change both variables, now its fine > sry for the noise > --- > dlls/msi/join.c | 5 +++-- > dlls/msi/msiquery.c | 4 ++-- > dlls/msi/storages.c | 2 +- > dlls/msi/streams.c | 2 +- > 4 files changed, 7 insertions(+), 6 deletions(-) > > > > + UINT_PTR i > + UINT row_value; Missing a semicolon? ;) Mike.

1 0

XkbSetDetectableAutoRepeat not working on some platforms
by F Capela July 20, 2009

July 20, 2009

XkbSetDetectableAutoRepeat is not working correctly on some distributions, including Ubuntu 9.04. In practical terms, this means programs running under Wine on such distros are unable to ignore autorepeat, doing strange things to games that use keyboard controls, for example. One way to work around the problem is to manually filter out the key release messages, which can be done with a few extra lines in dlls/winex11.drv/event.c . My question is, should I send a patch enabling this filtering? From what I have seen, patches to work around bugs that are clearly outside Wine are usually not accepted. I'm adding the patch inlined in this message for those that might want to look at it; if I feel the patch do have a chance to be accepted I will resend to wine-patches. Fábio Oliveira Schmidt Capela fabio.capela(a)yahoo.com diff --git a/dlls/winex11.drv/event.c b/dlls/winex11.drv/event.c index a965b9f..793b842 100644 --- a/dlls/winex11.drv/event.c +++ b/dlls/winex11.drv/event.c @@ -283,6 +283,13 @@ static enum event_merge_action merge_events( XEvent *prev, XEvent *next ) return MERGE_DISCARD; } break; + case KeyRelease: + if (prev->xany.window == next->xany.window && next->type == KeyPress && next->xkey.time == prev->xkey.time && next->xkey.keycode == prev->xkey.keycode && next->xkey.state == prev->xkey.state) + { + TRACE( "Discarding autorepeat key release for window %lx\n", prev->xany.window ); + return MERGE_DISCARD; + } + break; } return MERGE_HANDLE; }

2 1

appinstall update: Adobe Photoshop CS 2 installer test added
by Austin English July 20, 2009

July 20, 2009

For those interested, I've added a test for the Photoshop CS 2 demo installer (http://code.google.com/p/winezeug/source/detail?r=580 / http://winezeug.googlecode.com/svn/trunk/appinstall/scripts/photoshopcs2.ahk) Currently, the test only installs the demo and verifies the install. I'll add tests to make sure it starts up in the next day or so. Whlie it works, I found a few previously hidden bugs: http://bugs.winehq.org/show_bug.cgi?id=19342 - Adobe Photoshop CS2: Doesn't display license screen for installation (appinstall) http://bugs.winehq.org/show_bug.cgi?id=19343 - Adobe Photoshop CS2: Edit controls have wrong names (appinstall) http://bugs.winehq.org/show_bug.cgi?id=19344 - Adobe Photoshop CS2: Some files missing on installation (appinstall) If you've got a few minutes, check out those bugs and see if you have any information about them. Other than that, CS 2 seems to install fine. P.S., I'm working on some instructions for writing new tests, that I'll put on the wiki (later tomorrow-ish). While the testing framework is mostly complete, it's still pretty alpha/beta, and stuff may change, as I find different features that I need/bugs to workaround. It's at a point though that most of the 'core' stuff shouldn't change. If you'd like to help with adding tests, shoot me an e-mail if you've got more specific questions. -- -Austin

2 2

Re: Add stub for IoOpenDeviceRegistryKey
by James McKenzie July 19, 2009

July 19, 2009

Uwe Bonnes wrote: This needs a test case to demonstrate what will happen in Windows versus what you are proposing as a stub out that always returns SUCCESS. James McKenzie

