Add versioned podcast asset metadata schema and backfill script

backend/api/content_assets/router.py

@@ -5,7 +5,7 @@ API endpoints for managing unified content assets across all modules.
 
 from fastapi import APIRouter, Depends, HTTPException, Query, Body
 from sqlalchemy.orm import Session
-from typing import List, Optional, Dict, Any, Set
+from typing import List, Optional, Dict, Any
 from pydantic import BaseModel, Field
 from datetime import datetime
 
@@ -47,33 +47,6 @@ class AssetResponse(BaseModel):
         from_attributes = True
 
 
-
-
-def _parse_source_modules(source_module: Optional[List[str]]) -> Optional[List[AssetSource]]:
-    """Parse source_module query values from repeated params and/or comma-separated values."""
-    if not source_module:
-        return None
-
-    parsed_values: List[AssetSource] = []
-    seen: Set[AssetSource] = set()
-
-    for raw_value in source_module:
-        for value in raw_value.split(","):
-            normalized = value.strip().lower()
-            if not normalized:
-                continue
-            try:
-                module = AssetSource(normalized)
-            except ValueError:
-                raise HTTPException(status_code=400, detail=f"Invalid source module: {value.strip()}")
-
-            if module not in seen:
-                seen.add(module)
-                parsed_values.append(module)
-
-    return parsed_values or None
-
-
 class AssetListResponse(BaseModel):
     """Response model for asset list."""
     assets: List[AssetResponse]
@@ -85,7 +58,7 @@ class AssetListResponse(BaseModel):
 @router.get("/", response_model=AssetListResponse)
 async def get_assets(
     asset_type: Optional[str] = Query(None, description="Filter by asset type"),
-    source_module: Optional[List[str]] = Query(None, description="Filter by source module(s); supports repeated params and comma-separated values"),
+    source_module: Optional[str] = Query(None, description="Filter by source module"),
     search: Optional[str] = Query(None, description="Search query"),
     tags: Optional[str] = Query(None, description="Comma-separated tags"),
     favorites_only: bool = Query(False, description="Only favorites"),
@@ -116,7 +89,12 @@ async def get_assets(
             except ValueError:
                 raise HTTPException(status_code=400, detail=f"Invalid asset type: {asset_type}")
         
-        source_modules_enum = _parse_source_modules(source_module)
+        source_module_enum = None
+        if source_module:
+            try:
+                source_module_enum = AssetSource(source_module.lower())
+            except ValueError:
+                raise HTTPException(status_code=400, detail=f"Invalid source module: {source_module}")
         
         tags_list = None
         if tags:
@@ -148,7 +126,7 @@ async def get_assets(
         assets, total = service.get_user_assets(
             user_id=user_id,
             asset_type=asset_type_enum,
-            source_modules=source_modules_enum,
+            source_module=source_module_enum,
             search_query=search,
             tags=tags_list,
             favorites_only=favorites_only,
@@ -222,7 +200,7 @@ async def create_asset(
         asset = service.create_asset(
             user_id=user_id,
             asset_type=asset_type_enum,
-            source_modules=source_modules_enum,
+            source_module=source_module_enum,
             filename=asset_data.filename,
             file_url=asset_data.file_url,
             file_path=asset_data.file_path,

