* session aware process manager via store creator
* remove unused clear_session_id
* restore original scopes
* stateless test helpers
* more updates
* more updates
* more updates
* more updates
---------
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
* asset record based deletion
* fix order
* delete assets on session destroy
* fix style
* guard delete via file system check
* guard delete via file system check
* adjust responses a bit
* bump version
* test every workflow on every processor
* fix low fps bug (#1226)
* significant improve the test
* move partial logic to vision.py
* logic belongs to conditional_get_target_vision_frames
* use avg_frame_rate next to r_frame_rate for fps extraction
* update onnxruntime
* fix upload icon
* improve css
---------
Co-authored-by: Harisreedhar <46858047+harisreedhar@users.noreply.github.com>
* introduce content store for content analyser
* introduce content store for content analyser part2
* guard content store as well
* minor changes
* minor changes
* minor changes
* Goodbye tqdm
* Goodbye tqdm
* remove unused api
* show just file name
* support different mode
* support different mode
* minor polishing
* minor polishing
* switch to ProgressUnit to avoid confusion with ProcessMode
* improve naming
Add master's test_fix_audio_encoder and test_fix_video_encoder; v4 has
both functions and matches master's expected mappings. Kept v4's
sanitize_* and spawn_frames tests and path-isolation signatures.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01V7MdZEmd1GE8uSDTMyq34r
Add master's tests for the builders v4 gained via the ffmpeg reader swap
(seek_to, set_output_format, restrict_color_transfer, convert_color_space,
set_thread_count); kept v4's capture_video/enforce_pixel_format/
set_stream_quality tests for its stream builders.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01V7MdZEmd1GE8uSDTMyq34r
Take master's rewritten test_get_inference_pool (arena-leak cli/api pool
sharing) and new test_resolve_static_inference_providers (override/adjust
hooks), adapted from 'ui' to v4's 'api' app context. Both pass on v4.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01V7MdZEmd1GE8uSDTMyq34r
v4 inherited an old-base skipped stub; master implemented the real
job-list test post-fork. v4 has the job-list command, so the real test
passes. Kept v4's assert_helper/get_test_output_path and --workflow-mode.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01V7MdZEmd1GE8uSDTMyq34r