December 2020 - Wine-Devel - list.winehq.org

Re: [PATCH v2] po: Update Ukrainian translation.

by Lauri Kenttä

Hello, On 2020-12-25 21:05, Andrey Gusev wrote: > msgctxt "Solid" > -msgid "&o" > -msgstr "&Ні" > +msgid "&Sol" > +msgstr "&Суц" You can't change msgid, that's the English version. You can only change msgstr. In this case, msgctxt gives you the original meaning or context ("Solid") and msgid shows the current English version ("&o", apparently for some accelerator use case), and you should choose a similar one-letter translation for the accelerator, for example "&у" if it's reasonable. -- Lauri Kenttä

Dec. 30, 2020

[PATCH] readme: Update Finnish translation.

by Lauri Kenttä

Signed-off-by: Lauri Kenttä <lauri.kentta(a)gmail.com> --- documentation/README.fi | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/documentation/README.fi b/documentation/README.fi index 30311d82807..b2c6524548e 100644 --- a/documentation/README.fi +++ b/documentation/README.fi @@ -39,7 +39,7 @@ Winen kääntämiseen tarvitaan jokin seuraavista: FreeBSD 8.0 tai uudempi Solaris x86 9 tai uudempi NetBSD-current - Mac OS X 10.5 tai uudempi + Mac OS X 10.8 tai uudempi Wine vaatii kerneliltä tuen säikeille. Tämän takia toistaiseksi vain yllä mainittuja käyttöjärjestelmiä tuetaan; tulevaisuudessa saatetaan lisätä tuki @@ -59,8 +59,9 @@ Tietoa NetBSD:lle: USER_LDT, SYSVSHM, SYSVSEM ja SYSVMSG täytyy aktivoida kernelistä. Tietoa Mac OS X:lle: - Winen onnistuneeseen kääntämiseen x86:lla tarvitaan Xcode 2.4 tai uudempi. - Mac-ajuri vaatii OS X 10.6:n tai uudemman, ja sitä ei käännetä 10.5:ssä. + Winen kääntämiseen tarvitaan Xcode Command Line Tools tai Apple cctools. + Vähimmäisversiot ovat clang 3.8, MacOSX10.10.sdk ja mingw-w64 v8. + MacOSX10.14.sdk ja myöhemmät sopivat vain wine64:n kääntämiseen. Tuetut tiedostojärjestelmät: -- 2.30.0

Dec. 30, 2020

[PATCH] include: Add DBACCESSORFLAGSENUM/DBBINDSTATUSENUM values

by Alistair Leslie-Hughes

Signed-off-by: Alistair Leslie-Hughes <leslie_alistair(a)hotmail.com> --- include/access.idl | 21 +++++++++++++++++++++ 1 file changed, 21 insertions(+) diff --git a/include/access.idl b/include/access.idl index 5ade09b85ce..fe1f3baddd7 100644 --- a/include/access.idl +++ b/include/access.idl @@ -32,6 +32,27 @@ interface IAccessor : IUnknown typedef DWORD DBBINDSTATUS; + enum DBACCESSORFLAGSENUM + { + DBACCESSOR_INVALID = 0x00, + DBACCESSOR_PASSBYREF = 0x01, + DBACCESSOR_ROWDATA = 0x02, + DBACCESSOR_PARAMETERDATA = 0x04, + DBACCESSOR_OPTIMIZED = 0x08, + DBACCESSOR_INHERITED = 0x10, + }; + + enum DBBINDSTATUSENUM + { + DBBINDSTATUS_OK = 0, + DBBINDSTATUS_BADORDINAL = 1, + DBBINDSTATUS_UNSUPPORTEDCONVERSION = 2, + DBBINDSTATUS_BADBINDINFO = 3, + DBBINDSTATUS_BADSTORAGEFLAGS = 4, + DBBINDSTATUS_NOINTERFACE = 5, + DBBINDSTATUS_MULTIPLESTORAGE = 6, + }; + [local] HRESULT AddRefAccessor([in] HACCESSOR hAccessor, [in, out, unique, annotation("__out_opt")] DBREFCOUNT *pcRefCount); -- 2.29.2

Dec. 30, 2020

[PATCH 2/2] conhost: Ignore CONTROL_C_EXIT exceptions in Unix mode.

by Jacek Caban

Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=50304 Signed-off-by: Jacek Caban <jacek(a)codeweavers.com> --- programs/conhost/conhost.c | 10 ++++++++++ 1 file changed, 10 insertions(+)

Dec. 29, 2020

[PATCH 1/2] condrv: Don't requre write access for IOCTL_CONDRV_CTRL_EVENT ioctl.

by Jacek Caban

Fixes propagating CONTROL_C_EXIT exception from console server handle. Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=50304 Signed-off-by: Jacek Caban <jacek(a)codeweavers.com> --- include/wine/condrv.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)

