December 2019 - Wine-Devel - list.winehq.org

[PATCH 1/5] avicap32: Define NONAMELESSUNION in files that include objidl.h.

by Zebediah Figura

Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=47710 Signed-off-by: Zebediah Figura <z.figura12(a)gmail.com> --- This is not currently necessary, but will be with the prescribed solution for bug 47710, viz. defining the type _STGMEDIUM_UNION. dlls/avicap32/avicap32_main.c | 1 + 1 file changed, 1 insertion(+) diff --git a/dlls/avicap32/avicap32_main.c b/dlls/avicap32/avicap32_main.c index 9e2a99d4c7c..0178dc87edb 100644 --- a/dlls/avicap32/avicap32_main.c +++ b/dlls/avicap32/avicap32_main.c @@ -43,6 +43,7 @@ # include <unistd.h> #endif +#define NONAMELESSUNION #include "windef.h" #include "winbase.h" #include "wingdi.h" -- 2.24.0

Dec. 17, 2019

Re: [PATCH v3 2/2] ntdll: Start find_free_area from address_space_start instead of 0.

by Rémi Bernon

Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=47974 --- The bug is also about some other issue, but these patches should fix the main crash cause in X Rebirth, as discussed in this bug.

Dec. 17, 2019

Re: [PATCH] rpcrt4: Fix ITypeInfo ref count handling.

by Jacek Caban

On 12/16/19 10:52 PM, Zebediah Figura wrote: > On 12/16/19 3:19 PM, Jacek Caban wrote: >> From: Kevin Puetz <PuetzKevinA(a)JohnDeere.com> >> >> Signed-off-by: Jacek Caban <jacek(a)codeweavers.com> >> --- >> dlls/rpcrt4/ndr_typelib.c | 23 +++++++++++++++++++---- >> 1 file changed, 19 insertions(+), 4 deletions(-) >> >> > Hello Jacek, > > This feels to me like it's making things even more awkward. I'm > presuming the problem is that get_iface_info() is releasing a passed-in > ITypeInfo, in which case I feel like standard COM practice would just be > to not do that. Does the attached patch work for you? Sure, I considered doing the same. Signed-off-by: Jacek Caban <jacek(a)codeweavers.com> Thanks, Jacek

Dec. 17, 2019

[PATCH 5/5] ole32/tests: Remove trace noise from moniker tests.

by Nikolay Sivov

Signed-off-by: Nikolay Sivov <nsivov(a)codeweavers.com> --- dlls/ole32/tests/moniker.c | 23 +++++++---------------- 1 file changed, 7 insertions(+), 16 deletions(-) diff --git a/dlls/ole32/tests/moniker.c b/dlls/ole32/tests/moniker.c index 984f3a753c..7b4bf5c0fa 100644 --- a/dlls/ole32/tests/moniker.c +++ b/dlls/ole32/tests/moniker.c @@ -41,14 +41,13 @@ #define CHECK_EXPECTED_METHOD(method_name) \ do { \ - trace("%s\n", method_name); \ - ok(*expected_method_list != NULL, "Extra method %s called\n", method_name); \ - if (*expected_method_list) \ - { \ - ok(!strcmp(*expected_method_list, method_name), "Expected %s to be called instead of %s\n", \ - *expected_method_list, method_name); \ - expected_method_list++; \ - } \ + ok(*expected_method_list != NULL, "Extra method %s called\n", method_name); \ + if (*expected_method_list) \ + { \ + ok(!strcmp(*expected_method_list, method_name), "Expected %s to be called instead of %s\n", \ + *expected_method_list, method_name); \ + expected_method_list++; \ + } \ } while(0) static char const * const *expected_method_list; @@ -116,8 +115,6 @@ static ULONG WINAPI ExternalConnection_Release(IExternalConnection *iface) static DWORD WINAPI ExternalConnection_AddConnection(IExternalConnection *iface, DWORD extconn, DWORD reserved) { - trace("add connection\n"); - ok(extconn == EXTCONN_STRONG, "extconn = %d\n", extconn); ok(!reserved, "reserved = %x\n", reserved); return ++external_connections; @@ -126,8 +123,6 @@ static DWORD WINAPI ExternalConnection_AddConnection(IExternalConnection *iface, static DWORD WINAPI ExternalConnection_ReleaseConnection(IExternalConnection *iface, DWORD extconn, DWORD reserved, BOOL fLastReleaseCloses) { - trace("release connection\n"); - ok(extconn == EXTCONN_STRONG, "extconn = %d\n", extconn); ok(!reserved, "reserved = %x\n", reserved); @@ -838,7 +833,6 @@ static int count_moniker_matches(IBindCtx * pbc, IEnumMoniker * spEM) CoTaskMemFree(szDisplayn); } } - trace("Total number of monikers is %i\n", monCnt); return matchCnt; } @@ -1071,7 +1065,6 @@ todo_wine ok(hr == S_OK, "Failed to get interface, hr %#x.\n", hr); matchCnt = count_moniker_matches(pbc, spEM1); - trace("Number of matches is %i\n", matchCnt); grflags= grflags | ROTFLAGS_REGISTRATIONKEEPSALIVE; hr = IRunningObjectTable_Register(pprot, grflags, lpEM1, pmk1, &pdwReg1); @@ -1088,12 +1081,10 @@ todo_wine matchCnt = count_moniker_matches(pbc, spEM2); ok(matchCnt==2, "Number of matches should be equal to 2 not %i\n", matchCnt); - trace("IEnumMoniker::Clone\n"); IEnumMoniker_Clone(spEM2, &spEM3); matchCnt = count_moniker_matches(pbc, spEM3); ok(matchCnt==0, "Number of matches should be equal to 0 not %i\n", matchCnt); - trace("IEnumMoniker::Reset\n"); IEnumMoniker_Reset(spEM3); matchCnt = count_moniker_matches(pbc, spEM3); -- 2.24.0

Dec. 17, 2019

[PATCH 4/5] ole32/tests: Add some tests for combined antimonikers.

by Nikolay Sivov

Signed-off-by: Nikolay Sivov <nsivov(a)codeweavers.com> --- dlls/ole32/tests/moniker.c | 104 ++++++++++++++++++++++++++++++++++++- 1 file changed, 102 insertions(+), 2 deletions(-) diff --git a/dlls/ole32/tests/moniker.c b/dlls/ole32/tests/moniker.c index 51fad3ae2f..984f3a753c 100644 --- a/dlls/ole32/tests/moniker.c +++ b/dlls/ole32/tests/moniker.c @@ -1933,10 +1933,26 @@ todo_wine } } +static void stream_write_dword(IStream *stream, DWORD value) +{ + LARGE_INTEGER pos; + HRESULT hr; + + pos.QuadPart = 0; + hr = IStream_Seek(stream, pos, STREAM_SEEK_SET, NULL); + ok(hr == S_OK, "Failed to seek, hr %#x.\n", hr); + + hr = IStream_Write(stream, &value, sizeof(value), NULL); + ok(hr == S_OK, "Stream write failed, hr %#x.\n", hr); + + hr = IStream_Seek(stream, pos, STREAM_SEEK_SET, NULL); + ok(hr == S_OK, "Failed to seek, hr %#x.\n", hr); +} + static void test_anti_moniker(void) { + IMoniker *moniker, *moniker2, *moniker3; HRESULT hr; - IMoniker *moniker; DWORD moniker_type; DWORD hash; IBindCtx *bindctx; @@ -1944,6 +1960,8 @@ static void test_anti_moniker(void) IMoniker *inverse; IUnknown *unknown; static const WCHAR expected_display_name[] = { '\\','.','.',0 }; + IStream *stream; + WCHAR *name; hr = CreateAntiMoniker(&moniker); ok_ole_success(hr, CreateAntiMoniker); @@ -1997,8 +2015,90 @@ todo_wine hr = IMoniker_BindToStorage(moniker, bindctx, NULL, &IID_IUnknown, (void **)&unknown); ok(hr == E_NOTIMPL, "IMoniker_BindToStorage should return E_NOTIMPL, not 0x%08x\n", hr); - IBindCtx_Release(bindctx); + /* ComposeWith */ + hr = CreateAntiMoniker(&moniker2); + ok(hr == S_OK, "Failed to create moniker, hr %#x.\n", hr); + + moniker3 = moniker; + hr = IMoniker_ComposeWith(moniker, moniker2, TRUE, &moniker3); + ok(hr == MK_E_NEEDGENERIC, "Unexpected hr %#x.\n", hr); + ok(!moniker3, "Unexpected interface.\n"); + + hr = IMoniker_ComposeWith(moniker, moniker2, FALSE, &moniker3); + ok(hr == S_OK, "Failed to compose, hr %#x.\n", hr); + hr = IMoniker_IsSystemMoniker(moniker3, &moniker_type); + ok(hr == S_OK, "Failed to get moniker type, hr %#x.\n", hr); + ok(moniker_type == MKSYS_GENERICCOMPOSITE, "Unexpected moniker type %d.\n", moniker_type); + IMoniker_Release(moniker3); + + IMoniker_Release(moniker2); + + /* Load with composed number > 1. */ + hr = CreateStreamOnHGlobal(NULL, TRUE, &stream); + ok(hr == S_OK, "Failed to create a stream, hr %#x.\n", hr); + + stream_write_dword(stream, 2); + + hr = IMoniker_Load(moniker, stream); +todo_wine + ok(hr == S_OK, "Unexpected hr %#x.\n", hr); + + hr = IMoniker_Hash(moniker, &hash); + ok(hr == S_OK, "Failed to get hash value, hr %#x.\n", hr); +todo_wine + ok(hash == 0x80000002, "Unexpected hash value %#x.\n", hash); + /* Display name reflects anti combination. */ + hr = IMoniker_GetDisplayName(moniker, bindctx, NULL, &name); + ok(hr == S_OK, "Failed to get display name, hr %#x.\n", hr); +todo_wine + ok(!lstrcmpW(name, L"\\..\\.."), "Unexpected display name %s.\n", wine_dbgstr_w(name)); + CoTaskMemFree(name); + + /* Limit is at 0xfffff. */ + stream_write_dword(stream, 0xfffff); + + hr = IMoniker_Load(moniker, stream); +todo_wine + ok(hr == S_OK, "Unexpected hr %#x.\n", hr); + + hr = IMoniker_Hash(moniker, &hash); + ok(hr == S_OK, "Failed to get hash value, hr %#x.\n", hr); +todo_wine + ok(hash == 0x800fffff, "Unexpected hash value %#x.\n", hash); + + stream_write_dword(stream, 0xfffff + 1); + + hr = IMoniker_Load(moniker, stream); +todo_wine + ok(hr == E_INVALIDARG, "Unexpected hr %#x.\n", hr); + + hr = IMoniker_Hash(moniker, &hash); + ok(hr == S_OK, "Failed to get hash value, hr %#x.\n", hr); +todo_wine + ok(hash == 0x800fffff, "Unexpected hash value %#x.\n", hash); + + /* Zero combining counter is also valid. */ + stream_write_dword(stream, 0); + + hr = IMoniker_Load(moniker, stream); +todo_wine + ok(hr == S_OK, "Unexpected hr %#x.\n", hr); + + hr = IMoniker_Hash(moniker, &hash); + ok(hr == S_OK, "Failed to get hash value, hr %#x.\n", hr); +todo_wine + ok(hash == 0x80000000, "Unexpected hash value %#x.\n", hash); + + hr = IMoniker_GetDisplayName(moniker, bindctx, NULL, &name); + ok(hr == S_OK, "Failed to get display name, hr %#x.\n", hr); +todo_wine + ok(!lstrcmpW(name, L""), "Unexpected display name %s.\n", wine_dbgstr_w(name)); + CoTaskMemFree(name); + + IStream_Release(stream); + + IBindCtx_Release(bindctx); IMoniker_Release(moniker); } -- 2.24.0

Dec. 17, 2019

[PATCH 3/5] ole32/tests: Test inverse moniker type.

