* Viewer node could free image while it is being redrawn, viewer image
buffers now need acquire/release to be accessed as was already the
case for render results.
* The Composite node could free the image buffers outside of a lock,
also causing simultaneous redraw to crash.
* Especially on Windows, re-rendering could crash when drawing an image
that was freed. When RE_RenderInProgress was true it would access the
image buffer and simply return it while it could still contain a pointer
to a render result buffer that was already freed. I don't understand
why this case was there in the first place, so I've removed it.
Possibly fixes bugs #20174, #21418, #21391, #21394.
unrelated changes that ended up being more trouble to commit separate...
- removed BLI_split_dirfile(), was nasty, occasionaly modifying the source string, it could create directories and used the $CWD in some cases. was only used in 2 places in filesel.c, if this gives problems can address without bringing back this function.
renamed BLI_split_dirfile_basic --> BLI_split_dirfile
- view3d_operator_needs_opengl was being called for offscreen render when it wasnt needed.
rename BKE_get_image_export_path() to BKE_rebase_path() and take a filename arg rather then an image.
obj export file copy was also not working because of a missing import.
Note, this is not like GE ffmpg, but Blender Image Texture
display for GLSL materials. Speed can be disappointing,
use smaller images for realtime edits.
Implementation note: this was done by giving each Render a slot number,
and for every slot a new Render will be created. Not sure if this is
ideal, but it ensures that all passes, render info, etc are separate so
you can also compare render layers and passes, in 2.4x only whatever it
was currently displaying was backed up.
- added an option to BLI_convertstringframe and BLI_convertstringframe_range to add digits if not found.
- removed BLI_convertstringframe where its obviously not needed - such as loading movies and sounds.
Various internal fixes, also additional feature - can drag on the histogram to change scale
(0 key to reset).
Also fix [#20844] Color balance node (lift freeze)
do this on drawing instead since SCREEN_OT_animation_step isnt calling the notifier (assume this is to be more efficient?). this isnt slow so is ok to do on drawing.
rename BKE_image_user_calc_imanr to BKE_image_user_calc_frame
Now it's a bit more robust, tagging images with profiles when they're loaded,
which then get interpreted later on by conversion functions. Just Linear RGB
and sRGB profiles at the moment, same as before.
This commit fixes Martin's problem with EXRs and Multilayer images loading/
saving too dark, and it also makes the sequence editor work correctly with it too.
Also fixes:
[#19647] gamma correction with color management is reset when resetting Curve
[#19454] 2.5: Dither does not work when Color management is enabled
- scene.render_data.frame_path(frame=num), returns the output path for rending images of video.
- scene.render_data.file_extension, readonly attribute, gives the extension ".jpg", ".mov" etc
- player support was guessing names, use the above functions to get the actual names used, accounting for #'s replacing numbers.
* Convert all code to use new functions.
* Branch maintainers may want to skip this commit, and run this
conversion script instead, if they use a lot of math functions
in new code:
http://www.pasteall.org/9052/python
* Rendering twice or more could crash layer/pass buttons.
* Compositing would crash while drawing the image.
* Rendering animations could also crash drawing the image.
* Compositing could crash
* Starting to rendering while preview render / compo was
still running could crash.
* Exiting while rendering an animation would not abort the
renderer properly, making Blender seemingly freeze.
* Fixes theoretically possible issue with setting malloc
lock with nested threads.
* Drawing previews inside nodes could crash when those nodes
were being rendered at the same time.
There's more crashes, manipulating the scene data or undo can
still crash, this commit only focuses on making sure the image
buffer and render result access is thread safe.
Implementation:
* Rather than assuming the render result does not get freed
during render, which seems to be quite difficult to do given
that e.g. the compositor is allowed to change the size of
the buffer or output different passes, the render result is
now protected with a read/write mutex.
* The read/write mutex allows multiple readers (and pixel
writers) at the same time, but only allows one writer to
manipulate the data structure.
* Added BKE_image_acquire_ibuf/BKE_image_release_ibuf to access
images being rendered, cases where this is not needed (most
code) can still use BKE_image_get_ibuf.
* The job manager now allows only one rendering job at the same
time, rather than the G.rendering check which was not reliable.
* copied I/O scripts
* copied, modified rna_*_api.c and rna_*.c
I/O scripts not working yet due to slight BPY differences and RNA changes. Will fix them later.
Not merged changes:
* C unit testing integration, because it is clumsy
* scons cross-compiling, can be merged easily later
* The image panels in the image editor and texture buttons
should be more complete now, with working new/open,
refreshes, and using the layout engine.
* Paint panels in image editor are now consistent with the
ones in the 3d view toolbar.
* Curves panel also uses layout engine, and doesn't look
squashed anymore.
A couple of new functions:
BLF_width_and_height - Merge of BLF_width and BLF_height in one call to avoid freetype2 stuff.
BLF_buffer - Set the buffer, size and number of channel.
BLF_buffer_col - Set the text color (the alpha is not used right now).
BLF_draw_buffer - Draw the text in the current buffer.
Also tweak a little the boundbox and draw function to avoid access the freetype2 and use the cache info.
