Bugfix for [#22284] Blender cursor gets stuck in the timeline when scrubbing (jack transport).
Dirty hack fix for:
* [#22366] Cutting audio and meta strips with audio does not actually cut audio
* [#22639] Audio not clipped to meta bounds
Also fixed a seemingly symptomless bug in sequencer_edit.c
mostly the startup.blend was trailing behind. Also renamed B.blend.c.
* Lamp shadow buffer was Classical instead of Classical Halfway.
* Point Lamp was named "Spot".
* Render resolution is 50% 1080p.
* Scene and material bake/use tangent space normal maps.
* Remove empty text datablock.
* Enable auto ray bias on material.
* Change default material diffuse color to match new material.
* Mist start/depth from 0/0 to 5/25 so it does something.
* AO uses Add instead of Multiply.
* Change world colors for new world same as startup.blend.
* Default cube rotation was 0,-0,0 now 0,0,0.
* Enable relative/filter/hide files in user preferences.
* filebrowser now should respect User Preference for using relative paths
* Also set remap_relative default to 'True' for saving .blend file
(With 'hide dot files' set and 'relative paths' set in the User Preferences this should now behave as in the suggestion made by brecht - in the future we could remove the user preference and just use the default.)
- Okey sets the border in the display.
- Okey resets the frame offset in the sequencer timeline.
- ghost icon in the header can enable/disable.
- frame offset can be relative or absolute (lock icon)
Not very happy that this commit adds a call to BKE_animsys_evaluate_animdata(scene, ...) in do_build_seq_array_recursively()
without this the offset frames dont have fcurves applied.
Though we will need something like this for prefetch frames to work too.
it stores it and sets it later when the double click event is handled.
Decided to not reuse prevx but made prevclickx, because there may be
multiple mousemoves between the two clicks, and prevx is already used
for some other tricky things.
- blend load/save uses os message.
- image load gives os message. (remove check for slash at end of line, just let the os report an error)
- python api load image/font/text raise errors with message (was just retuning None for image and font)
- minor edits to py api errors.
This made new image operator and seperate image sequence call ED_undo_push_op() twice.
Tested with move to layer and python select pattern operators and it works ok.
including backtraces for the double calls just incase.
# first
#2 0x00000000009ff4c4 in ED_undo_push_op (C=0x20e1098, op=0x3ea13a8) at /media/data/blender_ideasman42/blender_trunk/source/blender/editors/util/undo.c:187
#3 0x00000000008b5fa1 in WM_operator_props_popup (C=0x20e1098, op=0x3ea13a8, event=0x3ea0d28) at /media/data/blender_ideasman42/blender_trunk/source/blender/windowmanager/intern/wm_operators.c:1032
#4 0x00000000008be6be in wm_operator_invoke (C=0x20e1098, ot=0x2408bd8, event=0x3ea0d28, properties=0x3e943d8, reports=0x0) at /media/data/blender_ideasman42/blender_trunk/source/blender/windowmanager/intern/wm_event_system.c:613
#5 0x00000000008bfa44 in wm_handler_operator_call (C=0x20e1098, handlers=0x25509a0, handler=0x25cb658, event=0x3ea0d28, properties=0x3e943d8) at /media/data/blender_ideasman42/blender_trunk/source/blender/windowmanager/intern/wm_event_system.c:1158
# second
#2 0x00000000009ff4c4 in ED_undo_push_op (C=0x20e1098, op=0x3ea13a8) at /media/data/blender_ideasman42/blender_trunk/source/blender/editors/util/undo.c:187
#3 0x00000000008bde8e in wm_operator_finished (C=0x20e1098, op=0x3ea13a8, repeat=0) at /media/data/blender_ideasman42/blender_trunk/source/blender/windowmanager/intern/wm_event_system.c:439
#4 0x00000000008be82a in wm_operator_invoke (C=0x20e1098, ot=0x2408bd8, event=0x3ea0d28, properties=0x3e943d8, reports=0x0) at /media/data/blender_ideasman42/blender_trunk/source/blender/windowmanager/intern/wm_event_system.c:640
#5 0x00000000008bfa44 in wm_handler_operator_call (C=0x20e1098, handlers=0x25509a0, handler=0x25cb658, event=0x3ea0d28, properties=0x3e943d8) at /media/data/blender_ideasman42/blender_trunk/source/blender/windowmanager/intern/wm_event_system.c:1158
* the default properties of the file operator now only contain the "filepath", which means only the complete path to a file is returned.
* "filename" and "directory" has been added to the link/append operator - the only place it was used.
* sequence operators still work on the "files", which was custom property passed to the file operator anyway.
* have tested sequence loading, image loading and append/link - please report if there are any issues
necessary due to the more accurate mouse move events that are useful for
sculpting and painting (at least on Linux/X11, not sure about other platforms).
If the update function takes a while to run, this in turn causes more mouse
move events to be accumulated, making things even slower, .. going into a spiral
of slower and slower redraws.
As a solution I've added a INBETWEEN_MOUSEMOVE event next to MOUSEMOVE. A
MOUSEMOVE event is automatically changed to INBETWEEN_MOUSEMOVE when a
MOUSEMOVE event is added after it. This new event type is only handled by
painting/sculpting operators, everything else can happily ignore it.