by Nikolay Sivov

Signed-off-by: Nikolay Sivov <nsivov(a)codeweavers.com> --- dlls/ole32/tests/moniker.c | 30 +++++++++++++++++++++++++----- 1 file changed, 25 insertions(+), 5 deletions(-) diff --git a/dlls/ole32/tests/moniker.c b/dlls/ole32/tests/moniker.c index a275671a69..51fad3ae2f 100644 --- a/dlls/ole32/tests/moniker.c +++ b/dlls/ole32/tests/moniker.c @@ -1609,7 +1609,10 @@ todo_wine IBindCtx_Release(bindctx); hr = IMoniker_Inverse(moniker, &inverse); - ok_ole_success(hr, IMoniker_Inverse); + ok(hr == S_OK, "Failed to get inverse, hr %#x.\n", hr); + hr = IMoniker_IsSystemMoniker(inverse, &moniker_type); + ok(hr == S_OK, "Failed to get moniker type, hr %#x.\n", hr); + ok(moniker_type == MKSYS_ANTIMONIKER, "Unexpected moniker type %d.\n", moniker_type); IMoniker_Release(inverse); IMoniker_Release(moniker); @@ -1617,8 +1620,9 @@ todo_wine static void test_file_moniker(WCHAR* path) { + IMoniker *moniker1 = NULL, *moniker2 = NULL, *inverse; + DWORD moniker_type; IStream *stream; - IMoniker *moniker1 = NULL, *moniker2 = NULL; IUnknown *unk; HRESULT hr; @@ -1634,6 +1638,13 @@ todo_wine IUnknown_Release(unk); } + hr = IMoniker_Inverse(moniker1, &inverse); + ok(hr == S_OK, "Failed to get inverse, hr %#x.\n", hr); + hr = IMoniker_IsSystemMoniker(inverse, &moniker_type); + ok(hr == S_OK, "Failed to get moniker type, hr %#x.\n", hr); + ok(moniker_type == MKSYS_ANTIMONIKER, "Unexpected moniker type %d.\n", moniker_type); + IMoniker_Release(inverse); + hr = CreateStreamOnHGlobal(NULL, TRUE, &stream); ok_ole_success(hr, CreateStreamOnHGlobal); @@ -1892,7 +1903,10 @@ todo_wine IBindCtx_Release(bindctx); hr = IMoniker_Inverse(moniker, &inverse); - ok_ole_success(hr, IMoniker_Inverse); + ok(hr == S_OK, "Failed to get inverse, hr %#x.\n", hr); + hr = IMoniker_IsSystemMoniker(inverse, &moniker_type); + ok(hr == S_OK, "Failed to get moniker type, hr %#x.\n", hr); + ok(moniker_type == MKSYS_ANTIMONIKER, "Unexpected moniker type %d.\n", moniker_type); IMoniker_Release(inverse); IMoniker_Release(moniker); @@ -2068,7 +2082,10 @@ todo_wine IBindCtx_Release(bindctx); hr = IMoniker_Inverse(moniker, &inverse); - ok_ole_success(hr, IMoniker_Inverse); + ok(hr == S_OK, "Failed to get inverse, hr %#x.\n", hr); + hr = IMoniker_IsSystemMoniker(inverse, &moniker_type); + ok(hr == S_OK, "Failed to get moniker type, hr %#x.\n", hr); + ok(moniker_type == MKSYS_GENERICCOMPOSITE, "Unexpected moniker type %d.\n", moniker_type); IMoniker_Release(inverse); IMoniker_Release(moniker); @@ -2150,7 +2167,10 @@ todo_wine moniker_type); hr = IMoniker_Inverse(moniker, &inverse); - ok_ole_success(hr, IMoniker_Inverse); + ok(hr == S_OK, "Failed to get inverse, hr %#x.\n", hr); + hr = IMoniker_IsSystemMoniker(inverse, &moniker_type); + ok(hr == S_OK, "Failed to get moniker type, hr %#x.\n", hr); + ok(moniker_type == MKSYS_ANTIMONIKER, "Unexpected moniker type %d.\n", moniker_type); IMoniker_Release(inverse); hr = CreateBindCtx(0, &bindctx); -- 2.24.0

Dec. 17, 2019

[PATCH 2/5] ole32/tests: Add some tests for item moniker equality.

by Nikolay Sivov

Signed-off-by: Nikolay Sivov <nsivov(a)codeweavers.com> --- dlls/ole32/tests/moniker.c | 38 ++++++++++++++++++++++++++++++++++++++ 1 file changed, 38 insertions(+) diff --git a/dlls/ole32/tests/moniker.c b/dlls/ole32/tests/moniker.c index 4fec233a57..a275671a69 100644 --- a/dlls/ole32/tests/moniker.c +++ b/dlls/ole32/tests/moniker.c @@ -1728,6 +1728,23 @@ static void test_item_moniker(void) { item_moniker_unicode_item_stream, sizeof(item_moniker_unicode_item_stream), L"!B" }, { item_moniker_unicode_delim_item_stream, sizeof(item_moniker_unicode_delim_item_stream), L"!C" }, }; + static const struct + { + const WCHAR *delim1; + const WCHAR *item1; + const WCHAR *delim2; + const WCHAR *item2; + HRESULT hr; + } isequal_tests[] = + { + { L"!", L"Item1", L"!", L"ITEM1", S_OK }, + { NULL, L"Item1", L"!", L"ITEM1", S_OK }, + { L"", L"Item1", L"!", L"ITEM1", S_OK }, + { L"&", L"Item1", L"!", L"ITEM1", S_OK }, + {L"&&", L"Item1", L"&", L"&Item1", S_FALSE }, + { NULL, L"Item1", NULL, L"Item2", S_FALSE }, + { NULL, L"Item1", NULL, L"ITEM1", S_OK }, + }; IMoniker *moniker, *moniker2; HRESULT hr; DWORD moniker_type, i; @@ -1879,6 +1896,27 @@ todo_wine IMoniker_Release(inverse); IMoniker_Release(moniker); + + /* IsEqual */ + for (i = 0; i < ARRAY_SIZE(isequal_tests); ++i) + { + hr = CreateItemMoniker(isequal_tests[i].delim1, isequal_tests[i].item1, &moniker); + ok(hr == S_OK, "Failed to create moniker, hr %#x.\n", hr); + + hr = CreateItemMoniker(isequal_tests[i].delim2, isequal_tests[i].item2, &moniker2); + ok(hr == S_OK, "Failed to create moniker, hr %#x.\n", hr); + + hr = IMoniker_IsEqual(moniker, moniker2); + todo_wine_if(i == 4 || i == 5) + ok(hr == isequal_tests[i].hr, "%d: unexpected result %#x.\n", i, hr); + + hr = IMoniker_IsEqual(moniker2, moniker); + todo_wine_if(i == 4 || i == 5) + ok(hr == isequal_tests[i].hr, "%d: unexpected result %#x.\n", i, hr); + + IMoniker_Release(moniker); + IMoniker_Release(moniker2); + } } static void test_anti_moniker(void) -- 2.24.0

Dec. 17, 2019

[PATCH 1/5] ole32/tests: Add tests for system moniker identification queries.

by Nikolay Sivov

Signed-off-by: Nikolay Sivov <nsivov(a)codeweavers.com> --- dlls/ole32/tests/moniker.c | 64 ++++++++++++++++++++++++++++++++++++-- 1 file changed, 61 insertions(+), 3 deletions(-) diff --git a/dlls/ole32/tests/moniker.c b/dlls/ole32/tests/moniker.c index d911ce57e9..4fec233a57 100644 --- a/dlls/ole32/tests/moniker.c +++ b/dlls/ole32/tests/moniker.c @@ -31,6 +31,7 @@ #include "ocidl.h" #include "comcat.h" #include "olectl.h" +#include "initguid.h" #include "wine/test.h" @@ -62,6 +63,13 @@ static const CLSID CLSID_TestMoniker = {0xb9, 0x3e, 0x2f, 0xf9, 0xc8, 0x32, 0x23, 0xd7} }; +DEFINE_OLEGUID(CLSID_FileMoniker, 0x303, 0, 0); +DEFINE_OLEGUID(CLSID_ItemMoniker, 0x304, 0, 0); +DEFINE_OLEGUID(CLSID_AntiMoniker, 0x305, 0, 0); +DEFINE_OLEGUID(CLSID_CompositeMoniker, 0x309, 0, 0); +DEFINE_OLEGUID(CLSID_ClassMoniker, 0x31a, 0, 0); +DEFINE_OLEGUID(CLSID_PointerMoniker, 0x306, 0, 0); + static LONG cLocks; static void LockModule(void) @@ -1543,7 +1551,15 @@ static void test_class_moniker(void) hr = CreateClassMoniker(&CLSID_StdComponentCategoriesMgr, &moniker); ok_ole_success(hr, CreateClassMoniker); - if (!moniker) return; + + hr = IMoniker_QueryInterface(moniker, &CLSID_ClassMoniker, (void **)&unknown); +todo_wine + ok(hr == S_OK, "Unexpected hr %#x.\n", hr); + if (SUCCEEDED(hr)) + { + ok(unknown == (IUnknown *)moniker, "Unexpected interface.\n"); + IUnknown_Release(unknown); + } test_moniker("class moniker", moniker, expected_class_moniker_marshal_data, sizeof(expected_class_moniker_marshal_data), @@ -1603,11 +1619,21 @@ static void test_file_moniker(WCHAR* path) { IStream *stream; IMoniker *moniker1 = NULL, *moniker2 = NULL; + IUnknown *unk; HRESULT hr; hr = CreateFileMoniker(path, &moniker1); ok_ole_success(hr, CreateFileMoniker); + hr = IMoniker_QueryInterface(moniker1, &CLSID_FileMoniker, (void **)&unk); +todo_wine + ok(hr == S_OK, "Unexpected hr %#x.\n", hr); + if (SUCCEEDED(hr)) + { + ok(unk == (IUnknown *)moniker1, "Unexpected interface.\n"); + IUnknown_Release(unk); + } + hr = CreateStreamOnHGlobal(NULL, TRUE, &stream); ok_ole_success(hr, CreateStreamOnHGlobal); @@ -1719,6 +1745,15 @@ static void test_item_moniker(void) hr = CreateItemMoniker(NULL, wszObjectName, &moniker); ok(hr == S_OK, "Failed to create item moniker, hr %#x.\n", hr); + hr = IMoniker_QueryInterface(moniker, &CLSID_ItemMoniker, (void **)&unknown); +todo_wine + ok(hr == S_OK, "Unexpected hr %#x.\n", hr); + if (SUCCEEDED(hr)) + { + ok(unknown == (IUnknown *)moniker, "Unexpected interface.\n"); + IUnknown_Release(unknown); + } + test_moniker("item moniker 2", moniker, expected_item_moniker_marshal_data2, sizeof(expected_item_moniker_marshal_data2), expected_item_moniker_saved_data2, sizeof(expected_item_moniker_saved_data2), @@ -1860,7 +1895,15 @@ static void test_anti_moniker(void) hr = CreateAntiMoniker(&moniker); ok_ole_success(hr, CreateAntiMoniker); - if (!moniker) return; + + hr = IMoniker_QueryInterface(moniker, &CLSID_AntiMoniker, (void **)&unknown); +todo_wine + ok(hr == S_OK, "Unexpected hr %#x.\n", hr); + if (SUCCEEDED(hr)) + { + ok(unknown == (IUnknown *)moniker, "Unexpected interface.\n"); + IUnknown_Release(unknown); + } test_moniker("anti moniker", moniker, expected_anti_moniker_marshal_data, sizeof(expected_anti_moniker_marshal_data), @@ -1932,6 +1975,13 @@ static void test_generic_composite_moniker(void) hr = CreateGenericComposite(moniker1, moniker2, &moniker); ok_ole_success(hr, CreateGenericComposite); + /* Generic composite is special, as it does not addref in this case. */ + hr = IMoniker_QueryInterface(moniker, &CLSID_CompositeMoniker, (void **)&unknown); +todo_wine + ok(hr == S_OK, "Unexpected hr %#x.\n", hr); + if (SUCCEEDED(hr)) + ok(unknown == (IUnknown *)moniker, "Unexpected interface.\n"); + test_moniker("generic composite moniker", moniker, expected_gc_moniker_marshal_data, sizeof(expected_gc_moniker_marshal_data), expected_gc_moniker_saved_data, sizeof(expected_gc_moniker_saved_data), @@ -2007,7 +2057,15 @@ static void test_pointer_moniker(void) hr = CreatePointerMoniker((IUnknown *)&Test_ClassFactory, &moniker); ok_ole_success(hr, CreatePointerMoniker); - if (!moniker) return; + + hr = IMoniker_QueryInterface(moniker, &CLSID_PointerMoniker, (void **)&unknown); +todo_wine + ok(hr == S_OK, "Unexpected hr %#x.\n", hr); + if (SUCCEEDED(hr)) + { + ok(unknown == (IUnknown *)moniker, "Unexpected interface.\n"); + IUnknown_Release(unknown); + } ok_more_than_one_lock(); -- 2.24.0

Dec. 17, 2019

[PATCH v2 2/2] winemac.drv: Support VK_EXT_metal_surface.

