December 2013 - Wine-Devel - list.winehq.org

Mobile version input methods (on-screen buttons, keyboards, virtual joysticks)

by WineBuG

How would cope with lack of the keyboard and mouse in the future ARM (Android) version of Wine? Will one be able to add virtual on-screen buttons and joysticks in some way? Are you going to analyze other mobile apps (for example, games) controls solutions? Regards, WineBuG

Dec. 7, 2013

Re: [PATCH] user32: Don't send message to get window text, avoid a deadlock.

by Qian Hong

Hi Dmitry, On Fri, Dec 6, 2013 at 11:18 PM, Dmitry Timoshkov <dmitry(a)baikal.ru> wrote: > This patch needs at least some tests. Thanks for reviewing, you're right, it needs more work, I'll investigate later. -- Regards, Qian Hong - http://www.codeweavers.com

Dec. 7, 2013

Launchpad project for Wine

by Tae Wong

There's a Launchpad project for Wine: https://launchpad.net/wine But there's a misspelling: ...a compatability layer... It says compatability instead of compatibility. The account https://launchpad.net/~seotaewong40 has been disabled. How to enable this account to file bugs? When trying to visit the page, it says: Page gone. There’s no page with this address in Launchpad. If this is blocking your work, let us know by asking a question at Launchpad Support . Otherwise, complain to the maintainers of the page that linked here.

Dec. 7, 2013

Re: [PATCH] user32: Don't send message to get window text, avoid a deadlock.

by Dmitry Timoshkov

Qian Hong <qhong(a)codeweavers.com> wrote: > - Fixed http://bugs.winehq.org/show_bug.cgi?id=27282, affects Arcania: > Gothic 4 and online bank of abchina and > - Makes Anastasius happy :p This patch needs at least some tests. -- Dmitry.

Dec. 6, 2013

Re: [PATCH 2/5] ole32: Simplify enumeration cloning using duplicated key

by Alexandre Julliard

Nikolay Sivov <nsivov(a)codeweavers.com> writes: > - new_this->IEnumCATEGORYINFO_iface = This->IEnumCATEGORYINFO_iface; > - new_this->ref = 1; > - new_this->lcid = This->lcid; > - /* FIXME: could we more efficiently use DuplicateHandle? */ > - open_classes_key(HKEY_CLASSES_ROOT, keyname, KEY_READ, &new_this->key); > - new_this->next_index = This->next_index; > + cloned->IEnumCATEGORYINFO_iface = This->IEnumCATEGORYINFO_iface; > + cloned->ref = 1; > + cloned->lcid = This->lcid; > > - *ppenum = &new_this->IEnumCATEGORYINFO_iface; > + if (!DuplicateHandle(GetCurrentProcess(), This->key, GetCurrentProcess(), (HANDLE*)&cloned->key, 0, FALSE, DUPLICATE_SAME_ACCESS)) > + { > + IEnumCATEGORYINFO_Release(&cloned->IEnumCATEGORYINFO_iface); > + return E_FAIL; > + } Using DuplicateHandle is not a good idea on registry keys, particularly for HKEY_CLASSES_ROOT keys. -- Alexandre Julliard julliard(a)winehq.org

Dec. 6, 2013

Re: shell32/tests: Use GetVersion() to skip Timeout test (try 2)

by Nikolay Sivov

On 12/6/2013 09:10, Zhenbo Li wrote: > I'm using win_skip() for Vista&Win2008 this time. > > --- > dlls/shell32/tests/shlfileop.c | 36 +++++++++++++++++++----------------- > 1 file changed, 19 insertions(+), 17 deletions(-) > > Hm, it's the same issue. Please don't use GetVersion().

