January 2015 - Wine-Devel - list.winehq.org

Re: dpnet: Implement IDirectPlay8Address GetComponentByIndex (try 5)
by Stefan Dösinger Jan. 26, 2015

Jan. 26, 2015

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi, > +static void add_component(IDirectPlay8AddressImpl *This, struct component *item) > +{ > + if(!This->comp_count) > + This->components = HeapAlloc(GetProcessHeap(), 0, sizeof(struct component *) * This->comp_count+1 ); You can use your heap_alloc() helper here. The same applies to the HeapFree calls added in the patch. Other than that I think error handling would be a good idea. You can check for NULL (as well as in the place where you allocate storage for the component in AddComponent) and return from AddComponent with E_OUTOFMEMORY. The patch has a few formatting issues. Since it looks like dplay & friends will be mostly your reign I'll leave it up to you on how to handle formatting. Do you want me to continue pointing them out? In that case please send me a mail what your intended formatting is. (Obviously it'd be even better if you described your intended formatting to git in the form of a git pre-commit hook containing some regexes ;-) ). Other than this the patch looks good. The following is unrelated to this patch, I just noticed it (again?) while reading through it: Am 2015-01-23 um 08:10 schrieb Alistair Leslie-Hughes: > + case DPNA_DATATYPE_STRING: > + memcpy(pvBuffer, &entry->data.string, entry->size); > + break; Please add a test for a case where AddComponent is called with DPNA_DATATYPE_STRING where (strlen(string) + 1) * sizeof(WCHAR) != dwDataSize. -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iQIcBAEBAgAGBQJUxS4iAAoJEN0/YqbEcdMwvFkP/RF6kH1o1C4fQgfa/J+4j6zV kwp3M7rqU5VTasKYV+G0pdAQ+wvBzHpMPSqLRD/HfBWGf3dAPCk94C9d659enkJN Qhy7hhO6aazsXKq7iWVAgQrGJy59oBoL14Vg+44hUJIeu7qWv3uuqihVgKUqiG2K Bl235p9cM/WYLGXFEF4d1UFHBKZb0bhRudwuzRbuN8r3RLwpEhrKbbKg8yXOPjLq Vejc4d219iib3ZpbT13pq0j/+a+ox+R62+O9S5wGbu/G8IGxVPTgsY20mpFyPsHX DoedE56wfSSe5Psb78NVmC32bWY8nmUzJA9YBb26a0KPO9QZNVnJkkYkOKIFr6P/ ouVLfWbO4QV2dOPRbYnh+40XCIbPeatGMq9vE7TDZPtxLBkRTEl1GudZ5usFwCK7 XgTPjb+BIu37UCbt0hpWA4IH8UeyYue08DnGCWkMkGS6ClH9KCG2xVqL8F1gr8Bw W5iKcVVHFZzHS6D947nN17Fl6Rg4+oLui0zkgtveU/AY/ew8xptHt4rW6It3QU5t HkNce4St+2w1LErIuYRITrvQV6lrndDDgvvlPcK1P1zt2D1a9gpqnsPgKM/tZLu+ fO8P8Yjl81KZHlAVWMUutiDzlelUOJvaXnVNX5VVkyOajHDI569MKy4qOETSdiim gwulxslghTkPCsAa+LpB =iY0u -----END PGP SIGNATURE-----

1 1

Re: [PATCH 5/5] atl100/tests: Added initial interface tests for AtlAxAttachControl.
by Qian Hong Jan. 26, 2015

Jan. 26, 2015

Please ignore this patch, I'm working on an improved version. On Fri, Jan 23, 2015 at 5:21 PM, Qian Hong <qhong(a)codeweavers.com> wrote: > Test cases based on dlls/mshtml/tests/activex.c, thanks Jacek Caban for the > previous work. > > --- > dlls/atl100/tests/Makefile.in | 2 +- > dlls/atl100/tests/atl.c | 446 > ++++++++++++++++++++++++++++++++++++++++++ > 2 files changed, 447 insertions(+), 1 deletion(-) > > > > > -- Regards, Qian Hong - http://www.codeweavers.com

1 0

Re: [3/3] cmd: implemented && and || functionality (try 2)
by Marvin Jan. 24, 2015

Jan. 24, 2015

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=11295 Your paranoid android. === build (build) === Patch failed to apply

