June 2022 - Wine-Devel - list.winehq.org

[PATCH 1/1] include: Emit CFI directives when building on Mac with gcc.

by Stefan Dösinger

From: Stefan Dösinger <stefan(a)codeweavers.com> Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=53157 Building with FSF gcc on mac is possible and the only way to get a --without-mingw build since clang does not support -mabi=ms. (Some hacks are needed to build winemac.drv.so though.) Signed-off-by: Stefan Dösinger <stefan(a)codeweavers.com> --- No, FSF gcc on macos does not define __GCC_HAVE_DWARF2_CFI_ASM. --- include/wine/asm.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/wine/asm.h b/include/wine/asm.h index 0547ee94b19..2c996ccd86b 100644 --- a/include/wine/asm.h +++ b/include/wine/asm.h @@ -35,7 +35,7 @@ # define __ASM_FASTCALL(name,args) __ASM_NAME("__fastcall_" name) #endif -#if defined(__GCC_HAVE_DWARF2_CFI_ASM) || (defined(__clang__) && defined(__GNUC__) && !defined(__SEH__)) +#if defined(__GCC_HAVE_DWARF2_CFI_ASM) || ((defined(__APPLE__) || defined(__clang__)) && defined(__GNUC__) && !defined(__SEH__)) # define __ASM_CFI(str) str #else # define __ASM_CFI(str) -- GitLab https://gitlab.winehq.org/wine/wine/-/merge_requests/249

June 15, 2022

[PATCH 0/1] MR249: include: Emit CFI directives when building on Mac with gcc.

by Stefan Dösinger

Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=53157 Building with FSF gcc on mac is possible and the only way to get a --without-mingw build since clang does not support -mabi=ms. (Some hacks are needed to build winemac.drv.so though.) Signed-off-by: Stefan Dösinger <stefan(a)codeweavers.com> --- No, FSF gcc on macos does not define __GCC_HAVE_DWARF2_CFI_ASM. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/249

June 15, 2022

Re: [PATCH v2 1/1] shell32: Partially implement IShellItemImageFactory (icon only, no thumbnail).