by Chip Davis

Prefer it to VK_MVK_macos_surface when present. MoltenVK has deprecated VK_MVK_macos_surface in favor of VK_EXT_metal_surface. It's likely that the former extension will vanish at some point. Signed-off-by: Chip Davis <cdavis(a)codeweavers.com> --- Notes: v2: Simplify. Use memmove(3) to avoid UB. Fix off-by-one error. Update comment. dlls/winemac.drv/vulkan.c | 62 +++++++++++++++++++++++++++++++-------- 1 file changed, 50 insertions(+), 12 deletions(-) diff --git a/dlls/winemac.drv/vulkan.c b/dlls/winemac.drv/vulkan.c index 21e93827c156..360dd02f7b8b 100644 --- a/dlls/winemac.drv/vulkan.c +++ b/dlls/winemac.drv/vulkan.c @@ -50,6 +50,9 @@ WINE_DECLARE_DEBUG_CHANNEL(fps); typedef VkFlags VkMacOSSurfaceCreateFlagsMVK; #define VK_STRUCTURE_TYPE_MACOS_SURFACE_CREATE_INFO_MVK 1000123000 +typedef VkFlags VkMetalSurfaceCreateFlagsEXT; +#define VK_STRUCTURE_TYPE_METAL_SURFACE_CREATE_INFO_EXT 1000217000 + struct wine_vk_surface { macdrv_metal_device device; @@ -65,9 +68,18 @@ typedef struct VkMacOSSurfaceCreateInfoMVK const void *pView; /* NSView */ } VkMacOSSurfaceCreateInfoMVK; +typedef struct VkMetalSurfaceCreateInfoEXT +{ + VkStructureType sType; + const void *pNext; + VkMetalSurfaceCreateFlagsEXT flags; + const void *pLayer; /* CAMetalLayer */ +} VkMetalSurfaceCreateInfoEXT; + static VkResult (*pvkCreateInstance)(const VkInstanceCreateInfo *, const VkAllocationCallbacks *, VkInstance *); static VkResult (*pvkCreateSwapchainKHR)(VkDevice, const VkSwapchainCreateInfoKHR *, const VkAllocationCallbacks *, VkSwapchainKHR *); static VkResult (*pvkCreateMacOSSurfaceMVK)(VkInstance, const VkMacOSSurfaceCreateInfoMVK*, const VkAllocationCallbacks *, VkSurfaceKHR *); +static VkResult (*pvkCreateMetalSurfaceEXT)(VkInstance, const VkMetalSurfaceCreateInfoEXT*, const VkAllocationCallbacks *, VkSurfaceKHR *); static void (*pvkDestroyInstance)(VkInstance, const VkAllocationCallbacks *); static void (*pvkDestroySurfaceKHR)(VkInstance, VkSurfaceKHR, const VkAllocationCallbacks *); static void (*pvkDestroySwapchainKHR)(VkDevice, VkSwapchainKHR, const VkAllocationCallbacks *); @@ -103,6 +115,7 @@ static BOOL WINAPI wine_vk_init(INIT_ONCE *once, void *param, void **context) LOAD_FUNCPTR(vkCreateInstance) LOAD_FUNCPTR(vkCreateSwapchainKHR) LOAD_FUNCPTR(vkCreateMacOSSurfaceMVK) + LOAD_FUNCPTR(vkCreateMetalSurfaceEXT) LOAD_FUNCPTR(vkDestroyInstance) LOAD_FUNCPTR(vkDestroySurfaceKHR) LOAD_FUNCPTR(vkDestroySwapchainKHR) @@ -159,7 +172,7 @@ static VkResult wine_vk_instance_convert_create_info(const VkInstanceCreateInfo */ if (!strcmp(src->ppEnabledExtensionNames[i], "VK_KHR_win32_surface")) { - enabled_extensions[i] = "VK_MVK_macos_surface"; + enabled_extensions[i] = pvkCreateMetalSurfaceEXT ? "VK_EXT_metal_surface" : "VK_MVK_macos_surface"; } else { @@ -239,7 +252,6 @@ static VkResult macdrv_vkCreateWin32SurfaceKHR(VkInstance instance, const VkAllocationCallbacks *allocator, VkSurfaceKHR *surface) { VkResult res; - VkMacOSSurfaceCreateInfoMVK create_info_host; struct wine_vk_surface *mac_surface; struct macdrv_win_data *data; @@ -279,12 +291,26 @@ static VkResult macdrv_vkCreateWin32SurfaceKHR(VkInstance instance, goto err; } - create_info_host.sType = VK_STRUCTURE_TYPE_MACOS_SURFACE_CREATE_INFO_MVK; - create_info_host.pNext = NULL; - create_info_host.flags = 0; /* reserved */ - create_info_host.pView = macdrv_view_get_metal_layer(mac_surface->view); + if (pvkCreateMetalSurfaceEXT) + { + VkMetalSurfaceCreateInfoEXT create_info_host; + create_info_host.sType = VK_STRUCTURE_TYPE_METAL_SURFACE_CREATE_INFO_EXT; + create_info_host.pNext = NULL; + create_info_host.flags = 0; /* reserved */ + create_info_host.pLayer = macdrv_view_get_metal_layer(mac_surface->view); + + res = pvkCreateMetalSurfaceEXT(instance, &create_info_host, NULL /* allocator */, &mac_surface->surface); + } + else + { + VkMacOSSurfaceCreateInfoMVK create_info_host; + create_info_host.sType = VK_STRUCTURE_TYPE_MACOS_SURFACE_CREATE_INFO_MVK; + create_info_host.pNext = NULL; + create_info_host.flags = 0; /* reserved */ + create_info_host.pView = macdrv_view_get_metal_layer(mac_surface->view); - res = pvkCreateMacOSSurfaceMVK(instance, &create_info_host, NULL /* allocator */, &mac_surface->surface); + res = pvkCreateMacOSSurfaceMVK(instance, &create_info_host, NULL /* allocator */, &mac_surface->surface); + } if (res != VK_SUCCESS) { ERR("Failed to create MoltenVK surface, res=%d\n", res); @@ -342,6 +368,7 @@ static VkResult macdrv_vkEnumerateInstanceExtensionProperties(const char *layer_ uint32_t *count, VkExtensionProperties* properties) { unsigned int i; + BOOL seen_surface = FALSE; VkResult res; TRACE("layer_name %s, count %p, properties %p\n", debugstr_a(layer_name), count, properties); @@ -353,8 +380,9 @@ static VkResult macdrv_vkEnumerateInstanceExtensionProperties(const char *layer_ return VK_ERROR_LAYER_NOT_PRESENT; } - /* We will return the same number of instance extensions reported by the host back to - * winevulkan. Along the way we may replace MoltenVK extensions with their win32 equivalents. + /* We will return at most the same number of instance extensions reported by the host back to + * winevulkan. Along the way we may replace MoltenVK extensions with their win32 equivalents, + * or remove redundant extensions outright. * Winevulkan will perform more detailed filtering as it knows whether it has thunks * for a particular extension. */ @@ -364,14 +392,24 @@ static VkResult macdrv_vkEnumerateInstanceExtensionProperties(const char *layer_ for (i = 0; i < *count; i++) { - /* For now the only MoltenVK extension we need to fixup. Long-term we may need an array. */ - if (!strcmp(properties[i].extensionName, "VK_MVK_macos_surface")) + /* For now the only MoltenVK extensions we need to fixup. Long-term we may need an array. */ + if (!strcmp(properties[i].extensionName, "VK_MVK_macos_surface") || + !strcmp(properties[i].extensionName, "VK_EXT_metal_surface")) { - TRACE("Substituting VK_MVK_macos_surface for VK_KHR_win32_surface\n"); + if (seen_surface) + { + /* If we've already seen a surface extension, just hide this one. */ + memmove(properties + i, properties + i + 1, (*count - i - 1) * sizeof(*properties)); + --*count; + --i; + continue; + } + TRACE("Substituting %s for VK_KHR_win32_surface\n", properties[i].extensionName); snprintf(properties[i].extensionName, sizeof(properties[i].extensionName), VK_KHR_WIN32_SURFACE_EXTENSION_NAME); properties[i].specVersion = VK_KHR_WIN32_SURFACE_SPEC_VERSION; + seen_surface = TRUE; } } -- 2.24.0

Dec. 17, 2019

[PATCH v2 1/2] winemac.drv: Pass the CAMetalLayer to MoltenVK.

by Chip Davis

It isn't safe to access the view object from any thread other than the main thread. In fact, if you try to call vkCreateMacOSSurfaceMVK() from any other thread, MoltenVK prints out a big, scary warning telling you not to do this! Instead, get the layer from the view ourselves and pass that to MoltenVK. Recent versions of MoltenVK can accept either the view or the layer. Signed-off-by: Chip Davis <cdavis(a)codeweavers.com> --- Notes: v2: Don't check for the main thread. dlls/winemac.drv/cocoa_window.m | 12 ++++++++++++ dlls/winemac.drv/macdrv_cocoa.h | 2 ++ dlls/winemac.drv/vulkan.c | 2 +- 3 files changed, 15 insertions(+), 1 deletion(-) diff --git a/dlls/winemac.drv/cocoa_window.m b/dlls/winemac.drv/cocoa_window.m index 877653ea0078..edbfab518e3c 100644 --- a/dlls/winemac.drv/cocoa_window.m +++ b/dlls/winemac.drv/cocoa_window.m @@ -3796,6 +3796,18 @@ macdrv_metal_view macdrv_view_create_metal_view(macdrv_view v, macdrv_metal_devi return (macdrv_metal_view)metalView; } +macdrv_metal_layer macdrv_view_get_metal_layer(macdrv_metal_view v) +{ + WineMetalView* view = (WineMetalView*)v; + __block CAMetalLayer* layer; + + OnMainThread(^{ + layer = (CAMetalLayer*)view.layer; + }); + + return (macdrv_metal_layer)layer; +} + void macdrv_view_release_metal_view(macdrv_metal_view v) { WineMetalView* view = (WineMetalView*)v; diff --git a/dlls/winemac.drv/macdrv_cocoa.h b/dlls/winemac.drv/macdrv_cocoa.h index 676adb435bb3..b02ad79f0259 100644 --- a/dlls/winemac.drv/macdrv_cocoa.h +++ b/dlls/winemac.drv/macdrv_cocoa.h @@ -140,6 +140,7 @@ #ifdef HAVE_METAL_METAL_H typedef struct macdrv_opaque_metal_device* macdrv_metal_device; typedef struct macdrv_opaque_metal_view* macdrv_metal_view; +typedef struct macdrv_opaque_metal_layer* macdrv_metal_layer; #endif typedef struct macdrv_opaque_status_item* macdrv_status_item; struct macdrv_event; @@ -587,6 +588,7 @@ extern void macdrv_set_window_color_key(macdrv_window w, CGFloat keyRed, CGFloat extern macdrv_metal_device macdrv_create_metal_device(void) DECLSPEC_HIDDEN; extern void macdrv_release_metal_device(macdrv_metal_device d) DECLSPEC_HIDDEN; extern macdrv_metal_view macdrv_view_create_metal_view(macdrv_view v, macdrv_metal_device d) DECLSPEC_HIDDEN; +extern macdrv_metal_layer macdrv_view_get_metal_layer(macdrv_metal_view v) DECLSPEC_HIDDEN; extern void macdrv_view_release_metal_view(macdrv_metal_view v) DECLSPEC_HIDDEN; #endif extern int macdrv_get_view_backing_size(macdrv_view v, int backing_size[2]) DECLSPEC_HIDDEN; diff --git a/dlls/winemac.drv/vulkan.c b/dlls/winemac.drv/vulkan.c index cc7f22f67880..21e93827c156 100644 --- a/dlls/winemac.drv/vulkan.c +++ b/dlls/winemac.drv/vulkan.c @@ -282,7 +282,7 @@ static VkResult macdrv_vkCreateWin32SurfaceKHR(VkInstance instance, create_info_host.sType = VK_STRUCTURE_TYPE_MACOS_SURFACE_CREATE_INFO_MVK; create_info_host.pNext = NULL; create_info_host.flags = 0; /* reserved */ - create_info_host.pView = mac_surface->view; + create_info_host.pView = macdrv_view_get_metal_layer(mac_surface->view); res = pvkCreateMacOSSurfaceMVK(instance, &create_info_host, NULL /* allocator */, &mac_surface->surface); if (res != VK_SUCCESS) -- 2.24.0

Dec. 17, 2019

Re: [wine-devel] Bugs To Fix?

by Thales

Thanks for the tip, Alan! ...John Alway On Monday, December 16, 2019, 06:40:17 PM CST, Alan W. Irwin <alan.w.irwin1234(a)gmail.com> wrote: On 2019-12-16 23:03-0000 Thales wrote: > Hello, > What are some of the most important bugs that need fixing? I would like to take a crack at one or more. I've been looking at Bugzilla and there's quite a bit there to sift through. > > Thanks for any feedback! > ...John Alway Hi John: My advice is work on a Wine bug that is blocking something of interest to you since that interest will provide important focus and motivation. For example, if you are interested in free software distributions for Windows like Cygwin or MSYS2, you might want to work on one or more bugs that are blocking those from working on Wine **after** reading a summary of the current situation at <https://github.com/msys2/MSYS2-packages/issues/682> (particularly the remarks starting at "<corinna> the old method up to XP/2K3 is not available anymore" .) Note, I prefer to stick to my own free software development rather than work on Wine, but I do follow the Cygwin and MSYS2 situation because as a Linux user I would like to test my free software using the free software libraries in those distributions for the Windows platform without paying a Microsoft tax. And there are others with similar interests as well such as "fracting" (Qian Hong) who got MSYS2 to work reasonably well under Wine-staging before the post-XP change to MSYS2 that he refers to in that discussion. However, there is obviously a huge range of software available on Windows besides the software (e.g., the installers) supporting those two software distributions so if your interest is in something else (e.g., some game) you should follow your interest rather than mine! Alan __________________________ Alan W. Irwin Programming affiliations with the FreeEOS equation-of-state implementation for stellar interiors (freeeos.sf.net) the Time Ephemerides project (timeephem.sf.net) PLplot scientific plotting software package (plplot.org) the libLASi project (unifont.org/lasi) the Loads of Linux Links project (loll.sf.net) and the Linux Brochure Project (lbproject.sf.net) __________________________ Linux-powered Science __________________________

