diffusers-workflow 0.9.0__tar.gz → 0.10.0__tar.gz

{diffusers_workflow-0.9.0/diffusers_workflow.egg-info → diffusers_workflow-0.10.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: diffusers-workflow
-Version: 0.9.0
+Version: 0.10.0
 Summary: Declarative workflow engine and web UI for Hugging Face Diffusers
 Author: dkackman
 License: Apache-2.0

{diffusers_workflow-0.9.0 → diffusers_workflow-0.10.0/diffusers_workflow.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: diffusers-workflow
-Version: 0.9.0
+Version: 0.10.0
 Summary: Declarative workflow engine and web UI for Hugging Face Diffusers
 Author: dkackman
 License: Apache-2.0

{diffusers_workflow-0.9.0 → diffusers_workflow-0.10.0}/diffusers_workflow.egg-info/SOURCES.txt

@@ -19,12 +19,15 @@ dw/audio_qc.py
 dw/cache_blocks.json
 dw/cache_blocks.py
 dw/content_types.py
+dw/devices.py
 dw/dissolve_frame_errors.py
 dw/download_watch.py
 dw/dsp.py
 dw/elision.py
 dw/events.py
 dw/for_each.py
+dw/guides.py
+dw/hold_audio.py
 dw/host_memory.py
 dw/host_memory_projection.py
 dw/hub_cache.py
@@ -39,6 +42,7 @@ dw/lora_hub.py
 dw/media.py
 dw/media_frames.py
 dw/media_types.py
+dw/outbound.py
 dw/output_extraction.py
 dw/path_redaction.py
 dw/pipeline_ownership.py
@@ -57,6 +61,7 @@ dw/run.py
 dw/runs.py
 dw/scalar_result_validation.py
 dw/schema.py
+dw/script_lines.py
 dw/security.py
 dw/serve.py
 dw/settings.py
@@ -80,6 +85,7 @@ dw/video_extensions.py
 dw/video_size_errors.py
 dw/vram_estimate.py
 dw/vram_inheritance.py
+dw/window_count_errors.py
 dw/worker.py
 dw/worker_manager.py
 dw/worker_protocol.py
@@ -89,6 +95,7 @@ dw/workflow_schema.json
 dw/workspace.py
 dw/writers.py
 dw/community_pipelines/pipeline_flux_rf_inversion.py
+dw/community_pipelines/pipeline_ltx2_refine.py
 dw/docs/ACCELERATION.md
 dw/docs/AGENT_LOOP.md
 dw/docs/ARCHITECTURE.md
@@ -113,6 +120,9 @@ dw/pipeline_processors/adapters.py
 dw/pipeline_processors/chain.py
 dw/pipeline_processors/components.py
 dw/pipeline_processors/config_objects.py
+dw/pipeline_processors/h3_guides.py
+dw/pipeline_processors/h3_hold.py
+dw/pipeline_processors/h3_rules.py
 dw/pipeline_processors/pipeline.py
 dw/pipeline_processors/placement.py
 dw/pipeline_processors/progress.py
@@ -132,13 +142,16 @@ dw/server/http_security.py
 dw/server/inline_media.py
 dw/server/job_history.py
 dw/server/job_record.py
+dw/server/job_results.py
 dw/server/jobs.py
 dw/server/mcp_mount.py
 dw/server/netinfo.py
 dw/server/observed_cost.py
 dw/server/outputs.py
+dw/server/pool.py
 dw/server/sysinfo.py
 dw/server/updater.py
+dw/server/worker_memory.py
 dw/server/routes/__init__.py
 dw/server/routes/assets.py
 dw/server/routes/files.py
@@ -163,45 +176,45 @@ dw/server/ui/assets/csharp-BKxtCVv1.js
 dw/server/ui/assets/csp-bTuwJoIa.js
 dw/server/ui/assets/css-DIMkf-bt.js
 dw/server/ui/assets/css.worker-B3ciXF_0.js
-dw/server/ui/assets/cssMode-C1mrKegs.js
+dw/server/ui/assets/cssMode-wXe774OW.js
 dw/server/ui/assets/cypher-CVaqCwHa.js
 dw/server/ui/assets/dart-onAF5SnQ.js
 dw/server/ui/assets/dockerfile-DZFCIeNp.js
 dw/server/ui/assets/ecl-D05T4iGw.js