Dec. 6, 2013

Re: shell32/tests: Use GetVersion() to skip Timeout test (try 2)

by Marvin

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 https://newtestbot.winehq.org/JobDetails.pl?Key=3671 Your paranoid android. === wvista (32 bit shlfileop) === Timeout === w2008s64 (32 bit shlfileop) === Timeout === w2008s64 (64 bit shlfileop) === Timeout

Dec. 6, 2013

Re: Build issues with winemac.drv and Mavericks

by Charles Davis

On Dec 2, 2013, at 1:49 PM, C.W. Betts wrote: > I’ve run into issues when building winemac.drv on Mavericks. Attached is the log. Ken's patch should've fixed the undeclared GL functions warnings, but it won't address the other warnings. > ../../../wine-git/dlls/winemac.drv/clipboard.c:320:78: warning: format specifies type 'const unichar *' (aka 'const unsigned short *') but the argument has type 'WCHAR *' (aka 'unsigned short *') [-Wformat] > registered_name_type_prefix, buffer); > ^~~~~~ This is a false positive, because a Mac OS unichar is the same as a Wine WCHAR (i.e. a UTF-16 character). You could probably shut Clang up with a cast. > ../../../wine-git/dlls/winemac.drv/opengl.c:2985:9: warning: 'gluCheckExtension' is deprecated: first deprecated in OS X 10.9 [-Wdeprecated-declarations] > if (gluCheckExtension((GLubyte*)"GL_ARB_color_buffer_float", (GLubyte*)gl_info.glExtensions)) > ^ > /System/Library/Frameworks/OpenGL.framework/Headers/glu.h:250:18: note: 'gluCheckExtension' declared here > extern GLboolean gluCheckExtension (const GLubyte *extName, const GLubyte *extString) OPENGL_DEPRECATED(10_2, 10_9); > ^ > ../../../wine-git/dlls/winemac.drv/opengl.c:2991:9: warning: 'gluCheckExtension' is deprecated: first deprecated in OS X 10.9 [-Wdeprecated-declarations] > if (gluCheckExtension((GLubyte*)"GL_ARB_multisample", (GLubyte*)gl_info.glExtensions)) > ^ > /System/Library/Frameworks/OpenGL.framework/Headers/glu.h:250:18: note: 'gluCheckExtension' declared here > extern GLboolean gluCheckExtension (const GLubyte *extName, const GLubyte *extString) OPENGL_DEPRECATED(10_2, 10_9); > ^ > ../../../wine-git/dlls/winemac.drv/opengl.c:2994:9: warning: 'gluCheckExtension' is deprecated: first deprecated in OS X 10.9 [-Wdeprecated-declarations] > if (gluCheckExtension((GLubyte*)"GL_ARB_framebuffer_sRGB", (GLubyte*)gl_info.glExtensions)) > ^ > /System/Library/Frameworks/OpenGL.framework/Headers/glu.h:250:18: note: 'gluCheckExtension' declared here > extern GLboolean gluCheckExtension (const GLubyte *extName, const GLubyte *extString) OPENGL_DEPRECATED(10_2, 10_9); > ^ > ../../../wine-git/dlls/winemac.drv/opengl.c:2997:9: warning: 'gluCheckExtension' is deprecated: first deprecated in OS X 10.9 [-Wdeprecated-declarations] > if (gluCheckExtension((GLubyte*)"GL_APPLE_pixel_buffer", (GLubyte*)gl_info.glExtensions)) > ^ > /System/Library/Frameworks/OpenGL.framework/Headers/glu.h:250:18: note: 'gluCheckExtension' declared here > extern GLboolean gluCheckExtension (const GLubyte *extName, const GLubyte *extString) OPENGL_DEPRECATED(10_2, 10_9); > ^ > ../../../wine-git/dlls/winemac.drv/opengl.c:3011:13: warning: 'gluCheckExtension' is deprecated: first deprecated in OS X 10.9 [-Wdeprecated-declarations] > if (gluCheckExtension((GLubyte*)"GL_ARB_texture_rectangle", (GLubyte*)gl_info.glExtensions) || > ^ > /System/Library/Frameworks/OpenGL.framework/Headers/glu.h:250:18: note: 'gluCheckExtension' declared here > extern GLboolean gluCheckExtension (const GLubyte *extName, const GLubyte *extString) OPENGL_DEPRECATED(10_2, 10_9); > ^ > ../../../wine-git/dlls/winemac.drv/opengl.c:3012:13: warning: 'gluCheckExtension' is deprecated: first deprecated in OS X 10.9 [-Wdeprecated-declarations] > gluCheckExtension((GLubyte*)"GL_EXT_texture_rectangle", (GLubyte*)gl_info.glExtensions)) > ^ > /System/Library/Frameworks/OpenGL.framework/Headers/glu.h:250:18: note: 'gluCheckExtension' declared here > extern GLboolean gluCheckExtension (const GLubyte *extName, const GLubyte *extString) OPENGL_DEPRECATED(10_2, 10_9); > ^ > ../../../wine-git/dlls/winemac.drv/opengl.c:3036:9: warning: 'gluCheckExtension' is deprecated: first deprecated in OS X 10.9 [-Wdeprecated-declarations] > if (gluCheckExtension((GLubyte*)"GL_EXT_framebuffer_sRGB", (GLubyte*)gl_info.glExtensions)) > ^ > /System/Library/Frameworks/OpenGL.framework/Headers/glu.h:250:18: note: 'gluCheckExtension' declared here > extern GLboolean gluCheckExtension (const GLubyte *extName, const GLubyte *extString) OPENGL_DEPRECATED(10_2, 10_9); > ^ > ../../../wine-git/dlls/winemac.drv/opengl.c:3039:9: warning: 'gluCheckExtension' is deprecated: first deprecated in OS X 10.9 [-Wdeprecated-declarations] > if (gluCheckExtension((GLubyte*)"GL_EXT_packed_float", (GLubyte*)gl_info.glExtensions)) > ^ > /System/Library/Frameworks/OpenGL.framework/Headers/glu.h:250:18: note: 'gluCheckExtension' declared here > extern GLboolean gluCheckExtension (const GLubyte *extName, const GLubyte *extString) OPENGL_DEPRECATED(10_2, 10_9); > ^ > ../../../wine-git/dlls/winemac.drv/opengl.c:3109:9: warning: 'gluCheckExtension' is deprecated: first deprecated in OS X 10.9 [-Wdeprecated-declarations] > if (gluCheckExtension((GLubyte*)"GL_APPLE_flush_render", (GLubyte*)gl_info.glExtensions)) > ^ > /System/Library/Frameworks/OpenGL.framework/Headers/glu.h:250:18: note: 'gluCheckExtension' declared here > extern GLboolean gluCheckExtension (const GLubyte *extName, const GLubyte *extString) OPENGL_DEPRECATED(10_2, 10_9); > ^ We should probably replace gluCheckExtension with our own extension checker. No modern app uses GLU anymore--that's why Apple deprecated it. (That, and they want you to use GLKit instead. Like that'll ever happen in Wine...) > ../../../wine-git/dlls/winemac.drv/opengl.c:75:5: warning: 'CGLPBufferObj' is deprecated: first deprecated in OS X 10.7 [-Wdeprecated-declarations] > CGLPBufferObj pbuffer; > ^ > /System/Library/Frameworks/OpenGL.framework/Headers/CGLTypes.h:21:41: note: 'CGLPBufferObj' declared here > typedef struct _CGLPBufferObject *CGLPBufferObj OPENGL_DEPRECATED(10_3, 10_7); > ^ As for this and the other pbuffer-related deprecations... Any Mac capable of running Mavericks (including my ridiculously old 2007 Santa Rosa MacBook Pro) or even Lion probably supports FBOs; we can define them away if the minimum version targeted is >=10.7. If we absolutely must (say, because some legacy Win app wants pbuffers and doesn't understand FBOs), we could probably emulate pbuffers with FBOs. The rest of the deprecations are of PFAs that for some reason have fallen out of favor with Apple. We can't set them at all when we add support for OpenGL 3/4--attempting to create a context with any of the deprecated PFAs set along with the profile PFA will make the creation fail. kCGLPFAWindow in particular is redundant--it's the default if you don't say you want one of the other kinds of rendering contexts. (The other kinds are all deprecated, too, by the way. If you want full-screen rendering, you're supposed to create a window covering the screen, and if you want off-screen rendering, you're supposed to render to an FBO.) Most of the other deprecated PFAs are largely useless (or so they tell us). > ../../../wine-git/dlls/winemac.drv/cocoa_event.m:466:9: warning: variable 'semaphore' is captured uninitialized whenever 'if' condition is false [-Wsometimes-uninitialized] > if (!queue) > ^~~~~~ > ../../../wine-git/dlls/winemac.drv/cocoa_event.m:473:23: note: variable is captured by block here > OnMainThreadAsync(^{ > ^~ > ../../../wine-git/dlls/winemac.drv/cocoa_event.m:466:5: note: remove the 'if' if its condition is always true > if (!queue) > ^~~~~~~~~~~ > ../../../wine-git/dlls/winemac.drv/cocoa_event.m:463:35: note: initialize the variable 'semaphore' to silence this warning > dispatch_semaphore_t semaphore; > ^ > = NULL This is another false positive. Clang apparently cannot see into the block, so it doesn't see that, even though `semaphore' is captured, it's only used under the same conditions that it's initialized under. Chip

