39d6f64885
d3d10core: Add a separate function for d3d10_geometry_shader initialization.
2009-10-30 11:29:01 +01:00
bf10dbe8a6
wined3d: Map D3DFMT_D24S8 to WINED3DFMT_D24_UNORM_S8_UINT.
...
The internal order of the depth and stencil bits is unspecified since the
format isn't lockable.
2009-10-30 11:28:52 +01:00
fbae08672b
wined3d: It's not an error to request multisampling.
2009-10-30 11:28:35 +01:00
a5f87a4731
ole32: Rename StorageImpl_CopyTo to StorageBaseImpl_CopyTo.
2009-10-30 11:28:22 +01:00
39e35cd7b4
ole32: Don't call internal functions from OLECONVERT_WriteOLE20ToBuffer.
2009-10-30 11:28:14 +01:00
6a36025713
ole32: Don't call internal functions from OLECONVERT_GetOLE20FromOLE10.
2009-10-30 11:28:07 +01:00
4f6e2453e9
ole32: Rename StorageImpl_CreateStorage to StorageBaseImpl_CreateStorage.
...
We don't just use the function for StorageImpl.
2009-10-30 11:27:43 +01:00
f82924ffb9
ole32: Take a StorageBaseImpl in updatePropertyChain.
...
The function is not always called with a top-level storage object, and the
signature should reflect that.
2009-10-30 11:27:29 +01:00
13b9666443
ole32: Convert adjustPropertyChain into real binary tree removal.
...
Finding the parent of a node in a binary tree is a detail that should
be handled inside the function.
2009-10-30 11:27:07 +01:00
fd993b7d9d
ole32: Remove directory entries from the tree before we free them.
...
The directory entry should still be valid as long as it's in the tree.
2009-10-30 11:26:50 +01:00
07b735682b
crypt32: Check CA certificates for the enhanced key usage extension.
2009-10-30 11:26:39 +01:00
60770fb011
crypt32: Only permit v1 or v2 CA certificates without a key usage extension if they're installed locally.
2009-10-30 11:26:30 +01:00
7b0297769d
crypt32: Use a helper function to find an existing cert by hash.
2009-10-30 11:26:21 +01:00
77472187c9
crypt32: Add key usage extension to chain4_0.
2009-10-30 11:26:15 +01:00
33a6235053
crypt32: Only permit v1 or v2 CA certificates without a basic constraints extension if they're installed locally.
2009-10-30 11:26:06 +01:00
552fec4002
crypt32: Add basic constraints to chain quality selection algorithm.
2009-10-30 11:24:23 +01:00
c310637f4f
crypt32: Remove redundant if clause.
2009-10-30 11:24:10 +01:00
9a13e1c70b
crypt32: Add basic constraints to chain15_0.
2009-10-30 11:23:58 +01:00
118374c081
crypt32: Add basic constraints to chain14_0.
2009-10-30 11:23:53 +01:00
0bd67b4c6f
crypt32: Add basic constraints and key usage to chain12_0.
2009-10-30 11:23:47 +01:00
002439e2f0
crypt32: Add basic constraints and key usage to chain8_0.
2009-10-30 11:23:41 +01:00
4557a8705b
crypt32: Add basic constraints and key usage to chain5_0.
2009-10-30 11:23:35 +01:00
86d6177215
crypt32: Add basic constraints and key usage to chain2_0.
2009-10-30 11:23:27 +01:00
6bf0e52011
crypt32: Add basic constraints and key usage to chain0_0.
2009-10-30 11:23:22 +01:00
36e56ad1aa
kernel32/tests: Fix FormatMessage test failures on various Windows versions.
2009-10-30 11:03:39 +01:00
0e26e741f7
kernel32: Replicate the MS bug with va_list argument counting in FormatMessage.
2009-10-30 11:03:39 +01:00
24a8e33fb6
kernel32/tests: Use __ms_va_list in the FormatMessage test.
2009-10-30 11:03:39 +01:00
b353401513
kernel32: Fix handling of width and precision arguments and remove assumptions about va_list implementation in FormatMessage.
2009-10-30 11:03:39 +01:00
100131d4ca
kernel32: Propagate the error result from RtlFindMessage in FormatMessage.
2009-10-30 11:03:39 +01:00
b381958f78
kernel32: Reimplement the formatting of inserts in FormatMessageW.
2009-10-30 11:03:21 +01:00
3307563a6c
kernel32: Reimplement the formatting of inserts in FormatMessageA.
2009-10-30 11:01:51 +01:00
1be99033b1
kernel32/tests: Add more tests for FormatMessageA.
2009-10-29 17:07:53 +01:00
290c5de8d9
wia: Add rudimentary wia headers.
2009-10-29 15:49:19 +01:00
0513f3c4e6
msi: Add support for msidbControlAttributesProgress95.
2009-10-29 15:48:38 +01:00
df8780356b
msi: Limit the progress to its maximal value, use MulDiv for better results.
2009-10-29 15:48:27 +01:00
37f3f3d92a
ole32: Write the new data in getFreeProperty and rename the function.
...
This is to avoid calling WriteProperty with an index to a directory entry
that has not yet been reserved in the file. We reserve the entry by writing
the new data to it.
2009-10-29 15:48:12 +01:00
20dd1886d5
ole32: Split the data conversion functions out of WriteProperty.
2009-10-29 15:47:55 +01:00
ac9d77cd10
ole32: Use the raw directory entry functions in getFreeProperty.
...
The meaning of the "index" argument in Read/WriteProperty will likely
change in the future, but getFreeProperty must work with real indexes
into the file's real directory stream.
2009-10-29 15:47:44 +01:00
e71c696b51
ole32: Move the low-level functionality of WriteProperty to a new function.
2009-10-29 15:47:30 +01:00
56622de951
ole32: Move the low-level functionality of ReadProperty to a new function.
2009-10-29 15:47:20 +01:00
5d550cdc45
ole32: Further simplify adjustPropertyChain and fold in findPlaceholder.
...
The logic is unchanged from the original version.
2009-10-29 15:47:03 +01:00
a7e69339dc
ole32: Reduce duplicated code in adjustPropertyChain.
...
Most of the work is the same regardless of relation type, so bring
those parts out of the relation type test.
2009-10-29 15:46:16 +01:00
d1f9561747
include: Add rowchg.idl.
2009-10-29 13:51:52 +01:00
4f6c58dcd8
include: Add row.idl.
2009-10-29 13:51:42 +01:00
846cc2b157
winhttp: Implement WinHttpQueryHeaders(WINHTTP_QUERY_VERSION).
2009-10-29 13:51:20 +01:00
39b1dbed7a
winhttp: Accept empty headers parameter in WinHttpSendRequest.
2009-10-29 13:51:14 +01:00
ebe6dabe27
ws2_32: Add a test for GetAddrInfoW.
2009-10-29 13:51:05 +01:00
140eede67d
ws2_32: Avoid accessing uninitialized memory in GetAddrInfoW.
...
Found by valgrind.
2009-10-29 13:50:58 +01:00
5e18596fc5
configure: preloader is Linux specific, don't build it elsewhere.
2009-10-29 13:10:40 +01:00
052d955567
wined3d: Don't depend on uninitialized data in the CreateSurface() thunks (Valgrind).
...
Perhaps IDirectDraw7_CreateSurface() should also always set "pSurface7" to
NULL on failure, but at least for our internal functions that's a silly thing
to rely on.
2009-10-29 13:09:45 +01:00
Henri Verbeet