 dw/server/ui/assets/editor-jjEx9u7D.css
-dw/server/ui/assets/editor.api-EZbUgQvM.js
+dw/server/ui/assets/editor.api-DqvgVCmU.js
 dw/server/ui/assets/editor.worker-q-txB4vs.js
 dw/server/ui/assets/elixir-6RTg0lbw.js
 dw/server/ui/assets/flow9-C5_-GSwl.js
-dw/server/ui/assets/freemarker2-Dth1K2Vy.js
+dw/server/ui/assets/freemarker2-DR8uhCda.js
 dw/server/ui/assets/fsharp-C8Ef5oNN.js
 dw/server/ui/assets/go-C-y9NEjX.js
 dw/server/ui/assets/graphql-fmXr3nnJ.js
-dw/server/ui/assets/handlebars-8Sswqnxq.js
+dw/server/ui/assets/handlebars-D1n_Fhy1.js
 dw/server/ui/assets/hcl-CpzslTdj.js
-dw/server/ui/assets/html-BYtYo7f1.js
+dw/server/ui/assets/html-BVJM2HQM.js
 dw/server/ui/assets/html.worker-C93Ht9o9.js
-dw/server/ui/assets/htmlMode-B_AiWoXc.js
-dw/server/ui/assets/index-B6Ujrvuj.js
-dw/server/ui/assets/index-BChhQc7I.css
+dw/server/ui/assets/htmlMode-Cb3bisrD.js
+dw/server/ui/assets/index-BxCxre0j.css
+dw/server/ui/assets/index-Bxcur2S3.js
 dw/server/ui/assets/ini-sBoK_t0W.js
 dw/server/ui/assets/java-BEtHBSE6.js
-dw/server/ui/assets/javascript-E4m083D3.js
+dw/server/ui/assets/javascript-DhyRtgFn.js
 dw/server/ui/assets/json.worker-B2V3pomh.js
-dw/server/ui/assets/jsonMode--h7ZikWy.js
+dw/server/ui/assets/jsonMode-IURul1-R.js
 dw/server/ui/assets/julia-Bri6UV-V.js
 dw/server/ui/assets/kotlin-BOotOW0E.js
 dw/server/ui/assets/less-B9JPFI3C.js
 dw/server/ui/assets/lexon-CfSJPG6W.js
-dw/server/ui/assets/liquid-BzbKdLRW.js
-dw/server/ui/assets/lspLanguageFeatures-CRa0bRFV.js
+dw/server/ui/assets/liquid-CdeVzoDf.js
+dw/server/ui/assets/lspLanguageFeatures-C7DVi-2z.js
 dw/server/ui/assets/lua-CsQS60Ue.js
 dw/server/ui/assets/m3-D-oSqn_W.js
 dw/server/ui/assets/markdown-Cimd5fb3.js
-dw/server/ui/assets/mdx-BJ4u6XQs.js
+dw/server/ui/assets/mdx-BDSAHpOx.js
 dw/server/ui/assets/mips-CIPQ_RoX.js
 dw/server/ui/assets/monaco--ixms01u.css
-dw/server/ui/assets/monaco-DwaeaVri.js
+dw/server/ui/assets/monaco-B8zQveVO.js
 dw/server/ui/assets/msdax-DauUninz.js
 dw/server/ui/assets/mysql-SOo6toE5.js
 dw/server/ui/assets/objective-c-FvmIjYaQ.js
@@ -216,10 +229,10 @@ dw/server/ui/assets/powerquery-D3hlyOfw.js
 dw/server/ui/assets/powershell-DmHpPYUd.js
 dw/server/ui/assets/protobuf-C531GsRP.js
 dw/server/ui/assets/pug-Z5eAx3Zn.js
-dw/server/ui/assets/python-CoX_gCAL.js
+dw/server/ui/assets/python-C6JhntrO.js
 dw/server/ui/assets/qsharp-DkqhCAOL.js
 dw/server/ui/assets/r-BwWrilGY.js
-dw/server/ui/assets/razor-BCTdG5SN.js
+dw/server/ui/assets/razor-DUoZWg4e.js
 dw/server/ui/assets/redis-ClamHrr6.js
 dw/server/ui/assets/redshift-DT7zqm-g.js
 dw/server/ui/assets/restructuredtext-BYgofb2h.js
@@ -239,31 +252,37 @@ dw/server/ui/assets/swift-Bxkupp3x.js
 dw/server/ui/assets/systemverilog-Bz4Y3fRF.js
 dw/server/ui/assets/tcl-DISqw1ZD.js
 dw/server/ui/assets/ts.worker-D7T1-Ig5.js
-dw/server/ui/assets/tsMode-B6BV7a76.js
+dw/server/ui/assets/tsMode-CxKbOt5l.js
 dw/server/ui/assets/twig-De2hgUGE.js
-dw/server/ui/assets/typescript-DJe7-fvZ.js
+dw/server/ui/assets/typescript-BYnQ0RjU.js
 dw/server/ui/assets/typespec-B8J7ngcE.js
 dw/server/ui/assets/vb-DV3o63ZY.js
 dw/server/ui/assets/wgsl-DpFanUEy.js