by Marvin

Hi, It looks like your patch introduced the new failures shown below. Please investigate and fix them if they are indeed new. Note that rare failures and failures with always changing text (e.g. because of memory addresses) can cause false positives. If this is what happened, then fixing those would really help. The tests also ran into some preexisting test failures. If you know how to fix them that would be helpful. See the TestBot job for the details: The full results can be found at: https://testbot.winehq.org/JobDetails.pl?Key=117022 Your paranoid android. === w1064v1809 (32 bit report) === shell32: shlfolder.c:4959: Test failed: CREATE: expected notification type 2, got: 40000 shlfolder.c:4966: Test failed: GetDisplayNameOf failed: 0x80070057 shlfolder.c:4976: Test failed: Didn't expect a WM_USER_NOTIFY message (event: 2) === w1064 (32 bit report) === shell32: shlfolder.c:4959: Test failed: CREATE: expected notification type 2, got: 40000 shlfolder.c:4966: Test failed: GetDisplayNameOf failed: 0x80070057 shlfolder.c:4976: Test failed: Didn't expect a WM_USER_NOTIFY message (event: 2) === w1064_tsign (32 bit report) === shell32: shlfolder.c:4959: Test failed: CREATE: expected notification type 2, got: 40000 shlfolder.c:4966: Test failed: GetDisplayNameOf failed: 0x80070057 shlfolder.c:4976: Test failed: Didn't expect a WM_USER_NOTIFY message (event: 2) === w1064v1809 (64 bit report) === shell32: shlfolder.c:4976: Test failed: Didn't expect a WM_USER_NOTIFY message (event: 3) === w1064 (64 bit report) === shell32: shlfolder.c:4976: Test failed: Didn't expect a WM_USER_NOTIFY message (event: 3) === w1064_2qxl (64 bit report) === shell32: shlfolder.c:4976: Test failed: Didn't expect a WM_USER_NOTIFY message (event: 3) === w1064_adm (64 bit report) === shell32: shlfolder.c:4976: Test failed: Didn't expect a WM_USER_NOTIFY message (event: 40000) === w1064_tsign (64 bit report) === shell32: shlfolder.c:4959: Test failed: CREATE: expected notification type 2, got: 40000 shlfolder.c:4966: Test failed: GetDisplayNameOf failed: 0x80070057 shlfolder.c:4976: Test failed: Didn't expect a WM_USER_NOTIFY message (event: 2) === w10pro64_en_AE_u8 (64 bit report) === shell32: shlfolder.c:4976: Test failed: Didn't expect a WM_USER_NOTIFY message (event: 3) shlfolder.c:4976: Test failed: Didn't expect a WM_USER_NOTIFY message (event: 3) shlfolder.c:4976: Test failed: Didn't expect a WM_USER_NOTIFY message (event: 3) shlfolder.c:4976: Test failed: Didn't expect a WM_USER_NOTIFY message (event: 3) shlfolder.c:4976: Test failed: Didn't expect a WM_USER_NOTIFY message (event: 3) shlfolder.c:4959: Test failed: RMDIR: expected notification type 10, got: 40000 shlfolder.c:4966: Test failed: GetDisplayNameOf failed: 0x80070057 shlfolder.c:4976: Test failed: Didn't expect a WM_USER_NOTIFY message (event: 3) === w10pro64_ar (64 bit report) === shell32: shlfolder.c:4976: Test failed: Didn't expect a WM_USER_NOTIFY message (event: 2) shlfolder.c:4959: Test failed: RMDIR: expected notification type 10, got: 40000 shlfolder.c:4966: Test failed: GetDisplayNameOf failed: 0x80070057 shlfolder.c:4976: Test failed: Didn't expect a WM_USER_NOTIFY message (event: 10) shlfolder.c:4976: Test failed: Didn't expect a WM_USER_NOTIFY message (event: 3) shlfolder.c:4976: Test failed: Didn't expect a WM_USER_NOTIFY message (event: 3) shlfolder.c:4976: Test failed: Didn't expect a WM_USER_NOTIFY message (event: 3) === w10pro64_ja (64 bit report) === shell32: shlfolder.c:4959: Test failed: RMDIR: expected notification type 10, got: 40000 shlfolder.c:4966: Test failed: GetDisplayNameOf failed: 0x80070057 shlfolder.c:4976: Test failed: Didn't expect a WM_USER_NOTIFY message (event: 10) shlfolder.c:4976: Test failed: Didn't expect a WM_USER_NOTIFY message (event: 3) shlfolder.c:4976: Test failed: Didn't expect a WM_USER_NOTIFY message (event: 3) shlfolder.c:4976: Test failed: Didn't expect a WM_USER_NOTIFY message (event: 3) shlfolder.c:4976: Test failed: Didn't expect a WM_USER_NOTIFY message (event: 3) shlfolder.c:4959: Test failed: CREATE: expected notification type 2, got: 40000 shlfolder.c:4966: Test failed: GetDisplayNameOf failed: 0x80070057 shlfolder.c:4976: Test failed: Didn't expect a WM_USER_NOTIFY message (event: 3) === w10pro64_zh_CN (64 bit report) === shell32: shlfolder.c:4976: Test failed: Didn't expect a WM_USER_NOTIFY message (event: 3)

June 15, 2022

[PATCH v2 1/1] comctl32: Always paint SS_BITMAP static control background with WM_CTLCOLORSTATIC brush.

