September 2017 - Wine-Devel

Re: Please change WINE logo

by Farhad Mohammadi Majd

On Wed, 2017-09-27 at 07:45 -0500, Aric Stewart wrote: > ...... > OK, I wait to see a good designed logo and icon, but just don't forget this issue, thanks and good bye. .

Sept. 30, 2017

Re: [PATCH] midimap: Add warning for user in case midi won't work

by Fabian Maurer

> Oh, if you're intentionally just checking for the prefix then it's > fine as it is. I wasn't sure how certain the name of the port is, so I just checked the prefix. I initially wanted to use StrStrI, but this would mean linking against shlwapi, and I didn't want to add that dependency. Regards, Fabian Maurer

Sept. 29, 2017

Re: [PATCH] midimap: Add warning for user in case midi won't work

by Matteo Bruni

2017-09-29 21:17 GMT+02:00 Fabian Maurer <dark.shadow4(a)web.de>: > Thanks, you mean I should check whether the whole string equals "midi through > port" or what it is? Oh, if you're intentionally just checking for the prefix then it's fine as it is.

Sept. 29, 2017

Re: [PATCH] midimap: Add warning for user in case midi won't work

by Fabian Maurer

Thanks, you mean I should check whether the whole string equals "midi through port" or what it is? Regards, Fabian Maurer

Sept. 29, 2017

Re: [PATCH 3/3] mshtml/tests: Run events tests in IE9 mode.

by Jacek Caban

On 29.09.2017 19:50, Alexandre Julliard wrote: > Jacek Caban <jacek(a)codeweavers.com> writes: > >> Signed-off-by: Jacek Caban <jacek(a)codeweavers.com> >> --- >> dlls/mshtml/tests/events.c | 132 >> +++++++++++++++++++++++++++++++++++---------- >> 1 file changed, 103 insertions(+), 29 deletions(-) > It doesn't work here: Sorry about that, I sent the right version. Jacek

Sept. 29, 2017

Re: [PATCH 3/3] mshtml/tests: Run events tests in IE9 mode.

by Alexandre Julliard

Jacek Caban <jacek(a)codeweavers.com> writes: > Signed-off-by: Jacek Caban <jacek(a)codeweavers.com> > --- > dlls/mshtml/tests/events.c | 132 > +++++++++++++++++++++++++++++++++++---------- > 1 file changed, 103 insertions(+), 29 deletions(-) It doesn't work here: ../../../tools/runtest -q -P wine -T ../../.. -M mshtml.dll -p mshtml_test.exe.so events && touch events.ok events.c:917: Test failed: pdp->cArgs = 1 events.c:954: Test failed: pdp->cArgs = 1 events.c:903: Test failed: pdp->cArgs = 1 events.c:917: Test failed: pdp->cArgs = 1 events.c:954: Test failed: pdp->cArgs = 1 events.c:903: Test failed: pdp->cArgs = 1 events.c:917: Test failed: pdp->cArgs = 1 events.c:954: Test failed: pdp->cArgs = 1 events.c:903: Test failed: pdp->cArgs = 1 events.c:917: Test failed: pdp->cArgs = 1 events.c:954: Test failed: pdp->cArgs = 1 events.c:903: Test failed: pdp->cArgs = 1 events.c:954: Test failed: pdp->cArgs = 1 events.c:903: Test failed: pdp->cArgs = 1 events.c:1242: Test failed: pdp->cArgs = 1 events.c:1218: Test failed: pdp->cArgs = 1 events.c:1242: Test failed: pdp->cArgs = 1 events.c:1218: Test failed: pdp->cArgs = 1 events.c:1242: Test failed: pdp->cArgs = 1 events.c:1074: Test failed: pdp->cArgs = 1 events.c:1060: Test failed: pdp->cArgs = 1 events.c:1085: Test failed: pdp->cArgs = 1 events.c:1074: Test failed: pdp->cArgs = 1 events.c:1085: Test failed: pdp->cArgs = 1 events.c:1074: Test failed: pdp->cArgs = 1 events.c:1074: Test failed: pdp->cArgs = 1 events.c:1074: Test failed: pdp->cArgs = 1 events.c:1151: Test failed: pdp->cArgs = 1 events.c:1074: Test failed: pdp->cArgs = 1 events.c:1060: Test failed: pdp->cArgs = 1 events.c:1151: Test failed: pdp->cArgs = 1 events.c:1074: Test failed: pdp->cArgs = 1 events.c:1151: Test failed: pdp->cArgs = 1 events.c:1074: Test failed: pdp->cArgs = 1 events.c:1151: Test failed: pdp->cArgs = 1 events.c:1074: Test failed: pdp->cArgs = 1 events.c:1191: Test failed: pdp->cArgs = 1 Makefile:267: recipe for target 'events.ok' failed make[1]: *** [events.ok] Error 37 -- Alexandre Julliard julliard(a)winehq.org

