August 2022 - Wine-gitlab - list.winehq.org

[PATCH v3 0/4] MR393: wmvcore: Asynchronous reader state transitions.
by Rémi Bernon 03 Aug '22

03 Aug '22

Before https://gitlab.winehq.org/wine/wine/-/merge_requests/140, and to avoid a situation in the future where we might decide to implement this, and where the callback would be called asynchronously as they are supposed to, without the ASF reader filter supporting it. -- v3: winegstreamer: Make IWMReader state transitions asynchronous. winegstreamer: Leave the stream CS while waiting for a sample. winegstreamer: Move IWMReaderCallbackAdvanced *callback_advanced to a local variable. wmvcore/tests: Add more IWMReader_(Open|Start|Stop|Close) async checks. https://gitlab.winehq.org/wine/wine/-/merge_requests/393

2 5

Re: [PATCH v3 0/11] MR557: mciqtz32: Correct video window behavior by creating default window.
by Akihiro Sagawa (＠sgwaki) 03 Aug '22

03 Aug '22

On Tue Aug 2 05:14:16 2022 +0000, **** wrote: > Zebediah Figura (she/her) replied on the mailing list: > ``` > On 7/31/22 06:53, Akihiro Sagawa wrote: > > @@ -1573,6 +1575,37 @@ static void test_video_window(void) > > todo_wine_if(i != 3) > > ok(style == expected, "hwnd %p: got %#lx, expected %#lx\n", > video_window, style, expected); > > > > + /* get source video size */ > > + err = mciSendCommandW(wDeviceID, MCI_WHERE, > MCI_DGV_WHERE_SOURCE, (DWORD_PTR)&parm); > > + ok(!err,"mciCommand where source returned %s\n", dbg_mcierr(err)); > > + src_rc = parm.where.rc; > This should just be equal to the whole video rect (0,0)-(64,32) [unless, > I suspect, it is set by MCI_DGV_PUT_SOURCE]; would you mind adding an > explicit test for that? > > + > > + /* test default video destination size */ > > + err = mciSendCommandW(wDeviceID, MCI_WHERE, > MCI_DGV_WHERE_DESTINATION, (DWORD_PTR)&parm); > > + ok(!err,"mciCommand where destination returned %s\n", dbg_mcierr(err)); > > + if (style & (WS_POPUP|WS_CHILD)) > > + rc = src_rc; > > + else > > + GetClientRect(video_window, &rc); > > + todo_wine_if(style & (WS_POPUP|WS_CHILD)) > > + ok(EqualRect(&parm.where.rc, &rc), "got %s, expected %s\n", > wine_dbgstr_rect(&parm.where.rc), wine_dbgstr_rect(&rc)); > And this should too. The test here is a bit awkward as such because the > destination and source rect are *always* equal, and separately, the > client rect will also be equal to the video size if the window is a > popup or child. > > + > > + /* test default video window size */ > > + rc = src_rc; > > + AdjustWindowRect(&rc, WS_OVERLAPPEDWINDOW, FALSE); /* sigh */ > > + OffsetRect(&rc, -rc.left, -rc.top); > > + rc.right = max(rc.right, GetSystemMetrics(SM_CXMIN)); > I gather "sigh" here means "Windows is broken", in the way spelled out > in 8/11, right? It wouldn't hurt to spell that out explicitly ;-) > _______________________________________________ > wine-gitlab mailing list -- wine-gitlab(a)winehq.org > To unsubscribe send an email to wine-gitlab-leave(a)winehq.org > ``` ``` > @@ -1573,6 +1575,37 @@ static void test_video_window(void) > todo_wine_if(i != 3) > ok(style == expected, "hwnd %p: got %#lx, expected %#lx\n", video_window, style, expected); > > + /* get source video size */ > + err = mciSendCommandW(wDeviceID, MCI_WHERE, MCI_DGV_WHERE_SOURCE, (DWORD_PTR)&parm); > + ok(!err,"mciCommand where source returned %s\n", dbg_mcierr(err)); > + src_rc = parm.where.rc; ``` > This should just be equal to the whole video rect (0,0)-(64,32) [unless, > I suspect, it is set by MCI_DGV_PUT_SOURCE]; would you mind adding an > explicit test for that? Not at all. Actually, video rect is (0,0)-(32,24). ``` > + > + /* test default video destination size */ > + err = mciSendCommandW(wDeviceID, MCI_WHERE, MCI_DGV_WHERE_DESTINATION, (DWORD_PTR)&parm); > + ok(!err,"mciCommand where destination returned %s\n", dbg_mcierr(err)); > + if (style & (WS_POPUP|WS_CHILD)) > + rc = src_rc; > + else > + GetClientRect(video_window, &rc); > + todo_wine_if(style & (WS_POPUP|WS_CHILD)) > + ok(EqualRect(&parm.where.rc, &rc), "got %s, expected %s\n", wine_dbgstr_rect(&parm.where.rc), wine_dbgstr_rect(&rc)); ``` > And this should too. The test here is a bit awkward as such because the > destination and source rect are *always* equal, and separately, the > client rect will also be equal to the video size if the window is a > popup or child. Minimum width of video window is SM_CXMIN. So, if the video width less than SM_CXMIN approximately, destination rect isn't identical to the source one. ``` > + > + /* test default video window size */ > + rc = src_rc; > + AdjustWindowRect(&rc, WS_OVERLAPPEDWINDOW, FALSE); /* sigh */ > + OffsetRect(&rc, -rc.left, -rc.top); > + rc.right = max(rc.right, GetSystemMetrics(SM_CXMIN)); ``` > I gather "sigh" here means "Windows is broken", in the way spelled out > in 8/11, right? It wouldn't hurt to spell that out explicitly ;-) Right. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/557#note_5586