Dec. 29, 2020

[PATCH] winedbg: Remove arm_disasm_branchreg.

by liumeo＠pku.edu.cn

Dec. 29, 2020

Re: [PATCH v2] winegstreamer: Split audio data to fit in the sample buffer.

by Zebediah Figura (she/her)

If necessary I'm willing to sign off on the patch now, because it's code freeze, but since there's probably still time I've inlined some comments. I've also replaced the contents of the below message with the output of "git diff --ignore-all-space" for clarity. On 12/24/20 9:12 PM, Anton Baskanov wrote: > Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=50355 > Signed-off-by: Anton Baskanov <baskanov(a)gmail.com> > > diff --git a/dlls/winegstreamer/gstdemux.c b/dlls/winegstreamer/gstdemux.c > index 9114e6610a3..676b7396515 100644 > --- a/dlls/winegstreamer/gstdemux.c > +++ b/dlls/winegstreamer/gstdemux.c > @@ -81,6 +81,7 @@ struct gstdemux_source > HANDLE caps_event, eos_event; > GstSegment *segment; > SourceSeeking seek; > + DWORD bytes_per_second; > }; > > static inline struct gstdemux *impl_from_strmbase_filter(struct strmbase_filter *iface) > @@ -801,6 +802,7 @@ static GstFlowReturn got_data_sink(GstPad *pad, GstObject *parent, GstBuffer *bu > BYTE *ptr = NULL; > IMediaSample *sample; > GstMapInfo info; > + gsize position = 0; > > TRACE("%p %p\n", pad, buf); > > @@ -809,43 +811,60 @@ static GstFlowReturn got_data_sink(GstPad *pad, GstObject *parent, GstBuffer *bu > return GST_FLOW_OK; > } > > + gst_buffer_map(buf, &info, GST_MAP_READ); > + > + while (position < info.size) > + { > + gsize advance; > + > hr = BaseOutputPinImpl_GetDeliveryBuffer(&pin->pin, &sample, NULL, NULL, 0); > > if (hr == VFW_E_NOT_CONNECTED) { > + gst_buffer_unmap(buf, &info); > gst_buffer_unref(buf); > return GST_FLOW_NOT_LINKED; > } > > if (FAILED(hr)) { > + gst_buffer_unmap(buf, &info); > gst_buffer_unref(buf); > ERR("Could not get a delivery buffer (%x), returning GST_FLOW_FLUSHING\n", hr); > return GST_FLOW_FLUSHING; > } > > - gst_buffer_map(buf, &info, GST_MAP_READ); > + advance = min(IMediaSample_GetSize(sample), info.size - position); > > - hr = IMediaSample_SetActualDataLength(sample, info.size); > + hr = IMediaSample_SetActualDataLength(sample, advance); > if(FAILED(hr)){ > + gst_buffer_unmap(buf, &info); > + gst_buffer_unref(buf); Nitpick, but this is a preëxisting bug and could be a separate patch. > WARN("SetActualDataLength failed: %08x\n", hr); > return GST_FLOW_FLUSHING; > } > > IMediaSample_GetPointer(sample, &ptr); > > - memcpy(ptr, info.data, info.size); > - > - gst_buffer_unmap(buf, &info); > + memcpy(ptr, &info.data[position], advance); > > if (GST_BUFFER_PTS_IS_VALID(buf)) { > - REFERENCE_TIME rtStart = gst_segment_to_running_time(pin->segment, GST_FORMAT_TIME, buf->pts); > + REFERENCE_TIME rtStart; > + GstClockTime ptsStart = buf->pts; > + if (position > 0 && pin->bytes_per_second) I don't exactly like this. The "position > 0" part should be redundant (and isn't worth including just to avoid an extra calculation), but what bothers me more is the overloaded meaning of "pin->bytes_per_second". In practice I don't think we should really be indiscriminately splitting samples like this. We can't split e.g. video frames, for one. It's a bit of a moot point, since a buffer too big to fit in one sample is a bug *anyway*, but if we need to be able to scale by bit rate, we should limit the splitting to circumstances where bit rate is actually valid, which I think means only raw audio. (I'm not sure if it's also possible for GStreamer to arbitrarily cut mp3 