This started off doing pointcache debugging but it's also very useful for users too.
Previously it was very hard to see the state of the system when you're working caches
such as physics point cache - is it baked? which frames are cached? is it out of date?
Now, for better feedback, cached frames are drawn for the active object at the bottom
of the timeline - a semitransparent area shows the entire cache extents, and more
solid blocks on top show the frames that are cached. Darker versions indicate it's
using a disk cache.
It can be disabled in general in the timeline View -> Caches menu, or by each individual
system that can be shown.
There's still a bit to do on this, behaviour needs to be clarified still eg. deciding what
shows when it's out of date, or when it's been played back but not cached, etc. etc.
Part of this is due to a lack of definition in the point cache system itself, so we should
try and clean up/clarify this behaviour and what it means to users, at the same time.
Also would be interested in extending this to other caches such as fluid cache,
sequencer memory cache etc. in the future, too.
path -> filepath (for rna and operators, as agreed on with elubie)
path -> data_path (for windowmanager context functions, this was alredy used in many places)
- added relative option to saving external multires data
- renamed multires external functiosn to have save / pack as suffix.
- added TODO's for file select operators that should support relative paths but dont.
- also disable openmp on linux cross compile, mingw currently isnt linking -lgomp
broke when including the blend path in the modules filename.
- new function BLI_path_basename(), matches pythons os.path.basename().
replace a number of cases where BLI_split_dirfile was being used to get the filename only.
* Now it displays the last report from the global list, not just from operators
* Rather than disappearing when a new operator is run, it stays until it times
out or a new report is added
* Fun animated transitions ;)
http://mke3.net/blender/devel/2.5/reports_header.mov
Now need to investigate report usage with popups. Ideally we can have most
reports non-blocking, so they're less intrusive, only popping up for dire errors.
Problem is many things in Blender right now are marked as RPT_ERROR
when probably RPT_WARNING is more appropriate. Should probably keep
RPT_ERROR for things that demand immediate attention.
Displays a global progress indicator in the application icon reflecting the total progress of all running jobs.
Currently fully implemented on OSX (Cocoa).
On other OSes that do not allow to redraw the app icon, this can be implemented as a [x%] display in the app title, so to appear in the taskbar.
Thanks to Matt for the windowmanager wrapper.
- fix for blend file thumbnails not being immediately visible in an external file manager (was writing the thumb before the blend)
- move overlay function from wm_files.c into thumbs_blend.c
After talking with Brecht he agreed on allowing Esc to be used as input for key input butons.
In order to let the user to cancel an input it can cancel it clicking outside the button.
Also replacing manual check by ui_mouse_inside_button in ui_text function
(patch reviewed (and helped) by Matt (but the change on ISHOTKEY, that's on my own risk ;))
from Riakiotakis Antonis (psy-fi)
with modifications.
notes:
- needed to remove quotes around filepaths for copying.
- WITH_JACK, doesnt link
- WITH_FFMPEG, doesnt build
- WITH_RAYOPTIMIZATION, gave an error for me. (no mmx support)
Now, rather than the bit-too-alarming stop sign, threaded wmJobs
display a progress indicator in the header. This is an optional feature
for each job type and still uses the same hardcoded ui template
(could use further work here...).
Currently implemented for:
Render - parts completed, then nodes comped
Compositor - nodes comped
Fluid Sim - frames simulated
Texture Bake - faces baked
Example: http://mke3.net/blender/devel/2.5/progress.mov
- uses same thumbnail system as image browser
- blend files show thumbnails in ubuntu/gnome (freedesktop spec)
- 128x128 images are embedded into the blend file header, a simple loader avoids reading the entire blend file to extract it when generating thumbnails in the file selector.
When the image browser reads a directory it loads images and creates thumbnails, blend files embedded images are treated just like loading an image.
- the thumbnail is created from the camera view in solid mode. (no camera == no thumbnal).
- readfile/writefile.c: had to use the 'TEST' code name to save thumbnails, anything else would segfault older blender versions on load. (its not used elsewhere).
tile cache code in imbuf, but it is not hooked up to the render engine.
Imbuf module: some small refactoring and removing a lot of unused or old code
(about 6.5k lines).
* Added a ImFileType struct with callbacks to make adding an file format type,
or making changes to the API easier.
* Move imbuf init/exit code into IMB_init()/IMB_exit() functions.
* Increased mipmap levels from 10 to 20, you run into this limit already with
a 2k image.
* Removed hamx, amiga, anim5 format support.
* Removed colormap saving, only simple colormap code now for reading tga.
* Removed gen_dynlibtiff.py, editing this is almost as much work as just
editing the code directly.
* Functions removed that were only used for sequencer plugin API:
IMB_anim_nextpic, IMB_clever_double, IMB_antialias, IMB_gamwarp,
IMB_scalefieldImBuf, IMB_scalefastfieldImBuf, IMB_onethird, IMB_halflace,
IMB_dit0, IMB_dit2, IMB_cspace
* Write metadata info into OpenEXR images. Can be viewed with the command
line utility 'exrheader'
For the image tile cache code, see this page:
http://wiki.blender.org/index.php/Dev:2.5/Source/Imaging/ImageTileCache
