refactor(section-a+c): decompose step_tasks.py into pipeline domain

Section A — 1172-line step_tasks.py split into 4 focused modules under backend/app/domains/pipeline/tasks/: - extract_metadata.py: process_step_file, reextract_cad_metadata, _auto_populate_materials_for_cad, bbox helpers + PipelineLogger - render_thumbnail.py: render_step_thumbnail, regenerate_thumbnail + PipelineLogger instrumentation - render_order_line.py: dispatch_order_line_render, render_order_line_task (full still+turntable paths) + PipelineLogger step tracking - export_glb.py: generate_gltf_geometry_task, generate_gltf_production_task All task names preserved (app.tasks.step_tasks.*) for backward compat. step_tasks.py is now a 23-line import-only shim. Section C — celery_app.py task routing: - app.domains.pipeline.tasks.* → step_processing (primary route) - All 4 new modules added to include list for Celery discovery - app.tasks.step_tasks.* kept as legacy fallback for in-flight tasks Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-03-08 20:43:56 +01:00
parent 07e3d1e026
commit c6556434d6
8 changed files with 1307 additions and 1175 deletions
@@ -0,0 +1,156 @@
+"""Thumbnail rendering tasks.
+
+Covers:
+- render_step_thumbnail  — render thumbnail for a freshly-processed STEP file
+- regenerate_thumbnail   — re-render thumbnail with updated per-part colours
+"""
+import logging
+from pathlib import Path
+
+from app.tasks.celery_app import celery_app
+from app.core.task_logs import log_task_event
+from app.core.pipeline_logger import PipelineLogger
+
+logger = logging.getLogger(__name__)
+
+
+@celery_app.task(bind=True, name="app.tasks.step_tasks.render_step_thumbnail", queue="thumbnail_rendering")
+def render_step_thumbnail(self, cad_file_id: str):
+    """Render the thumbnail for a freshly-processed STEP file.
+
+    Runs on the dedicated thumbnail_rendering queue (concurrency=1) so the
+    blender-renderer service is never overwhelmed by concurrent requests.
+    On success, also auto-populates materials and marks the CadFile as completed.
+    """
+    pl = PipelineLogger(task_id=self.request.id)
+    pl.step_start("render_step_thumbnail", {"cad_file_id": cad_file_id})
+    logger.info(f"Rendering thumbnail for CAD file: {cad_file_id}")
+
+    # Compute and persist STEP file hash for STL cache lookups
+    try:
+        from sqlalchemy import create_engine
+        from sqlalchemy.orm import Session
+        from app.config import settings as app_settings
+        from app.models.cad_file import CadFile
+        from app.domains.products.cache_service import compute_step_hash
+
+        sync_url = app_settings.database_url.replace("+asyncpg", "")
+        _eng = create_engine(sync_url)
+        with Session(_eng) as _sess:
+            _cad = _sess.get(CadFile, cad_file_id)
+            if _cad and _cad.stored_path and not _cad.step_file_hash:
+                _hash = compute_step_hash(_cad.stored_path)
+                _cad.step_file_hash = _hash
+                _sess.commit()
+                logger.info(f"Saved step_file_hash for {cad_file_id}: {_hash[:12]}…")
+        _eng.dispose()
+    except Exception:
+        logger.warning(f"step_file_hash computation failed for {cad_file_id} (non-fatal)")
+
+    try:
+        from app.services.step_processor import regenerate_cad_thumbnail
+        pl.info("render_step_thumbnail", "Calling regenerate_cad_thumbnail")
+        success = regenerate_cad_thumbnail(cad_file_id, part_colors={})
+        if not success:
+            raise RuntimeError("regenerate_cad_thumbnail returned False")
+    except Exception as exc:
+        pl.step_error("render_step_thumbnail", f"Thumbnail render failed: {exc}", exc)
+        logger.error(f"Thumbnail render failed for {cad_file_id}: {exc}")
+        raise self.retry(exc=exc, countdown=30, max_retries=2)
+
+    # Extract bounding box from the thumbnail GLB generated by the renderer.
+    # GLB bbox via trimesh is fast and avoids re-parsing the STEP file.
+    # Falls back to cadquery STEP re-parse if GLB is not found.