-dw/server/ui/assets/workers-DCzLJTPE.js
-dw/server/ui/assets/xml-CNvWfUrB.js
-dw/server/ui/assets/yaml-ByARP6WE.js
+dw/server/ui/assets/workers-CqfSFzdX.js
+dw/server/ui/assets/xml-BTej-yKG.js
+dw/server/ui/assets/yaml-DLu_25Zn.js
 dw/tasks/assess.py
 dw/tasks/audio_dynamics.py
 dw/tasks/audio_transcription.py
 dw/tasks/audio_utils.py
 dw/tasks/background_remover.py
+dw/tasks/beats.py
 dw/tasks/borders.py
 dw/tasks/compose_text.py
 dw/tasks/concat_videos.py
+dw/tasks/cuts.py
 dw/tasks/depth_estimator.py
 dw/tasks/diffusion_upscale.py
 dw/tasks/dissolve_videos.py
+dw/tasks/face_track.py
+dw/tasks/finish.py
+dw/tasks/fit.py
 dw/tasks/format_messages.py
 dw/tasks/gather.py
 dw/tasks/grade.py
 dw/tasks/h3_latent_upscale.py
 dw/tasks/h3_latent_upscaler_model.py
+dw/tasks/image_ops.py
 dw/tasks/image_to_text.py
 dw/tasks/image_utils.py
 dw/tasks/interpolate_frames.py
@@ -271,11 +290,14 @@ dw/tasks/join_into_song.py
 dw/tasks/joins.py
 dw/tasks/judge.py
 dw/tasks/loop_bed.py
+dw/tasks/lut.py
 dw/tasks/model_cache.py
 dw/tasks/pair_audio.py
 dw/tasks/qr_code.py
+dw/tasks/registry.py
 dw/tasks/restore_faces.py
 dw/tasks/rife_model.py
+dw/tasks/script_check.py
 dw/tasks/segment.py
 dw/tasks/select.py
 dw/tasks/speech_generation.py
@@ -284,9 +306,11 @@ dw/tasks/task.py
 dw/tasks/tensor_image.py
 dw/tasks/text_generation.py
 dw/tasks/text_sections.py
+dw/tasks/trim.py
 dw/tasks/upscale.py
 dw/tasks/video_utils.py
 dw/tasks/voice_attribution.py
+dw/tasks/windows.py
 dw/tasks/zoe_depth.py
 dw/workflows/h3_context_ir.json
 dw/workflows/test.json
@@ -311,6 +335,8 @@ dw_mcp/tools_jobs.py
 dw_mcp/tools_media.py
 dw_mcp/workspaces.py
 tests/test_admission.py
+tests/test_admission_capacity.py
+tests/test_analyze_beats.py
 tests/test_api_contract.py
 tests/test_arch_metrics.py
 tests/test_arch_report.py
@@ -353,13 +379,23 @@ tests/test_dsp.py
 tests/test_elision.py
 tests/test_events.py
 tests/test_examples.py
+tests/test_face_track.py
 tests/test_find_loop_bed.py
+tests/test_finish.py
+tests/test_fit_templates.py
+tests/test_fit_to_model.py
 tests/test_for_each.py
 tests/test_gather.py
 tests/test_grade.py
+tests/test_guide_chain_validation.py
+tests/test_guides_validation.py
 tests/test_h3_adapters.py
 tests/test_h3_context_ir.py
+tests/test_h3_guides.py
+tests/test_h3_hold_audio.py
 tests/test_h3_latent_upscale.py
+tests/test_h3_refine.py
+tests/test_h3_rules.py
 tests/test_h3_schedule.py
 tests/test_h3_upscaler_source.py
 tests/test_h3_vram_ceiling.py
@@ -367,17 +403,21 @@ tests/test_hf_placement.py
 tests/test_host_memory.py
 tests/test_host_memory_projection.py
 tests/test_hub_cache.py
+tests/test_image_ops.py
 tests/test_image_to_text.py
 tests/test_image_utils.py
+tests/test_ingredients_grid.py
 tests/test_integration.py
 tests/test_interpolate_frames.py
 tests/test_introspection.py
+tests/test_job_history_device.py
 tests/test_job_history_wal.py
 tests/test_job_progress.py
 tests/test_job_warnings.py
 tests/test_jobs_listing.py
 tests/test_jobs_reused.py
 tests/test_join_into_song.py
+tests/test_join_windows.py
 tests/test_judge.py
 tests/test_kernel_availability.py