frames, but if it is, and we need to worry about overlarge buffers, we'll need to find another way to calculate the bit rate.) Splitting the code like that would probably encourage the separation of a helper function to do some of the work of filling and sending an individual DirectShow sample. That of course would be a separate patch if introduced. Note that in that case the "bytes_per_second" field should probably go away, and be replaced with direct access of nAvgBytesPerSec from got_data_sink(). > + ptsStart = buf->pts + gst_util_uint64_scale(position, GST_SECOND, pin->bytes_per_second); > + rtStart = gst_segment_to_running_time(pin->segment, GST_FORMAT_TIME, ptsStart); > if (rtStart >= 0) > rtStart /= 100; > > if (GST_BUFFER_DURATION_IS_VALID(buf)) { > - REFERENCE_TIME tStart = buf->pts / 100; > - REFERENCE_TIME tStop = (buf->pts + buf->duration) / 100; > REFERENCE_TIME rtStop; > - rtStop = gst_segment_to_running_time(pin->segment, GST_FORMAT_TIME, buf->pts + buf->duration); > + REFERENCE_TIME tStart; > + REFERENCE_TIME tStop; > + GstClockTime ptsStop = buf->pts + buf->duration; > + if (position + advance < info.size && pin->bytes_per_second) > + ptsStop = buf->pts + gst_util_uint64_scale(position + advance, GST_SECOND, pin->bytes_per_second); > + tStart = ptsStart / 100; > + tStop = ptsStop / 100; > + rtStop = gst_segment_to_running_time(pin->segment, GST_FORMAT_TIME, ptsStop); > if (rtStop >= 0) > rtStop /= 100; > TRACE("Current time on %p: %i to %i ms\n", pin, (int)(rtStart / 10000), (int)(rtStop / 10000)); > @@ -860,9 +879,11 @@ static GstFlowReturn got_data_sink(GstPad *pad, GstObject *parent, GstBuffer *bu > IMediaSample_SetMediaTime(sample, NULL, NULL); > } > > - IMediaSample_SetDiscontinuity(sample, GST_BUFFER_FLAG_IS_SET(buf, GST_BUFFER_FLAG_DISCONT)); > + IMediaSample_SetDiscontinuity(sample, > + !position && GST_BUFFER_FLAG_IS_SET(buf, GST_BUFFER_FLAG_DISCONT)); This looks correct, but... > IMediaSample_SetPreroll(sample, GST_BUFFER_FLAG_IS_SET(buf, GST_BUFFER_FLAG_LIVE)); > - IMediaSample_SetSyncPoint(sample, !GST_BUFFER_FLAG_IS_SET(buf, GST_BUFFER_FLAG_DELTA_UNIT)); > + IMediaSample_SetSyncPoint(sample, > + !position && !GST_BUFFER_FLAG_IS_SET(buf, GST_BUFFER_FLAG_DELTA_UNIT)); this looks wrong, or at least weird. It doesn't make sense in the first place to mark a raw audio sample as a delta unit, but if hypothetically there were a splittable sample which was a delta unit, I think it'd be true that all of the pieces are delta units—none of them can be decoded alone. > > if (!pin->pin.pin.peer) > hr = VFW_E_NOT_CONNECTED; > @@ -871,14 +892,27 @@ static GstFlowReturn got_data_sink(GstPad *pad, GstObject *parent, GstBuffer *bu > > TRACE("sending sample returned: %08x\n", hr); > > - gst_buffer_unref(buf); > IMediaSample_Release(sample); > > if (hr == VFW_E_NOT_CONNECTED) > + { > + gst_buffer_unmap(buf, &info); > + gst_buffer_unref(buf); > return GST_FLOW_NOT_LINKED; > + } > > if (FAILED(hr)) > + { > + gst_buffer_unmap(buf, &info); > + gst_buffer_unref(buf); > return GST_FLOW_FLUSHING; > + } > + > + position += advance; > + } > + > + gst_buffer_unmap(buf, &info); > + gst_buffer_unref(buf); > > return GST_FLOW_OK; > } > @@ -2104,6 +2138,8 @@ static HRESULT WINAPI GSTOutPin_DecideBufferSize(struct strmbase_source *iface, > { > WAVEFORMATEX *format = (WAVEFORMATEX *)pin->pin.pin.mt.pbFormat; > buffer_size = format->nAvgBytesPerSec; > + > + pin->bytes_per_second = format->nAvgBytesPerSec; > } > > props->cBuffers = max(props->cBuffers, 1);