Dec. 17, 2019

Re: Bugs To Fix?

by Thales

Thanks, Stefan! Sounds like there is wide latitude. ...John On Monday, December 16, 2019, 05:47:42 PM CST, Stefan Dösinger <stefandoesinger(a)gmail.com> wrote: Hi, There are plenty of bugs, we have enough to keep everyone busy :-) . Other than "it would be great to have this bug fixed", other things to keep in mind is how difficult the bug is to fix, how much investigative effort has been put into it and what your area of interest is. Stefan Am 17.12.19 um 00:03 schrieb Thales: > Hello, > > What are some of the most important bugs that need fixing? I would > like to take a crack at one or more. I've been looking at Bugzilla and > there's quite a bit there to sift through. > > > Thanks for any feedback! > > ...John Alway

Dec. 17, 2019

[PATCH] oledb32/tests: Add PromptNew interactive test.

by Alistair Leslie-Hughes

Signed-off-by: Alistair Leslie-Hughes <leslie_alistair(a)hotmail.com> --- dlls/oledb32/tests/database.c | 20 ++++++++++++++++++++ 1 file changed, 20 insertions(+) diff --git a/dlls/oledb32/tests/database.c b/dlls/oledb32/tests/database.c index 8ff55b9d2a..3a6a582f56 100644 --- a/dlls/oledb32/tests/database.c +++ b/dlls/oledb32/tests/database.c @@ -939,6 +939,26 @@ static void test_dslocator(void) hr = IDataSourceLocator_QueryInterface(dslocator, &IID_IRpcOptions, (void **)&opts); ok(hr == E_NOINTERFACE, "got %08x\n", hr); + + if (winetest_interactive) + { + IDispatch *disp = NULL; + + hr = IDataSourceLocator_PromptNew(dslocator, NULL); + ok(hr == E_INVALIDARG, "got %08x\n", hr); + + hr = IDataSourceLocator_PromptNew(dslocator, &disp); + if (hr == S_OK) + { + ok(disp != NULL, "got %08x\n", hr); + IDispatch_Release(disp); + } + else + { + ok(hr == S_FALSE, "got %08x\n", hr); + ok(!disp, "got %08x\n", hr); + } + } IDataInitialize_Release(datainit2); IDataInitialize_Release(datainit); -- 2.17.1

Dec. 17, 2019

Re: [PATCH 2/2] winemac.drv: Support VK_EXT_metal_surface.

by Ken Thomases

On Dec 13, 2019, at 10:57 AM, Chip Davis <cdavis(a)codeweavers.com> wrote: > > Prefer it to VK_MVK_macos_surface when present. > > MoltenVK has deprecated VK_MVK_macos_surface in favor of > VK_EXT_metal_surface. It's likely that this extension will vanish at > some point. > > Signed-off-by: Chip Davis <cdavis(a)codeweavers.com> > --- > dlls/winemac.drv/vulkan.c | 66 +++++++++++++++++++++++++++++++++------ > 1 file changed, 57 insertions(+), 9 deletions(-) > > diff --git a/dlls/winemac.drv/vulkan.c b/dlls/winemac.drv/vulkan.c > index 21e93827c15..60f3dcda6af 100644 > --- a/dlls/winemac.drv/vulkan.c > +++ b/dlls/winemac.drv/vulkan.c > […] > @@ -341,7 +367,7 @@ static void macdrv_vkDestroySwapchainKHR(VkDevice device, VkSwapchainKHR swapcha > static VkResult macdrv_vkEnumerateInstanceExtensionProperties(const char *layer_name, > uint32_t *count, VkExtensionProperties* properties) > { > - unsigned int i; > + unsigned int i, mvk_surface = -1, ext_surface = -1; > VkResult res; > > TRACE("layer_name %s, count %p, properties %p\n", debugstr_a(layer_name), count, properties); > @@ -364,14 +390,36 @@ static VkResult macdrv_vkEnumerateInstanceExtensionProperties(const char *layer_ > > for (i = 0; i < *count; i++) > { > - /* For now the only MoltenVK extension we need to fixup. Long-term we may need an array. */ > + /* For now the only MoltenVK extensions we need to fixup. Long-term we may need an array. */ > if (!strcmp(properties[i].extensionName, "VK_MVK_macos_surface")) > { > + if (ext_surface != -1) > + { > + /* If we've already seen EXT_metal_surface, just hide this one. */ > + memcpy(properties + i, properties + i + 1, (--(*count) - i) * sizeof(VkExtensionProperties)); This and the memcpy() below should be memmove(), strictly speaking. Use sizeof(*properties) instead of sizeof(VkExtensionProperties). Also, please decrement *count in a separate statement. And you need to decrement i, too, or you skip the extension that was moved to properties[i]. > + continue; > + } > TRACE("Substituting VK_MVK_macos_surface for VK_KHR_win32_surface\n"); > > snprintf(properties[i].extensionName, sizeof(properties[i].extensionName), > VK_KHR_WIN32_SURFACE_EXTENSION_NAME); > properties[i].specVersion = VK_KHR_WIN32_SURFACE_SPEC_VERSION; > + mvk_surface = i; > + } > + if (!strcmp(properties[i].extensionName, "VK_EXT_metal_surface")) > + { > + if (mvk_surface != -1) > + { > + /* If we've already seen MVK_macos_surface, just hide this one. */ > + memcpy(properties + i, properties + i + 1, (--(*count) - i) * sizeof(VkExtensionProperties)); > + continue; > + } > + TRACE("Substituting VK_EXT_metal_surface for VK_KHR_win32_surface\n"); > + > + snprintf(properties[i].extensionName, sizeof(properties[i].extensionName), > + VK_KHR_WIN32_SURFACE_EXTENSION_NAME); > + properties[i].specVersion = VK_KHR_WIN32_SURFACE_SPEC_VERSION; > + ext_surface = i; > } These two "if" statements are so similar they can and should be combined. There can be a single flag for "seen either surface extension". (No need to track the index.) Finally, there's a comment above the call to the host vkEnumerateInstanceExtensionProperties() that asserts this function will return the same count. That's no longer true. -Ken

Dec. 17, 2019

Re: [PATCH 1/2] winemac.drv: Pass the CAMetalLayer to MoltenVK.

by Ken Thomases

On Dec 13, 2019, at 10:57 AM, Chip Davis <cdavis(a)codeweavers.com> wrote: > > It isn't safe to access the view object from any thread other than the > main thread. In fact, if you try to call vkCreateMacOSSurfaceMVK() from > any other thread, MoltenVK prints out a big, scary warning telling you > not to do this! Instead, get the layer from the view ourselves and pass > that to MoltenVK. Recent versions of MoltenVK can accept either the view > or the layer. > > Signed-off-by: Chip Davis <cdavis(a)codeweavers.com> > --- > dlls/winemac.drv/cocoa_window.m | 14 ++++++++++++++ > dlls/winemac.drv/macdrv_cocoa.h | 2 ++ > dlls/winemac.drv/vulkan.c | 2 +- > 3 files changed, 17 insertions(+), 1 deletion(-) > > diff --git a/dlls/winemac.drv/cocoa_window.m b/dlls/winemac.drv/cocoa_window.m > index 877653ea007..6ac30843025 100644 > --- a/dlls/winemac.drv/cocoa_window.m > +++ b/dlls/winemac.drv/cocoa_window.m > @@ -3796,6 +3796,20 @@ macdrv_metal_view macdrv_view_create_metal_view(macdrv_view v, macdrv_metal_devi > return (macdrv_metal_view)metalView; > } > > +macdrv_metal_layer macdrv_view_get_metal_layer(macdrv_metal_view v) > +{ > + WineMetalView* view = (WineMetalView*)v; > + __block CAMetalLayer* layer; > + > + if ([NSThread isMainThread]) > + layer = (CAMetalLayer*)view.layer; This will never be called from the main thread. > + else OnMainThread(^{ > + layer = (CAMetalLayer*)view.layer; > + }); > + > + return (macdrv_metal_layer)layer; > +} > + > void macdrv_view_release_metal_view(macdrv_metal_view v) > { > WineMetalView* view = (WineMetalView*)v;

Dec. 17, 2019

Re: [wine-devel] Bugs To Fix?

by Alan W. Irwin

On 2019-12-16 23:03-0000 Thales wrote: > Hello, > What are some of the most important bugs that need fixing? I would like to take a crack at one or more. I've been looking at Bugzilla and there's quite a bit there to sift through. > > Thanks for any feedback! > ...John Alway Hi John: My advice is work on a Wine bug that is blocking something of interest to you since that interest will provide important focus and motivation. For example, if you are interested in free software distributions for Windows like Cygwin or MSYS2, you might want to work on one or more bugs that are blocking those from working on Wine **after** reading a summary of the current situation at <https://github.com/msys2/MSYS2-packages/issues/682> (particularly the remarks starting at "<corinna> the old method up to XP/2K3 is not available anymore" .) Note, I prefer to stick to my own free software development rather than work on Wine, but I do follow the Cygwin and MSYS2 situation because as a Linux user I would like to test my free software using the free software libraries in those distributions for the Windows platform without paying a Microsoft tax. And there are others with similar interests as well such as "fracting" (Qian Hong) who got MSYS2 to work reasonably well under Wine-staging before the post-XP change to MSYS2 that he refers to in that discussion. However, there is obviously a huge range of software available on Windows besides the software (e.g., the installers) supporting those two software distributions so if your interest is in something else (e.g., some game) you should follow your interest rather than mine! Alan __________________________ Alan W. Irwin Programming affiliations with the FreeEOS equation-of-state implementation for stellar interiors (freeeos.sf.net) the Time Ephemerides project (timeephem.sf.net) PLplot scientific plotting software package (plplot.org) the libLASi project (unifont.org/lasi) the Loads of Linux Links project (loll.sf.net) and the Linux Brochure Project (lbproject.sf.net) __________________________ Linux-powered Science __________________________

Dec. 17, 2019

[PATCH 3/3] ddraw/tests: Skip caps test if certain caps not present.

by Jeff Smith

Signed-off-by: Jeff Smith <whydoubt(a)gmail.com> --- dlls/ddraw/tests/ddraw1.c | 7 +++++++ dlls/ddraw/tests/ddraw2.c | 7 +++++++ dlls/ddraw/tests/ddraw4.c | 7 +++++++ dlls/ddraw/tests/ddraw7.c | 7 +++++++ 4 files changed, 28 insertions(+) diff --git a/dlls/ddraw/tests/ddraw1.c b/dlls/ddraw/tests/ddraw1.c index a03d0881c5..30069738b0 100644 --- a/dlls/ddraw/tests/ddraw1.c +++ b/dlls/ddraw/tests/ddraw1.c @@ -13018,6 +13018,13 @@ static void test_caps(void) "Got unexpected caps %#x, expected %#x.\n", hel_caps.ddsOldCaps.dwCaps, hel_caps.ddsCaps.dwCaps); + if (hal_caps.ddsCaps.dwCaps && (~hal_caps.ddsCaps.dwCaps & caps_always)) + { + skip("Failed to find sufficient hardware caps, skipping test.\n"); + IDirectDraw_Release(ddraw); + return; + } + no3d = !(hal_caps.ddsCaps.dwCaps & DDSCAPS_3DDEVICE); if (hal_caps.ddsCaps.dwCaps) { diff --git a/dlls/ddraw/tests/ddraw2.c b/dlls/ddraw/tests/ddraw2.c index 688cb532bd..48c00e2220 100644 --- a/dlls/ddraw/tests/ddraw2.c +++ b/dlls/ddraw/tests/ddraw2.c @@ -13850,6 +13850,13 @@ static void test_caps(void) "Got unexpected caps %#x, expected %#x.\n", hel_caps.ddsOldCaps.dwCaps, hel_caps.ddsCaps.dwCaps); + if (hal_caps.ddsCaps.dwCaps && (~hal_caps.ddsCaps.dwCaps & caps_always)) + { + skip("Failed to find sufficient hardware caps, skipping test.\n"); + IDirectDraw_Release(ddraw); + return; + } + no3d = !(hal_caps.ddsCaps.dwCaps & DDSCAPS_3DDEVICE); if (hal_caps.ddsCaps.dwCaps) { diff --git a/dlls/ddraw/tests/ddraw4.c b/dlls/ddraw/tests/ddraw4.c index 96d8e2a701..3c2143e5bf 100644 --- a/dlls/ddraw/tests/ddraw4.c +++ b/dlls/ddraw/tests/ddraw4.c @@ -16502,6 +16502,13 @@ static void test_caps(void) "Got unexpected caps %#x, expected %#x.\n", hel_caps.ddsOldCaps.dwCaps, hel_caps.ddsCaps.dwCaps); + if (hal_caps.ddsCaps.dwCaps && (~hal_caps.ddsCaps.dwCaps & caps_always)) + { + skip("Failed to find sufficient hardware caps, skipping test.\n"); + IDirectDraw_Release(ddraw); + return; + } + no3d = !(hal_caps.ddsCaps.dwCaps & DDSCAPS_3DDEVICE); if (hal_caps.ddsCaps.dwCaps) { diff --git a/dlls/ddraw/tests/ddraw7.c b/dlls/ddraw/tests/ddraw7.c index 8044e49a60..2ca73d2d66 100644 --- a/dlls/ddraw/tests/ddraw7.c +++ b/dlls/ddraw/tests/ddraw7.c @@ -16476,6 +16476,13 @@ static void test_caps(void) "Got unexpected caps %#x, expected %#x.\n", hel_caps.ddsOldCaps.dwCaps, hel_caps.ddsCaps.dwCaps); + if (hal_caps.ddsCaps.dwCaps && (~hal_caps.ddsCaps.dwCaps & caps_always)) + { + skip("Failed to find sufficient hardware caps, skipping test.\n"); + IDirectDraw_Release(ddraw); + return; + } + no3d = !(hal_caps.ddsCaps.dwCaps & DDSCAPS_3DDEVICE); if (hal_caps.ddsCaps.dwCaps) { -- 2.23.0