backend/api/podcast/handlers/audio.py

@@ -26,6 +26,7 @@ from services.database import get_db
 from middleware.auth_middleware import get_current_user, get_current_user_with_query_token
 from api.story_writer.utils.auth import require_authenticated_user
 from utils.asset_tracker import save_asset_to_library
+from models.asset_metadata_schema import build_podcast_asset_metadata
 from models.story_models import StoryAudioResult
 from loguru import logger
 from ..constants import get_podcast_audio_service, get_podcast_media_dir
@@ -217,11 +218,11 @@ async def upload_podcast_audio(
                     title=f"Uploaded Audio - {project_id}",
                     description="Uploaded podcast audio/voice sample",
                     tags=["podcast", "audio", "upload", project_id],
-                    asset_metadata={
-                        "project_id": project_id,
-                        "type": "uploaded_audio",
-                        "status": "completed",
-                    },
+                    asset_metadata=build_podcast_asset_metadata(
+                        asset_role="uploaded_audio",
+                        project_id=project_id,
+                        origin="podcast.audio.upload",
+                    ),
                 )
             except Exception as e:
                 logger.warning(f"[Podcast] Failed to save audio asset: {e}")
@@ -455,11 +456,12 @@ async def generate_podcast_audio(
                 provider=result.get("provider"),
                 model=result.get("model"),
                 cost=result.get("cost"),
-                asset_metadata={
-                    "scene_id": request.scene_id,
-                    "scene_title": request.scene_title,
-                    "status": "completed",
-                },
+                asset_metadata=build_podcast_asset_metadata(
+                    asset_role="podcast_audio",
+                    project_id=request.project_id,
+                    origin="podcast.audio.generate",
+                    extras={"scene_id": request.scene_id, "scene_title": request.scene_title},
+                ),
             )
     except Exception as e:
         logger.warning(f"[Podcast] Failed to save audio asset: {e}")
@@ -621,13 +623,12 @@ async def combine_podcast_audio(
                     title=f"Combined Podcast - {request.project_id}",
                     description=f"Combined podcast audio from {len(request.scene_ids)} scenes",
                     tags=["podcast", "audio", "combined", request.project_id],
-                    asset_metadata={
-                        "project_id": request.project_id,
-                        "scene_ids": request.scene_ids,
-                        "scene_count": len(request.scene_ids),
-                        "total_duration": total_duration,
-                        "status": "completed",
-                    },
+                    asset_metadata=build_podcast_asset_metadata(
+                        asset_role="combined_podcast_audio",
+                        project_id=request.project_id,
+                        origin="podcast.audio.combine",
+                        extras={"scene_ids": request.scene_ids, "scene_count": len(request.scene_ids), "total_duration": total_duration},
+                    ),
                 )
             except Exception as e:
                 logger.warning(f"[Podcast] Failed to save combined audio asset: {e}")

backend/api/podcast/handlers/avatar.py

@@ -18,6 +18,7 @@ from api.story_writer.utils.auth import require_authenticated_user
 from services.llm_providers.main_image_generation import generate_image
 from services.llm_providers.main_image_editing import edit_image
 from utils.asset_tracker import save_asset_to_library
+from models.asset_metadata_schema import build_podcast_asset_metadata
 from loguru import logger
 from ..constants import get_podcast_media_dir, PODCAST_AVATARS_SUBDIR
 from ..presenter_personas import choose_persona_id, get_persona
@@ -111,11 +112,11 @@ async def upload_podcast_avatar(
                     title=f"Podcast Presenter Avatar - {project_id}",
                     description="Podcast presenter avatar image",
                     tags=["podcast", "avatar", project_id],
-                    asset_metadata={
-                        "project_id": project_id,
-                        "type": "presenter_avatar",
-                        "status": "completed",
-                    },
+                    asset_metadata=build_podcast_asset_metadata(
+                        asset_role="presenter_avatar",
+                        project_id=project_id,
+                        origin="podcast.avatar.upload",
+                    ),
                 )
             except Exception as e:
                 logger.warning(f"[Podcast] Failed to save avatar asset (non-fatal): {e}")
@@ -223,12 +224,12 @@ async def make_avatar_presentable(
                     tags=["podcast", "avatar", "presenter", "transformed", project_id],
                     provider=result.provider,
                     model=result.model,
-                    asset_metadata={
-                        "project_id": project_id,
-                        "type": "transformed_presenter",
-                        "original_avatar_url": avatar_url,
-                        "status": "completed",
-                    },
+                    asset_metadata=build_podcast_asset_metadata(
+                        asset_role="transformed_presenter",
+                        project_id=project_id,
+                        origin="podcast.avatar.make_presentable",
+                        extras={"original_avatar_url": avatar_url},
+                    ),
                 )
             except Exception as e:
                 logger.warning(f"[Podcast] Failed to save transformed avatar asset: {e}")
@@ -404,14 +405,12 @@ async def generate_podcast_presenters(
                         tags=["podcast", "avatar", "presenter", project_id],
                         provider=result.provider,
                         model=result.model,
-                        asset_metadata={
-                            "project_id": project_id,
-                            "speaker_number": i + 1,
-                            "type": "generated_presenter",
-                            "status": "completed",
-                        "persona_id": selected_persona_id,
-                        "seed": seed,
-                        },
+                        asset_metadata=build_podcast_asset_metadata(
+                            asset_role="generated_presenter",
+                            project_id=project_id,
+                            origin="podcast.avatar.generate",
+                            extras={"speaker_number": i + 1, "persona_id": selected_persona_id, "seed": seed},
+                        ),
                     )
                 except Exception as e:
                     logger.warning(f"[Podcast] Failed to save presenter asset: {e}")

