August 2014 - Wine-Devel - list.winehq.org

Re: [PATCH 1/5] comdlg32/tests: Add some IFileDialogEvents tests.
by Marvin Aug. 31, 2014

Aug. 31, 2014

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=8724 Your paranoid android. === wvistau64 (32 bit itemdlg) === itemdlg.c:1210: Test failed: Got 3 === wvistau64 (64 bit itemdlg) === itemdlg.c:1210: Test failed: Got 5

1 0

Re: [PATCH] mshtml: Add IHTMLTableCell::bgColor property implementation.
by Jacek Caban Aug. 29, 2014

Aug. 29, 2014

Hi Jactry, On 08/29/14 17:17, Jactry Zeng wrote: > --- a/dlls/mshtml/htmltablecell.c > +++ b/dlls/mshtml/htmltablecell.c > @@ -174,15 +174,47 @@ static HRESULT WINAPI HTMLTableCell_get_vAlign(IHTMLTableCell *iface, BSTR *p) > static HRESULT WINAPI HTMLTableCell_put_bgColor(IHTMLTableCell *iface, VARIANT v) > { > HTMLTableCell *This = impl_from_IHTMLTableCell(iface); > - FIXME("(%p)->(%s)\n", This, debugstr_variant(&v)); > - return E_NOTIMPL; > + nsAString strColor; > + nsresult nsres; > + > + TRACE("(%p)->(%s)\n", This, debugstr_variant(&v)); > + > + nsAString_InitDepend(&strColor, V_BSTR(&v)); > + variant_to_nscolor(&v, &strColor); Please handle variant_to_nscolor errors here. Jacek

2 1

Implementation of IDwriteFactory::CreateCustomFontCollection
by Guillaume Charifi Aug. 29, 2014

Aug. 29, 2014

Hello, I heard you are working on IDwriteFactory::CreateCustomFontCollection implementation, so I attached my current work, if it can help. It's almost done, just this part, retrieving the font family name to effectively save it, is to be implemented (even if Adobe Creative Cloud works with this changes): + if(!This->families) { + This->count = 1; + This->alloc = 1; + This->families = heap_alloc(This->alloc * sizeof(WCHAR *)); + This->families[0] = heap_strdupW((const WCHAR[]){'d','u','m','m','y',0}); + } else { + add_family_syscollection(This, (const WCHAR[]){'d','u','m','m','y',0}); + } In the hope it could be useful, Regards

1 0

Re: [2/2] winemenubuilder: ignore icons in unsupported PNG format
by Indrek Altpere Aug. 29, 2014

Aug. 29, 2014

Ah, indeed, my bad for not digging deep enough to the windowscodecs code and jumping to assumptions based on BITMAPINFO. Thanks for catching that, and as I can see on patches page the invalid patch is already marked as rejected. Regards, Indrek

1 0

Re: include: add EncryptedFileRaw functions and defines (try 2)
by Michael Stefaniuc Aug. 28, 2014

Aug. 28, 2014

On 08/28/2014 12:37 AM, Austin English wrote: > Splitting from https://source.winehq.org/patches/data/106010 for easier > review And the patch? bye michael

2 1

Re: [2/2] winemenubuilder: ignore icons in unsupported PNG format
by Damjan Jovanovic Aug. 28, 2014

Aug. 28, 2014

Please don't commit this patch, it's unnecessary. PNG support is not "missing": winemenubuilder supports PNG icons just fine and has for many years (IIRC, from .ico files and "valid" .exes/.dlls): it's implemented by using windowscodecs (starting from IWICImagingFactory_CreateDecoderFromStream() called from convert_to_native_icon()) which auto-detects the file format. In the case of PNG, that BITMAPINFO *pIcon pointer may not point to a valid BITMAPINFO structure, but who cares, the fields of that structure are never accessed, only opaquely memcpy()'ed en-masse. Using void* or BYTE* instead of BITMAPINFO* might be clearer, but the semantics are the same. I've tested Stepmania 5, one of the problematic apps from #19241, and winemenubuilder creates working 256x256 PNG icons when the previous patch is applied and this one isn't. Otherwise thank you for your contribution Indrek. Regards Damjan On Thu, Aug 28, 2014 at 6:09 PM, Indrek Altpere <efbiaiinzinz(a)hotmail.com> wrote: > While fixing https://bugs.winehq.org/show_bug.cgi?id=19241 I saw that > winemenubuilder does not handle/ignore png icon entries currently. > > As per MSDN blog, icon resources can contain raw PNG information instead of > regular BITMAPINFO, but due to weird decisions, only way to differentiate > between them is to check if the resource starts with PNG header bytes. > http://blogs.msdn.com/b/oldnewthing/archive/2010/10/22/10079192.aspx > Made the winemenubuilder ignore such entries for now (with fixme notices), > since using PNG raw bytes as BITMAPINFO data is definitely invalid and is > due to cause other issues/crashes when the best matching size icon happens > to be PNG. A new bug should be opened for the missing PNG support. > > Regards, > Indrek > > >