by Jinoh Kang

From: Jinoh Kang <jinoh.kang.kr(a)gmail.com> This is consistent with the behaviour of other static control styles. Signed-off-by: Jinoh Kang <jinoh.kang.kr(a)gmail.com> --- dlls/comctl32/static.c | 7 ++++--- dlls/comctl32/tests/static.c | 1 - 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/dlls/comctl32/static.c b/dlls/comctl32/static.c index 3151cda6552..c09655da903 100644 --- a/dlls/comctl32/static.c +++ b/dlls/comctl32/static.c @@ -849,13 +849,16 @@ static void STATIC_PaintBitmapfn(HWND hwnd, HDC hdc, HBRUSH hbrush, DWORD style { HDC hMemDC; HBITMAP hBitmap, oldbitmap; + RECT rcClient; + + GetClientRect( hwnd, &rcClient ); + FillRect( hdc, &rcClient, hbrush ); if ((hBitmap = STATIC_GetImage( hwnd, IMAGE_BITMAP, style )) && (GetObjectType(hBitmap) == OBJ_BITMAP) && (hMemDC = CreateCompatibleDC( hdc ))) { BITMAP bm; - RECT rcClient; LOGBRUSH brush; BLENDFUNCTION blend = { AC_SRC_OVER, 0, 255, AC_SRC_ALPHA }; struct static_extra_info *extra = get_extra_ptr( hwnd, FALSE ); @@ -870,10 +873,8 @@ static void STATIC_PaintBitmapfn(HWND hwnd, HDC hdc, HBRUSH hbrush, DWORD style if (brush.lbStyle == BS_SOLID) SetBkColor(hdc, brush.lbColor); } - GetClientRect(hwnd, &rcClient); if (style & SS_CENTERIMAGE) { - FillRect( hdc, &rcClient, hbrush ); rcClient.left = (rcClient.right - rcClient.left)/2 - bm.bmWidth/2; rcClient.top = (rcClient.bottom - rcClient.top)/2 - bm.bmHeight/2; rcClient.right = rcClient.left + bm.bmWidth; diff --git a/dlls/comctl32/tests/static.c b/dlls/comctl32/tests/static.c index dbd71539a57..756fed86fe0 100644 --- a/dlls/comctl32/tests/static.c +++ b/dlls/comctl32/tests/static.c @@ -118,7 +118,6 @@ static void test_updates(int style) { HDC hdc = GetDC(hStatic); COLORREF colour = GetPixel(hdc, 10, 10); - todo_wine ok(colour == 0, "Unexpected pixel color.\n"); ReleaseDC(hStatic, hdc); } -- GitLab https://gitlab.winehq.org/wine/wine/-/merge_requests/229

June 15, 2022

[PATCH v2 0/1] MR229: comctl32: Always paint SS_BITMAP static control background with WM_CTLCOLORSTATIC brush.

by Jinoh Kang (＠iamahuman)

This is consistent with the behaviour of other static control styles. Signed-off-by: Jinoh Kang <jinoh.kang.kr(a)gmail.com> -- v2: comctl32: Always paint SS_BITMAP static control background with WM_CTLCOLORSTATIC brush. https://gitlab.winehq.org/wine/wine/-/merge_requests/229

June 15, 2022

[PATCH v2 1/1] shell32: Partially implement IShellItemImageFactory (icon only, no thumbnail).