backend/api/podcast/handlers/images.py

@@ -16,6 +16,7 @@ from middleware.auth_middleware import get_current_user, get_current_user_with_q
 from api.story_writer.utils.auth import require_authenticated_user
 from services.llm_providers.main_image_generation import generate_image, generate_character_image
 from utils.asset_tracker import save_asset_to_library
+from models.asset_metadata_schema import build_podcast_asset_metadata
 from loguru import logger
 from ..constants import get_podcast_media_dir
 from ..models import PodcastImageRequest, PodcastImageResponse
@@ -417,11 +418,12 @@ async def generate_podcast_scene_image(
                 tags=["podcast", "scene", request.scene_id],
                 provider=result.provider,
                 model=result.model,
-                asset_metadata={
-                    "scene_id": request.scene_id,
-                    "scene_title": request.scene_title,
-                    "status": "completed",
-                },
+                asset_metadata=build_podcast_asset_metadata(
+                    asset_role="podcast_scene_image",
+                    project_id=request.project_id,
+                    origin="podcast.images.generate",
+                    extras={"scene_id": request.scene_id, "scene_title": request.scene_title},
+                ),
             )
         except Exception as e:
             logger.warning(f"[Podcast] Failed to save image asset: {e}")

backend/models/asset_metadata_schema.py

@@ -0,0 +1,76 @@
+"""Shared schema/builders for content asset metadata."""
+
+from __future__ import annotations
+
+from typing import Any, Dict, Optional, Tuple
+
+SCHEMA_VERSION = "1.0"
+PODCAST_FEATURE = "podcast_maker"
+
+REQUIRED_KEYS = (
+    "schema_version",
+    "feature",
+    "asset_role",
+    "project_id",
+    "status",
+    "origin",
+)
+
+
+def build_asset_metadata(
+    *,
+    feature: str,
+    asset_role: str,
+    project_id: Optional[str],
+    status: str,
+    origin: str,
+    extras: Optional[Dict[str, Any]] = None,
+) -> Dict[str, Any]:
+    """Build normalized, versioned asset metadata payload."""
+    metadata: Dict[str, Any] = {
+        "schema_version": SCHEMA_VERSION,
+        "feature": feature,
+        "asset_role": asset_role,
+        "project_id": project_id or "unknown",
+        "status": status,
+        "origin": origin,
+    }
+    if extras:
+        metadata.update({k: v for k, v in extras.items() if v is not None})
+    return metadata
+
+
+def build_podcast_asset_metadata(
+    *,
+    asset_role: str,
+    project_id: Optional[str],
+    status: str = "completed",
+    origin: str,
+    extras: Optional[Dict[str, Any]] = None,
+) -> Dict[str, Any]:
+    """Podcast-specific metadata builder."""
+    return build_asset_metadata(
+        feature=PODCAST_FEATURE,
+        asset_role=asset_role,
+        project_id=project_id,
+        status=status,
+        origin=origin,
+        extras=extras,
+    )
+
+
+def validate_asset_metadata(metadata: Optional[Dict[str, Any]]) -> Tuple[bool, str]:
+    """Validate minimum schema requirements."""
+    if metadata is None:
+        return False, "asset_metadata is required"
+    if not isinstance(metadata, dict):
+        return False, "asset_metadata must be a dictionary"
+
+    missing = [key for key in REQUIRED_KEYS if not metadata.get(key)]
+    if missing:
+        return False, f"asset_metadata missing required keys: {', '.join(missing)}"
+
+    if str(metadata.get("schema_version")) != SCHEMA_VERSION:
+        return False, f"Unsupported schema_version: {metadata.get('schema_version')}"
+
+    return True, "ok"