 tests/test_library_path.py
@@ -392,7 +432,9 @@ tests/test_lora_library.py
 tests/test_ltx2_diffusion_decode.py
 tests/test_ltx2_extend_clip.py
 tests/test_ltx2_ic_loras.py
+tests/test_ltx2_refine_pipeline.py
 tests/test_ltx_prompt_library.py
+tests/test_lut.py
 tests/test_mcp_assets.py
 tests/test_mcp_authoring.py
 tests/test_mcp_catalog.py
@@ -421,8 +463,10 @@ tests/test_modular_progress.py
 tests/test_mps_fallback_warning.py
 tests/test_mps_rope_precision.py
 tests/test_null_media.py
+tests/test_numeric_coercion.py
 tests/test_observed_cost.py
 tests/test_offload_placement.py
+tests/test_outbound.py
 tests/test_output_references.py
 tests/test_pair_audio.py
 tests/test_pair_audio_fit.py
@@ -432,6 +476,7 @@ tests/test_phase_events.py
 tests/test_pipeline_caching.py
 tests/test_pipeline_components.py
 tests/test_plan.py
+tests/test_plan_cuts.py
 tests/test_plugin_skills.py
 tests/test_prepare_pipeline.py
 tests/test_previous_result_walker_parity.py
@@ -441,6 +486,7 @@ tests/test_prompt_parser.py
 tests/test_prompt_references.py
 tests/test_prompt_weighting.py
 tests/test_prompts.py
+tests/test_read_frame_range.py
 tests/test_real_models.py
 tests/test_realize.py
 tests/test_recenter_crop.py
@@ -450,6 +496,7 @@ tests/test_reference_skip_sets.py
 tests/test_references.py
 tests/test_rerun_new_seed.py
 tests/test_resize_bucket.py
+tests/test_restore_faces_video.py
 tests/test_result.py
 tests/test_result_fps.py
 tests/test_result_output_naming.py
@@ -458,6 +505,7 @@ tests/test_runs.py
 tests/test_scalar_result_validation.py
 tests/test_scheduler_config.py
 tests/test_schema.py
+tests/test_script_check.py
 tests/test_security.py
 tests/test_security_auth.py
 tests/test_security_decoder_bombs.py
@@ -469,6 +517,7 @@ tests/test_security_web_headers.py
 tests/test_segment.py
 tests/test_select.py
 tests/test_select_validation.py
+tests/test_separate_stems.py
 tests/test_serve_main.py
 tests/test_server.py
 tests/test_server_assess.py
@@ -496,6 +545,8 @@ tests/test_subfolders.py
 tests/test_task.py
 tests/test_task_discovery.py
 tests/test_task_domains.py
+tests/test_task_public_imports.py
+tests/test_task_registry_rules.py
 tests/test_task_signature_errors.py
 tests/test_template_defaults.py
 tests/test_template_subfolders.py
@@ -503,7 +554,9 @@ tests/test_tensor_image.py
 tests/test_text_generation.py
 tests/test_text_sections.py
 tests/test_transcribe_audio.py
+tests/test_trim_video.py
 tests/test_type_helpers.py
+tests/test_type_helpers_concurrency.py
 tests/test_ui_twins.py
 tests/test_updater.py
 tests/test_validate_arguments.py
@@ -515,12 +568,16 @@ tests/test_video_size_errors.py
 tests/test_video_utils.py
 tests/test_voice_attribution.py
 tests/test_vram_estimate.py
+tests/test_vram_guides.py
 tests/test_vram_inheritance.py
+tests/test_window_count_errors.py
+tests/test_window_video.py
 tests/test_worker.py
 tests/test_worker_execute.py
 tests/test_worker_job_peak_rss.py
 tests/test_worker_manager.py
 tests/test_worker_messages.py
+tests/test_worker_pool.py
 tests/test_workflow.py
 tests/test_workflow_snapshot.py
 tests/test_workflow_step_cache.py

