August 2011 - Wine-devel - list.winehq.org

Re: About cmd patches
by Nowres Rafid 19 Aug '11

19 Aug '11

On 19/08/2011 03:14, Dan Kegel wrote: > Hi Nowres, > it's awesome that you're digging into cmd like this! > > The rule with Wine is that > Each patch you submit shall pass tests. > > That means that your patch series > [1/2] cmd: fixing an error with redirection operators > [2/2] cmd/tests: fixing an error with redirection operators (updating > test results) > > is incorrect, since tests marked todo_wine will fail after the first patch. > > Better to send things like that as a single merged patch in the future. > > Thanks, > Dan Ok Dan, i queued these patch in my email client, but when I sent them they were sent in bad order. Nowres,

2 1

d3dx9: CloneMesh test and improvements
by Michael Mc Donnell 19 Aug '11

19 Aug '11

Hi I've been working on a test and improvements for CloneMesh as part of GSoC 2011. I would be grateful if anyone could comment on my work. There are five test cases in the attached patch: 0. Basic mesh cloning. Declaration has position and normal, and the new declaration is the same as the original. 1. Same as test 0, but with 16-bit indices. 2. Shows that the vertex buffer can be widened so that it has room for a new vertex component. The declaration has position and normal, and the new declaration has texture coordinates after position and normal. 3. Same as test 2, but the new declaration has the texture coordinates in between position and normal. 4. Shows that a vertex component can be converted into another type if the new declaration specifies another type. The declaration has position and normal. The normal is a FLOAT2 and in the new declaration it is instead FLOAT16_2. Test 0 and 1 tests what is already known to work in the current implementation. Test 2, 3 and 4 do not work in the current implementation because the original and new declarations are not the same. I've also improved the current CloneMesh implementation so that it also passes test 2, 3, and 4. The patch is not complete because it does not support converting all the possible types. I expect to implement conversion for the remaining types during this week. Again any comments or ideas are welcome. Thanks, Michael Mc Donnell

5 15

re: mscoree: COM cleanup for the ICLRRuntimeInfo iface.
by Dan Kegel 19 Aug '11

19 Aug '11

Seems to cause a test failure here? make[1]: Entering directory `/home/dank/wineslave.dir/sandbox/slave/runtests/build/dlls/mscoree/tests' ../../../tools/runtest -q -P wine -M mscoree.dll -T ../../.. -p mscoree_test.exe.so metahost.c && touch metahost.ok metahost.c:135: Test failed: got unexpected version L"v1.1.4322" fixme:mscoree:CLRMetaHost_GetRuntime Unrecognized version L"v2.0.0" Command exited with non-zero status 1 0.04user 0.03system 0:00.09elapsed 82%CPU (0avgtext+0avgdata 27104maxresident)k 0inputs+0outputs (0major+5905minor)pagefaults 0swaps make[1]: *** [metahost.ok] Error 1 ../../../tools/runtest -q -P wine -M mscoree.dll -T ../../.. -p mscoree_test.exe.so mscoree.c && touch mscoree.ok mscoree.c:140: Test failed: version is L"v1.1.4322" , expected L"v2.0.50727" mscoree.c:145: Test failed: version is L"v1.1.4322" , expected L"v2.0.50727" fixme:mscoree:CreateConfigStream ((null), 0x32fbb4): stub fixme:mscoree:CreateConfigStream (L"Z:\\home\\dank\\wineslave.dir\\sandbox\\slave\\runtests\\build\\dlls\\mscoree\\tests\\conf.xml", (nil)): stub fixme:mscoree:CreateConfigStream ((null), (nil)): stub fixme:mscoree:CreateConfigStream (L"nonexist.xml", 0x32fbb4): stub fixme:mscoree:CreateConfigStream (L"Z:\\home\\dank\\wineslave.dir\\sandbox\\slave\\runtests\\build\\dlls\\mscoree\\tests\\conf.xml", 0x32fbb4): stub Command exited with non-zero status 2 0.05user 0.02system 0:00.09elapsed 81%CPU (0avgtext+0avgdata 27296maxresident)k 0inputs+8outputs (0major+5969minor)pagefaults 0swaps make[1]: *** [mscoree.ok] Error 2 make[1]: Target `test' not remade because of errors. (Log at http://buildbot.kegel.com:8010/builders/runtests/builds/32 for a day or so.)