1 0

Wine 1.7.35 source code scanning (Cppcheck)
by Игорь Чебаненко Jan. 24, 2015

Jan. 24, 2015

1 0

Re: [PATCH 3/5] atl: Added stub AtlIPersistPropertyBag_Save.
by Jacek Caban Jan. 23, 2015

Jan. 23, 2015

On 01/23/15 10:18, Qian Hong wrote: > +HRESULT WINAPI AtlIPersistPropertyBag_Save(LPPROPERTYBAG pPropBag, BOOL fClearDirty, BOOL fSaveAllProperties, > + ATL_PROPMAP_ENTRY *pMap, void *pThis, IUnknown *pUnk) > +{ > + > + FIXME("(%p, %x, %x, %p, %p, %p)\n", pPropBag, fClearDirty, fSaveAllProperties, pMap, pThis, pUnk); > + return S_OK; > +} It's a minor detail, but after two patches in series removing unnecessary empty lines, you add a new one here ;) Cheers, Jacek

2 1

Re: [PATCH 4/5] atl100: Implemented semi stub AtlAxCreateControlLic and AtlAxCreateControlLicEx.
by Jacek Caban Jan. 23, 2015

Jan. 23, 2015

Hi Qian, On 01/23/15 10:19, Qian Hong wrote: > diff --git a/dlls/atl/atl_ax.c b/dlls/atl/atl_ax.c > index 15b0550..656165b 100644 > --- a/dlls/atl/atl_ax.c > +++ b/dlls/atl/atl_ax.c > @@ -1435,8 +1435,9 @@ INT_PTR WINAPI AtlAxDialogBoxA(HINSTANCE hInstance, LPCSTR lpTemplateName, HWND > */ > HRESULT WINAPI AtlAxCreateControlLic(const WCHAR *lpTricsData, HWND hwnd, IStream *stream, IUnknown **container, BSTR lic) > { > - FIXME("(%s %p %p %p %s)\n", debugstr_w(lpTricsData), hwnd, stream, container, debugstr_w(lic)); > - return E_NOTIMPL; > + if (lic) > + FIXME("semi stub (%s %p %p %p %s)\n", debugstr_w(lpTricsData), hwnd, stream, container, debugstr_w(lic)); > + return AtlAxCreateControl(lpTricsData, hwnd, stream, container); > } > > /*********************************************************************** > @@ -1446,9 +1447,10 @@ HRESULT WINAPI AtlAxCreateControlLic(const WCHAR *lpTricsData, HWND hwnd, IStrea > HRESULT WINAPI AtlAxCreateControlLicEx(const WCHAR *lpTricsData, HWND hwnd, IStream *stream, > IUnknown **container, IUnknown **control, REFIID iidSink, IUnknown *punkSink, BSTR lic) > { > - FIXME("(%s %p %p %p %p %s %p %s)\n", debugstr_w(lpTricsData), hwnd, stream, container, control, > + if (lic) > + FIXME("semi stub (%s %p %p %p %p %s %p %s)\n", debugstr_w(lpTricsData), hwnd, stream, container, control, > debugstr_guid(iidSink), punkSink, debugstr_w(lic)); > - return E_NOTIMPL; > + return AtlAxCreateControlEx(lpTricsData, hwnd, stream, container, control, iidSink, punkSink); > } > > #endif Please forward AtlAxCreateControlEx to AtlAxCreateControlLicEx and add FIXME about unsupported argument instead of the other way around. Otherwise we'd have to do that later to support the argument anyway. Also AtlAxCreateControlLic would probably be better forwarded to AtlAxCreateControlLicEx. Cheers, Jacek

2 1

Re: [PATCH 5/5] wined3d: Use the core version of the shader object functions.
by Henri Verbeet Jan. 23, 2015

Jan. 23, 2015

On 21 January 2015 at 18:32, Matteo Bruni <mbruni(a)codeweavers.com> wrote: > + typedef void (WINE_GLAPI *gl_shader_source_type)(GLuint, GLsizei, const GLchar * const *, const GLint*); > + ... > + #define MAP_GL_FUNCTION_CAST(core_func, cast, ext_func) \ > + do \ > + { \ > + if (!gl_info->gl_ops.ext.p_##core_func) \ > + gl_info->gl_ops.ext.p_##core_func = (cast)gl_info->gl_ops.ext.p_##ext_func; \ > + } while (0) ... > + MAP_GL_FUNCTION_CAST(glShaderSource, gl_shader_source_type, glShaderSourceARB); I'm not sure that's really much better than just adding a (void *) cast.