backend/scripts/backfill_podcast_asset_metadata.py

@@ -0,0 +1,63 @@
+"""Backfill recent podcast assets with normalized metadata schema."""
+
+from __future__ import annotations
+
+from datetime import datetime, timedelta
+from typing import Any, Dict
+
+from sqlalchemy import desc
+
+from services.database import SessionLocal
+from models.content_asset_models import ContentAsset, AssetSource
+from models.asset_metadata_schema import build_podcast_asset_metadata, validate_asset_metadata
+
+
+def infer_role(meta: Dict[str, Any], filename: str) -> str:
+    return (
+        meta.get("asset_role")
+        or meta.get("type")
+        or ("podcast_audio" if filename.lower().endswith((".mp3", ".wav", ".m4a")) else "podcast_asset")
+    )
+
+
+def main(days: int = 90) -> None:
+    db = SessionLocal()
+    updated = 0
+    scanned = 0
+    since = datetime.utcnow() - timedelta(days=days)
+    try:
+        assets = (
+            db.query(ContentAsset)
+            .filter(ContentAsset.source_module == AssetSource.PODCAST_MAKER)
+            .filter(ContentAsset.created_at >= since)
+            .order_by(desc(ContentAsset.created_at))
+            .all()
+        )
+
+        for asset in assets:
+            scanned += 1
+            meta = asset.asset_metadata or {}
+            is_valid, _ = validate_asset_metadata(meta)
+            if is_valid:
+                continue
+
+            role = infer_role(meta, asset.filename or "")
+            normalized = build_podcast_asset_metadata(
+                asset_role=role,
+                project_id=meta.get("project_id"),
+                status=meta.get("status", "completed"),
+                origin=meta.get("origin", "migration.backfill_podcast_asset_metadata"),
+                extras=meta,
+            )
+            asset.asset_metadata = normalized
+            db.add(asset)
+            updated += 1
+
+        db.commit()
+        print(f"Scanned={scanned} Updated={updated} Since={since.isoformat()}")
+    finally:
+        db.close()
+
+
+if __name__ == "__main__":
+    main()

backend/services/content_asset_service.py

@@ -107,7 +107,6 @@ class ContentAssetService:
         user_id: str,
         asset_type: Optional[AssetType] = None,
         source_module: Optional[AssetSource] = None,
-        source_modules: Optional[List[AssetSource]] = None,
         search_query: Optional[str] = None,
         tags: Optional[List[str]] = None,
         favorites_only: bool = False,
@@ -126,7 +125,6 @@ class ContentAssetService:
             user_id: Clerk user ID
             asset_type: Filter by asset type (optional)
             source_module: Filter by source module (optional)
-            source_modules: Filter by multiple source modules (optional)
             search_query: Search in title, description, prompt (optional)
             tags: Filter by tags (optional)
             favorites_only: Only return favorites (optional)
@@ -144,9 +142,7 @@ class ContentAssetService:
             if asset_type:
                 query = query.filter(ContentAsset.asset_type == asset_type)
             
-            if source_modules:
-                query = query.filter(ContentAsset.source_module.in_(source_modules))
-            elif source_module:
+            if source_module:
                 query = query.filter(ContentAsset.source_module == source_module)
             
             if favorites_only:

backend/tests/api/test_content_assets_router.py

@@ -1,31 +0,0 @@
-import importlib.util
-from pathlib import Path
-from fastapi import HTTPException
-
-ROOT = Path(__file__).resolve().parents[3]
-ROUTER_PATH = ROOT / 'backend' / 'api' / 'content_assets' / 'router.py'
-MODELS_PATH = ROOT / 'backend' / 'models' / 'content_asset_models.py'
-
-models_spec = importlib.util.spec_from_file_location('content_asset_models', MODELS_PATH)
-models = importlib.util.module_from_spec(models_spec)
-models_spec.loader.exec_module(models)
-AssetSource = models.AssetSource
-
-router_spec = importlib.util.spec_from_file_location('content_assets_router', ROUTER_PATH)
-router = importlib.util.module_from_spec(router_spec)
-router_spec.loader.exec_module(router)
-
-
-def test_parse_source_modules_supports_repeated_and_csv_values():
-    parsed = router._parse_source_modules(["blog_writer", "youtube,podcast"])
-    assert parsed == [AssetSource.BLOG_WRITER, AssetSource.YOUTUBE, AssetSource.PODCAST]
-
-
-def test_parse_source_modules_raises_for_invalid_values():
-    try:
-        router._parse_source_modules(["blog_writer,unknown"])
-    except HTTPException as exc:
-        assert exc.status_code == 400
-        assert "Invalid source module" in exc.detail
-    else:
-        raise AssertionError("Expected HTTPException for invalid source module")

