- Do not cast RGBQUAD to PALETTEENTRY or the other way around, you idiot.
- Report success even when we're asked fo 0 ScanLines in NtGdiGetDIBitsInternal.
svn path=/branches/reactos-yarotows/; revision=48417
- NtGdiGetBitmapBits : Bits we're given are 16 bits aligned.
- NtGdiGetDIBitsInternal : we must not stretch, for this there is NtGdiStretchDIBitsInternal. Use lower level functions to do the work, there is no need to create HDCs etc...
Now icons are back.
svn path=/branches/reactos-yarotows/; revision=48416
- Use correct offset in UserLoadImage, create the bitmap using display DC.
- BITMAP_GetObject : Report BI_BITFIELDS compression when we should.
- Use a 0 compatible DC when none is given in NtGdiCreateDIBitmapInternal.
This fixes some wine tests. Enjoy.
svn path=/branches/reactos-yarotows/; revision=48415
- Apply a better fix for correctly report 16 bits alignment for DDBs, and keep them 32 bits aligned as it should be.
- Reapply Pigglesworth patch, which was correct since mine was not.
Dedicated to tkreuzer, PigglesWorth and lassy, the three guys still on IRC to watch my commits at 3:30 AM.
svn path=/branches/reactos-yarotows/; revision=48408
- DDB are 16 bits aligned.
- Reset hdc field of the unselected bitmap.
- Lock it too, so it's not messed with when we unselect it.
- Move Pattern creation of IntGdiCreateDIBBrush to DIB_CreateDIBSection.
svn path=/branches/reactos-yarotows/; revision=48406
- define REGISTERCALL in order to replace __attribute__((regparm(3))) and fix compilation with msvc. It was used only as an optimization so it is safe to be defined as FASTCALL for msvc. For gcc it is still defined as __attribute__((regparm(3)))
svn path=/trunk/; revision=48397
- Revert unwanted change and get back to PAL_BGR for default DIB palette mode.
Now yarotows looks like an OS again.
svn path=/branches/reactos-yarotows/; revision=48390
- Finally get rid of ProbeAndConvertToBitmapV5Info
- Rewrite renderBITMAPfromDIB, and a good bunch of DIB related functions accordingly.
- Rewrite BITMAP_CopyBitmap into something simpler.
- Use already existing DIB functions in IntGdiCreateDIBBrush
- Use DIB sections in NtGdiStretchDIBitsInternal and NtGdiSetDIBits.
- Use Bitmap hdc for NtGdiGetDIBitsInternal if there is one.
svn path=/branches/reactos-yarotows/; revision=48389
- Update Wine-specific headers to Wine-1.3.
- Remove unneeded 16 bit header, and add a bit of stuff into another legacy header.
- Indicate in config.h that there is a freetype library present.
svn path=/trunk/; revision=48388
- Get back to old/compatible way of declaring dummy unions and structs for IMAGE_RESOURCE_DIRECTORY_ENTRY. Also, second nameless structure is called DUMMYSTRUCTNAME3 to be compatible with Wine's code which, for some reason, expects WORD Id wrapped into yet another struct, depending on endiannes.
svn path=/trunk/; revision=48384
- Patch by Carlo Bramini reviewed by lassy:
The CF must be tested when reading time because the legacy DS12x87
chip shows a condition where it's updating its time registers and if it's the
case then the call should be retried later. Now the time is valid when both CF flags for date and time are good at the same time.
It's better to send the correct value of CF with the proper field to the REGS structure instead of hoping that the flag won't be changed by compiled code.
In the Int386() function I did the fix for acquiring the status flags to send to the software interrupt.
I also think it is also a good idea to always clear the direction flag before
doing movsb/movsw/movsd in this function. Since we are working with DOS stuff,
the state of the direction register could be unknown, who knows what happens
after those INT opcodes. Afterall I don't think that adding these two "cld"
would be a big problem.
See issue #2786 for more details.
svn path=/trunk/; revision=48383
- Move all includes into a common header file.
- Create a (still incomplete) well-known-sid lookup table and use it in LsarLookupSids.
- Add missing file headers.
svn path=/trunk/; revision=48378
- keep the hacky declaration of SHANDLE_PTR and make it work with msvc as well
- Make BEGIN_MSG_MAP compatible with msvc
svn path=/trunk/; revision=48376
- Cast Colors to RGBQUAD in EngCreateBitmap, since that's what they are.
- Copy selected palette of the DC when creating a DIBSection with DIB_PAL_COLORS.
- Do not try to get Colors in NtGdiGetDIBitsInternal when bpp == 0, this doesn't make any sense.
- Get rid of SURFACE::dsBitfields and SURFACE::biClrUsed.
- Use RGB as default palette mod when creating the DIB Palette.
svn path=/branches/reactos-yarotows/; revision=48369
- Make the caption the same as the app title so that bring to focus can work as expected in Slovak and other languages as well. Add a note for translators.
See issue #5533 for more details.
svn path=/trunk/; revision=48367
- rewrite UserLoadImage so that it uses information from the BITMAPFILEHEADER and probes the right buffer.
svn path=/branches/reactos-yarotows/; revision=48364
- Rewrite NtGdiGetDIBitsInternal, with WINE as a reference.
- Get back DIB Section creation to classic BITMAPINFO.
This si the beginnig of the end for all this BITMAPV5INFO stuff.
It is horrible, BITMAPCOREINFO->BITMAPINFO is already made in user mode, and V4 and V5 features are not used so often.
svn path=/branches/reactos-yarotows/; revision=48359
-define SHANDLE_PTR
[browseui]
-do not define SHANDLE_PTR locally. Using ms psdk isn't the same with compiling with msvc
svn path=/trunk/; revision=48355