3 4

Re: shell32: Implement IApplicationAssociationRegistration, QueryCurrentDefault (try 6)
by Nikolay Sivov Jan. 23, 2015

Jan. 23, 2015

On 23.01.2015 10:24, Alistair Leslie-Hughes wrote: > + ret = RegGetValueW(hkey, NULL, NULL, RRF_RT_REG_SZ, &keytype, NULL, &size); > + if(ret == ERROR_SUCCESS) > + { > + *association = CoTaskMemAlloc(size * sizeof(WCHAR)); Does it really return length in characters? I thought it's always in bytes. > + if(*association) > + { > + ret = RegGetValueW(hkey, NULL, NULL, RRF_RT_REG_SZ, &keytype, *association, &size); > + if(ret == ERROR_SUCCESS) > + hr = S_OK; > + } > + } In case of !*association please return earlier with E_OUTOFMEMORY.

1 0

Re: dpnet: Implement IDirectPlay8Address GetComponentByIndex (try 5)
by Marvin Jan. 23, 2015

Jan. 23, 2015

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=11271 Your paranoid android. === w8 (32 bit address) === The test timed out

1 0

Re: WineHQ zh_CN Translation Patch Up
by Bruno Jesus Jan. 22, 2015

Jan. 22, 2015

On Thu, Jan 22, 2015 at 2:10 PM, Jeff Bai <jeffbaichina(a)gmail.com> wrote: > Hello fellow developers > > Here is a copy of the patch of changed zh_CN localization from AOSC (Anthon > Open Source Community) L10n project. We've fixed the msgid width problem > which was addressed last time. The patch is large so we put this on our > public PasteBin site, as follows > > https://pastebin.anthonos.org/view/a0c1c1a3 > > Please mail back with any issue over this patch. Hi, please send the email again with the patch as an attachment and I think a more proper subject would be: po: Update Simplified Chinese translation > Keep up the good work! > > Jeff Bai > AOSC Best regards, Bruno

1 0

Re: WineHQ zh_CN Translation Patch Up
by Alex Henrie Jan. 22, 2015

Jan. 22, 2015

Jeff, The best way to submit patches is with `git format-patch` and `git send-email`. It looks like Alexandre won't accept patches submitted via pastebin. -Alex

1 0

Re: ws2_32: Better trace setsockopt/getsockopt
by Bruno Jesus Jan. 22, 2015

Jan. 22, 2015

On Tue, Jan 6, 2015 at 12:49 AM, Bruno Jesus <00cpxxx(a)gmail.com> wrote: > The intention of this patch is to make it easier to read winsock logs, > for setsockopt it even displays the value if its size is less than 5 > bytes. Hi, any complains about this patch? It's pending for 2 weeks now and I think this is a good waiting time to ask. Best Regards, Bruno

2 1

Re: [PATCH 3/5] wined3d: Initialize the last two components of R16G16_SINT attributes.
by Henri Verbeet Jan. 22, 2015

Jan. 22, 2015

On 21 January 2015 at 18:32, Matteo Bruni <mbruni(a)codeweavers.com> wrote: > - GL_EXTCALL(glVertexAttrib4svARB(index, ptr)); > + { > + GLshort s[4] = {((const GLshort *)ptr)[0], ((const GLshort *)ptr)[1], 0, 1}; > + GL_EXTCALL(glVertexAttrib4svARB(index, s)); > break; > + } Perhaps I'm missing something, but wouldn't it be much easier to just use glVertexAttrib2svARB() instead?

1 0

Re: [PATCH 6/6] kernel32: Invalid console handles for new processes are 0, not INVALID_HANDLE_VALUE.
by Alistair Leslie-Hughes Jan. 21, 2015

Jan. 21, 2015

On 21/01/15 08:51, Erich E. Hoover wrote: > Several applications rely on the TEB value for fresh invalid console > handles to be zero instead of INVALID_HANDLE_VALUE (fixes Bug #36216 > and Bug #37914). > > +····if·(!params->hStdOutput·||·params->hStdOutput·==·INVALID_HANDLE_VALUE) > +········params->hStdOutput·=·0; The !params->hStdOutput check is redundant since it's already 0. Applies to the other places as well. Best Regards Alistair Leslie-Hughes