backend/tests/services/test_content_asset_service.py

@@ -1,50 +0,0 @@
-import importlib.util
-from pathlib import Path
-
-ROOT = Path(__file__).resolve().parents[3]
-SERVICE_PATH = ROOT / 'backend' / 'services' / 'content_asset_service.py'
-MODELS_PATH = ROOT / 'backend' / 'models' / 'content_asset_models.py'
-
-models_spec = importlib.util.spec_from_file_location('content_asset_models', MODELS_PATH)
-models = importlib.util.module_from_spec(models_spec)
-models_spec.loader.exec_module(models)
-AssetSource = models.AssetSource
-
-service_spec = importlib.util.spec_from_file_location('content_asset_service', SERVICE_PATH)
-service_module = importlib.util.module_from_spec(service_spec)
-service_spec.loader.exec_module(service_module)
-ContentAssetService = service_module.ContentAssetService
-
-
-class DummyQuery:
-    def __init__(self):
-        self.filters = []
-
-    def filter(self, expr):
-        self.filters.append(expr)
-        return self
-
-    def count(self): return 0
-    def order_by(self, *_args, **_kwargs): return self
-    def limit(self, *_args, **_kwargs): return self
-    def offset(self, *_args, **_kwargs): return self
-    def all(self): return []
-
-
-class DummyDB:
-    def __init__(self): self.query_obj = DummyQuery()
-    def query(self, *_args, **_kwargs): return self.query_obj
-
-
-def test_get_user_assets_accepts_multiple_source_modules_filter():
-    db = DummyDB()
-    service = ContentAssetService(db)
-
-    assets, total = service.get_user_assets(
-        user_id="user-1",
-        source_modules=[AssetSource.BLOG_WRITER, AssetSource.YOUTUBE],
-    )
-
-    assert assets == []
-    assert total == 0
-    assert len(db.query_obj.filters) >= 2

backend/utils/asset_tracker.py

@@ -11,6 +11,8 @@ import logging
 import re
 from urllib.parse import urlparse
 
+from models.asset_metadata_schema import validate_asset_metadata
+
 logger = logging.getLogger(__name__)
 
 # Maximum file size (100MB)
@@ -140,6 +142,12 @@ def save_asset_to_library(
             if len(title) > 200:
                 title = title[:197] + '...'
         
+        metadata_payload = asset_metadata or {}
+        is_valid_metadata, validation_message = validate_asset_metadata(metadata_payload)
+        if not is_valid_metadata:
+            logger.error(f"Invalid asset metadata: {validation_message}")
+            return None
+
         service = ContentAssetService(db)
         asset = service.create_asset(
             user_id=user_id,
@@ -154,7 +162,7 @@ def save_asset_to_library(
             description=description,
             prompt=prompt,
             tags=tags or [],
-            asset_metadata=asset_metadata or {},
+            asset_metadata=metadata_payload,
             provider=provider,
             model=model,
             cost=cost,

frontend/src/hooks/__tests__/useContentAssets.test.ts

@@ -1,35 +0,0 @@
-import { renderHook, waitFor } from '@testing-library/react';
-import { useContentAssets } from '../useContentAssets';
-
-const getTokenMock = jest.fn();
-
-jest.mock('@clerk/clerk-react', () => ({
-  useAuth: () => ({ getToken: getTokenMock }),
-}));
-
-describe('useContentAssets', () => {
-  beforeEach(() => {
-    getTokenMock.mockResolvedValue('test-token');
-    global.fetch = jest.fn().mockResolvedValue({
-      ok: true,
-      json: async () => ({ assets: [], total: 0, limit: 100, offset: 0 }),
-    } as Response);
-  });
-
-  afterEach(() => {
-    jest.clearAllMocks();
-  });
-
-  it('sends all source_module values as repeated query params', async () => {
-    renderHook(() =>
-      useContentAssets({ source_module: ['blog_writer', 'youtube'], limit: 50, offset: 0 })
-    );
-
-    await waitFor(() => expect(global.fetch).toHaveBeenCalled());
-
-    const calledUrl = (global.fetch as jest.Mock).mock.calls[0][0] as string;
-    const params = new URL(calledUrl).searchParams;
-
-    expect(params.getAll('source_module')).toEqual(['blog_writer', 'youtube']);
-  });
-});

frontend/src/hooks/useContentAssets.ts

@@ -29,7 +29,7 @@ export interface ContentAsset {
 
 export interface AssetFilters {
   asset_type?: 'text' | 'image' | 'video' | 'audio';
-  source_module?: string | string[]; // Supports single or multiple source modules
+  source_module?: string | string[]; // Support single or multiple source modules
   search?: string;
   tags?: string[];
   favorites_only?: boolean;
@@ -146,10 +146,8 @@ export const useContentAssets = (filters: AssetFilters = {}) => {
       if (currentFilters.source_module) {
         // Handle both string and array cases
         if (Array.isArray(currentFilters.source_module)) {
-          // Send every selected source module as repeated query params
-          currentFilters.source_module.forEach((module) => {
-            params.append('source_module', module);
-          });
+          // For arrays, use the first value (backend doesn't support multiple yet)
+          params.append('source_module', currentFilters.source_module[0]);
         } else {
           params.append('source_module', currentFilters.source_module);
         }