Dec. 6, 2013

Re: wined3d: Use BOOL type where appropriate

by Henri Verbeet

On 5 December 2013 23:14, Frédéric Delanoy <frederic.delanoy(a)gmail.com> wrote: > Note: in surface.c, used "?1:0" construct to avoid implicit BOOL arithmetic > --- > dlls/wined3d/arb_program_shader.c | 19 +++++++++---------- > dlls/wined3d/context.c | 20 ++++++++++---------- > dlls/wined3d/device.c | 2 +- > dlls/wined3d/gl_compat.c | 4 ++-- > dlls/wined3d/shader.c | 20 ++++++++++---------- > dlls/wined3d/stateblock.c | 22 +++++++++++----------- > dlls/wined3d/surface.c | 8 ++++---- > dlls/wined3d/utils.c | 8 ++++---- > 8 files changed, 51 insertions(+), 52 deletions(-) > I don't think so, most of these are bitfields, and the surface.c change is actually making the code worse.

Dec. 6, 2013

Re: [PATCH 3/3] ws2_32: Add an interactive test for interface change notifications (resend 2).

by Marvin

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 https://newtestbot.winehq.org/JobDetails.pl?Key=3670 Your paranoid android. === wxppro (32 bit sock) === sock.c:520: Test failed: oob_server (4c0): unexpectedly at the OOB mark: 0 === w2008s64 (32 bit sock) === sock.c:520: Test failed: oob_server (750): unexpectedly at the OOB mark: 0 sock: unhandled exception c0000005 at 757C9799 === w2008s64 (64 bit sock) === sock.c:520: Test failed: oob_server (744): unexpectedly at the OOB mark: 0 No test summary line found === w864 (64 bit sock) === sock.c:1741: Test failed: getservbyname could not retrieve information for domain: 11004 sock.c:1741: Test failed: sock.c:1741: Test failed: getservbyname could not retrieve information for domain: 11004 sock.c:1741: Test failed: getservbyname could not retrieve information for telnet: 11004 sock.c:1741: Test failed: getservbyname could not retrieve information for domain: 11004 sock.c:1741: Test failed: getservbyname could not retrieve information for telnet: 11004 sock.c:1741: Test failed: getservbyname could not retrieve information for domain: 11004 sock.c:1741: Test failed: getservbyname could not retrieve information for telnet: 11004 sock.c:1741: Test failed: getservbyname could not retrieve information for domain: 11004 sock.c:1741: Test failed: getservbyname could not retrieve information for telnet: 11004 sock.c:1741: Test failed: getservbyname could not retrieve information for domain: 11004 sock.c:1741: Test failed: getservbyname could not retrieve information for telnet: 11004 sock.c:1741: Test failed: getservbyname could not retrieve information for domain: 11004 sock.c:1741: Test failed: getservbyname could not retrieve information for telnet: 11004 sock.c:1741: Test failed: getservbyname could not retrieve information for domain: 11004 sock.c:1741: Test failed: getservbyname could not retrieve information for telnet: 11004 sock.c:1741: Test failed: getservbyname could not retrieve information for domain: 11004 sock.c:1748: Test failed: getservbyname: winsock resized servent buffer when not necessary

Dec. 5, 2013