Dec. 17, 2019

[PATCH 2/3] ddraw/tests: Allow for broken surface pitch.

by Jeff Smith

Signed-off-by: Jeff Smith <whydoubt(a)gmail.com> --- dlls/ddraw/tests/ddraw1.c | 3 ++- dlls/ddraw/tests/ddraw2.c | 3 ++- dlls/ddraw/tests/ddraw4.c | 3 ++- dlls/ddraw/tests/ddraw7.c | 3 ++- 4 files changed, 8 insertions(+), 4 deletions(-) diff --git a/dlls/ddraw/tests/ddraw1.c b/dlls/ddraw/tests/ddraw1.c index b1045e2854..a03d0881c5 100644 --- a/dlls/ddraw/tests/ddraw1.c +++ b/dlls/ddraw/tests/ddraw1.c @@ -5509,7 +5509,8 @@ static void test_create_surface_pitch(void) "Test %u: Got unexpected pitch %u, expected %u.\n", i, U1(surface_desc).lPitch, test_data[i].pitch_out64); else - ok(U1(surface_desc).lPitch == test_data[i].pitch_out32, + ok(U1(surface_desc).lPitch == test_data[i].pitch_out32 || broken( + U1(surface_desc).lPitch == 0xfc && (test_data[i].caps & DDSCAPS_VIDEOMEMORY)), "Test %u: Got unexpected pitch %u, expected %u.\n", i, U1(surface_desc).lPitch, test_data[i].pitch_out32); } diff --git a/dlls/ddraw/tests/ddraw2.c b/dlls/ddraw/tests/ddraw2.c index 5a26d4eec4..688cb532bd 100644 --- a/dlls/ddraw/tests/ddraw2.c +++ b/dlls/ddraw/tests/ddraw2.c @@ -6452,7 +6452,8 @@ static void test_create_surface_pitch(void) "Test %u: Got unexpected pitch %u, expected %u.\n", i, U1(surface_desc).lPitch, test_data[i].pitch_out64); else - ok(U1(surface_desc).lPitch == test_data[i].pitch_out32, + ok(U1(surface_desc).lPitch == test_data[i].pitch_out32 || broken( + U1(surface_desc).lPitch == 0xfc && (test_data[i].caps & DDSCAPS_VIDEOMEMORY)), "Test %u: Got unexpected pitch %u, expected %u.\n", i, U1(surface_desc).lPitch, test_data[i].pitch_out32); } diff --git a/dlls/ddraw/tests/ddraw4.c b/dlls/ddraw/tests/ddraw4.c index 33e3db259a..96d8e2a701 100644 --- a/dlls/ddraw/tests/ddraw4.c +++ b/dlls/ddraw/tests/ddraw4.c @@ -8288,7 +8288,8 @@ static void test_create_surface_pitch(void) "Test %u: Got unexpected pitch %u, expected %u.\n", i, U1(surface_desc).lPitch, test_data[i].pitch_out64); else - ok(U1(surface_desc).lPitch == test_data[i].pitch_out32, + ok(U1(surface_desc).lPitch == test_data[i].pitch_out32 || broken( + U1(surface_desc).lPitch == 0xfc && (test_data[i].caps & DDSCAPS_VIDEOMEMORY)), "Test %u: Got unexpected pitch %u, expected %u.\n", i, U1(surface_desc).lPitch, test_data[i].pitch_out32); } diff --git a/dlls/ddraw/tests/ddraw7.c b/dlls/ddraw/tests/ddraw7.c index 8ad1e859fa..8044e49a60 100644 --- a/dlls/ddraw/tests/ddraw7.c +++ b/dlls/ddraw/tests/ddraw7.c @@ -8135,7 +8135,8 @@ static void test_create_surface_pitch(void) "Test %u: Got unexpected pitch %u, expected %u.\n", i, U1(surface_desc).lPitch, test_data[i].pitch_out64); else - ok(U1(surface_desc).lPitch == test_data[i].pitch_out32, + ok(U1(surface_desc).lPitch == test_data[i].pitch_out32 || broken( + U1(surface_desc).lPitch == 0xfc && (test_data[i].caps & DDSCAPS_VIDEOMEMORY)), "Test %u: Got unexpected pitch %u, expected %u.\n", i, U1(surface_desc).lPitch, test_data[i].pitch_out32); } -- 2.23.0

Dec. 17, 2019

[PATCH 1/3] ddraw/tests: Allow for no-texture-hardware error.

by Jeff Smith

Signed-off-by: Jeff Smith <whydoubt(a)gmail.com> --- This series cleans up failures seen on the XP test machine using the "VMware SVGA II" driver, which seems to lack various features, such as 3D texture support. dlls/ddraw/tests/ddraw1.c | 3 +++ dlls/ddraw/tests/ddraw2.c | 3 +++ dlls/ddraw/tests/ddraw4.c | 3 +++ dlls/ddraw/tests/ddraw7.c | 3 +++ 4 files changed, 12 insertions(+) diff --git a/dlls/ddraw/tests/ddraw1.c b/dlls/ddraw/tests/ddraw1.c index b513121fbb..b1045e2854 100644 --- a/dlls/ddraw/tests/ddraw1.c +++ b/dlls/ddraw/tests/ddraw1.c @@ -5488,6 +5488,9 @@ static void test_create_surface_pitch(void) } if ((test_data[i].caps & DDSCAPS_VIDEOMEMORY) && hr == DDERR_NODIRECTDRAWHW) continue; + if (((test_data[i].caps & (DDSCAPS_VIDEOMEMORY | DDSCAPS_TEXTURE)) == + (DDSCAPS_VIDEOMEMORY | DDSCAPS_TEXTURE)) && hr == DDERR_NOTEXTUREHW) + continue; ok(hr == test_data[i].hr, "Test %u: Got unexpected hr %#x, expected %#x.\n", i, hr, test_data[i].hr); if (FAILED(hr)) continue; diff --git a/dlls/ddraw/tests/ddraw2.c b/dlls/ddraw/tests/ddraw2.c index 541362c748..5a26d4eec4 100644 --- a/dlls/ddraw/tests/ddraw2.c +++ b/dlls/ddraw/tests/ddraw2.c @@ -6430,6 +6430,9 @@ static void test_create_surface_pitch(void) } if ((test_data[i].caps & DDSCAPS_VIDEOMEMORY) && hr == DDERR_NODIRECTDRAWHW) continue; + if (((test_data[i].caps & (DDSCAPS_VIDEOMEMORY | DDSCAPS_TEXTURE)) == + (DDSCAPS_VIDEOMEMORY | DDSCAPS_TEXTURE)) && hr == DDERR_NOTEXTUREHW) + continue; ok(hr == test_data[i].hr, "Test %u: Got unexpected hr %#x, expected %#x.\n", i, hr, test_data[i].hr); if (FAILED(hr)) continue; diff --git a/dlls/ddraw/tests/ddraw4.c b/dlls/ddraw/tests/ddraw4.c index 8cadfae52a..33e3db259a 100644 --- a/dlls/ddraw/tests/ddraw4.c +++ b/dlls/ddraw/tests/ddraw4.c @@ -8266,6 +8266,9 @@ static void test_create_surface_pitch(void) } if ((test_data[i].caps & DDSCAPS_VIDEOMEMORY) && hr == DDERR_NODIRECTDRAWHW) continue; + if (((test_data[i].caps & (DDSCAPS_VIDEOMEMORY | DDSCAPS_TEXTURE)) == + (DDSCAPS_VIDEOMEMORY | DDSCAPS_TEXTURE)) && hr == DDERR_NOTEXTUREHW) + continue; ok(hr == test_data[i].hr, "Test %u: Got unexpected hr %#x, expected %#x.\n", i, hr, test_data[i].hr); if (FAILED(hr)) continue; diff --git a/dlls/ddraw/tests/ddraw7.c b/dlls/ddraw/tests/ddraw7.c index f4a709711b..8ad1e859fa 100644 --- a/dlls/ddraw/tests/ddraw7.c +++ b/dlls/ddraw/tests/ddraw7.c @@ -8110,6 +8110,9 @@ static void test_create_surface_pitch(void) } if ((test_data[i].caps & DDSCAPS_VIDEOMEMORY) && hr == DDERR_NODIRECTDRAWHW) continue; + if (((test_data[i].caps & (DDSCAPS_VIDEOMEMORY | DDSCAPS_TEXTURE)) == + (DDSCAPS_VIDEOMEMORY | DDSCAPS_TEXTURE)) && hr == DDERR_NOTEXTUREHW) + continue; if (is_ddraw64 && (test_data[i].caps & DDSCAPS_TEXTURE) && SUCCEEDED(test_data[i].hr)) todo_wine ok(hr == E_NOINTERFACE, "Test %u: Got unexpected hr %#x.\n", i, hr); else -- 2.23.0

Dec. 17, 2019

Re: Bugs To Fix?

by Stefan Dösinger

Hi, There are plenty of bugs, we have enough to keep everyone busy :-) . Other than "it would be great to have this bug fixed", other things to keep in mind is how difficult the bug is to fix, how much investigative effort has been put into it and what your area of interest is. Stefan Am 17.12.19 um 00:03 schrieb Thales: > Hello, > > What are some of the most important bugs that need fixing? I would > like to take a crack at one or more. I've been looking at Bugzilla and > there's quite a bit there to sift through. > > > Thanks for any feedback! > > ...John Alway

Dec. 16, 2019

Bugs To Fix?

by Thales

Hello, What are some of the most important bugs that need fixing? I would like to take a crack at one or more. I've been looking at Bugzilla and there's quite a bit there to sift through. Thanks for any feedback! ...John Alway

Dec. 16, 2019

Re: [PATCH] rpcrt4: Fix ITypeInfo ref count handling.

by Zebediah Figura

On 12/16/19 3:19 PM, Jacek Caban wrote: > > From: Kevin Puetz <PuetzKevinA(a)JohnDeere.com> > > Signed-off-by: Jacek Caban <jacek(a)codeweavers.com> > --- > dlls/rpcrt4/ndr_typelib.c | 23 +++++++++++++++++++---- > 1 file changed, 19 insertions(+), 4 deletions(-) > > Hello Jacek, This feels to me like it's making things even more awkward. I'm presuming the problem is that get_iface_info() is releasing a passed-in ITypeInfo, in which case I feel like standard COM practice would just be to not do that. Does the attached patch work for you?