1 0

SVG Logo from the website
by Joel Holdsworth July 19, 2009

July 19, 2009

Quick question: Does anyone have an SVG of the wine logo as used on the wine website? I'd like to use it as part of my graphics refresh to improve idb_wine.bmp as shown here: http://www.airwebreathe.org.uk/wine-icon/

3 4

Re: winealsa.drv: Re-allow fallback to waveout (regression in dsound)
by Nikolay Sivov July 19, 2009

July 19, 2009

On Sun, Jul 19, 2009 at 9:55 AM, Erich Hoover<ehoover(a)mines.edu> wrote: > Real Name: > Erich Hoover > Description: > Commit 'f27d88e16fe0292d3efa7dac8966a0b612e62e6d' resulted in a > regression in the support of dsound, causing C&C3 and C&C3: Kane's > Wrath to crash on load (Bug #19369). This patch reverts the fallback > to waveout that was removed by the commit (code unrelated to the rest > of the patch). > Changelog: > winealsa.drv: Re-allow fallback to waveout (regression in dsound) > > > Hi, Eric. You forgot a patch.

2 1

Re: Initial italian translation for taskmgr
by Nikolay Sivov July 18, 2009

July 18, 2009

On Sat, Jul 18, 2009 at 9:25 PM, Davide Pizzetti<erpizzo(a)alice.it> wrote: > Here's the initial translation for taskmgr > > > > Same here. Add file first with 'git add' then 'git commit' changes, and send 'git format-patch' output.

1 0

Re: Initial italian translation for reg
by Nikolay Sivov July 18, 2009

July 18, 2009

On Sat, Jul 18, 2009 at 9:23 PM, Davide Pizzetti<erpizzo(a)alice.it> wrote: > Here's the initial italian translation for reg > Your patch should create new file It.rc, it can't be applied now.

1 0

Re: kernel32: Update French translation of error messages.
by Paul Vriens July 18, 2009

July 18, 2009

Frédéric Delanoy wrote: > Added translation for missing string. > FWIW the Wine translation stats page (http://source.winehq.org/transl/) doesn't mention non-translated strings in kernel32/dlls/winerr_XXX.mc > Maybe this could be added... > > Frédéric > > > ------------------------------------------------------------------------ > > Hi Frédéric, I think that the issue is with wrc. Nonetheless, still worth fixing of course. -- Cheers, Paul.

1 0

Re: patch: fixed Dutch translation
by Paul Vriens July 18, 2009

July 18, 2009

On 17 jul 2009, at 23:33, Vincent Beers <vincentbeers(a)gmail.com> wrote: > This is my first patch, ever, so please bear with me and tell me if > I made any mistakes in generating and submitting the patch. :) > > This is a simple patch that improves the Dutch translation in winecfg. > > Vincent Beers You could start by sending the patch ;) Paul

1 0

re: Doing better than barely keeping up with bug reports - Bug Day this Monday (July 20)
by Dan Kegel July 18, 2009

July 18, 2009

> DELETED BugDay 12:33 Informations DmitryTimoshkov #01 Not useful, eveone works on bugs or features at its own preference, at its own time Dmitry's being overly negative, I think, and it's just plain rude to delete wiki pages like that. Let's put it back. That said, it might be wise to post proposals for events and let them percolate for a couple days before announcing them. (In this case, seeing a few enthusiastic replies might have made Dmitry think twice before doing what he did.) - Dan

6 8

Re: Stub implementation for ftd2xx.dll
by Jeff Zaroyko July 17, 2009

July 17, 2009

On Sat, Jul 18, 2009 at 12:50 AM, Uwe Bonnes<bon(a)elektron.ikp.physik.tu-darmstadt.de> wrote: > +/* > + * FTD2XX.dll > + * > + * Generated from FTD2XX.dll by winedump. > + * > + * DO NOT SUBMIT GENERATED DLLS FOR INCLUSION INTO WINE! > + * > + */ > + You'll probably want to fix this by inserting the appropriate copyright notice and description of the dll following the style of others. Jeff

1 0