2 1

Re: [PATCH v3 0/11] MR557: mciqtz32: Correct video window behavior by creating default window.
by Akihiro Sagawa (＠sgwaki) 03 Aug '22

03 Aug '22

On Tue Aug 2 05:14:13 2022 +0000, **** wrote: > Zebediah Figura (she/her) replied on the mailing list: > ``` > On 7/31/22 06:53, Akihiro Sagawa wrote: > > +static BOOL MCIQTZ_CreateWindow(WINE_MCIQTZ* wma, DWORD dwFlags, > LPMCI_DGV_OPEN_PARMSW lpParms) > > +{ > > + HWND parent = NULL; > > + DWORD style = (WS_OVERLAPPEDWINDOW | WS_CLIPCHILDREN) & ~WS_MAXIMIZEBOX; > > + LONG width, height; > > + RECT rc; > > + HRESULT hr; > > + > > + if (dwFlags & MCI_DGV_OPEN_PARENT) parent = lpParms->hWndParent; > > + if (dwFlags & MCI_DGV_OPEN_WS) style = lpParms->dwStyle; > > + > > + hr = IBasicVideo_GetVideoSize(wma->vidbasic, &width, &height); > > + if (hr == E_NOINTERFACE) > > + return TRUE; /* audio file */ > > + else if (FAILED(hr)) > > + return FALSE; > > + > > + /* FIXME: Use video size here */ > > + width = 640; > > + height = 480; > Since this is kind of a regression otherwise (and since patch 8/11 is > pretty much a one-liner) I think it makes more sense just to merge that > patch into this one. Apologies if I said the opposite last review... > > + > > + wma->window = CreateWindowW(MCIQTZ_CLASS, > lpParms->lpstrElementName, style, > > + CW_USEDEFAULT, CW_USEDEFAULT, > > + width, height, > > + parent, 0, MCIQTZ_hInstance, > > + 0); > _______________________________________________ > wine-gitlab mailing list -- wine-gitlab(a)winehq.org > To unsubscribe send an email to wine-gitlab-leave(a)winehq.org > ``` I'll do. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/557#note_5585

1 0

Re: [PATCH v3 0/11] MR557: mciqtz32: Correct video window behavior by creating default window.
by Akihiro Sagawa (＠sgwaki) 03 Aug '22