Dec. 16, 2019

[PATCH] rpcrt4: Fix ITypeInfo ref count handling.

by Jacek Caban

From: Kevin Puetz <PuetzKevinA(a)JohnDeere.com> Signed-off-by: Jacek Caban <jacek(a)codeweavers.com> --- dlls/rpcrt4/ndr_typelib.c | 23 +++++++++++++++++++---- 1 file changed, 19 insertions(+), 4 deletions(-)

Dec. 16, 2019

Re: [PATCH 2/2] kernel32: Forwrad OpenProcessToken() to kernelbase.

by Jeff Smith

Typo in the title. On Mon, Dec 16, 2019 at 12:06 PM Chip Davis <cdavis(a)codeweavers.com> wrote: > > Signed-off-by: Chip Davis <cdavis(a)codeweavers.com> > --- > dlls/kernel32/kernel32.spec | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/dlls/kernel32/kernel32.spec b/dlls/kernel32/kernel32.spec > index 466ef5e52a16..c7a8e2024431 100644 > --- a/dlls/kernel32/kernel32.spec > +++ b/dlls/kernel32/kernel32.spec > @@ -1127,7 +1127,7 @@ > # @ stub OpenPrivateNamespaceA > # @ stub OpenPrivateNamespaceW > @ stdcall -import OpenProcess(long long long) > -# @ stub OpenProcessToken > +@ stdcall -import OpenProcessToken(long long ptr) > @ stdcall OpenProfileUserMapping() > @ stdcall OpenSemaphoreA(long long str) > @ stdcall -import OpenSemaphoreW(long long wstr) > -- > 2.24.0 > >

Dec. 16, 2019

[PATCH] dswave/tests: Use wide-char string literals

by Michael Stefaniuc

Signed-off-by: Michael Stefaniuc <mstefani(a)winehq.org> --- dlls/dswave/tests/dswave.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/dlls/dswave/tests/dswave.c b/dlls/dswave/tests/dswave.c index d9bdc85ea7..62a76f1806 100644 --- a/dlls/dswave/tests/dswave.c +++ b/dlls/dswave/tests/dswave.c @@ -190,7 +190,6 @@ static void test_parsedescriptor(void) IStream *stream; DMUS_OBJECTDESC desc = {0}; HRESULT hr; - const WCHAR s_inam[] = {'I','N','A','M','\0'}; const FOURCC alldesc[] = { FOURCC_RIFF, mmioFOURCC('W','A','V','E'), DMUS_FOURCC_CATEGORY_CHUNK, FOURCC_LIST, @@ -279,7 +278,7 @@ static void test_parsedescriptor(void) ok(hr == S_OK, "ParseDescriptor failed: %08x, expected S_OK\n", hr); ok(desc.dwValidData == DMUS_OBJ_NAME, "Got valid data %#x, expected DMUS_OBJ_NAME\n", desc.dwValidData); - ok(!memcmp(desc.wszName, s_inam, sizeof(s_inam)), "Got name '%s', expected 'INAM'\n", + ok(!lstrcmpW(desc.wszName, L"INAM"), "Got name '%s', expected 'INAM'\n", wine_dbgstr_w(desc.wszName)); IStream_Release(stream); @@ -291,7 +290,7 @@ static void test_parsedescriptor(void) ok(desc.dwValidData == (DMUS_OBJ_OBJECT | DMUS_OBJ_NAME | DMUS_OBJ_VERSION), "Got valid data %#x, expected DMUS_OBJ_OBJECT | DMUS_OBJ_NAME | DMUS_OBJ_VERSION\n", desc.dwValidData); - ok(!memcmp(desc.wszName, s_inam, sizeof(s_inam)), "Got name '%s', expected 'INAM'\n", + ok(!lstrcmpW(desc.wszName, L"INAM"), "Got name '%s', expected 'INAM'\n", wine_dbgstr_w(desc.wszName)); IStream_Release(stream); -- 2.23.0

Dec. 16, 2019

[PATCH] dmstyle/tests: Use wide-char string literals

by Michael Stefaniuc

Signed-off-by: Michael Stefaniuc <mstefani(a)winehq.org> --- dlls/dmstyle/tests/dmstyle.c | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-) diff --git a/dlls/dmstyle/tests/dmstyle.c b/dlls/dmstyle/tests/dmstyle.c index a63d1dbd1c..5130af773c 100644 --- a/dlls/dmstyle/tests/dmstyle.c +++ b/dlls/dmstyle/tests/dmstyle.c @@ -475,8 +475,6 @@ static void test_parsedescriptor(void) DMUS_OBJECTDESC desc = {0}; HRESULT hr; DWORD valid; - const WCHAR s_inam[] = {'I','N','A','M','\0'}; - const WCHAR s_unam[] = {'U','N','A','M','\0'}; const FOURCC alldesc[] = { FOURCC_RIFF, DMUS_FOURCC_STYLE_FORM, DMUS_FOURCC_CATEGORY_CHUNK, FOURCC_LIST, @@ -550,7 +548,7 @@ static void test_parsedescriptor(void) wine_dbgstr_guid(&desc.guidClass)); ok(IsEqualGUID(&desc.guidObject, &GUID_NULL), "Got object guid %s, expected GUID_NULL\n", wine_dbgstr_guid(&desc.guidClass)); - ok(!memcmp(desc.wszName, s_unam, sizeof(s_unam)), "Got name '%s', expected 'UNAM'\n", + ok(!lstrcmpW(desc.wszName, L"UNAM"), "Got name '%s', expected 'UNAM'\n", wine_dbgstr_w(desc.wszName)); IStream_Release(stream); @@ -562,7 +560,7 @@ static void test_parsedescriptor(void) ok(hr == S_OK, "ParseDescriptor failed: %08x, expected S_OK\n", hr); ok(desc.dwValidData == (DMUS_OBJ_CLASS | DMUS_OBJ_NAME), "Got valid data %#x, expected DMUS_OBJ_CLASS | DMUS_OBJ_NAME\n", desc.dwValidData); - ok(!memcmp(desc.wszName, s_inam, sizeof(s_inam)), "Got name '%s', expected 'INAM'\n", + ok(!lstrcmpW(desc.wszName, L"INAM"), "Got name '%s', expected 'INAM'\n", wine_dbgstr_w(desc.wszName)); IStream_Release(stream); @@ -583,7 +581,7 @@ static void test_parsedescriptor(void) ok(hr == S_OK, "ParseDescriptor failed: %08x, expected S_OK\n", hr); valid = DMUS_OBJ_OBJECT|DMUS_OBJ_CLASS|DMUS_OBJ_NAME|DMUS_OBJ_VERSION; ok(desc.dwValidData == valid, "Got valid data %#x, expected %#x\n", desc.dwValidData, valid); - ok(!memcmp(desc.wszName, s_inam, sizeof(s_inam)), "Got name '%s', expected 'INAM'\n", + ok(!lstrcmpW(desc.wszName, L"INAM"), "Got name '%s', expected 'INAM'\n", wine_dbgstr_w(desc.wszName)); IStream_Release(stream); -- 2.23.0

Dec. 16, 2019

[PATCH] dmime/tests: Use wide-char string literals

by Michael Stefaniuc

Signed-off-by: Michael Stefaniuc <mstefani(a)winehq.org> --- dlls/dmime/tests/dmime.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/dlls/dmime/tests/dmime.c b/dlls/dmime/tests/dmime.c index 5bb3a8b5b3..99295f391b 100644 --- a/dlls/dmime/tests/dmime.c +++ b/dlls/dmime/tests/dmime.c @@ -956,8 +956,6 @@ static void test_parsedescriptor(void) HRESULT hr; DWORD valid; unsigned int i; - const WCHAR s_unam[] = {'U','N','A','M','\0'}; - const WCHAR s_inam[] = {'I','\0'}; /* fourcc ~0 will be replaced later on */ FOURCC alldesc[] = { @@ -1092,7 +1090,7 @@ static void test_parsedescriptor(void) "Got version %u.%u, expected 5.8\n", desc.vVersion.dwVersionMS, desc.vVersion.dwVersionLS); if (forms[i].form != mmioFOURCC('W','A','V','E')) - ok(!memcmp(desc.wszName, s_unam, sizeof(s_unam)), "Got name '%s', expected 'UNAM'\n", + ok(!lstrcmpW(desc.wszName, L"UNAM"), "Got name '%s', expected 'UNAM'\n", wine_dbgstr_w(desc.wszName)); IStream_Release(stream); @@ -1130,7 +1128,7 @@ static void test_parsedescriptor(void) valid |= DMUS_OBJ_NAME; ok(desc.dwValidData == valid, "Got valid data %#x, expected %#x\n", desc.dwValidData, valid); if (forms[i].form == mmioFOURCC('W','A','V','E')) - ok(!memcmp(desc.wszName, s_inam, sizeof(s_inam)), "Got name '%s', expected 'I'\n", + ok(!lstrcmpW(desc.wszName, L"I"), "Got name '%s', expected 'I'\n", wine_dbgstr_w(desc.wszName)); IStream_Release(stream); -- 2.23.0

Dec. 16, 2019

[PATCH v2] dmloader: Release objects on error paths (Coverity)

by Michael Stefaniuc

From: Alistair Leslie-Hughes <leslie_alistair(a)hotmail.com> CID: 1448587 CID: 1449363 CID: 1448601 Signed-off-by: Alistair Leslie-Hughes <leslie_alistair(a)hotmail.com> Signed-off-by: Michael Stefaniuc <mstefani(a)winehq.org> --- v2: - Add missing {} to an if - Whitespace fixes dlls/dmloader/loader.c | 12 ++++++++++++ dlls/dmloader/loaderstream.c | 7 ++++--- 2 files changed, 16 insertions(+), 3 deletions(-) diff --git a/dlls/dmloader/loader.c b/dlls/dmloader/loader.c index 66c0f92b1b..c5f5094fab 100644 --- a/dlls/dmloader/loader.c +++ b/dlls/dmloader/loader.c @@ -384,18 +384,24 @@ static HRESULT WINAPI IDirectMusicLoaderImpl_GetObject(IDirectMusicLoader8 *ifac /* create object */ result = CoCreateInstance (&pDesc->guidClass, NULL, CLSCTX_INPROC_SERVER, &IID_IDirectMusicObject, (LPVOID*)&pObject); if (FAILED(result)) { + IStream_Release(pStream); ERR(": could not create object\n"); return result; } /* acquire PersistStream interface */ result = IDirectMusicObject_QueryInterface (pObject, &IID_IPersistStream, (LPVOID*)&pPersistStream); if (FAILED(result)) { + IStream_Release(pStream); + IDirectMusicObject_Release(pObject); ERR("failed to Query\n"); return result; } /* load */ result = IPersistStream_Load (pPersistStream, pStream); if (result != S_OK) { + IStream_Release(pStream); + IPersistStream_Release(pPersistStream); + IDirectMusicObject_Release(pObject); WARN(": failed to (completely) load object (%08x)\n", result); return result; } @@ -408,6 +414,9 @@ static HRESULT WINAPI IDirectMusicLoaderImpl_GetObject(IDirectMusicLoader8 *ifac lstrcpyW (GotDesc.wszFileName, pDesc->wszFileName); } if (FAILED(result)) { + IStream_Release(pStream); + IPersistStream_Release(pPersistStream); + IDirectMusicObject_Release(pObject); ERR(": failed to get descriptor\n"); return result; } @@ -524,6 +533,7 @@ static HRESULT WINAPI IDirectMusicLoaderImpl_SetObject(IDirectMusicLoader8 *ifac /* create object */ hr = CoCreateInstance (&pDesc->guidClass, NULL, CLSCTX_INPROC_SERVER, &IID_IDirectMusicObject, (LPVOID*)&pObject); if (FAILED(hr)) { + IStream_Release(pStream); ERR("Object creation of %s failed 0x%08x\n", debugstr_guid(&pDesc->guidClass),hr); return DMUS_E_LOADER_FAILEDOPEN; } @@ -537,6 +547,8 @@ static HRESULT WINAPI IDirectMusicLoaderImpl_SetObject(IDirectMusicLoader8 *ifac /* hmph... due to some trouble I had with certain tests, we store current position and then set it back */ DM_STRUCT_INIT(&Desc); if (FAILED(IDirectMusicObject_ParseDescriptor (pObject, pStream, &Desc))) { + IStream_Release(pStream); + IDirectMusicObject_Release(pObject); ERR(": couldn't parse descriptor\n"); return DMUS_E_LOADER_FORMATNOTSUPPORTED; } diff --git a/dlls/dmloader/loaderstream.c b/dlls/dmloader/loaderstream.c index 855203ec3e..150a53524a 100644 --- a/dlls/dmloader/loaderstream.c +++ b/dlls/dmloader/loaderstream.c @@ -663,10 +663,11 @@ static HRESULT WINAPI IDirectMusicLoaderGenericStream_IStream_Clone (LPSTREAM if TRACE("(%p, %p)\n", iface, ppstm); result = DMUSIC_CreateDirectMusicLoaderGenericStream ((LPVOID*)&pOther); if (FAILED(result)) return result; - - if (FAILED(IStream_Clone (This->pStream, &pLowLevel))) + + if (FAILED(IStream_Clone (This->pStream, &pLowLevel))) { + IStream_Release(pOther); return E_FAIL; - + } IDirectMusicLoaderGenericStream_Attach (pOther, pLowLevel, This->pLoader); TRACE(": succeeded\n"); -- 2.23.0