{diffusers_workflow-0.9.0 → diffusers_workflow-0.10.0}/dw/adapter_compatibility.py

@@ -30,7 +30,7 @@ names and the partition each one denoises against are diffusers' own, and
 import logging
 
 from . import references
-from .for_each import MEMBER_SEPARATOR, render_path
+from .references import MEMBER_SEPARATOR, render_path
 
 logger = logging.getLogger("dw")
 

{diffusers_workflow-0.9.0 → diffusers_workflow-0.10.0}/dw/argument_media.py

@@ -2,6 +2,7 @@
 {media_type, location} reference, an `image`/`video` argument's path or URL, and
 the frame rate and shots a loaded video carries."""
 
+import contextlib
 import io
 import os
 import logging
@@ -16,7 +17,8 @@ from .security import (
     ALLOWED_IMAGE_EXTENSIONS,
     ALLOWED_VIDEO_EXTENSIONS,
 )
-from .locations import safe_get, validate_media_path
+from .locations import is_http_url, validate_media_path
+from .outbound import safe_get
 
 logger = logging.getLogger("dw")
 
@@ -81,10 +83,14 @@ def fetch_image_with_context(v, base_dir, key):
         ) from error
 
 
-def fetch_video_with_context(v, base_dir, key):
+def fetch_video_with_context(v, base_dir, key, with_audio=False):
     """fetch_video, with the same argument-key context as
-    fetch_image_with_context."""
+    fetch_image_with_context. `with_audio` reads a file's soundtrack along with
+    its frames, as an AudioVideo - an H3 guide's `"audio": true` beside its
+    `video` (#649), where frames alone leave the guide no audio to hold."""
     try:
+        if with_audio:
+            return _fetch_audio_video(v, base_dir)
         return fetch_video(v, base_dir)
     except ValueError as error:
         raise ValueError(
@@ -94,7 +100,30 @@ def fetch_video_with_context(v, base_dir, key):
         ) from error
 
