Upload 7 files

databases.py

@@ -340,6 +340,47 @@ def get_audiodescription(sha1sum: str, version: str) -> Optional[sqlite3.Row]:
         return None
 
 
+def get_videos_from_audiodescriptions() -> List[Dict[str, Any]]:
+    """Retorna vídeos disponibles segons audiodescriptions.db.
+
+    Llegeix demo/temp/db/audiodescriptions.db i retorna una llista de
+    diccionaris amb claus com a mínim: sha1sum i video_name.
+
+    - Si la taula té columna video_name, es fa DISTINCT(sha1sum, video_name).
+    - Si no la té, es fa DISTINCT(sha1sum) i s'utilitza sha1sum com a
+      video_name de manera per defecte.
+    """
+
+    try:
+        with _connect_audiodescriptions_db() as conn:
+            cur = conn.cursor()
+
+            # Detectar si existeix la columna video_name
+            cols = [r[1] for r in cur.execute("PRAGMA table_info(audiodescriptions)")]
+            has_video_name = "video_name" in cols
+
+            if has_video_name:
+                cur.execute("SELECT DISTINCT sha1sum, video_name FROM audiodescriptions")
+                rows = cur.fetchall()
+                result: List[Dict[str, Any]] = []
+                for r in rows:
+                    sha1 = r["sha1sum"]
+                    vname = r["video_name"] or r["sha1sum"]
+                    result.append({"sha1sum": sha1, "video_name": vname})
+                return result
+            else:
+                cur.execute("SELECT DISTINCT sha1sum FROM audiodescriptions")
+                rows = cur.fetchall()
+                result = []
+                for r in rows:
+                    sha1 = r["sha1sum"]
+                    result.append({"sha1sum": sha1, "video_name": sha1})
+                return result
+    except sqlite3.OperationalError:
+        # Taula o BD encara no creada
+        return []
+
+
 def get_audiodescription_history(sha1sum: str, version: str) -> list[sqlite3.Row]:
     """Retorna només la darrera fila d'audiodescriptions per sha1sum+version.
 

page_modules/analyze_audiodescriptions.py

@@ -7,6 +7,7 @@ import io
 from pathlib import Path
 from typing import Dict, Optional
 import hashlib
+import runpy
 
 import streamlit as st
 import yaml
@@ -14,7 +15,7 @@ import yaml
 from utils import save_bytes
 from persistent_data_gate import ensure_media_for_video
 from databases import (
-    get_accessible_videos_with_sha1,
+    get_videos_from_audiodescriptions,
     insert_demo_feedback_row,
     get_audiodescription,
     get_audiodescription_history,
@@ -152,11 +153,10 @@ def load_eval_values(vid_dir: Path, version: str, eval_content: Optional[str] =
 def render_analyze_audiodescriptions_page(api, permissions: Dict[str, bool]) -> None:
     st.header("Analitzar audiodescripcions")
 
-    # Llista de vídeos accessibles des de demo/temp/videos.db
-    session_id = st.session_state.get("session_id")
-    accessible_rows = get_accessible_videos_with_sha1(session_id)
+    # Llista de vídeos disponibles segons demo/temp/db/audiodescriptions.db
+    accessible_rows = get_videos_from_audiodescriptions()
 
-    # DEBUG: llistar parelles <sha1sum, video_name> extretes de videos.db
+    # DEBUG: llistar parelles <sha1sum, video_name> extretes de audiodescriptions.db
     try:
         debug_pairs = [(row.get("sha1sum"), row.get("video_name")) for row in accessible_rows]
         print(f"[DEBUG] accessible_rows (sha1sum, video_name) des de videos.db: {debug_pairs}")
@@ -167,18 +167,10 @@ def render_analyze_audiodescriptions_page(api, permissions: Dict[str, bool]) ->
     base_dir = Path(__file__).resolve().parent.parent
     base_media_dir = base_dir / "temp" / "media"
 
-    # Filtrar vídeos que tinguin carpeta de media disponible
-    filtered_rows = []
-    for row in accessible_rows:
-        sha1 = row.get("sha1sum")
-        if not sha1:
-            continue
-        media_dir = base_media_dir / sha1
-        if media_dir.exists() and media_dir.is_dir():
-            filtered_rows.append(row)
-
+    # Ja no filtrem per media: mostrem tots els vídeos presents a audiodescriptions.db
+    filtered_rows = accessible_rows
     if not filtered_rows:
-        st.info("No hi ha cap vídeo amb media disponible a temp/media per analitzar.")
+        st.info("No hi ha cap vídeo disponible per analitzar a audiodescriptions.db.")
         st.stop()
 
     # DEBUG: llistar subcarpetes actuals sota demo/temp/media