Update sozo_gen.py

sozo_gen.py

@@ -26,6 +26,7 @@ from google import genai
 import requests
 # In sozo_gen.py, near the other google imports
 from google.genai import types as genai_types
+import math # Add this import at the top of your sozo_gen.py file
 
 # --- Configuration ---
 logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(levelname)s - [%(funcName)s] - %(message)s')
@@ -359,9 +360,10 @@ def prepare_plot_data(spec: ChartSpecification, df: pd.DataFrame):
     return df[spec.x_col]
 
 # UPDATED: animate_chart now uses blit=False for accurate timing
+
 def animate_chart(spec: ChartSpecification, df: pd.DataFrame, dur: float, out: Path, fps: int = FPS) -> str:
     plot_data = prepare_plot_data(spec, df)
-    frames = max(10, int(dur * fps))
+    frames = math.ceil(dur * fps) # Use math.ceil to always round up frames
     fig, ax = plt.subplots(figsize=(WIDTH / 100, HEIGHT / 100), dpi=100)
     plt.tight_layout(pad=3.0)
     ctype = spec.chart_type
@@ -664,6 +666,8 @@ def generate_video_from_project(df: pd.DataFrame, raw_md: str, data_context: Dic
     logging.info(f"Generating video for project {project_id} with voice {voice_model}")
     llm = ChatGoogleGenerativeAI(model="gemini-2.5-flash", google_api_key=API_KEY, temperature=0.2)
     
+    domain = detect_dataset_domain(df)
+
     story_prompt = f"""
     Based on the following report, create a script for a {VIDEO_SCENES}-scene video.
     1. The first scene MUST be an "Introduction". It must contain narration and a stock video tag like: <generate_stock_video: "search query">.
@@ -673,10 +677,11 @@ def generate_video_from_project(df: pd.DataFrame, raw_md: str, data_context: Dic
     Report: {raw_md}
     Only output the script, no extra text.
     """
-    script = llm.invoke(story_prompt).content
+    script = llm.invoke(story_prompt).content.strip()
     scenes = [s.strip() for s in script.split("[SCENE_BREAK]") if s.strip()]
     video_parts, audio_parts, temps = [], [], []
     total_audio_duration = 0.0
+    conclusion_video_path = None
 
     for i, sc in enumerate(scenes):
         mp4 = Path(tempfile.gettempdir()) / f"{uuid.uuid4()}.mp4"
@@ -701,40 +706,54 @@ def generate_video_from_project(df: pd.DataFrame, raw_md: str, data_context: Dic
         video_dur = audio_dur + 1.5 
         
         try:
-            # --- Primary Visual Generation ---
-            chart_descs = extract_chart_tags(sc)
-            pexels_descs = extract_pexels_tags(sc)
-            
-            if pexels_descs:
-                logging.info(f"Scene {i+1}: Primary attempt with Pexels.")
-                query = extract_keywords_for_query(narrative, llm)
-                video_path = search_and_download_pexels_video(query, video_dur, mp4)
-                if not video_path: raise ValueError("Pexels search returned no results.")
+            primary_query = None
+            narration_lower = narrative.lower()
+            is_conclusion_scene = any(k in narration_lower for k in ["conclusion", "summary", "in closing", "final thoughts"])
+
+            if any(k in narration_lower for k in ["introduction", "welcome", "let's begin"]):
+                primary_query = f"abstract technology background {domain}"
+            elif is_conclusion_scene:
+                primary_query = f"future strategy business meeting {domain}"
+
+            if primary_query:
+                logging.info(f"Scene {i+1}: Pre-emptive guard triggered. Query: '{primary_query}'")
+                video_path = search_and_download_pexels_video(primary_query, video_dur, mp4)
+                if not video_path: raise ValueError("Pexels search failed for guarded query.")
                 video_parts.append(video_path)
-            elif chart_descs:
-                logging.info(f"Scene {i+1}: Primary attempt with animated chart.")
-                safe_chart(chart_descs[0], df, video_dur, mp4, data_context)
-                video_parts.append(str(mp4))
+                if is_conclusion_scene:
+                    conclusion_video_path = video_path
             else:
-                raise ValueError("No visual tag found in scene.")
+                chart_descs = extract_chart_tags(sc)
+                if chart_descs:
+                    logging.info(f"Scene {i+1}: Primary attempt with animated chart.")
+                    safe_chart(chart_descs[0], df, video_dur, mp4, data_context)
+                    video_parts.append(str(mp4))
+                else:
+                    raise ValueError("No chart tag found in a middle scene.")
         except Exception as e:
-            # --- Fallback Visual Generation ---
-            logging.warning(f"Scene {i+1}: Primary visual failed ({e}). Triggering fallback.")
-            try:
-                fallback_query = "abstract technology background"
-                video_path = search_and_download_pexels_video(fallback_query, video_dur, mp4)
-                if not video_path: raise ValueError("Fallback Pexels search failed.")
-                video_parts.append(video_path)
-            except Exception as fallback_e:
-                # --- Final Failsafe ---
-                logging.error(f"Scene {i+1}: Fallback visual also failed ({fallback_e}). Using placeholder.")
-                placeholder = placeholder_img()
-                placeholder.save(str(mp4).replace(".mp4", ".png"))
-                animate_image_fade(cv2.imread(str(mp4).replace(".mp4", ".png")), video_dur, mp4)
-                video_parts.append(str(mp4))
+            logging.warning(f"Scene {i+1}: Primary visual failed ({e}). Marking for fallback.")
+            video_parts.append("FALLBACK_NEEDED")
 
         temps.append(mp4)
 
+    # Post-processing loop to apply the conclusion video as a fallback
+    if not conclusion_video_path: # Failsafe if conclusion scene itself failed
+        logging.warning("No conclusion video was generated; creating a generic one for fallbacks.")
+        fallback_mp4 = Path(tempfile.gettempdir()) / f"{uuid.uuid4()}.mp4"
+        conclusion_video_path = search_and_download_pexels_video(f"data visualization abstract {domain}", 5.0, fallback_mp4)
+        if conclusion_video_path: temps.append(fallback_mp4)
+
+    final_video_parts = []
+    for part in video_parts:
+        if part == "FALLBACK_NEEDED":
+            if conclusion_video_path:
+                logging.info("Applying conclusion video as fallback for a failed scene.")
+                final_video_parts.append(conclusion_video_path)
+            else:
+                logging.error("Cannot apply fallback; no conclusion video available.")
+        else:
+            final_video_parts.append(part)
+
     with tempfile.NamedTemporaryFile(suffix=".mp4", delete=False) as temp_vid, \
             tempfile.NamedTemporaryFile(suffix=".mp3", delete=False) as temp_aud, \
             tempfile.NamedTemporaryFile(suffix=".mp4", delete=False) as final_vid:
@@ -743,7 +762,7 @@ def generate_video_from_project(df: pd.DataFrame, raw_md: str, data_context: Dic
         audio_mix_path = Path(temp_aud.name)
         final_vid_path = Path(final_vid.name)
         
-        concat_media(video_parts, silent_vid_path)
+        concat_media(final_video_parts, silent_vid_path)
         concat_media(audio_parts, audio_mix_path)
         
         cmd = [