Update app.py

app.py

@@ -90,10 +90,6 @@ def process_pdfs(uploaded_files, use_example, chunk_size, chunk_overlap, min_chu
         current_chunk = len(chunks) - 1
 
     chunk_text = chunks[current_chunk] if chunks else "No chunks available."    
-    # use_example_update = gr.update(
-    #     value=False,
-    #     interactive=uploaded_files is None or len(uploaded_files) == 0
-    #     )
         
     return pdf_chunks_dict, selected_pdfs, chunk_text, current_chunk#, use_example_update
 
@@ -151,18 +147,35 @@ def generate_synthetic_records(api_key, pdf_chunks_dict, num_records):
     }
 
     df_in = pd.DataFrame()
-    documents = list(pdf_chunks_dict.keys())
-    all_chunks = [(doc, chunk) for doc in documents for chunk in pdf_chunks_dict[doc]]
+    try:
+        documents = list(pdf_chunks_dict.keys())
+        all_chunks = [(doc, chunk) for doc in documents for chunk in pdf_chunks_dict[doc]]
+
+        for _ in range(num_records):
+            doc, chunk = random.choice(all_chunks)
+            df_doc = pd.DataFrame({'document': [doc], 'text': [chunk]})
+            df_in = pd.concat([df_in, df_doc], ignore_index=True)
+
+        df = navigator.edit(PROMPT, seed_data=df_in, **GENERATE_PARAMS)
+        df = df.drop(columns=['text'])
 
-    for _ in range(num_records):
-        doc, chunk = random.choice(all_chunks)
-        df_doc = pd.DataFrame({'document': [doc], 'text': [chunk]})
-        df_in = pd.concat([df_in, df_doc], ignore_index=True)
+        csv_file = os.path.join(output_dir, "synthetic_qa.csv")
+        df.to_csv(csv_file, index=False)
 
-    df = navigator.edit(PROMPT, seed_data=df_in, **GENERATE_PARAMS)
-    df = df.drop(columns=['text'])
+        return gr.update(value=df, visible=True), csv_file
 
-    return gr.update(value=df, visible=True)
+    except:
+        return gr.update(value=None, visible=False), None
+
+# def generate_and_show_download_button(api_key, pdf_chunks_dict, num_records):
+#     df = generate_synthetic_records(api_key, pdf_chunks_dict, num_records)
+#     csv_file = download_dataframe(df)
+#     return df, gr.update(visible=True), csv_file
+
+def download_dataframe(df):
+    csv_file = os.path.join(output_dir, "synthetic_qa.csv")
+    df.to_csv(csv_file, index=False)
+    return csv_file
 
 # CSS styling to center the logo and prevent right-click download
 css = """
@@ -273,6 +286,7 @@ with gr.Blocks() as demo:
             num_records = gr.Number(label="Number of Records", value=10)
 
             generate_button = gr.Button("Generate Synthetic Records", interactive=False)
+            download_link = gr.File(label="Download Link", visible=False)
 
             # Validate API key on input change and update button interactivity
             api_key_input.change(
@@ -283,10 +297,14 @@ with gr.Blocks() as demo:
 
             output_df = gr.Dataframe(headers=["document", "topic", "user_profile", "question", "answer", "context"], wrap=True, visible=True)
 
+            def generate_and_prepare_download(api_key, pdf_chunks_dict, num_records):
+                df, csv_file = generate_synthetic_records(api_key, pdf_chunks_dict, num_records)
+                return df, gr.update(value=csv_file, visible=df['value']!=None)
+
             generate_button.click(
-                fn=generate_synthetic_records,
+                fn=generate_and_prepare_download,
                 inputs=[api_key_input, pdf_chunks_dict, num_records],
-                outputs=[output_df]
+                outputs=[output_df, download_link]
             )
 
 demo.launch()