Update app.py

app.py

@@ -9,21 +9,34 @@ from src.format_script import format_script
 from src.eleven import generate_podcast as generate_audio
 
 def process_file(pdf_file, progress=gr.Progress()):
+    """Process uploaded PDF through the entire pipeline"""
+    
     # Create temp directory for intermediate files
     os.makedirs('temp', exist_ok=True)
     
     try:
         # Step 1: Process PDF
-        progress(0.1, desc="Cleaning PDF...")
-        with tempfile.NamedTemporaryFile(delete=False, suffix='.pdf') as tmp_pdf:
-            tmp_pdf.write(pdf_file.read())
-            pdf_text = extract_text_from_pdf(tmp_pdf.name)
+        progress(0.1, desc="Reading PDF...")
+        if pdf_file is None:
+            raise gr.Error("Please upload a PDF file")
+            
+        # Read PDF content
+        pdf_content = pdf_file.read()
+        
+        # Extract and clean text
+        progress(0.2, desc="Extracting text...")
+        pdf_text = extract_text_from_pdf(pdf_content)
+        if not pdf_text:
+            raise gr.Error("Failed to extract text from PDF")
             
-        # Clean and process text
+        # Process text in chunks
         chunks = create_screenplay_chunks(pdf_text)
         cleaned_text = ""
+        chunk_count = len(chunks)
+        
         for i, chunk in enumerate(chunks):
-            progress(0.1 + (0.2 * (i/len(chunks))), desc=f"Processing chunk {i+1}/{len(chunks)}...")
+            progress(0.2 + (0.2 * (i/chunk_count)), 
+                    desc=f"Processing chunk {i+1}/{chunk_count}...")
             processed = process_screenplay_chunk(chunk, i)
             cleaned_text += processed + "\n\n"
             
@@ -32,26 +45,26 @@ def process_file(pdf_file, progress=gr.Progress()):
             f.write(cleaned_text)
             
         # Step 2: Generate Initial Podcast Script
-        progress(0.3, desc="Generating podcast script...")
+        progress(0.4, desc="Generating podcast script...")
         initial_script = generate_initial_podcast(cleaned_text)
         with open("temp/generated_podcast.pkl", "wb") as f:
             pickle.dump(initial_script, f)
             
         # Step 3: Dramatize Script
-        progress(0.5, desc="Dramatizing script...")
+        progress(0.6, desc="Dramatizing script...")
         dramatized_script = dramatize_podcast(initial_script)
-        with open("temp/dramatized_podcast.pkl", "wb") as f:
-            pickle.dump(dramatized_script, f)
-        
+        with open("temp/dramatized_podcast.txt", "w") as f:
+            f.write(dramatized_script)
+            
         # Step 4: Format Script
-        progress(0.7, desc="Formatting script...")
+        progress(0.8, desc="Formatting script...")
         format_script("temp/dramatized_podcast.txt", "temp/formatted_podcast.txt")
         with open("temp/formatted_podcast.txt", "r") as f:
             formatted_script = f.read()
         
         # Step 5: Generate Audio
         progress(0.9, desc="Generating audio...")
-        audio_file = generate_audio()  # This will read formatted_podcast.txt
+        audio_file = generate_audio()
         
         progress(1.0, desc="Complete!")
         
@@ -65,10 +78,6 @@ def process_file(pdf_file, progress=gr.Progress()):
     
     except Exception as e:
         raise gr.Error(f"Error processing file: {str(e)}")
-    finally:
-        # Cleanup temp files if needed
-        if os.path.exists(tmp_pdf.name):
-            os.unlink(tmp_pdf.name)
 
 # Create Gradio interface
 with gr.Blocks(title="PDF to Podcast Converter") as demo:
@@ -140,15 +149,6 @@ with gr.Blocks(title="PDF to Podcast Converter") as demo:
         outputs=outputs,
         api_name="convert"
     )
-    
-    # Example section
-    gr.Examples(
-        examples=["example.pdf"],
-        inputs=pdf_input,
-        outputs=outputs,
-        fn=process_file,
-        cache_examples=True
-    )
 
 if __name__ == "__main__":
     demo.launch()