 
-def fetch_image(img_spec, base_dir=None):
+def _fetch_audio_video(video_spec, base_dir=None):
+    """A video file - a path, URL or {"location": ...} - loaded with its audio
+    as an AudioVideo. A deferred reference stays for later resolution, and a
+    value already loaded passes through, so a second realization keeps it."""
+    from .tasks.video_utils import is_video_location, load_audio_video
+
+    if isinstance(video_spec, str) and references.is_ref(
+        references.LAZY_MEDIA, video_spec
+    ):
+        return video_spec
+    if is_video_location(video_spec):
+        return load_audio_video(video_spec, base_dir)
+    return video_spec
+
+
+def _rgb_keeping_alpha(image):
+    """RGBA when the file carries transparency (an alpha band, or a palette
+    or greyscale image's transparency key), plain RGB otherwise."""
+    if "A" in image.getbands() or "transparency" in image.info:
+        return image.convert("RGBA")
+    return image.convert("RGB")
+
+
+def fetch_image(img_spec, base_dir=None, keep_alpha=False):
     """
     Load image from file path or URL with security validation.
 
@@ -102,6 +131,10 @@ def fetch_image(img_spec, base_dir=None):
         img_spec: Image specification (file path, URL, dict with 'location' key, PIL Image, or list of any of these)
         base_dir: Directory relative file paths are resolved against - the
             workflow file's directory. Defaults to the process working directory
+        keep_alpha: Load a file with transparency as RGBA instead of
+            flattening it to RGB. A pipeline input wants RGB; the finishing
+            commands (grade, sharpen, film_grain, apply_lut) put the alpha
+            back after their op, so a cutout stays a cutout (#775)
 
     Returns:
         Loaded PIL Image, list of PIL Images, or None if img_spec is None
@@ -116,7 +149,7 @@ def fetch_image(img_spec, base_dir=None):
     # Handle lists of images (recursively process each)
     if isinstance(img_spec, list):
         logger.debug(f"Loading list of {len(img_spec)} images")
-        return [fetch_image(img, base_dir) for img in img_spec]
+        return [fetch_image(img, base_dir, keep_alpha) for img in img_spec]
 
     # If already a PIL Image, return as-is (allows multiple realize_args calls)
     if hasattr(img_spec, "mode") and hasattr(img_spec, "size"):
@@ -140,6 +173,7 @@ def fetch_image(img_spec, base_dir=None):
         return img_spec
 
     logger.debug(f"Loading image from: {img_spec}")
+    convert_method = _rgb_keeping_alpha if keep_alpha else None
 
     try:
         # Check if it's a URL
@@ -150,7 +184,7 @@ def fetch_image(img_spec, base_dir=None):
             # without re-checking them; load_image still does the EXIF
             # transpose and RGB conversion on the decoded result
             response = safe_get(img_spec, "an image argument", timeout=60)
-            return load_image(Image.open(io.BytesIO(response.content)))
+            return load_image(Image.open(io.BytesIO(response.content)), convert_method)
         else:
             # Treat as file path, relative to the workflow file, and confined
             # to the directories this workflow may read (dw/locations.py)
@@ -164,7 +198,7 @@ def fetch_image(img_spec, base_dir=None):
                     f"Image file extension not allowed: {ext} - a video file "
                     "must go through video_frames first"
                 )
-            return load_image(validated_path)
+            return load_image(validated_path, convert_method)
 
     except SecurityError:
         raise
@@ -209,24 +243,41 @@ def _declared_fps(path):
         return None
 
 
-def _fetch_remote_video(url):
-    """A video URL's frames, fetched through `safe_get` and decoded from a
-    temporary file. `load_video` would fetch the URL itself and follow its
-    redirects unchecked; handed a path, it only decodes. The suffix comes
-    from the URL, as `load_video`'s own download names it, since a `.gif`
-    decodes differently."""
-    from .tasks.video_utils import FrameList
+@contextlib.contextmanager
+def local_media_file(location, what, default_suffix=""):
+    """`location` as a path a loader opens: a path as it is, an http(s) URL
+    fetched through `safe_get` into a temporary file that lasts as long as
+    the `with` block.
 