1 0

Re: wpcap: Load libpcap dynamically
by Alexandre Julliard Aug. 28, 2014

Aug. 28, 2014

André Hentschel <nerv(a)dawncrow.de> writes: > This is intended for packagers, without this patch libpcap easily becomes a fixed dependency of wine. > Something we don't want i guess. It's only used for wpcap, which can't work without it anyway, so I don't see any need to make it dynamic. -- Alexandre Julliard julliard(a)winehq.org

7 10

Re: winemenubuilder: fix crash caused by invalid icon entries and avoid future crashes by ignoring unhandled png entries (resend)
by Akihiro Sagawa Aug. 28, 2014

Aug. 28, 2014

On Thu, 28 Aug 2014 09:30:34 +0300, Indrek Altpere wrote: > Forgot to link to the bug before, also added more detailed description. > > Fixes https://bugs.winehq.org/show_bug.cgi?id=19241 > > For the InnoSetup 5 crash (and likely other similar reported crashes), the > issue seems to be GRPICONDIRENTRY with invalid information. (...) > As per MSDN blog, icon resources can contain raw PNG information instead of > regular BITMAPINFO, but due to weird decisions, only way to differentiate > between them is to check if the resource starts with PNG header bytes. > http://blogs.msdn.com/b/oldnewthing/archive/2010/10/22/10079192.aspx > Made the winemenubuilder ignore such entries for now (with fixme notices), > since using PNG raw bytes as BITMAPINFO data is definitely invalid and is > due to cause other issues/crashes when the best matching size icon happens > to be PNG. A new bug should be opened for the missing PNG support. Hello Indrek, I think your patch looks nice, but needs to split into two or more parts. This means one for invalid GRPICONDIRENTRY issue and one for PNG entry's enhancement. If you do this, it will be easier to read and easier to pass. For details, please read our Patch Submission Guidelines[1], especially Code Guidelines section. [1] http://wiki.winehq.org/SubmittingPatches Regards, Akihiro Sagawa

1 0

Re: [1/2] netcfgx: Add stub for INetCfgComponent
by Alistair Leslie-Hughes Aug. 28, 2014

Aug. 28, 2014

On 28/08/2014 6:39 AM, André Hentschel wrote: > --- > dlls/netcfgx/netcfg.c | 148 ++++++++++++++++++++++++++++++++++++++++++++++++++ > 1 file changed, 148 insertions(+) > Your missing tests to show that this interface exist. Same goes for INetCfgClassSetup2. Best Regards Alistair Leslie-Hughes

1 0

Re: [PATCH 2/5] dxgi: Free the wined3d object when device creation fails (Valgrind).
by Henri Verbeet Aug. 27, 2014

Aug. 27, 2014

On 27 August 2014 18:40, Matteo Bruni <mbruni(a)codeweavers.com> wrote: > --- > dlls/dxgi/device.c | 8 +++++++- > 1 file changed, 7 insertions(+), 1 deletion(-) > > diff --git a/dlls/dxgi/device.c b/dlls/dxgi/device.c > index 837c57f..0f3ae7d 100644 > --- a/dlls/dxgi/device.c > +++ b/dlls/dxgi/device.c > @@ -350,7 +350,7 @@ HRESULT dxgi_device_init(struct dxgi_device *device, struct dxgi_device_layer *l > IWineDXGIDeviceParent *dxgi_device_parent; > IWineDXGIAdapter *wine_adapter; > UINT adapter_ordinal; > - struct wined3d *wined3d; > + struct wined3d *wined3d = NULL; > void *layer_base; > HRESULT hr; > WINED3DCAPS caps; > @@ -430,6 +430,12 @@ fail: > wined3d_device_decref(device->wined3d_device); > LeaveCriticalSection(&dxgi_cs); > } > + else if (wined3d) > + { > + EnterCriticalSection(&dxgi_cs); > + wined3d_decref(wined3d); > + LeaveCriticalSection(&dxgi_cs); > + } I don't think this works as intended. For example, if IDXGIAdapter_QueryInterface() fails, you'd call wined3d_decref() twice. There's also a wined3d_decref() after wined3d_device_create(), although it's indeed missing for IWineDXGIDevice_QueryInterface() and wined3d_get_device_caps().

2 1