by Jinoh Kang

From: Jinoh Kang <jinoh.kang.kr(a)gmail.com> Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=52673 Signed-off-by: Jinoh Kang <jinoh.kang.kr(a)gmail.com> --- dlls/shell32/Makefile.in | 2 +- dlls/shell32/shellitem.c | 160 ++++++++++++++++++++++++++++++++- dlls/shell32/tests/shlfolder.c | 1 - 3 files changed, 159 insertions(+), 4 deletions(-) diff --git a/dlls/shell32/Makefile.in b/dlls/shell32/Makefile.in index 9e2395126fc..95afa482de5 100644 --- a/dlls/shell32/Makefile.in +++ b/dlls/shell32/Makefile.in @@ -1,7 +1,7 @@ MODULE = shell32.dll IMPORTLIB = shell32 IMPORTS = uuid shlwapi user32 gdi32 advapi32 -DELAYIMPORTS = ole32 oleaut32 shdocvw version comctl32 gdiplus +DELAYIMPORTS = ole32 oleaut32 shdocvw version comctl32 gdiplus windowscodecs C_SRCS = \ appbar.c \ diff --git a/dlls/shell32/shellitem.c b/dlls/shell32/shellitem.c index 3e8bd24a429..e388b4d85e2 100644 --- a/dlls/shell32/shellitem.c +++ b/dlls/shell32/shellitem.c @@ -31,9 +31,13 @@ #include "pidl.h" #include "shell32_main.h" #include "debughlp.h" +#include "wincodec.h" +#include "commoncontrols.h" WINE_DEFAULT_DEBUG_CHANNEL(shell); +HRESULT WINAPI WICCreateImagingFactory_Proxy(UINT, IWICImagingFactory**); + typedef struct _ShellItem { IShellItem2 IShellItem2_iface; LONG ref; @@ -565,17 +569,169 @@ static ULONG WINAPI ShellItem_IShellItemImageFactory_Release(IShellItemImageFact return IShellItem2_Release(&This->IShellItem2_iface); } +static HICON get_icon_via_cache(LPCWSTR icon_file, INT source_index, DWORD flags, SIZE size) +{ + IImageList *icon_list = NULL; + int i, ic_index; + HICON icon; + HRESULT hr; + + if (source_index == -1) + source_index = 0; /* special case for some control panel applications */ + + for (i = 0; i <= SHIL_LAST; i++) + { + IImageList *cur_list; + int cx, cy; + + hr = SHGetImageList(i, &IID_IImageList, (void **)&cur_list); + if (SUCCEEDED(hr)) + { + hr = IImageList_GetIconSize(cur_list, &cx, &cy); + if (SUCCEEDED(hr) && cx == size.cx && cy == size.cy) + { + icon_list = cur_list; + break; + } + IImageList_Release(cur_list); + } + } + + if (!icon_list) return NULL; + + ic_index = SIC_GetIconIndex(icon_file, source_index, flags); + if (ic_index != -1) + { + hr = IImageList_GetIcon(icon_list, ic_index, ILD_NORMAL, &icon); + if (FAILED(hr)) icon = NULL; + } + IImageList_Release(icon_list); + + return icon; +} + +static HRESULT ShellItem_get_icon(ShellItem *This, SIZE size, HICON *icon) +{ + IExtractIconW *ei; + WCHAR icon_file[MAX_PATH]; + INT source_index; + UINT gil_out_flags; + HICON cached_icon; + HRESULT hr; + + hr = IShellItem2_BindToHandler(&This->IShellItem2_iface, NULL, &BHID_SFUIObject, + &IID_IExtractIconW, (void **)&ei); + if (FAILED(hr)) goto done; + + hr = IExtractIconW_GetIconLocation(ei, 0, icon_file, ARRAY_SIZE(icon_file), + &source_index, &gil_out_flags); + if (FAILED(hr)) goto free_ei; + + if (!(gil_out_flags & (GIL_NOTFILENAME | GIL_DONTCACHE)) && + (cached_icon = get_icon_via_cache(icon_file, source_index, 0, size))) + { + *icon = cached_icon; + hr = S_OK; + } + else + { + INT iconsize = min(size.cx, size.cy); + if (iconsize <= 0 || iconsize > 0x7fff) + iconsize = 0x7fff; + hr = IExtractIconW_Extract(ei, icon_file, source_index, icon, NULL, MAKELONG(0, iconsize)); + } + +free_ei: + IExtractIconW_Release(ei); +done: + return hr; +} + +static HRESULT convert_wicbitmapsource_to_gdi(IWICImagingFactory *imgfactory, + IWICBitmapSource *bitmapsource, HBITMAP *gdibitmap) +{ + IWICBitmapSource *newsrc; + BITMAPINFOHEADER bmi; + UINT width, height; + HBITMAP bm; + void *bits; + HRESULT hr; + + *gdibitmap = NULL; + + hr = WICConvertBitmapSource(&GUID_WICPixelFormat32bppBGRA, bitmapsource, &newsrc); + if (FAILED(hr)) goto done; + + hr = IWICBitmapSource_GetSize(newsrc, &width, &height); + if (FAILED(hr)) goto free_newsrc; + + memset(&bmi, 0, sizeof(bmi)); + bmi.biSize = sizeof(bmi); + bmi.biWidth = width; + bmi.biHeight = -height; + bmi.biPlanes = 1; + bmi.biBitCount = 32; + bmi.biCompression = BI_RGB; + + bm = CreateDIBSection(NULL, (const BITMAPINFO *)&bmi, DIB_RGB_COLORS, &bits, NULL, 0); + if (!bm) + { + WARN("Cannot create bitmap.\n"); + hr = E_FAIL; + goto free_newsrc; + } + + hr = IWICBitmapSource_CopyPixels(newsrc, NULL, width * 4, width * height * 4, bits); + if (FAILED(hr)) + { + DeleteObject(bm); + goto free_newsrc; + } + + hr = S_OK; + *gdibitmap = bm; + +free_newsrc: + IWICBitmapSource_Release(newsrc); +done: + return hr; +} + static HRESULT WINAPI ShellItem_IShellItemImageFactory_GetImage(IShellItemImageFactory *iface, SIZE size, SIIGBF flags, HBITMAP *phbm) { ShellItem *This = impl_from_IShellItemImageFactory(iface); + IWICImagingFactory *imgfactory; + IWICBitmap *bitmap = NULL; static int once; + HICON hicon; + HRESULT hr; if (!once++) - FIXME("%p ({%lu, %lu} %d %p): stub\n", This, size.cx, size.cy, flags, phbm); + FIXME("%p ({%lu, %lu} %d %p): partial stub\n", This, size.cx, size.cy, flags, phbm); *phbm = NULL; - return E_NOTIMPL; + + if (flags != SIIGBF_BIGGERSIZEOK) return E_NOTIMPL; + + hr = WICCreateImagingFactory_Proxy(WINCODEC_SDK_VERSION, &imgfactory); + if (FAILED(hr)) return hr; + + hr = ShellItem_get_icon(This, size, &hicon); + if (SUCCEEDED(hr)) + { + hr = IWICImagingFactory_CreateBitmapFromHICON(imgfactory, hicon, &bitmap); + DeleteObject(hicon); + + if (SUCCEEDED(hr)) + { + hr = convert_wicbitmapsource_to_gdi(imgfactory, (IWICBitmapSource *)bitmap, phbm); + IWICBitmap_Release(bitmap); + } + } + IWICImagingFactory_Release(imgfactory); + + return hr; } static const IShellItemImageFactoryVtbl ShellItem_IShellItemImageFactory_Vtbl = { diff --git a/dlls/shell32/tests/shlfolder.c b/dlls/shell32/tests/shlfolder.c index da606a9e707..49614aed3c4 100644 --- a/dlls/shell32/tests/shlfolder.c +++ b/dlls/shell32/tests/shlfolder.c @@ -4456,7 +4456,6 @@ static void test_IShellItemImageFactory(void) ret = IShellItemImageFactory_GetImage(siif, size, SIIGBF_BIGGERSIZEOK, &hbm); IShellItemImageFactory_Release(siif); - todo_wine ok(ret == S_OK, "GetImage returned %lx\n", ret); ok(FAILED(ret) == !hbm, "result = %lx but bitmap = %p\n", ret, hbm); -- GitLab https://gitlab.winehq.org/wine/wine/-/merge_requests/190