Sept. 29, 2017

Re: [PATCH 1/3] mshtml: Inherit document mode from parent document.

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://testbot.winehq.org/JobDetails.pl?Key=33334 Your paranoid android. === w7u (32 bit htmldoc) === htmldoc.c:2682: Test failed: unexpected call GetOverrideKeyPath htmldoc.c:5092: Test failed: unexpected call {d9befc84-f21e-4166-874b-a8d3010dc764}

Sept. 29, 2017

Re: mshtml: Inherit document mode from parent document.

by Jacek Caban

On 28.09.2017 21:11, Gijs Vermeulen wrote: > Hi, > > 2017-09-28 19:50 GMT+02:00 Jacek Caban <jacek(a)codeweavers.com > <mailto:jacek(a)codeweavers.com>>: > > + TRACE("parnet mode %u\n", parent_mode); > > > There seems to be a typo here. Good catch, thanks. Jacek

Sept. 29, 2017

Re: [v7 1/5] server: Named pipe query SD from pipe object

by Jacek Caban

Hi Jonathan, On 29.09.2017 14:35, Jonathan Doron wrote: > Hi Jackek, > > I went over your patch but I believe it might have few issues that i > wanted to raise before we change anything > 1. For some reason when you are doing the "ok" test after > CreateNamedPipe you do against != NULL should != INVALID_HANDLE_VALUE > this is critical... since the 2nd server might have failed with > OBJECT_NAME_ALREADY_EXISTS > 2. When you create the 2nd server instance against the "ok" check is > against NULL but not only that, you check against "server" and not > "server2" Good catch, thanks. > The reason I'm raising these issues is because from my understanding > of Windows, there can be only 1 instance for a given named pipe instance, There may be as many instances as you specify in CreateNamedPipe call. I fixed tests to handle that properly (see attachment). And, indeed, it looks like SD is shared between all pipe instances. Thanks, Jacek

Sept. 29, 2017

Re: [v7 1/5] server: Named pipe query SD from pipe object

by Jonathan Doron

Hi Jackek, I went over your patch but I believe it might have few issues that i wanted to raise before we change anything 1. For some reason when you are doing the "ok" test after CreateNamedPipe you do against != NULL should != INVALID_HANDLE_VALUE this is critical... since the 2nd server might have failed with OBJECT_NAME_ALREADY_EXISTS 2. When you create the 2nd server instance against the "ok" check is against NULL but not only that, you check against "server" and not "server2" The reason I'm raising these issues is because from my understanding of Windows, there can be only 1 instance for a given named pipe instance, Please see that once you change these things, my logic remains correct... and that simply server2 was never created and got INVALID_HANDLE_VALUE with a should be GLE of 231 aka ERROR_PIPE_BUSY. Please let me know how it goes so we will know how to continue. Thanks and I hope you have a great weekend, -- Jon. On Wed, Sep 27, 2017 at 6:33 PM, Jacek Caban <jacek(a)codeweavers.com> wrote: > Hi Jonathan, > > Your tests still don't really show intended behaviour. Your calls that > set security descriptor use new SD that has the same tested values as > before the call, then what's the point in setting it? Anyway, I cleaned > it up and changed a bit to see how it's really supposed to work. See the > attachment. > > It looks like client and served should indeed share security descriptor, > but pipe object should not, so that part is wrong. > > Thanks, > Jacek >

