- Fix BRUSHOBJ_pvGetRbrush, it's supposed to provide a realization, when there is none. For this reason call ENRUSHOBJ_bRealizeBrush, that will call Eng/DrvRealizeBrush to create a realisation. The old implementation never did this and I wonder how 3rd party display drivers were expected to work correctly without this.
- Implement EBRUSHOBJ_pvGetEngBrush working similar to BRUSHOBJ_pvGetRbrush, but providing the engbrush as GDI's realization. It's currently implemented as a handle to a translated pattern and not used anywhere yet.
- Rename EBRUSHOBJ_vUnrealizeBrus to EBRUSHOBJ_vCleanup. This function now needs to be called to free the resources for a BRUSHOBJECT that GDI or a driver might have allocated. This is done in DC_Cleanup before deleting the DC and in EBRUSHOBJ_vUpdate before newly initializing it.
- Get rid of the EBRUSHOBJ's XlateObject, instead for now create new XLATEOBJ's whenever needed. Don't worry these will be gone, as soon as we make use of the engbrush.
- Pass a pointer to the DC to EBRUSHOBJ_vInit, as this allows us to save stuff like the back and fore color.
- Fix a bitmap leak: CallDibStretchBlt returned before releasing the reference to the pattern surface.
- Some FASTCALL -> NTAPI changes
- tested with VBE, VMWare-Tools and VBox driver
- I hope I didn't forget anything
svn path=/trunk/; revision=41855
- Rubber works as color replacer when used with right mouse button
(mouse handling code reorganized)
- Bug occuring when painting to neg. coordinates finally fixed
- Polygon-tool added
- Bezier-tool added
svn path=/trunk/; revision=41851
- Implement InitXlateImpl(), that currently does ... nothing :)
- Don't pass a palette type to IntEngCreateXlate, when already passing a valid palette.
svn path=/trunk/; revision=41847
- The pin creation parameters are now placed right after the reference string. Handle this case in KsValidateConnectRequest
- Only store the passed create items, no need to copy the passed create items. This is required for enabling dynamic audio devices
- Fix copying of the object class in KsAllocateObjectHeader
- Check for KSCREATE_ITEM_IRP_STORAGE flag in KsCreate
[PORTCLS]
- Reduce initial audio buffer to about 1/3 second which greatly improves performance
[SYSAUDIO]
- Remove concept of audio subdevices
- Remove imported KsCreatePin
- Replace OpenDevice loop hack
svn path=/trunk/; revision=41842
- Add a new file which will hold the executive layer above the pool allocator (the Ex* routines).
- This should eventually be moved to \ex.
- Note that the current ReactOS pool Ex* routines are in \mm, which is dead wrong.
- Define the POOL_DESCRIPTOR and POOL_HEADER structures for the NT 5.2 pool.
- Define, create, and implement the routine to initialize a pool vector.
- For now, we assume you want the nonpaged pool vector/descriptor.
- Call this routine from the ARM3 initialization routine.
- No changes to system behavior other than the bugfix above.
svn path=/trunk/; revision=41824
Notorious ass-clown Arch Blackmann forgot to pop the trylevel that handled an exception. Subsequent throws would keep being caught by the same trylevel over and over again, and infinite loops are indeed possible. Eat shit and die, Arch
If test_bug_4663 test in the PSEH2 test suite is actually representative of the bug, the bug itself may have been fixed
svn path=/trunk/; revision=41815
Don't use ok() when we mean trace()
Fix the stack after corruption of the registration list
Revised figures: "only" FOURTY of ninety-three tests fail with a corrupted registration list
svn path=/trunk/; revision=41811
Test case for bug #4663 (still unsolved)
New sanity check for correct unlinking of the frame. !!!EIGHTY TESTS OUT OF NINETY-THREE CURRENTLY FAIL THIS!!!
svn path=/trunk/; revision=41810
- Return immediately if any bitmap size is smaller than zero
- Also check calculated input bitmap size against source size
- Remove outdated FIXME
- Fixes three gdi32 bitmap winetests
svn path=/trunk/; revision=41800
- Fix a printf case, where the exponent is zero
- Implement exponent increment/decrement for printf in scientific notation
My changes <grschneider AT gmail DOT com>:
- Change some lessequal/greaterequal to equal
- Exchange OR with AND operation
- Modify patch identation to match current code, remove FIXME
- Skipped the ecvt part of the patch (not needed, sprint has the bugs)
- Fixes five msvcrt printf winetests, see bug #4584 for more information
svn path=/trunk/; revision=41799
