Some points of the refactor not sorted by importance:
* Fixed immutability of readers and factories (there are exceptions...)
* Fixed copy constructors and = operators
* Removed messaging system
* Removed reader types
* Added const where possible
* Using initalisers when possible
* Avoided use of pointers when possible
* Removed AUD_NEW and AUD_DELETE macros
* Removed useless NULL pointer checks
* Fixed exception catching
* Fixed some yet unknown bugs
* Lots of other stuff
* Renamed get_c_device() to device()
* Made the threshold parameter of Sound.square() optional
* Added a sample rate parameter for Sound.sine()
* Example updates
Error was cased by undefined UINT64_C in stdint.h for c++
programs. As I understand from ffmpeg mailing list the simpliest
solution is to define __STDC_CONSTANT_MACROS before including
ffmpeg libraries. There already was some amout of such definitions in
the code and I've tested this with newer and older ffmpeg
libraries -- this should work and should be safe, but maybe
there is better way of fixing this problem.
* Division by zero fix for TNT SVD code.
* Sound fix, in case ffmpeg decode fails, don't use the samples.
* Fix for incorrect bounds of transformed objects in new raytracing code.
* Gave memory arena's a name used for allocations for easier memory
usage debugging.
* Dupligroup no_draw option was using layers but not restrict view/render
setting. (not a bugfix exactly but would do display list context switching
while drawing for no reason).
* Fix objects instanced on hair particles not giving consistent results
when the object is transformed.
* New math functions: madd_v4_v4fl, len_squared_v3v3, interp_v4_v4v4v4,
mul_v4_m4v4, SH and form factor functions, box_minmax_bounds_m4.
* mul_m4_m4m4 and mul_m3_m3m3 now accept the same pointers for multiple
arguments.
* endjob callback for WM jobs system.
* Geometry node uv/color layer now has search list/autocomplete.
* Various small buildsystem tweaks, not strictly needed yet in trunk.
- Python script to crossfade two sound strips in the sequencer
- Fix for the libsamplerate code producing awful audio when resampling sequencer strips
- Changed default resampler to a linear one (as temporary workaround for a bug that seems to be in the samplerate code)
- Fix for the OpenAL device to return a more accurate playback position
- recode of the whole sequencer audio handling
- encode audio flag removed, instead you choose None as audio codec, added None for video codec too
- ffmpeg formats/codecs: enabled: theora, ogg, vorbis; added: matroska, flac (not working, who can fix?), mp3, wav
- sequencer wave drawing
- volume animation (now also working when mixing down to a file!)
- made sequencer strip position and length values unanimatable
- transform default scale was too hight, calls to random were inconsistant. (fault of own modif's)
- cmake openal include was added twice on recent commit.
* Refactored the whole audaspace library to use float as sample format over all readers.
* Added new Readers like the linear resampler, envelope, lowpass, highpass and butterworth.
* Note: The butterworth filter isn't working correctly, some bug in there... Maybe also true for the envelope.
* Added a sound to f-curve operator that behaves mostly like the soundtracker script of technoestupido.
* Added GLEW_STATIC where necessary to make Blender compile again
* First attempt at compiling blenderplayer again - compiles, but doesn't link yet
* removed deprecated SND_ functions from blenderplayer
* Threading buxfix letting MSVC Debug builds crash because of corrupted std::lists
* Adopted two property ranges
* Changed the mixdown volume to set the device volume instead of the volume of every sound.
I also removed the private redefinition of m_logicmgr in SCA_BasicEventManager, which was already defined protected in the parent class SCA_EventManager and thus caused a bug letting GE crash here because of an uninitialized pointer.
* bring back 'player' libtype, after investigation with ideasman.
scons/mingw works nicely, for some reason msvc fails to link still, will look further into it.
Note that With libsndfile also need libflac and libogg here (Linux), right
now I just add this two librarys to the NAN_SNDFILELIBS, but maybe
it's better split this ? (NAN_FLAC/NAN_OGG)
I hoped for it to resolve the bug of loading sounds with relative paths didn't work, but Main.name isn't set before the sounds are loaded, so the bug resists!
Someone who is into file loading should please fix this!
This add support for: sdl, ffmpeg, openal, jack and sndfile.
We have new options:
WITH_OPENAL, default false
WITH_JACK, default false
WITH_SNDFILE, default false
and with this:
NAN_OPENAL, default /usr
NAN_JACK, default /usr
NAN_JACKCFLAGS, default -I$(NAN_JACK)/include/jack
NAN_JACKLIBS, default $(NAN_JACK)/lib/libjack.a
NAN_SNDFILE, default /usr
NAN_SNDFILECFLAGS, default -I$(NAN_SNDFILE)/include
NAN_SNDFILELIBS, default $(NAN_SNDFILE)/lib/libsndfile.
Also add two new option on source/Makefile for jack and sndfile libs,
but only for linux, so let me know for other OS.
* add jack support for audio system.
Note that for OSX jack support is still off by default, since
I'm not sure about what goes where. OSX maintainers, please check and fix.