Dec. 16, 2019

[PATCH] dmband/tests: Use wide-char string literals

by Michael Stefaniuc

Signed-off-by: Michael Stefaniuc <mstefani(a)winehq.org> --- dlls/dmband/tests/dmband.c | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-) diff --git a/dlls/dmband/tests/dmband.c b/dlls/dmband/tests/dmband.c index dce8a47ba4..4b8aaf6120 100644 --- a/dlls/dmband/tests/dmband.c +++ b/dlls/dmband/tests/dmband.c @@ -392,8 +392,6 @@ static void test_parsedescriptor(void) DMUS_OBJECTDESC desc = {0}; HRESULT hr; DWORD valid; - const WCHAR s_inam[] = {'I','N','A','M','\0'}; - const WCHAR s_unam[] = {'U','N','A','M','\0'}; const FOURCC alldesc[] = { FOURCC_RIFF, DMUS_FOURCC_BAND_FORM, DMUS_FOURCC_CATEGORY_CHUNK, FOURCC_LIST, @@ -477,7 +475,7 @@ static void test_parsedescriptor(void) wine_dbgstr_guid(&desc.guidClass)); ok(IsEqualGUID(&desc.guidObject, &GUID_NULL), "Got object guid %s, expected GUID_NULL\n", wine_dbgstr_guid(&desc.guidClass)); - ok(!memcmp(desc.wszName, s_unam, sizeof(s_unam)), "Got name '%s', expected 'UNAM'\n", + ok(!lstrcmpW(desc.wszName, L"UNAM"), "Got name '%s', expected 'UNAM'\n", wine_dbgstr_w(desc.wszName)); ok(!desc.ftDate.dwHighDateTime && !desc.ftDate.dwLowDateTime, "Got file time %08x %08x, expected 0\n", desc.ftDate.dwHighDateTime, @@ -492,7 +490,7 @@ static void test_parsedescriptor(void) ok(hr == S_OK, "ParseDescriptor failed: %08x, expected S_OK\n", hr); ok(desc.dwValidData == (DMUS_OBJ_CLASS | DMUS_OBJ_NAME), "Got valid data %#x, expected DMUS_OBJ_CLASS | DMUS_OBJ_NAME\n", desc.dwValidData); - ok(!memcmp(desc.wszName, s_inam, sizeof(s_inam)), "Got name '%s', expected 'INAM'\n", + ok(!lstrcmpW(desc.wszName, L"INAM"), "Got name '%s', expected 'INAM'\n", wine_dbgstr_w(desc.wszName)); IStream_Release(stream); @@ -513,7 +511,7 @@ static void test_parsedescriptor(void) ok(hr == S_OK, "ParseDescriptor failed: %08x, expected S_OK\n", hr); valid = DMUS_OBJ_OBJECT|DMUS_OBJ_CLASS|DMUS_OBJ_NAME|DMUS_OBJ_CATEGORY|DMUS_OBJ_VERSION|DMUS_OBJ_DATE; ok(desc.dwValidData == valid, "Got valid data %#x, expected %#x\n", desc.dwValidData, valid); - ok(!memcmp(desc.wszName, s_inam, sizeof(s_inam)), "Got name '%s', expected 'INAM'\n", + ok(!lstrcmpW(desc.wszName, L"INAM"), "Got name '%s', expected 'INAM'\n", wine_dbgstr_w(desc.wszName)); IStream_Release(stream); -- 2.23.0

Dec. 16, 2019

[PATCH] jscript/tests: Allow GetTypeInfo calls.

by Jacek Caban

It's called by some win10 versions. Signed-off-by: Jacek Caban <jacek(a)codeweavers.com> --- dlls/jscript/tests/activex.c | 4 +--- dlls/jscript/tests/caller.c | 4 +--- 2 files changed, 2 insertions(+), 6 deletions(-)

Dec. 16, 2019

Re: [PATCH 2/2] kernel32: Forwrad OpenProcessToken() to kernelbase.

by Marvin

Hi, While running your changed tests, 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=62223 Your paranoid android. === debian10 (32 bit report) === kernel32: debugger.c:320: Test failed: GetThreadContext failed: 5

Dec. 16, 2019

Code freeze

by Alexandre Julliard

Folks, As you know, we are now in code freeze towards 5.0. This means that patches will only be accepted if they are small enough, and fix an actual bug. Please make sure to mention in the patch which bug you are fixing, so that I can assess the benefit/risk ratio of the changes. Patches that can't go in will be marked as "deferred", and will have to be resubmitted once 5.0 is out. Changes to the tests are still welcome as long as they don't touch the code outside the tests, so this is a good opportunity to work on making the test results look better... As usual, I'll be making rc releases every week until the final 5.0, which I expect to be ready in the second half of January. -- Alexandre Julliard julliard(a)winehq.org

Dec. 16, 2019

Re: [PATCH 1/2] kernel32: Forward OpenThreadToken() to kernelbase.

by Marvin

Hi, While running your changed tests, 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=62222 Your paranoid android. === debian10 (32 bit report) === kernel32: comm.c:918: Test failed: OutQueue should not be empty debugger: Timeout === debian10 (32 bit Chinese:China report) === kernel32: change.c:320: Test failed: should be ready debugger.c:320: Test failed: GetThreadContext failed: 5

Dec. 16, 2019

Re: [PATCH 5/5] adsldp/tests: Add some tests for LDAPNamespace.

by Alexandre Julliard

Zebediah Figura <z.figura12(a)gmail.com> writes: > On 12/16/19 6:10 AM, Francois Gouget wrote: >> On Mon, 16 Dec 2019, Dmitry Timoshkov wrote: >> >>> Marvin <testbot(a)winehq.org> wrote: >>> >>>> === debian10 (32 bit report) === >>>> >>>> adsldp: >>>> ldap.c:45: Test failed: got 0x80040154 >>>> Unhandled exception: page fault on read access to 0x00000000 in 32-bit code (0x004015c0). >>>> >>>> Report errors: >>>> adsldp:ldap crashed (c0000005) >>> >>> testbot is broken and doesn't update Wine prefix. >> >> The TestBot recreates the WinePrefixes from scratch with the daily Wine >> commits: >> >> https://source.winehq.org/git/tools.git/blob/HEAD:/testbot/bin/build/WineRe… >> >> But it does not recreate the wineprefixes from scratch with each patch. >> For those it just runs Wine which is supposed to update the wineprefix >> automatically. >> >> Not updating the wineprefix automatically would force the users to >> delete and recreate their wineprefixes, including reinstalling >> applications, every time they upgrade Wine, which I would argue is a >> bug. >> > > I'd agree. Unfortunately, we don't have a way of automatically detecting > when a new WINE_REGISTRY resource is added to an existing DLL. I'm not > sure what would be a good solution there (manually updating wine.inf.in?) wine.inf is updated on every release, so it's not an issue for users. Developers are expected to be able to figure out such issues and update manually when necessary (which happens rarely in practice). The testbot could probably update on every patch, or update and retry when a test failed. -- Alexandre Julliard julliard(a)winehq.org

Dec. 16, 2019

[PATCH 2/2] kernel32: Forwrad OpenProcessToken() to kernelbase.

by Chip Davis

Signed-off-by: Chip Davis <cdavis(a)codeweavers.com> --- dlls/kernel32/kernel32.spec | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/dlls/kernel32/kernel32.spec b/dlls/kernel32/kernel32.spec index 466ef5e52a16..c7a8e2024431 100644 --- a/dlls/kernel32/kernel32.spec +++ b/dlls/kernel32/kernel32.spec @@ -1127,7 +1127,7 @@ # @ stub OpenPrivateNamespaceA # @ stub OpenPrivateNamespaceW @ stdcall -import OpenProcess(long long long) -# @ stub OpenProcessToken +@ stdcall -import OpenProcessToken(long long ptr) @ stdcall OpenProfileUserMapping() @ stdcall OpenSemaphoreA(long long str) @ stdcall -import OpenSemaphoreW(long long wstr) -- 2.24.0

Dec. 16, 2019

[PATCH 1/2] kernel32: Forward OpenThreadToken() to kernelbase.

by Chip Davis

Signed-off-by: Chip Davis <cdavis(a)codeweavers.com> --- dlls/kernel32/kernel32.spec | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/dlls/kernel32/kernel32.spec b/dlls/kernel32/kernel32.spec index 408ea415401a..466ef5e52a16 100644 --- a/dlls/kernel32/kernel32.spec +++ b/dlls/kernel32/kernel32.spec @@ -1132,7 +1132,7 @@ @ stdcall OpenSemaphoreA(long long str) @ stdcall -import OpenSemaphoreW(long long wstr) @ stdcall -import OpenThread(long long long) -# @ stub OpenThreadToken +@ stdcall -import OpenThreadToken(long long long ptr) @ stdcall -i386 OpenVxDHandle(long) @ stdcall OpenWaitableTimerA(long long str) @ stdcall -import OpenWaitableTimerW(long long wstr) -- 2.24.0

Dec. 16, 2019

Re: [PATCH v3 2/3] qedit/samplegrabber: Consistently the "pMemInputPin" member of struct strmbase_source.

by Zebediah Figura

On 12/16/19 2:36 AM, Sven Baars wrote: > Hey Zeb, > > I think you forgot a verb in the commit message. > > Sven > Thanks for catching that, it should be "Consistently use the "pMemInputPin" member of struct strmbase_source." Alexandre, would you mind fixing that up, please?

Dec. 16, 2019

Re: [PATCH 5/5] adsldp/tests: Add some tests for LDAPNamespace.

by Zebediah Figura

On 12/16/19 6:10 AM, Francois Gouget wrote: > On Mon, 16 Dec 2019, Dmitry Timoshkov wrote: > >> Marvin <testbot(a)winehq.org> wrote: >> >>> === debian10 (32 bit report) === >>> >>> adsldp: >>> ldap.c:45: Test failed: got 0x80040154 >>> Unhandled exception: page fault on read access to 0x00000000 in 32-bit code (0x004015c0). >>> >>> Report errors: >>> adsldp:ldap crashed (c0000005) >> >> testbot is broken and doesn't update Wine prefix. > > The TestBot recreates the WinePrefixes from scratch with the daily Wine > commits: > > https://source.winehq.org/git/tools.git/blob/HEAD:/testbot/bin/build/WineRe… > > But it does not recreate the wineprefixes from scratch with each patch. > For those it just runs Wine which is supposed to update the wineprefix > automatically. > > Not updating the wineprefix automatically would force the users to > delete and recreate their wineprefixes, including reinstalling > applications, every time they upgrade Wine, which I would argue is a > bug. > I'd agree. Unfortunately, we don't have a way of automatically detecting when a new WINE_REGISTRY resource is added to an existing DLL. I'm not sure what would be a good solution there (manually updating wine.inf.in?)

Dec. 16, 2019

Re: [PATCH 2/5] activeds: Implement ADsOpenObject.

by Zebediah Figura

On 12/16/19 3:21 AM, Dmitry Timoshkov wrote: > Nikolay Sivov <bunglehead(a)gmail.com> wrote: > >>>>>>> + hr = >>> IADsOpenDSObject_OpenDSObject(adsopen, >>>>> (BSTR)path, (BSTR)user, (BSTR)password, reserved, &disp); >>>>>>> + if (hr == S_OK) >>>>>> You cannot cast to a BSTR, it will corrupt the data structure. >>>>> >>>>> I depends wheather the callee uses SysStringLen() and friends, in this >>>>> particular case it works just fine, there are many other places in the >>> code >>>>> when it's not necessary to create intermediate copies of the strings. >>>>> >>>> >>>> It does not matter if it works in this case, arguments should use correct >>>> types. If we have similar violations somewhere else, those need fixing as >>>> well. >>> >>> That's a pretty incorrect assertion, there's nothing wrong in casting >>> BSTR to LPWSTR and vice versa. >>> >> >> What's incorrect about it? You can use BSTR as WCHAR * argument, obviously. >> Not the other way around, even if it happens to work. > > You are contradicting yourself: if it works that means that's correct > behaviour. If you are going to further argue, please show the code that > won't work for this particular implementation. > I think it's just a matter of following correct and documented behaviour on principle. You wouldn't intentionally use an object after calling Release() just because it happens to work (relying on the fact that the object is static, or that the heap memory just doesn't get overwritten). Native might already have used something like SysStringLen() internally and have corrupted data structures that simply aren't causing any problems yet with this test. It also might not, but it costs very little to just do the documented thing.

Dec. 16, 2019

Re: [PATCH 5/5] adsldp/tests: Add some tests for LDAPNamespace.

by Dmitry Timoshkov

Sven Baars <sven.wine(a)gmail.com> wrote: > On 16-12-2019 15:05, Dmitry Timoshkov wrote: > > Sven Baars <sven.wine(a)gmail.com> wrote: > > > >>> +START_TEST(ldap) > >>> +{ > >>> + HRESULT hr; > >>> + > >>> + hr = CoInitialize(NULL); > >>> + ok(hr == S_OK, "got %#x\n", hr); > >>> + > >>> + test_LDAP(); > >>> +} > >>> > >> I think you forgot to call CoUninitialize here. > > > > The process is going to terminate anyway, so it's a waste of time. > > > In that case, you might as well remove all *_Release calls from your > tests... *_Release calls in the test help to ensure that reference counting works correctly, and I do check that when looking at the logs. However calling CoUninitialize() is meaningless for the test results. -- Dmitry.

Dec. 16, 2019

Re: [PATCH 5/5] adsldp/tests: Add some tests for LDAPNamespace.

by Sven Baars

On 16-12-2019 15:05, Dmitry Timoshkov wrote: > Sven Baars <sven.wine(a)gmail.com> wrote: > >>> +START_TEST(ldap) >>> +{ >>> + HRESULT hr; >>> + >>> + hr = CoInitialize(NULL); >>> + ok(hr == S_OK, "got %#x\n", hr); >>> + >>> + test_LDAP(); >>> +} >>> >> I think you forgot to call CoUninitialize here. > > The process is going to terminate anyway, so it's a waste of time. > In that case, you might as well remove all *_Release calls from your tests...