June 15, 2022

[PATCH v2 0/1] MR190: shell32: Partially implement IShellItemImageFactory (icon only, no thumbnail).

by Jinoh Kang (＠iamahuman)

Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=52673 Signed-off-by: Jinoh Kang <jinoh.kang.kr(a)gmail.com> -- v2: shell32: Partially implement IShellItemImageFactory (icon only, no thumbnail). https://gitlab.winehq.org/wine/wine/-/merge_requests/190

June 15, 2022

Re: Gitlab experiment wrapup

by Huw Davies

On Tue, Jun 14, 2022 at 10:01:23PM +0200, Alexandre Julliard wrote: > Another option would be to require the submitter to explicitly approve > their own MR if it contains a patch written by someone else; that would > be similar to a reviewer pushing fixups and then approving the MR. It > would require some mechanism to detect missing approvals though, so it > may not be worth the trouble. Could the (yet to be written) auto-add-reviewer script check whether the submitter and author are different and in that case auto add the submitter as a reviewer (as well as any other appropriate reviewers)? Then the approval check just becomes the same as for regular reviewers. Huw.

June 15, 2022

Re: Gitlab experiment wrapup

by Alexandre Julliard

Giovanni Mascellani <gmascellani(a)codeweavers.com> writes: > Hi, > > Il 14/06/22 22:01, Alexandre Julliard ha scritto: >> Signed-off-by is convenient because the git tools have direct support >> for it, but the real motivation for us was to keep track of who >> submitted the patch, to support sending a patch written by someone else. >> In that case the convention with email is to set From: to the original >> author, but the committer info is lost, hence the Signed-off-by. >> With merge requests, the information of who submitted the merge >> request >> is preserved independently of the commit contents, so the easiest option >> is to simply get rid of signoffs completely. > > If that was the only intended meaning of the SOBs, then yes, I would > say than now we have a better options and we could get rid of them. > > Right now the SOBs also help keeping a trace of who had a look at the > patch and said it was good, which might be an interesting information > to keep tracking. My point is that we shouldn't be using SOB for that, > because its standard meaning is another one. With Gitlab, this is already replaced by approvals. They are stored in the merge request, and in the "Approved-by" field in the git notes. -- Alexandre Julliard julliard(a)winehq.org

June 15, 2022

Re: [PATCH v2 0/1] MR248: winhttp: Release object on allocation failure. - closed

by Hans Leidekker (＠hans)

This merge request was closed by Hans Leidekker. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/248

June 15, 2022