Sept. 29, 2017

Re: [PATCH] midimap: Add warning for user in case midi won't work

by Matteo Bruni

2017-09-23 22:52 GMT+02:00 Fabian Maurer <dark.shadow4(a)web.de>: > When no midi port or only the midi-through port > is found, we can inform the user that midi > won't work as expected. > > Signed-off-by: Fabian Maurer <dark.shadow4(a)web.de> > --- > dlls/midimap/midimap.c | 8 ++++++++ > 1 file changed, 8 insertions(+) > > diff --git a/dlls/midimap/midimap.c b/dlls/midimap/midimap.c > index 64b4dc8546..c80834515b 100644 > --- a/dlls/midimap/midimap.c > +++ b/dlls/midimap/midimap.c > @@ -553,6 +553,7 @@ static LRESULT MIDIMAP_drvOpen(void) > { > MIDIOUTCAPSW moc; > unsigned dev, i; > + WCHAR throughportW[] = {'M','i','d','i',' ','T','h','r','o','u','g','h',0}; > > if (midiOutPorts) > return 0; > @@ -578,6 +579,13 @@ static LRESULT MIDIMAP_drvOpen(void) > } > } > > + if(numMidiOutPorts == 0) > + MESSAGE("Midi sound output won't work - No midi port found.\n"); > + else if(numMidiOutPorts == 1 && midiOutPorts[0].loaded == -1) > + MESSAGE("Midi sound output won't work - No working midi port found.\n"); > + else if(numMidiOutPorts == 1 && strncmpW(midiOutPorts[0].name, throughportW, strlenW(throughportW)) == 0) > + MESSAGE("Midi sound output probably won't work - Only 'midi through' port found. Make sure your setup is correct.\n"); > + > return 1; > } Since you have to resend anyway... The string comparison doesn't look correct to me, I think that breaks if the midi port name has a prefix matching the throughportW string. I guess it's not going to be a problem in practice but it seems better to do it right anyway. There should be a space between 'if' and the '(', like elsewhere in the function and source file.

Sept. 28, 2017

Re: mshtml: Inherit document mode from parent document.

by Gijs Vermeulen

Hi, 2017-09-28 19:50 GMT+02:00 Jacek Caban <jacek(a)codeweavers.com>: > > + TRACE("parnet mode %u\n", parent_mode); > > There seems to be a typo here. Sincerely, Gijs

Sept. 28, 2017

Re: mshtml: Inherit document mode from parent document.

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://testbot.winehq.org/JobDetails.pl?Key=33325 Your paranoid android. === w7u (32 bit htmldoc) === htmldoc.c:2682: Test failed: unexpected call GetOverrideKeyPath htmldoc.c:5092: Test failed: unexpected call {d9befc84-f21e-4166-874b-a8d3010dc764}

Sept. 28, 2017

Re: [PATCH] midimap: Add warning for user in case midi won't work

by Alexandre Julliard

Fabian Maurer <dark.shadow4(a)web.de> writes: > When no midi port or only the midi-through port > is found, we can inform the user that midi > won't work as expected. This should probably use the winediag debug channel. -- Alexandre Julliard julliard(a)winehq.org

Sept. 28, 2017

Re: [PATCH v3 2/3] hlink/tests: Add tests for navigating from a string reference.

by Zebediah Figura

On 09/22/2017 07:19 AM, Jacek Caban wrote: > Hi Zebediah, > > This test leaves iexplore instance running on Wine. > > Jacek > > Thanks; I think I forgot to rerun the tests after changing something. On Windows, IHlinkTarget::Navigate() is evidently supposed to check the HTTP status code before actually opening the browser. I don't know what the proper method of doing this in Wine is. Would it be acceptable just to disable those tests for now?

Sept. 28, 2017

Re: [PATCH 2/9] sapi: Implement SpDataKey::SetKey().

by Huw Davies

On Thu, Sep 28, 2017 at 02:45:09AM -0500, Marvin wrote: > 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://testbot.winehq.org/JobDetails.pl?Key=33314 > > Your paranoid android. > > > === build (build) === > Recreation of tests/Makefile failed Looks like testbot needs to be taught about make_makefiles. Here's a run with the diff of the entire series (including configure.ac changes this time). https://testbot.winehq.org/JobDetails.pl?Key=33317 Huw.