Dec. 16, 2019

Re: [PATCH 5/5] adsldp/tests: Add some tests for LDAPNamespace.

by Dmitry Timoshkov

Sven Baars <sven.wine(a)gmail.com> wrote: > > +START_TEST(ldap) > > +{ > > + HRESULT hr; > > + > > + hr = CoInitialize(NULL); > > + ok(hr == S_OK, "got %#x\n", hr); > > + > > + test_LDAP(); > > +} > > > I think you forgot to call CoUninitialize here. The process is going to terminate anyway, so it's a waste of time. -- Dmitry.

Dec. 16, 2019

[PATCH 2/2] kernelbase: Use translated Unicode default char to check invalid chars.

by Akihiro Sagawa

Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=48288 Signed-off-by: Akihiro Sagawa <sagawa.aki(a)gmail.com> --- dlls/kernel32/tests/codepage.c | 3 +++ dlls/kernelbase/locale.c | 6 +++--- 2 files changed, 6 insertions(+), 3 deletions(-)

Dec. 16, 2019

[PATCH 1/2] kernel32/tests: Refactor MB_ERR_INVALID_CHARS tests.

by Akihiro Sagawa

Signed-off-by: Akihiro Sagawa <sagawa.aki(a)gmail.com> --- dlls/kernel32/tests/codepage.c | 20 ++++++++++---------- 1 file changed, 10 insertions(+), 10 deletions(-)

Dec. 16, 2019

Re: [PATCH 5/5] adsldp/tests: Add some tests for LDAPNamespace.

by Sven Baars

On 16-12-2019 02:42, Dmitry Timoshkov wrote: > + > +START_TEST(ldap) > +{ > + HRESULT hr; > + > + hr = CoInitialize(NULL); > + ok(hr == S_OK, "got %#x\n", hr); > + > + test_LDAP(); > +} > I think you forgot to call CoUninitialize here.

Dec. 16, 2019

Re: [PATCH 5/5] adsldp/tests: Add some tests for LDAPNamespace.

by Francois Gouget

On Mon, 16 Dec 2019, Dmitry Timoshkov wrote: > Marvin <testbot(a)winehq.org> wrote: > > > === debian10 (32 bit report) === > > > > adsldp: > > ldap.c:45: Test failed: got 0x80040154 > > Unhandled exception: page fault on read access to 0x00000000 in 32-bit code (0x004015c0). > > > > Report errors: > > adsldp:ldap crashed (c0000005) > > testbot is broken and doesn't update Wine prefix. The TestBot recreates the WinePrefixes from scratch with the daily Wine commits: https://source.winehq.org/git/tools.git/blob/HEAD:/testbot/bin/build/WineRe… But it does not recreate the wineprefixes from scratch with each patch. For those it just runs Wine which is supposed to update the wineprefix automatically. Not updating the wineprefix automatically would force the users to delete and recreate their wineprefixes, including reinstalling applications, every time they upgrade Wine, which I would argue is a bug. -- Francois Gouget <fgouget(a)codeweavers.com>

Dec. 16, 2019

[PATCH] dmloader: Release objects on error paths (Coverity)

by Alistair Leslie-Hughes

CID: 1448587 CID: 1449363 CID: 1448601 Signed-off-by: Alistair Leslie-Hughes <leslie_alistair(a)hotmail.com> --- dlls/dmloader/loader.c | 12 ++++++++++++ dlls/dmloader/loaderstream.c | 1 + 2 files changed, 13 insertions(+) diff --git a/dlls/dmloader/loader.c b/dlls/dmloader/loader.c index cd7f6ff7ee3..2af4de9fc80 100644 --- a/dlls/dmloader/loader.c +++ b/dlls/dmloader/loader.c @@ -383,18 +383,24 @@ static HRESULT WINAPI IDirectMusicLoaderImpl_GetObject(IDirectMusicLoader8 *ifac /* create object */ result = CoCreateInstance (&pDesc->guidClass, NULL, CLSCTX_INPROC_SERVER, &IID_IDirectMusicObject, (LPVOID*)&pObject); if (FAILED(result)) { + IStream_Release (pStream); ERR(": could not create object\n"); return result; } /* acquire PersistStream interface */ result = IDirectMusicObject_QueryInterface (pObject, &IID_IPersistStream, (LPVOID*)&pPersistStream); if (FAILED(result)) { + IStream_Release (pStream); + IDirectMusicObject_Release (pObject); ERR("failed to Query\n"); return result; } /* load */ result = IPersistStream_Load (pPersistStream, pStream); if (result != S_OK) { + IStream_Release (pStream); + IPersistStream_Release (pPersistStream); + IDirectMusicObject_Release (pObject); WARN(": failed to (completely) load object (%08x)\n", result); return result; } @@ -407,6 +413,9 @@ static HRESULT WINAPI IDirectMusicLoaderImpl_GetObject(IDirectMusicLoader8 *ifac lstrcpyW (GotDesc.wszFileName, pDesc->wszFileName); } if (FAILED(result)) { + IStream_Release (pStream); + IPersistStream_Release (pPersistStream); + IDirectMusicObject_Release (pObject); ERR(": failed to get descriptor\n"); return result; } @@ -523,6 +532,7 @@ static HRESULT WINAPI IDirectMusicLoaderImpl_SetObject(IDirectMusicLoader8 *ifac /* create object */ hr = CoCreateInstance (&pDesc->guidClass, NULL, CLSCTX_INPROC_SERVER, &IID_IDirectMusicObject, (LPVOID*)&pObject); if (FAILED(hr)) { + IStream_Release (pStream); ERR("Object creation of %s failed 0x%08x\n", debugstr_guid(&pDesc->guidClass),hr); return DMUS_E_LOADER_FAILEDOPEN; } @@ -536,6 +546,8 @@ static HRESULT WINAPI IDirectMusicLoaderImpl_SetObject(IDirectMusicLoader8 *ifac /* hmph... due to some trouble I had with certain tests, we store current position and then set it back */ DM_STRUCT_INIT(&Desc); if (FAILED(IDirectMusicObject_ParseDescriptor (pObject, pStream, &Desc))) { + IStream_Release (pStream); + IDirectMusicObject_Release (pObject); ERR(": couldn't parse descriptor\n"); return DMUS_E_LOADER_FORMATNOTSUPPORTED; } diff --git a/dlls/dmloader/loaderstream.c b/dlls/dmloader/loaderstream.c index 855203ec3e8..b306672771f 100644 --- a/dlls/dmloader/loaderstream.c +++ b/dlls/dmloader/loaderstream.c @@ -665,6 +665,7 @@ static HRESULT WINAPI IDirectMusicLoaderGenericStream_IStream_Clone (LPSTREAM if if (FAILED(result)) return result; if (FAILED(IStream_Clone (This->pStream, &pLowLevel))) + IStream_Release(pOther); return E_FAIL; IDirectMusicLoaderGenericStream_Attach (pOther, pLowLevel, This->pLoader); -- 2.24.0

Dec. 16, 2019

[PATCH] msado15: Implement _Recordset get_CursorLocation and put_CursorLocation.

by Hans Leidekker

From: Alistair Leslie-Hughes <leslie_alistair(a)hotmail.com> Signed-off-by: Alistair Leslie-Hughes <leslie_alistair(a)hotmail.com> Signed-off-by: Hans Leidekker <hans(a)codeweavers.com> --- dlls/msado15/recordset.c | 20 ++++++++++++++++---- dlls/msado15/tests/msado15.c | 6 ++++++ 2 files changed, 22 insertions(+), 4 deletions(-) diff --git a/dlls/msado15/recordset.c b/dlls/msado15/recordset.c index 1d28d6b957..1f16add4c3 100644 --- a/dlls/msado15/recordset.c +++ b/dlls/msado15/recordset.c @@ -43,6 +43,7 @@ struct recordset LONG allocated; LONG index; VARIANT *data; + CursorLocationEnum cursor_location; }; struct fields @@ -1229,14 +1230,24 @@ static HRESULT WINAPI recordset_CancelBatch( _Recordset *iface, AffectEnum affec static HRESULT WINAPI recordset_get_CursorLocation( _Recordset *iface, CursorLocationEnum *cursor_loc ) { - FIXME( "%p, %p\n", iface, cursor_loc ); - return E_NOTIMPL; + struct recordset *recordset = impl_from_Recordset( iface ); + + TRACE( "%p, %p\n", iface, cursor_loc ); + + *cursor_loc = recordset->cursor_location; + + return S_OK; } static HRESULT WINAPI recordset_put_CursorLocation( _Recordset *iface, CursorLocationEnum cursor_loc ) { - FIXME( "%p, %u\n", iface, cursor_loc ); - return E_NOTIMPL; + struct recordset *recordset = impl_from_Recordset( iface ); + + TRACE( "%p, %u\n", iface, cursor_loc ); + + recordset->cursor_location = cursor_loc; + + return S_OK; } static HRESULT WINAPI recordset_NextRecordset( _Recordset *iface, VARIANT *records_affected, _Recordset **record_set ) @@ -1524,6 +1535,7 @@ HRESULT Recordset_create( void **obj ) recordset->ISupportErrorInfo_iface.lpVtbl = &recordset_supporterrorinfo_vtbl; recordset->refs = 1; recordset->index = -1; + recordset->cursor_location = adUseServer; *obj = &recordset->Recordset_iface; TRACE( "returning iface %p\n", *obj ); diff --git a/dlls/msado15/tests/msado15.c b/dlls/msado15/tests/msado15.c index afdf0dfb6a..917d131063 100644 --- a/dlls/msado15/tests/msado15.c +++ b/dlls/msado15/tests/msado15.c @@ -65,6 +65,7 @@ static void test_Recordset(void) Field *field; LONG refs, count, state; VARIANT missing, val, index; + CursorLocationEnum location; BSTR name; HRESULT hr; @@ -132,6 +133,11 @@ static void test_Recordset(void) ok( hr == S_OK, "got %08x\n", hr ); ok( state == adStateClosed, "got %d\n", state ); + location = -1; + hr = _Recordset_get_CursorLocation( recordset, &location ); + ok( hr == S_OK, "got %08x\n", hr ); + ok( location == adUseServer, "got %d\n", location ); + VariantInit( &missing ); hr = _Recordset_AddNew( recordset, missing, missing ); ok( hr == MAKE_ADO_HRESULT( adErrObjectClosed ), "got %08x\n", hr ); -- 2.20.1

Dec. 16, 2019

Re: [PATCH 2/5] activeds: Implement ADsOpenObject.

by Dmitry Timoshkov

Nikolay Sivov <bunglehead(a)gmail.com> wrote: > > > > > > + hr = > > IADsOpenDSObject_OpenDSObject(adsopen, > > > > (BSTR)path, (BSTR)user, (BSTR)password, reserved, &disp); > > > > > > + if (hr == S_OK) > > > > > You cannot cast to a BSTR, it will corrupt the data structure. > > > > > > > > I depends wheather the callee uses SysStringLen() and friends, in this > > > > particular case it works just fine, there are many other places in the > > code > > > > when it's not necessary to create intermediate copies of the strings. > > > > > > > > > > It does not matter if it works in this case, arguments should use correct > > > types. If we have similar violations somewhere else, those need fixing as > > > well. > > > > That's a pretty incorrect assertion, there's nothing wrong in casting > > BSTR to LPWSTR and vice versa. > > > > What's incorrect about it? You can use BSTR as WCHAR * argument, obviously. > Not the other way around, even if it happens to work. You are contradicting yourself: if it works that means that's correct behaviour. If you are going to further argue, please show the code that won't work for this particular implementation. -- Dmitry.

Dec. 16, 2019

Re: [PATCH 2/5] activeds: Implement ADsOpenObject.

by Nikolay Sivov

On Mon, Dec 16, 2019 at 11:52 AM Dmitry Timoshkov <dmitry(a)baikal.ru> wrote: > Nikolay Sivov <bunglehead(a)gmail.com> wrote: > > > > > > + hr = > IADsOpenDSObject_OpenDSObject(adsopen, > > > (BSTR)path, (BSTR)user, (BSTR)password, reserved, &disp); > > > > > + if (hr == S_OK) > > > > You cannot cast to a BSTR, it will corrupt the data structure. > > > > > > I depends wheather the callee uses SysStringLen() and friends, in this > > > particular case it works just fine, there are many other places in the > code > > > when it's not necessary to create intermediate copies of the strings. > > > > > > > It does not matter if it works in this case, arguments should use correct > > types. If we have similar violations somewhere else, those need fixing as > > well. > > That's a pretty incorrect assertion, there's nothing wrong in casting > BSTR to LPWSTR and vice versa. > What's incorrect about it? You can use BSTR as WCHAR * argument, obviously. Not the other way around, even if it happens to work. > > -- > Dmitry. > >

Dec. 16, 2019