+    try:
+        from sqlalchemy import create_engine
+        from sqlalchemy.orm import Session
+        from app.config import settings as _cfg2
+        from app.models.cad_file import CadFile as _CadFile2
+        from app.domains.pipeline.tasks.extract_metadata import _bbox_from_glb, _bbox_from_step_cadquery
+
+        _sync_url2 = _cfg2.database_url.replace("+asyncpg", "")
+        _eng2 = create_engine(_sync_url2)
+        with Session(_eng2) as _sess2:
+            _cad2 = _sess2.get(_CadFile2, cad_file_id)
+            _step_path = _cad2.stored_path if _cad2 else None
+        _eng2.dispose()
+
+        if _step_path and not (_cad2.mesh_attributes or {}).get("dimensions_mm"):
+            _step = Path(_step_path)
+            _glb = _step.parent / f"{_step.stem}_thumbnail.glb"
+            bbox_data = _bbox_from_glb(str(_glb)) or _bbox_from_step_cadquery(_step_path)
+            if bbox_data:
+                _eng2 = create_engine(_sync_url2)
+                with Session(_eng2) as _sess2:
+                    _cad2 = _sess2.get(_CadFile2, cad_file_id)
+                    if _cad2:
+                        _cad2.mesh_attributes = {**(  _cad2.mesh_attributes or {}), **bbox_data}
+                        _sess2.commit()
+                        dims = bbox_data["dimensions_mm"]
+                        logger.info(
+                            f"bbox for {cad_file_id}: "
+                            f"{dims['x']}×{dims['y']}×{dims['z']} mm"
+                        )
+                _eng2.dispose()
+    except Exception:
+        logger.exception(f"bbox extraction failed for {cad_file_id} (non-fatal)")
+
+    # Auto-populate materials now that parsed_objects are available
+    try:
+        from app.domains.pipeline.tasks.extract_metadata import _auto_populate_materials_for_cad
+        _auto_populate_materials_for_cad(cad_file_id)
+    except Exception:
+        logger.exception(
+            f"Auto material population failed for cad_file {cad_file_id} (non-fatal)"
+        )
+
+    # Broadcast WebSocket event for live UI updates
+    try:
+        from sqlalchemy import create_engine, select as sql_select2
+        from sqlalchemy.orm import Session as _Session
+        from app.config import settings as _cfg
+        from app.models.cad_file import CadFile as _CadFile
+        _sync_url = _cfg.database_url.replace("+asyncpg", "")
+        _eng = create_engine(_sync_url)
+        with _Session(_eng) as _s:
+            _cad = _s.get(_CadFile, cad_file_id)
+            _tid = str(_cad.tenant_id) if _cad and _cad.tenant_id else None
+        _eng.dispose()
+        if _tid:
+            from app.core.websocket import publish_event_sync
+            publish_event_sync(_tid, {
+                "type": "cad_processing_complete",
+                "cad_file_id": cad_file_id,
+                "status": "completed",
+            })
+    except Exception:
+        logger.debug("WebSocket publish for CAD complete skipped (non-fatal)")
+
+    # Auto-generate geometry GLB so the 3D viewer is ready without manual trigger
+    try:
+        from app.domains.pipeline.tasks.export_glb import generate_gltf_geometry_task
+        generate_gltf_geometry_task.delay(cad_file_id)
+        pl.info("render_step_thumbnail", f"Queued generate_gltf_geometry_task for {cad_file_id}")
+        logger.info("render_step_thumbnail: queued generate_gltf_geometry_task for %s", cad_file_id)
+    except Exception:
+        logger.debug("Could not queue generate_gltf_geometry_task (non-fatal)")
+
+    pl.step_done("render_step_thumbnail")
+
+
+@celery_app.task(bind=True, name="app.tasks.step_tasks.regenerate_thumbnail", queue="thumbnail_rendering")
+def regenerate_thumbnail(self, cad_file_id: str, part_colors: dict):
+    """Regenerate thumbnail with per-part colours."""
+    pl = PipelineLogger(task_id=self.request.id)
+    pl.step_start("regenerate_thumbnail", {"cad_file_id": cad_file_id})
+    logger.info(f"Regenerating thumbnail for CAD file: {cad_file_id}")
+    try:
+        from app.services.step_processor import regenerate_cad_thumbnail
+        success = regenerate_cad_thumbnail(cad_file_id, part_colors)
+        if not success:
+            raise RuntimeError("regenerate_cad_thumbnail returned False")
+    except Exception as exc:
+        pl.step_error("regenerate_thumbnail", f"Thumbnail regeneration failed: {exc}", exc)
+        logger.error(f"Thumbnail regeneration failed for {cad_file_id}: {exc}")
+        raise self.retry(exc=exc, countdown=30, max_retries=2)
+    pl.step_done("regenerate_thumbnail")