Sept. 28, 2017

Re: [PATCH 2/2] fonts: Increase the Em size to fix a build failure with freetype 2.8.1

by Michael Stefaniuc

On 09/28/2017 10:07 AM, Dmitry Timoshkov wrote: > Michael Stefaniuc <mstefani(a)winehq.org> wrote: > >>> Did you test that after this patch font and glyph metrics still stay the same? >> To find the relevant changes I have verified that the numbers increased >> by the factor of 2048 divided by the old Em size (11 or 13 or 15). Of >> course as 11, 13, 15 aren't factors of 2048 there are rounding errors. >> >> But it doesn't matter as: >> a) Those are bitmap only fonts. >> b) Only the *.fon files are installed but not the corresponding *.ttf files. > > That's true, but it's sometimes useful to be able to load and test .ttf > files as well, although it looks like after some Windows update intermediate > Wine bitmap-only .ttf fonts is no longer possible to load under Windows, and > since freetype 2.8.1 Linux started to behave same way. Seems the Em increase fixed that on Windows too: <nsivov> puk, with fixed em field ttfs open now on windows too >> c) The generated *.fon files are bit by bit identical (sha1sum) before >> and after this commit. > > Great, that's the key point. Right, that's why I have included that info in my patch submission ;) https://www.winehq.org/pipermail/wine-patches/2017-September/165357.html "With this two patches applied I always get the exact same fon files for all possible combinations: - freetype 2.6.5 and unpatched Wine - freetype 2.6.5 and this two patches applied - freetype 2.8.1 and this two patches applied" bye michael

Sept. 28, 2017

Re: [PATCH 2/2] fonts: Increase the Em size to fix a build failure with freetype 2.8.1

by Dmitry Timoshkov

Michael Stefaniuc <mstefani(a)winehq.org> wrote: > > Did you test that after this patch font and glyph metrics still stay the same? > To find the relevant changes I have verified that the numbers increased > by the factor of 2048 divided by the old Em size (11 or 13 or 15). Of > course as 11, 13, 15 aren't factors of 2048 there are rounding errors. > > But it doesn't matter as: > a) Those are bitmap only fonts. > b) Only the *.fon files are installed but not the corresponding *.ttf files. That's true, but it's sometimes useful to be able to load and test .ttf files as well, although it looks like after some Windows update intermediate Wine bitmap-only .ttf fonts is no longer possible to load under Windows, and since freetype 2.8.1 Linux started to behave same way. > c) The generated *.fon files are bit by bit identical (sha1sum) before > and after this commit. Great, that's the key point. -- Dmitry.

Sept. 28, 2017

Re: [PATCH 9/9] sapi: Add a partial implementation of SpObjectTokenCategory::EnumTokens().

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://testbot.winehq.org/JobDetails.pl?Key=33316 Your paranoid android. === build (build) === Recreation of tests/Makefile failed

Sept. 28, 2017

Re: [PATCH 6/9] sapi: Add a partial implementation of SpObjectTokenEnum::Next().

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://testbot.winehq.org/JobDetails.pl?Key=33315 Your paranoid android. === build (build) === Recreation of tests/Makefile failed

Sept. 28, 2017

Re: [PATCH 2/9] sapi: Implement SpDataKey::SetKey().

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://testbot.winehq.org/JobDetails.pl?Key=33314 Your paranoid android. === build (build) === Recreation of tests/Makefile failed

Sept. 28, 2017

Re: [PATCH 2/2] fonts: Increase the Em size to fix a build failure with freetype 2.8.1