-    response = safe_get(url, "a video argument", timeout=300)
-    suffix = os.path.splitext(unquote(urlparse(url).path))[1] or ".mp4"
+    A loader handed a URL fetches it itself - diffusers' `load_video`,
+    `load_image` and its H3 references' `from_file` all call requests
+    directly, resolving the host a second time and following redirects
+    unchecked, with no cap on the body. Handed a path, they only decode.
+    The suffix comes from the URL, as those loaders' own downloads name
+    it, since a container is often told apart by it. The loader must have
+    read the file before the block ends.
+    """
+    if not is_http_url(location):
+        yield location
+        return
+    response = safe_get(location, what, timeout=300)
+    suffix = os.path.splitext(unquote(urlparse(location).path))[1] or default_suffix
     handle = tempfile.NamedTemporaryFile(suffix=suffix, delete=False)
     try:
         with handle:
             handle.write(response.content)
-        frames = load_video(handle.name)
-        fps = _declared_fps(handle.name)
+        yield handle.name
     finally:
         os.remove(handle.name)
+
+
+def _fetch_remote_video(url):
+    """A video URL's frames, decoded from `local_media_file`'s copy."""
+    from .tasks.video_utils import FrameList
+
+    with local_media_file(url, "a video argument", default_suffix=".mp4") as path:
+        frames = load_video(path)
+        fps = _declared_fps(path)
     # A URL has no run beside it, so it carries no shots
     return FrameList(frames, fps, None) if fps else frames
 

{diffusers_workflow-0.9.0 → diffusers_workflow-0.10.0}/dw/arguments.py

@@ -23,6 +23,7 @@ from .argument_media import (
     fetch_image_with_context,
     fetch_video,
     fetch_video_with_context,
+    local_media_file,
 )
 from PIL import Image
 from .security import (
@@ -139,7 +140,10 @@ def realize_args(arg, base_dir=None, apply_key_conventions=True):
             # Handle video loading for keys ending in '_video' or exactly 'video'
             elif apply_key_conventions and (k.endswith("_video") or k == "video"):
                 logger.debug(f"Loading video for key: {k}")
-                arg[k] = fetch_video_with_context(v, base_dir, k)
+                # A guide holding its soundtrack (#649) needs the file's audio too
+                arg[k] = fetch_video_with_context(
+                    v, base_dir, k, with_audio=arg.get("audio") is True
+                )
             # Handle type references, and the keys that only look like one
             elif apply_key_conventions and (
                 k.endswith("_type") or k.endswith("_dtype") or k == "dtype"
@@ -511,9 +515,11 @@ def realize_object(value, base_dir=None):
         k: v for k, v in value.items() if k not in (FROM_FILE_KEY, type_keys[0])
     }
     accepted, overrides = split_from_file_arguments(object_type, arguments)
-    return apply_field_overrides(
-        object_type.from_file(location, **accepted), overrides, object_type
-    )
+    # A URL is fetched here, through the host policy, and the type handed a
+    # path: diffusers' from_file downloads a URL with requests itself
+    with local_media_file(location, f"'{FROM_FILE_KEY}' media") as path:
+        built = object_type.from_file(path, **accepted)
+    return apply_field_overrides(built, overrides, object_type)
 
 
 def split_from_file_arguments(object_type, arguments):
@@ -906,7 +912,9 @@ def resolve_relative_path(path, base_dir):
 # get_frame and its two fixed-index siblings, and the assessment probes
 # (dw/tasks/assess.py), which stream the file themselves - decoding it to a
 # frame list first dropped the soundtrack they measure and failed every probe
-# on an asset:/output: video (#387) - see _realize_lazy_frame_arguments
+# on an asset:/output: video (#387) - and window_video, which cuts the
+# source's soundtrack with its frames and so reads the file with its audio
+# (#601) - and trim_video, which cuts a soundtrack the same way (#627) - see _realize_lazy_frame_arguments
 _LAZY_FRAME_COMMANDS = frozenset(
     {
         "get_frame",
@@ -915,6 +923,8 @@ _LAZY_FRAME_COMMANDS = frozenset(
         "analyze_shots",
         "analyze_seams",
         "analyze_sync_drift",
+        "window_video",
+        "trim_video",
     }
 )
 