03 Aug '22

On Tue Aug 2 05:14:10 2022 +0000, **** wrote: > Zebediah Figura (she/her) replied on the mailing list: > ``` > On 7/31/22 06:53, Akihiro Sagawa wrote: > > @@ -183,9 +183,13 @@ static BOOL MCIQTZ_CreateWindow(WINE_MCIQTZ* wma, > DWORD dwFlags, LPMCI_DGV_OPEN_ > > else if (FAILED(hr)) > > return FALSE; > > > > - /* FIXME: Use video size here */ > > - width = 640; > > - height = 480; > > + /* Native always assumes an overlapped window > > + when calculating default video window size. */ > > + SetRect(&rc, 0, 0, width, height); > > + AdjustWindowRect(&rc, WS_OVERLAPPEDWINDOW, FALSE); > > + min_width = GetSystemMetrics(SM_CXMIN); > > + width = max(rc.right - rc.left, min_width); > CreateWindow() should automatically enforce the minimum width, so there > should be no need to use SM_CXMIN here. > _______________________________________________ > wine-gitlab mailing list -- wine-gitlab(a)winehq.org > To unsubscribe send an email to wine-gitlab-leave(a)winehq.org > ``` Got it. Good tip. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/557#note_5584

1 0

[PATCH 0/1] MR579: dwrite: Fix stretch value validation in CreateTextFormat().
by Nikolay Sivov (＠nsivov) 03 Aug '22

03 Aug '22

Signed-off-by: Nikolay Sivov <nsivov(a)codeweavers.com> -- https://gitlab.winehq.org/wine/wine/-/merge_requests/579

2 1

[PATCH v4 0/4] MR565: Implement NCryptEncrypt and BCryptEncrypt
by Santino Mazza (＠tati1454) 03 Aug '22

03 Aug '22

Some installers, like the visual studio installer, also make use of this function, instead of just NCryptVerifySignature. -- v4: ncrypt: Initial implementation for NCryptEncrypt. bcrypt: Initial RSA encryption implementation. bcrypt/tests: Test for BCryptEncrypt with RSA keys. https://gitlab.winehq.org/wine/wine/-/merge_requests/565

3 5

[PATCH v3 0/2] MR511: webservices: Support error strings and faults in error.
by Shaun Ren (＠shaunren) 03 Aug '22

03 Aug '22

-- v3: webservices: Support faults in error. webservices: Implement error strings. https://gitlab.winehq.org/wine/wine/-/merge_requests/511

3 4

[PATCH 0/19] MR578: d2d1: Add initial command list recording implementation.
by Nikolay Sivov (＠nsivov) 03 Aug '22

03 Aug '22

Signed-off-by: Nikolay Sivov <nsivov(a)codeweavers.com> -- This merge request has too many patches to be relayed via email. Please visit the URL below to see the contents of the merge request. https://gitlab.winehq.org/wine/wine/-/merge_requests/578

1 0

Re: [PATCH v5 0/4] MR406: win32u: Introduce new add_modes device manager callback.
by Rémi Bernon 03 Aug '22

03 Aug '22

On Mon Aug 1 07:56:19 2022 +0000, Zhiyi Zhang wrote: > Hi, R��mi. This is causing a new Coverity report. In > NtUserChangeDisplaySettings(), default_mode is passed to > read_registry_settings(), which eventually calls read_adapter_mode() and > then writes to the mode + 1. This is out of bound access because > default_mode is not an array. I think the correct fix is to not set > dmDriverExtra for ENUM_REGISTRY_SETTINGS. It writes only if `dmDriverExtra` is not zero. I think in both these two locations it is initialized to 0. It is used to read full modes for the available mode list. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/406#note_5556

1 0

[PATCH v2 0/2] MR571: ntdll: Fix floating point registers not being restored under Wow64.
by Brendan Shanks (＠bshanks) 03 Aug '22

03 Aug '22

This bug showed up on macOS Wow64 where XMM registers weren't being restored after an exception. -- v2: ntdll: Fix floating point and extended registers not being restored under Wow64. wow64: Copy floating point and extended registers in copy_context_64to32. https://gitlab.winehq.org/wine/wine/-/merge_requests/571

2 3