Persist OCR captions in gallery

core/database.py

@@ -276,6 +276,7 @@ class GalleryImage(TimestampMixin, Base):
     id         = Column(String, primary_key=True, index=True)
     filename   = Column(String, nullable=False, unique=True)
     prompt     = Column(Text, nullable=False, default="")
+    caption    = Column(Text, nullable=True, default="")
     model      = Column(String, nullable=True)
     size       = Column(String, nullable=True)
     quality    = Column(String, nullable=True)
@@ -1182,6 +1183,29 @@ def _migrate_add_multiuser_owner_columns():
     _migrate_add_owner_to_table("documents", "ix_documents_owner")
 
 
+def _migrate_add_gallery_caption_column():
+    """Add OCR/vision caption storage for gallery images."""
+    import sqlite3
+    db_path = DATABASE_URL.replace("sqlite:///", "")
+    if not os.path.exists(db_path):
+        return
+    conn = None
+    try:
+        conn = sqlite3.connect(db_path)
+        columns = [row[1] for row in conn.execute("PRAGMA table_info(gallery_images)").fetchall()]
+        if columns and "caption" not in columns:
+            conn.execute("ALTER TABLE gallery_images ADD COLUMN caption TEXT DEFAULT ''")
+            conn.commit()
+            logging.getLogger(__name__).info("Migrated: added caption column to gallery_images")
+    except Exception as e:
+        logging.getLogger(__name__).warning(f"Migration gallery caption column failed: {e}")
+    finally:
+        try:
+            conn.close()
+        except Exception:
+            pass
+
+
 def _migrate_add_api_token_scopes_column():
     """Add API token scopes for existing installs.
 
@@ -1812,6 +1836,7 @@ def init_db():
     _migrate_add_token_columns()
     _migrate_add_mode_column()
     _migrate_add_multiuser_owner_columns()
+    _migrate_add_gallery_caption_column()
     _migrate_add_api_token_scopes_column()
     _migrate_backfill_document_owner_from_session()
     _migrate_assign_legacy_owner()

routes/gallery_helpers.py

@@ -99,6 +99,7 @@ def _image_to_dict(img: GalleryImage, session_name: str = None) -> Dict[str, Any
         "filename": img.filename,
         "url": f"/api/generated-image/{img.filename}",
         "prompt": img.prompt,
+        "caption": img.caption or "",
         "model": img.model,
         "size": img.size,
         "quality": img.quality,

routes/gallery_routes.py

@@ -508,6 +508,7 @@ def setup_gallery_routes() -> APIRouter:
                 from sqlalchemy import or_
                 q = q.filter(or_(
                     GalleryImage.prompt.ilike(term),
+                    GalleryImage.caption.ilike(term),
                     GalleryImage.tags.ilike(term),
                     GalleryImage.ai_tags.ilike(term),
                 ))

routes/upload_routes.py

@@ -267,6 +267,32 @@ def setup_upload_routes(upload_handler):
         os.makedirs(cache_dir, exist_ok=True)
         return os.path.join(cache_dir, file_id + ".txt")
 
+    def _sync_gallery_caption_for_upload(info: dict | None, owner: str | None, text: str) -> None:
+        """Copy upload OCR/vision text onto the promoted gallery image row."""
+        if not info:
+            return
+        file_hash = info.get("hash")
+        if not file_hash:
+            return
+        db = SessionLocal()
+        try:
+            q = db.query(GalleryImage).filter(
+                GalleryImage.file_hash == file_hash,
+                GalleryImage.is_active == True,  # noqa: E712
+            )
+            if owner:
+                q = q.filter(GalleryImage.owner == owner)
+            img = q.first()
+            if not img:
+                return
+            img.caption = (text or "").strip()
+            db.commit()
+        except Exception as e:
+            db.rollback()
+            logger.warning("Failed to sync OCR caption to gallery image: %s", e)
+        finally:
+            db.close()
+
     @router.get("/{file_id}/vision")
     async def get_vision_text(request: Request, file_id: str, force: int = 0):
         """Return the vision-model OCR/description for an uploaded image.
@@ -293,7 +319,9 @@ def setup_upload_routes(upload_handler):
         if not force and os.path.exists(cache_path):
             try:
                 with open(cache_path, encoding="utf-8") as f:
-                    return {"text": f.read(), "cached": True}
+                    cached_text = f.read()
+                _sync_gallery_caption_for_upload(info, file_owner or current_user, cached_text)
+                return {"text": cached_text, "cached": True}
             except Exception as e:
                 logger.warning(f"Vision cache read failed for {file_id}: {e}")
         from src.document_processor import analyze_image_with_vl
@@ -307,6 +335,7 @@ def setup_upload_routes(upload_handler):
                 f.write(text)
         except Exception as e:
             logger.warning(f"Vision cache write failed for {file_id}: {e}")
+        _sync_gallery_caption_for_upload(info, file_owner or current_user, text)
         return {"text": text, "cached": False}
 
     @router.put("/{file_id}/vision")
@@ -334,6 +363,7 @@ def setup_upload_routes(upload_handler):
             raise HTTPException(400, "text must be a string")
         with open(_vision_cache_path(file_id), "w", encoding="utf-8") as f:
             f.write(text)
+        _sync_gallery_caption_for_upload(info, file_owner or current_user, text)
         return {"ok": True}
 
     async def periodic_rate_limit_cleanup():

static/js/gallery.js

@@ -1201,7 +1201,7 @@ function _renderGrid() {
       .replace(/\.[^.]+$/, '')       // drop extension
       .replace(/[_-]+/g, ' ')
       .trim();
-    const labelText = (img.prompt || '').trim() || fallbackName || 'Photo';
+    const labelText = (img.caption || '').trim() || (img.prompt || '').trim() || fallbackName || 'Photo';
     const promptPreview = labelText.length > 60 ? labelText.substring(0, 58) + '...' : labelText;
     const favCls = img.favorite ? ' gallery-fav-active' : '';
     html += `
@@ -1410,6 +1410,7 @@ function _openDetail(img) {
             <svg class="gallery-name-enter" width="13" height="13" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2" stroke-linecap="round" stroke-linejoin="round" aria-hidden="true"><polyline points="9 10 4 15 9 20"/><path d="M20 4v7a4 4 0 0 1-4 4H4"/></svg>
           </div>
         </div>
+        ${img.caption ? `<div class="gallery-detail-section"><label>OCR Caption</label><div class="gallery-detail-prompt">${_esc(img.caption)}</div></div>` : ''}
         ${img.prompt && img.model !== 'imported' ? `<div class="gallery-detail-section"><label>Prompt</label><div class="gallery-detail-prompt">${_esc(img.prompt)}</div></div>` : ''}
         <div class="gallery-detail-section gallery-detail-section-date">
           <label>Date</label>