ColorConversion: Add cast to silence -Wclass-memaccess.

interface: add colorspaces support for RGB48 and RGBA64

these colorspaces are packed as RGB or RGBA, not BGR or BGRA.
RGB48_BIG and RGBA64 only differ in the endianess of the channel the 2-byte value.

this is a big difference with RGB24_BIG and RGBA32_BIG, in which case _BIG
means the order is RGB (BGR) and not BGR (BGRA).
BGR48, BGRA64 could indeed be added, if needed.

I chose 0x11 and 0x12 arbitrarily, but given the order of channels 0x1011
and 0x1012 might make more sense. This would mean using another bit for "real"
bigendian colorspaces.

Only the color conversion to 32-bits is implemented.

Tested with the RAWTranslator modified to output 16bpp with success.

Found some references in enum AVPixelFormat in libavutil/pixfmt.h.

Change-Id: I4b023dec85d01f1e63e1b053139e5bb5d263a0e0
Reviewed-on: https://review.haiku-os.org/c/haiku/+/4468
Tested-by: Commit checker robot <no-reply+buildbot@haiku-os.org>
Reviewed-by: Adrien Destugues <pulkomandy@gmail.com>

kernel/OS.h: Include sys/types.h instead of pthread.h.

This avoids polluting the namespace significantly. Also adjust
all files which depended on this behavior to include pthread.h
directly.

Fix drawing of B_GRAY1 bitmaps.

- Colors were reversed
- Padding was not handled properly because of a roundeing error

Add a test that shows the issue (behavior confirmed against BeOS)

Change-Id: I4c6e954fb6bdab92ad4e0e96897e78b26eb4727b
Reviewed-on: https://review.haiku-os.org/c/1025
Reviewed-by: Stephan Aßmus <superstippi@gmx.de>

interface: Remove some unreachable "break"s.

Spotted by Clang.

* Introduced PaletteConverter::IndexForRGBA32() which checks if the alpha
component of a 32 bit color is below 128 and returns the special CMAP8 index
for indicating transparency.
* Changed the WriteCMAP() color conversion function to expext a 32bit RGBA
value and use IndexForRGBA32().
* Adapted B_CMAP8 target color space case of ConvertBits() to the new
semantics.

This allows transparency in bitmaps when converting B_RGBA32 bitmaps to
B_CMAP8, tested only for this case and working as expected.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@36549 a95241bf-73f2-0310-859d-f6bbb57e9c96

Use pthread_once() to guard the initialization of sPaletteConverter. This
changes the semantics slightly: Previously after a failed initialization
another invocation of InitializeDefault() could theoretically initialize the
converter. Since the only error conditions are out of memory and broken app
server connection, this shouldn't really matter, though.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@34383 a95241bf-73f2-0310-859d-f6bbb57e9c96

Should fix bug #3401. Put the parenthesis at the wrong spot resulting in wrong offsets.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@29125 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Resolve further warnings on GCC4.
* Enable -Werror on GCC4 builds as well (limited to the same selected targets).


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@29118 a95241bf-73f2-0310-859d-f6bbb57e9c96

The system palette was defined in two files. Moved to a private header,
included by the two.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22660 a95241bf-73f2-0310-859d-f6bbb57e9c96

Static variables should have the s prefix.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22509 a95241bf-73f2-0310-859d-f6bbb57e9c96

PaletteConverter now exports a static InitializeDefault() method, which
can be used (and is, by _init_interface_kit()) to initialize the ...
er... default instance of
PaletteConverter using the system color map, thus avoid building the
list of colors, which takes some time. Fixes bug #505.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22508 a95241bf-73f2-0310-859d-f6bbb57e9c96

ConvertBits() now checks the out of range case before reading/writing the buffers. Previously it was possible to read/write one line after the buffers end because the values were not checked after line advancement. This fixes the crashing bug #850. It however reveals another bug that brings the app_server into an endless loop when using the TextFader screensaver.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@19626 a95241bf-73f2-0310-859d-f6bbb57e9c96

Added missing return statement

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@18326 a95241bf-73f2-0310-859d-f6bbb57e9c96

Some more GCC 4 and PPC fixes.
* Mesa doesn't compile yet, as some PPC specific stuff seems to be
missing, Philippe?
* Cortex and some other stuff has been marked x86-only, although
it's more of a "GCC 2.95.3"-only.
* I'm not sure if it's a bug in GCC 4, or if that's what the C
standard demands, but sizeof(some_type::some_field) is not
valid anymore :-/


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@17515 a95241bf-73f2-0310-859d-f6bbb57e9c96

* BPrivate::ConvertBits() did not take the source or destination bits length
into account when copying to the same color space (new code isn't that fast,
but it works).
* BBitmap::ImportBits() did not take the specified offset into account at all.
This fixes bug #313.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@16812 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Fixed ServerBitmap::ImportBits() to use the correct width/height
* Extended ConvertBits() to properly handle negative offsets and overlapping lines
* Implemented blitting the software cursor to the bitmap returned from ReadBitmap()