2 1

Wine Gecko 2.36-beta1
by Jacek Caban Jan. 21, 2015

Jan. 21, 2015

Hi all, I just uploaded new Wine Gecko beta packages. This time it's based on Firefox 36 beta branch, so it should be pretty stable. To test it please use the attached patch. That's all you need, Wine will download Gecko automatically. Packages are also available on Wine SourceForge if you prefer to download it manually. All testing is appreciated. Regressions and build problems are most important, but all feedback is welcomed. I'm also sending this announcement mingw-w64 mailing list. I know that some distro package maintainers that maintain Wine Gecko packages read it. Beta releases are a good chance to make sure that it builds with toolchains available in distros. Note that the package requires very recent mingw-w64 Git version. A new release on mingw-w64 is planned, so hopefully it will be out before final Wine Gecko release (which is expected in 5-6 weeks). Thanks, Jacek

1 0

Re: msi: Add a partial implementation of MsiProvideComponent.
by Marvin Jan. 21, 2015

Jan. 21, 2015

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=11245 Your paranoid android. === wxppro (32 bit msi) === The test timed out === w7u (32 bit msi) === The test timed out

1 0

Re: msxml3/tests: Actually test the return values (PVS-Studio)
by Michael Stefaniuc Jan. 21, 2015

Jan. 21, 2015

Please ignore, wrong resend. On 01/21/2015 10:48 AM, Michael Stefaniuc wrote: > --- > dlls/msxml3/tests/domdoc.c | 1 + > dlls/msxml3/tests/schema.c | 1 + > 2 files changed, 2 insertions(+) > > diff --git a/dlls/msxml3/tests/domdoc.c b/dlls/msxml3/tests/domdoc.c > index a23fb66..e669de4 100644 > --- a/dlls/msxml3/tests/domdoc.c > +++ b/dlls/msxml3/tests/domdoc.c > @@ -11085,6 +11085,7 @@ static void test_xmlns_attribute(void) > V_VT(&v) = VT_BSTR; > V_BSTR(&v) = _bstr_("urn:schemas-microsoft-com:datatypes"); > hr = IXMLDOMAttribute_put_nodeValue(pAttribute, v); > + ok(hr == S_OK, "ret %08x\n", hr ); > > hr = IXMLDOMElement_setAttributeNode(root, pAttribute, NULL); > ok(hr == S_OK, "ret %08x\n", hr ); > diff --git a/dlls/msxml3/tests/schema.c b/dlls/msxml3/tests/schema.c > index f5d6fcd..be18fcf 100644 > --- a/dlls/msxml3/tests/schema.c > +++ b/dlls/msxml3/tests/schema.c > @@ -1292,6 +1292,7 @@ static void test_XDR_datatypes(void) > > VariantClear(&type); > hr = IXMLDOMNode_get_nodeTypedValue(node, &type); > + EXPECT_HR(hr, S_OK); > ok(V_VT(&type) == VT_BSTR, "got variant type %i\n", V_VT(&v)); > VariantClear(&type); > IXMLDOMNode_Release(node); >

1 0

Re: msxml3/tests: Remove an unused assignment (PVS-Studio)
by Michael Stefaniuc Jan. 21, 2015

Jan. 21, 2015

*SIGH* please ignore this one as this is already commited. Pushed out the wrong mailbox. On 01/21/2015 10:43 AM, Michael Stefaniuc wrote: > --- dlls/msxml3/tests/domdoc.c | 1 - 1 file changed, 1 deletion(-) > > diff --git a/dlls/msxml3/tests/domdoc.c > b/dlls/msxml3/tests/domdoc.c index 1a73771..a23fb66 100644 --- > a/dlls/msxml3/tests/domdoc.c +++ b/dlls/msxml3/tests/domdoc.c @@ > -4903,7 +4903,6 @@ static void test_xmlTypes(void) > > doc = create_document(&IID_IXMLDOMDocument); > > - pNextChild = (void*)0xdeadbeef; hr = > IXMLDOMDocument_get_nextSibling(doc, NULL); ok(hr == E_INVALIDARG, > "ret %08x\n", hr );

1 0

Re: msxml3/tests: Actually test the return values (PVS-Studio)
by Marvin Jan. 21, 2015

Jan. 21, 2015

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=11239 Your paranoid android. === build (build) === Patch failed to apply