by Michael Stefaniuc

On 09/28/2017 06:53 AM, Dmitry Timoshkov wrote: > Michael Stefaniuc <mstefani(a)winehq.org> wrote: > >> -UnderlinePosition: -100 >> -UnderlineWidth: 50 >> -Ascent: 11 >> -Descent: 2 >> +UnderlinePosition: -15753.8 >> +UnderlineWidth: 7876.92 >> +Ascent: 1733 >> +Descent: 315 >> +InvalidEm: 0 >> LayerCount: 2 >> Layer: 0 0 "Back" 1 >> Layer: 1 0 "Fore" 0 >> @@ -21,7 +22,7 @@ OS2_UseTypoMetrics: 0 >> PfmFamily: 49 >> TTFWeight: 400 >> TTFWidth: 5 >> -LineGap: 1 >> +LineGap: 158 >> VLineGap: 0 >> Panose: 2 0 6 9 0 0 0 0 0 0 >> OS2TypoAscent: 0 >> @@ -56,770 +57,770 @@ BeginChars: 65536 498 >> >> StartChar: space >> Encoding: 32 32 0 >> -Width: 8 >> +Width: 1260 >> Flags: W >> LayerCount: 2 >> EndChar >> >> StartChar: exclam >> Encoding: 33 33 1 >> -Width: 8 >> +Width: 1260 >> Flags: W >> LayerCount: 2 >> EndChar > > Did you test that after this patch font and glyph metrics still stay the same? To find the relevant changes I have verified that the numbers increased by the factor of 2048 divided by the old Em size (11 or 13 or 15). Of course as 11, 13, 15 aren't factors of 2048 there are rounding errors. But it doesn't matter as: a) Those are bitmap only fonts. b) Only the *.fon files are installed but not the corresponding *.ttf files. c) The generated *.fon files are bit by bit identical (sha1sum) before and after this commit. bye michael

Sept. 28, 2017

Re: WineTest does not run the vcomp tests

by Francois Gouget

On Wed, 27 Sep 2017, Vincent Povirk wrote: > Could we try to run the .exe without specifying a test, and if it > fails due to a missing dll (can we determine that?), skip the test? Your idea is to not check for missing dlls and instead just run the executable as a way to check that all the required dlls are present? The problem is that if a dll is missing the executable remains stuck while Windows displays a dialog saying the dll is missing. Also, killing the process does not dismiss the windows dialog. Maybe we could take a page out of the shlexec test and wait for the executable for 5 seconds and do a FindWindowA("#32770", "Windows") to click on the OK ourselves. But I believe on Windows 8+ we had problems due to Windows popping up a window offering to install the missing dll but we could not intercept that one. That said the dlls would be installed on machines that run WineTest regularly anyway. -- Francois Gouget <fgouget(a)free.fr> http://fgouget.free.fr/ Dieu dit: "M-x Lumière". Et la lumière fut.

Sept. 28, 2017

RE: user.exe16: Support Windows 1.x and 2.x Menus

by Martin Payne

Hi Alexandre, > Mainly I find it interesting that Windows 2 apps still work, it probably hasn't been tested in 20 years. Apart from the menu issue, are other things looking OK? A few applications crash at startup with a null pointer dereference. I'd need to read up some more on Wine debugging to understand why. Icon and cursor resources don't load, but like with the menus it's just because the format is slightly different. It shouldn't be difficult to add support for that as most of the required code is already in Wine. Other than that, it's surprising how ~30 year old applications can still work on a modern OS! Regards, Martin

Sept. 28, 2017

Re: [PATCH 2/2] fonts: Increase the Em size to fix a build failure with freetype 2.8.1

by Dmitry Timoshkov

Michael Stefaniuc <mstefani(a)winehq.org> wrote: > -UnderlinePosition: -100 > -UnderlineWidth: 50 > -Ascent: 11 > -Descent: 2 > +UnderlinePosition: -15753.8 > +UnderlineWidth: 7876.92 > +Ascent: 1733 > +Descent: 315 > +InvalidEm: 0 > LayerCount: 2 > Layer: 0 0 "Back" 1 > Layer: 1 0 "Fore" 0 > @@ -21,7 +22,7 @@ OS2_UseTypoMetrics: 0 > PfmFamily: 49 > TTFWeight: 400 > TTFWidth: 5 > -LineGap: 1 > +LineGap: 158 > VLineGap: 0 > Panose: 2 0 6 9 0 0 0 0 0 0 > OS2TypoAscent: 0 > @@ -56,770 +57,770 @@ BeginChars: 65536 498 > > StartChar: space > Encoding: 32 32 0 > -Width: 8 > +Width: 1260 > Flags: W > LayerCount: 2 > EndChar > > StartChar: exclam > Encoding: 33 33 1 > -Width: 8 > +Width: 1260 > Flags: W > LayerCount: 2 > EndChar Did you test that after this patch font and glyph metrics still stay the same? -- Dmitry.

Sept. 28, 2017