By default the font size is 12, the UI still need work to allow change the font and size.
files but only manipulates paths. It produces both ablsolute and relative paths. COLLADA exporter can now use it.
- re-wrote unit test for it, this is now more compact and readable
- RNA API Image.get_export_path takes a boolean arg to indicate whether a relative or absolute path should be returned. Python scripts
can now use it.
Various fixes for painting, sculpting and particle edit, still
much to be done...
* Move RNA paint and sculpt structs into rna_sculpt_paint.c,
* Added Particle Edit RNA.
* Some tweaks to existing Paint RNA.
* Put texture paint and particle edit object in context.
* Fix some errors in the brush layout, properly doing None
checks, fixing some wrong property identifiers.
* Added tool enum for texture paint and particle edit in panels.
* Allow editing brush textures in the texture buttons, still with
a stupid toggle, ideas for how to make the connection better are
welcome.
- freeing strings returned by RNA struct functions in RNA_parameter_list_free
Unit tests:
- check that BKE_export_image actually creates a file. This test is becoming dangerous: it creates and deletes files under /tmp.
Having written this complicated test function I now realize it's much easier to write tests in a scripted language, which gives more
freedom in expressions and need not be compiled.
- replaced BKE_copy_images with BKE_export_image, now it handles only one image at a time, this is better since for exporters it is
easier to export one image at a time writing new image path to a file
- exposing BKE_export_image in RNA as Image.export, interestingly, RNA allowed me to define a function with PROP_STRING return type
although it doesn't free memory, will fix that in the next commit
- removed bpy.util.copy_images
Unit tests:
- re-wrote a test for BKE_export_image, it's more compact now
- moved unit tests to the creator module to avoid another executable, now running tests with `blender --test`
- as before, unit tests are built only if WITH_BF_UNIT_TEST is non 0
Unit tests:
- add a check that BKE_copy_images produces NULL filepath for images having type other than IMA_TYPE_IMAGE
- also expect NULL filepath for images having empty filename
Enhanced BKE_copy_images so the tests pass.
* UI layout for scene buttons has quite some changes, I tried to
better organize things according to the pipeline, and also showing
important properties by default, and collapsing less important ones.
Some changes compared to 2.4x:
* Panorama is now a Camera property.
* Sequence and Compositing are now enabled by default, but will only
do something when there is a node tree using nodes, or a strip in the
sequence editor.
* Enabling Full Sample now automatically enables Save Buffers too.
* Stamp option to include info in file is removed, it now simply always
does this if one of the stamp infos is enabled.
* Xvid, H.264 and Ogg Theora are now directly in the file format menu,
but still using FFMPEG. Unfortunately Ogg is broken at the moment
(also in 2.4x), so that's disabled. And Xvid crashes on 64bit linux,
maybe solvable by upgrading extern/xvidcore/, using ubuntu libs makes
it work.
* Organized file format menu by image/movie types.
Added:
* Render layers RNA wrapped, operatorized, layouted.
* FFMPEG format/codec options are now working.
Defaults changed:
* Compositing & Sequencer enabled.
* Tiles set to 8x8.
* Time/Date/Frame/Scene/Camera/Filename enabled for stamp.
Render usability:
- Option back to render to imagewindow, or fullscreen.
The latter is default. Setting is stored in Scene.
- Added button in output panel, the option "to new window" will follow!
- F11 again toggles render view
(moved MS Windows "full screen" to shift+F11 for now)
I needed this to make sure that BKE_copy_images works properly,
probably will be useful in future.
Using Check framework (http://check.sourceforge.net/doc/check.html/index.html).
WITH_BF_UNIT_TEST option builds 'alltest' program under [BUILDDIR]/bin,
which, when executed, runs unit tests, currently only 1.
Example output:
----------------------------------------------------------------------
Running suite(s): Image
0%: Checks: 1, Failures: 1, Errors: 0
tests/alltest.c:74:F:Core:test_copy_images:0: Expected //bar/image.png to be translated to /tmp/bar/image.png, got /tmp/bar/image.pn.
----------------------------------------------------------------------
Spent lots of time (a couple of days actually :) to figure out how to
link the test program with Blender libraries. As it turned out there
are circular dependencies among Blender libraries. GCC by default
doesn't expect circular dependencies - dependant libs should precede
libs they depend on.
The magical --start-group linker option helped to solve this
(http://stephane.carrez.free.fr/doc/ld_2.html#IDX122).
Also:
- added bpy.util module. bpy.sys.* functions will move here later
- added bpy.util.copy_images that uses BKE_copy_images
- export_obj.py uses bpy.util.copy_images
Image Window
* Unpack operator now works.
* Some small layout code tweaks.
Info Window Header
* Moved to python UI code.
* template_running_jobs, template_operator_search added.
* Ported external data operators: pack/unpack all, make
paths relative/absolute, find/report missing files.
Also
* Report RPT_INFO too, not only warnings and errors.
* Run UI handle functions after RNA and Operators.
* Rename particle system add/remove operators, to not
include "slot", that's only there for materials because
that's what they are called now in RNA.