Note: In the future we will have to directly use the final graphics buffer for ReadBitmap() if we want DirectWindow output too (R5 includes it). I don't know how R5 handles the hardware cursor though.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@16766 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Fixed a swapping bug for 16 bit colorspaces
* Added some comments
* Added some checking to avoid noop shifts
* Added buffer length checks
* Implemented (as Stephan suggested) a version of ConvertBits() that takes offsets.
This new version allows to move a region of the source into a region (possibly not at the same point) on the dest while converting colorspaces on the fly.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@16592 a95241bf-73f2-0310-859d-f6bbb57e9c96

Wrote a new color conversion engine. It's 2-5 times faster than the old one (depending on colorspaces) and about half in size.
I've put it into ColorConversion.cpp and moved PaletteConverter there too, cleaning up Bitmap.cpp.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@16580 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Introduced PaletteConverter::IndexForRGBA32() which checks if the alpha
component of a 32 bit color is below 128 and returns the special CMAP8 index
for indicating transparency.
* Changed the WriteCMAP() color conversion function to expext a 32bit RGBA
value and use IndexForRGBA32().
* Adapted B_CMAP8 target color space case of ConvertBits() to the new
semantics.

This allows transparency in bitmaps when converting B_RGBA32 bitmaps to
B_CMAP8, tested only for this case and working as expected.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@36549 a95241bf-73f2-0310-859d-f6bbb57e9c96

Use pthread_once() to guard the initialization of sPaletteConverter. This
changes the semantics slightly: Previously after a failed initialization
another invocation of InitializeDefault() could theoretically initialize the
converter. Since the only error conditions are out of memory and broken app
server connection, this shouldn't really matter, though.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@34383 a95241bf-73f2-0310-859d-f6bbb57e9c96

Should fix bug #3401. Put the parenthesis at the wrong spot resulting in wrong offsets.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@29125 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Resolve further warnings on GCC4.
* Enable -Werror on GCC4 builds as well (limited to the same selected targets).


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@29118 a95241bf-73f2-0310-859d-f6bbb57e9c96

The system palette was defined in two files. Moved to a private header,
included by the two.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22660 a95241bf-73f2-0310-859d-f6bbb57e9c96

Static variables should have the s prefix.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22509 a95241bf-73f2-0310-859d-f6bbb57e9c96

PaletteConverter now exports a static InitializeDefault() method, which
can be used (and is, by _init_interface_kit()) to initialize the ...
er... default instance of
PaletteConverter using the system color map, thus avoid building the
list of colors, which takes some time. Fixes bug #505.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22508 a95241bf-73f2-0310-859d-f6bbb57e9c96

ConvertBits() now checks the out of range case before reading/writing the buffers. Previously it was possible to read/write one line after the buffers end because the values were not checked after line advancement. This fixes the crashing bug #850. It however reveals another bug that brings the app_server into an endless loop when using the TextFader screensaver.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@19626 a95241bf-73f2-0310-859d-f6bbb57e9c96

Added missing return statement

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@18326 a95241bf-73f2-0310-859d-f6bbb57e9c96

Some more GCC 4 and PPC fixes.
* Mesa doesn't compile yet, as some PPC specific stuff seems to be
missing, Philippe?
* Cortex and some other stuff has been marked x86-only, although
it's more of a "GCC 2.95.3"-only.
* I'm not sure if it's a bug in GCC 4, or if that's what the C
standard demands, but sizeof(some_type::some_field) is not
valid anymore :-/


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@17515 a95241bf-73f2-0310-859d-f6bbb57e9c96

* BPrivate::ConvertBits() did not take the source or destination bits length
into account when copying to the same color space (new code isn't that fast,
but it works).
* BBitmap::ImportBits() did not take the specified offset into account at all.
This fixes bug #313.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@16812 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Fixed ServerBitmap::ImportBits() to use the correct width/height
* Extended ConvertBits() to properly handle negative offsets and overlapping lines
* Implemented blitting the software cursor to the bitmap returned from ReadBitmap()

Note: In the future we will have to directly use the final graphics buffer for ReadBitmap() if we want DirectWindow output too (R5 includes it). I don't know how R5 handles the hardware cursor though.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@16766 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Fixed a swapping bug for 16 bit colorspaces
* Added some comments
* Added some checking to avoid noop shifts
* Added buffer length checks
* Implemented (as Stephan suggested) a version of ConvertBits() that takes offsets.
This new version allows to move a region of the source into a region (possibly not at the same point) on the dest while converting colorspaces on the fly.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@16592 a95241bf-73f2-0310-859d-f6bbb57e9c96

Wrote a new color conversion engine. It's 2-5 times faster than the old one (depending on colorspaces) and about half in size.
I've put it into ColorConversion.cpp and moved PaletteConverter there too, cleaning up Bitmap.cpp.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@16580 a95241bf-73f2-0310-859d-f6bbb57e9c96