Dec. 29, 2020

[PATCH] devenum: Avoid initializing nameless unions in designated initializers.

by Zebediah Figura

Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=50423 Signed-off-by: Zebediah Figura <z.figura12(a)gmail.com> --- dlls/devenum/mediacatenum.c | 15 ++++++--------- 1 file changed, 6 insertions(+), 9 deletions(-) diff --git a/dlls/devenum/mediacatenum.c b/dlls/devenum/mediacatenum.c index ab675b5656f..f097c580261 100644 --- a/dlls/devenum/mediacatenum.c +++ b/dlls/devenum/mediacatenum.c @@ -137,17 +137,10 @@ static HRESULT WINAPI property_bag_Read(IPropertyBag *iface, } else if (!wcscmp(name, L"FilterData")) { - REGFILTERPINS2 reg_pins[2] = {{0}}; - REGFILTER2 reg_filter = - { - .dwVersion = 2, - .dwMerit = MERIT_NORMAL + 0x800, - .cPins2 = 2, - .rgPins2 = reg_pins, - }; - unsigned int count = 1, input_count, output_count, i; DMO_PARTIAL_MEDIATYPE *types = NULL, *new_array; + REGFILTERPINS2 reg_pins[2] = {{0}}; + REGFILTER2 reg_filter = {0}; REGPINTYPES *reg_types; HRESULT hr; @@ -192,6 +185,10 @@ static HRESULT WINAPI property_bag_Read(IPropertyBag *iface, reg_pins[1].cInstances = 1; reg_pins[1].nMediaTypes = output_count; reg_pins[1].lpMediaType = reg_types + count; + reg_filter.dwVersion = 2; + reg_filter.dwMerit = MERIT_NORMAL + 0x800, + reg_filter.cPins2 = 2; + reg_filter.rgPins2 = reg_pins; hr = create_filter_data(var, &reg_filter); free(reg_types); -- 2.29.2

Dec. 29, 2020

[PATCH v2] user32: Add stub for SetDisplayConfig().

by Paul Gofman

Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=50420 Signed-off-by: Paul Gofman <pgofman(a)codeweavers.com> --- v2: - add missing WINAPI and flags dump. dlls/user32/sysparams.c | 12 ++++++++++++ dlls/user32/user32.spec | 1 + 2 files changed, 13 insertions(+) diff --git a/dlls/user32/sysparams.c b/dlls/user32/sysparams.c index a620116053a..55d6d7e53a7 100644 --- a/dlls/user32/sysparams.c +++ b/dlls/user32/sysparams.c @@ -4794,3 +4794,15 @@ LONG WINAPI DisplayConfigGetDeviceInfo(DISPLAYCONFIG_DEVICE_INFO_HEADER *packet) return ERROR_INVALID_PARAMETER; } } + +/*********************************************************************** + * SetDisplayConfig (USER32.@) + */ +LONG WINAPI SetDisplayConfig(UINT32 path_info_count, DISPLAYCONFIG_PATH_INFO *path_info, UINT32 mode_info_count, + DISPLAYCONFIG_MODE_INFO *mode_info, UINT32 flags) +{ + FIXME("path_info_count %u, path_info %p, mode_info_count %u, mode_info %p, flags %#x stub.\n", + path_info_count, path_info, mode_info_count, mode_info, flags); + + return ERROR_SUCCESS; +} diff --git a/dlls/user32/user32.spec b/dlls/user32/user32.spec index 8051ae3f767..4ef75247d71 100644 --- a/dlls/user32/user32.spec +++ b/dlls/user32/user32.spec @@ -663,6 +663,7 @@ @ stdcall SetDeskWallPaper(str) # @ stub SetDeskWallpaper # @ stub SetDesktopBitmap +@ stdcall SetDisplayConfig(long ptr long ptr long) @ stdcall SetDlgItemInt(long long long long) @ stdcall SetDlgItemTextA(long long str) @ stdcall SetDlgItemTextW(long long wstr) -- 2.29.2

Dec. 29, 2020

Re: [PATCH] user32: Add stub for SetDisplayConfig().

by Paul Gofman

On 12/29/20 17:01, Zhiyi Zhang wrote: > > On 12/29/20 9:39 PM, Paul Gofman wrote: >> Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=50420 >> Signed-off-by: Paul Gofman <pgofman(a)codeweavers.com> >> --- >> dlls/user32/sysparams.c | 12 ++++++++++++ >> dlls/user32/user32.spec | 1 + >> 2 files changed, 13 insertions(+) >> >> diff --git a/dlls/user32/sysparams.c b/dlls/user32/sysparams.c >> index a620116053a..d9c383ba36f 100644 >> --- a/dlls/user32/sysparams.c >> +++ b/dlls/user32/sysparams.c >> @@ -4794,3 +4794,15 @@ LONG WINAPI DisplayConfigGetDeviceInfo(DISPLAYCONFIG_DEVICE_INFO_HEADER *packet) >> return ERROR_INVALID_PARAMETER; >> } >> } >> + >> +/*********************************************************************** >> + * SetDisplayConfig (USER32.@) >> + */ >> +LONG SetDisplayConfig(UINT32 path_info_count, DISPLAYCONFIG_PATH_INFO *path_info, UINT32 mode_info_count, >> + DISPLAYCONFIG_MODE_INFO *mode_info, UINT32 flags) >> +{ > Hi Paul, > > I think you forgot to add WINAPI to the function declaration and dump 'flags'. > > Thanks, > Zhiyi Oh yeah, indeed, thanks.

Dec. 29, 2020