1 0

Re: msxml3/tests: Remove an unused assignment (PVS-Studio)
by Marvin Jan. 21, 2015

Jan. 21, 2015

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=11238 Your paranoid android. === build (build) === Patch failed to apply

1 0

Re: user32/tests: Don't test function directly when reporting GetLastError()
by Marvin Jan. 21, 2015

Jan. 21, 2015

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=11234 Your paranoid android. === w7u (32 bit msg) === msg.c:12108: Test failed: unexpected hwnd 0014013C

1 0

About Patch 108345 Pending ( gdiplus: fix get_clip_hrgn when window origin point changed (try 2) )
by Changhui Liu Jan. 19, 2015

Jan. 19, 2015

Hi, all, I reviewed this path http://source.winehq.org/patches/data/108345, I only found a code style problem: this: + if ( status == Ok && graphics->hdc)should :+ if (status == Ok && graphics->hdc) Could you help me point out the other mistake ? Thanks . ------------------ Regards, Changhui.

2 1

Re: gdi32: Do not enumerate fonts with names that are too long
by Dmitry Timoshkov Jan. 19, 2015

Jan. 19, 2015

Aric Stewart <aric(a)codeweavers.com> wrote: > Both the FullName and the FamilyName need to fit in the LOGFONT > structure. What happens with such a font under Windows? I'd expect the name be simply truncated instead of the font being skipped. In the latter case the font should not be added to the list at all. -- Dmitry.

2 1

Re: [PATCH 8/9] wined3d: Filter messages generated by focus handling.
by Henri Verbeet Jan. 19, 2015

Jan. 19, 2015

On 4 December 2014 at 21:51, Stefan Dösinger <stefan(a)codeweavers.com> wrote: > We may want to consider moving mode changes, window setup and focus loss > handling into the client libs. With this patch and the next one we have > 3 wined3d flags just for focus loss handling. Furthermore, dxgi yet > again has different behavior - it doesn't hook the window procedure and Yeah, possibly, although I'd also like to keep the amount of Win32 specific stuff in the public wined3d API to a minimum, so I'd rather not have the client libs e.g. handle window messages directly. > diff --git a/dlls/wined3d/device.c b/dlls/wined3d/device.c > index ae6600e..89ec00a 100644 > --- a/dlls/wined3d/device.c > +++ b/dlls/wined3d/device.c > @@ -4961,7 +4961,7 @@ void device_invalidate_state(const struct wined3d_device *device, DWORD state) > LRESULT device_process_message(struct wined3d_device *device, HWND window, BOOL unicode, > UINT message, WPARAM wparam, LPARAM lparam, WNDPROC proc) > { > - if (device->filter_messages) > + if (device->filter_messages && message != WM_DISPLAYCHANGE) That seems pretty specific. I'd be curious how other broadcast messages are handled.

2 2

Re: [PATCH 9/9] wined3d: Pass WM_SYSCOMMAND(SC_RESTORE) to DefWindowProc.
by Henri Verbeet Jan. 19, 2015

Jan. 19, 2015

On 4 December 2014 at 21:51, Stefan Dösinger <stefan(a)codeweavers.com> wrote: > diff --git a/dlls/wined3d/device.c b/dlls/wined3d/device.c > index 89ec00a..51dabd4 100644 > --- a/dlls/wined3d/device.c > +++ b/dlls/wined3d/device.c > @@ -4992,6 +4992,16 @@ LRESULT device_process_message(struct wined3d_device *device, HWND window, BOOL > > device->device_parent->ops->activate(device->device_parent, wparam); > } > + else if (message == WM_SYSCOMMAND) > + { > + if (wparam == SC_RESTORE && !(device->wined3d->flags & WINED3D_NO_SYSCOMMAND_HANDLING)) I'm not really a fan of this kind of double negative. The way the flag is used, it also seems to be about restores in particular instead of syscommands in general. > + { > + if (unicode) > + DefWindowProcW(window, message, wparam, lparam); > + else > + DefWindowProcA(window, message, wparam, lparam); > + } > + } > > if (unicode) > return CallWindowProcW(proc, window, message, wparam, lparam); Is this really how it's supposed to work? In particular, should DefWindowProc really always get the message before the application does, or should we give the application a chance first, and only pass to DefWindowProc if it doesn't handle the message?

2 2