2 2

Re: winemenubuilder: appbundle patch 20110618
by Per Johansson 19 Aug '11

19 Aug '11

Hi Steven, list, I tried both AppleScript and shell script versions and AppleScript doesn't really add anything, so it's probably better to go with shell script and generate the bundle manually. I'll see if I can help smoothing it out. -- Pelle Johansson 17 aug 2011 kl. 00.06 skrev Steven Edwards: > You might be interested in this. It allows icons and localized > resources. I've just not gotten around to doing all the polish. > > > ---------- Forwarded message ---------- > From: Steven Edwards <winehacker(a)gmail.com> > Date: Sat, Jun 18, 2011 at 5:43 AM > Subject: winemenubuilder: appbundle patch 20110618 > To: Wine Developers List <wine-devel(a)winehq.com> > > > Hi, > I've finally gotten back around to hacking the *.app bundle support in > winemenubuilder and come close to having something that mostly works. > There are still things that need to be cleaned up and further > abstracted, and a bit more logic to work out. I wanted to go ahead and > send this out in case anyone else wants to play around with it, or if > for some reason I am unable to finish hackin on it. > > Some notes: > > - The first icon/bundle always seems to get the wrong icon, and uses > the default wine icon. I need to take some time to understand the code > a bit better and see why this is. > > - Bundles are created in $HOME/Applications/wine and icon's are > generated using the icns code. Support for building bundles for > shortcuts on $HOME/Desktop has not been hacked on yet. > > - There are still plenty of bugs to resolve such as making sure we are > actually using the correct icon, and properly setting various values > in the app *.plist. There are a couple of alternative ideas I looked > in to, such as doing some sort of spawn/fork to invoke > /usr/libexec/PlistBuddy rather than using the CoreFoundation API, but > I worry about creating race conditions and other performance problems > when installing software that creates many icons, such as Microsoft > Office. > > - I've tried to make this un-invasive to the existing xdg code path > (as one can run a FreeDesktop on Darwin/OS X) but have not had a > chance to actually test supporting both menu systems at the same time. > > I welcome your feedback. > Thanks > -- > Steven Edwards > > "There is one thing stronger than all the armies in the world, and > that is an idea whose time has come." - Victor Hugo > > > > -- > Steven Edwards > > "There is one thing stronger than all the armies in the world, and > that is an idea whose time has come." - Victor Hugo > <appbundles_20110617.diff>

1 0

re: cmd/tests: fixing an issue with extra whitespaces after cd parameter
by Dan Kegel 18 Aug '11

18 Aug '11

That patch will fail unless your earlier patch is applied, right? If that's the case, you have to submit the two patches together, either as a single patch, or as a two patch series ( see http://wiki.winehq.org/SubmittingPatches for more info, in particular, how to format a two patch series ).

1 0

re: Subject: winemenubuilder: Use start.exe to open URLs to respect URL handlers in registry.
by Dan Kegel 18 Aug '11

18 Aug '11

It doesn't apply here, looks like whitespace was corrupted. Try attaching the patch instead of sending it inline? patching file programs/winemenubuilder/winemenubuilder.c Hunk #1 FAILED at 2962. Hunk #2 FAILED at 3004. Hunk #3 FAILED at 3063. 3 out of 3 hunks FAILED -- saving rejects to file programs/winemenubuilder/winemenubuilder.c.rej (Log at http://buildbot.kegel.com:8010/builders/runtests/builds/22/steps/git/logs/s… for a few days.)

1 0

re: cmd: fixing an issue with extra whitespaces after cd parameter (retry)
by Dan Kegel 18 Aug '11

18 Aug '11

Frédéric Delanoy wote: > Nowres Rafid wrote: >> how can I generate the expected file (.exp)? > It already exists in the source tree I think Nowres meant, how can he update the .exp file to match the expected output of his new test. The answer is: run the new test, save its output, and add it to the .exp file by hand. It's kind of a pain.

2 2

Re: xlive: stub some functions (2/3)
by Austin English 18 Aug '11

18 Aug '11

On Wed, Aug 17, 2011 at 12:33, Giovanni Ongaro <giovanni.nicola(a)ticino.com> wrote: > I am working oon a xlive patch can you tell me how to finf dll ordinal > functions by using winedump i am tryiing specifically to find how much > parameters xlive 5282 has > > Can you help me? >From the patch: +/* This DLL is completely undocumented. DLL ordinal functions were found using winedump + and number of parameters determined using Python ctypes. See http://docs.python.org/library/ctypes.html */ -- -Austin

1 0

buildbot status
by Dan Kegel 17 Aug '11

17 Aug '11

I brought up the proof-of-concept buildbot-based patchwatcher on two old PCs here; its status page is http://buildbot.kegel.com/ It has caught one real problem so far this morning. to-do list: - make cleaner script and doc for adding a new slave - make it start automatically after reboot - sandbox it - turn on build result emails - add a bit of parallelism into the test runner, so e.g. tests that don't need the display or network run in parallel with those that do, probably in a separate wineprefix - add valgrind slave - add windows slave (like testbot, but with real graphics and sound) - add x86_64 slave - move master to winehq.org once it seems ready Here are the tests I had to blacklist so far: # http://bugs.winehq.org/show_bug.cgi?id=12053 touch dlls/user32/tests/msg.ok touch dlls/user32/tests/win.ok touch dlls/user32/tests/input.ok # http://bugs.winehq.org/show_bug.cgi?id=28038 touch dlls/wininet/tests/urlcache.ok # http://bugs.winehq.org/show_bug.cgi?id=28109 touch dlls/winmm/tests/capture.ok # http://bugs.winehq.org/show_bug.cgi?id=28048 touch dlls/winmm/tests/wave.ok # Blacklist until http://www.winehq.org/pipermail/wine-patches/2011-August/105358.html in touch dlls/winhttp/tests/winhttp.ok # http://bugs.winehq.org/show_bug.cgi?id=28102 touch dlls/ws2_32/tests/sock.ok # http://bugs.winehq.org/show_bug.cgi?id=28108 touch dlls/urlmon/tests/url.ok

1 0

Re: cmd: fixing an issue with extra whitespaces after cd parameter (retry)
by Frédéric Delanoy 17 Aug '11

17 Aug '11

2011/8/17 Nowres Rafid <nowres.rafed(a)gmail.com>: > On 17/08/2011 14:02, Frédéric Delanoy wrote: >> >> On Fri, Aug 12, 2011 at 19:14, Nowres Rafid<nowres.rafed(a)gmail.com> >> wrote: >>> >>> fixing cd ..(a)space@ >>> >>> please ignore the previous patch. >> >> You didn't include any test in your patch. >> All patches including a fix must have tests confirming the bug is fixed > > ok, > > how can I generate the expected file (.exp)? It already exists in the source tree

1 0