{diffusers_workflow-0.9.0 → diffusers_workflow-0.10.0}/dw/assessment_rules.py

@@ -88,7 +88,10 @@ RULES = (
         "threshold": -50.0,
         "severity": "warn",
         "says": "the track drops out at the join while both sides are voiced",
-        "unless": f"either side's rms is at or below {HOLE_VOICED_DBFS} dBFS",
+        "unless": (
+            f"either side's rms is at or below {HOLE_VOICED_DBFS} dBFS, or the"
+            " incoming shot is marked seam_fade_ms - a fade the caller asked for"
+        ),
     },
     {
         "name": "seam_frame_jump",

{diffusers_workflow-0.9.0 → diffusers_workflow-0.10.0}/dw/audio_qc.py

@@ -172,8 +172,9 @@ def warn_if_written_above_full_scale(
         f"{name} decodes at {peak:+.2f} dBFS - above full scale, so it "
         f"clips on playback. {cause}, so the fix is more headroom before "
         f"the file is written: a 'normalize_audio' step at 'peak_dbfs: -3' "
-        f"ahead of the step that saves it. A mux into a video needs more "
-        f"of it than an audio file does.",
+        f"ahead of the step that saves it (a template's intermediate shot has "
+        f"no such step to edit: set the join's 'match_levels' instead). A "
+        f"mux into a video needs more of it than an audio file does.",
         kind="audio_clipped",
         file=name,
         peak_dbfs=round(peak, 2),
@@ -291,7 +292,8 @@ def remeasure_shots_after_mux(artifact, probed_info, output_path, video_fps):
         round(probed_info["audio_stream_seconds"] * probed_info["sample_rate"])
     )
     measured_num_samples(artifact.shots, written_samples)
-    frame_count = len(getattr(artifact, "frames", []) or [])
+    frames = getattr(artifact, "frames", None)
+    frame_count = 0 if frames is None else len(frames)
     fps = video_fps(artifact)
     if not (frame_count and fps):
         return
@@ -340,10 +342,12 @@ def written_peak_already_warned(content_type, consumed_by_normalizer, headroom_w
 
     Only a plain audio save suppresses the post-write check; a lossless one
     only when a normalizer consumed it, a lossy one also when the pre-write
-    warning fired. A video always gets the ground-truth check.
+    warning fired. A video gets the ground-truth check, unless a normalizer
+    or a level-matching join consumes it: that step resets the level it
+    ships at, so the file's own written peak is not the deliverable's (#671).
     """
     if not content_type.startswith("audio"):
-        return False
+        return consumed_by_normalizer
     if content_type not in LOSSY_AUDIO_CONTENT_TYPES:
         return consumed_by_normalizer
     return headroom_warned or consumed_by_normalizer
@@ -414,7 +418,16 @@ def check_written_media(
     # that predicted risk but measured merely close-but-clean says
     # nothing. That is the file's own ground truth, not a threshold
     # bug.
-    if is_video and headroom_warned and written_peak is None:
+    #
+    # A video a normalizer or level-matching join consumes skips both:
+    # that step resets the level it ships at, so neither the written
+    # peak nor the prediction is the deliverable's (#671)
+    if (
+        is_video
+        and headroom_warned
+        and not consumed_by_normalizer
+        and written_peak is None
+    ):
         warn_held_prediction(output_path, predicted_peak_dbfs)
     source_mean_dbfs = getattr(artifact, "source_mean_dbfs", None)
     warn_if_written_near_silent(