Update app.py

app.py

@@ -35,7 +35,7 @@ def render_pdf_page(page, max_resolution=1540, scale=2.77):
 def process_pdf(pdf_path, num_pages=1):
     pdf = pdfium.PdfDocument(pdf_path)
     total_pages = len(pdf)
-    pages_to_process = min(num_pages, total_pages, 5)
+    pages_to_process = min(int(num_pages), total_pages, 5)
     images = []
     
     for i in range(pages_to_process):
@@ -60,7 +60,7 @@ def process_input(file_input, temperature, num_pages):
     
     if file_path.lower().endswith('.pdf'):
         try:
-            images_to_process, total_pages = process_pdf(file_path, num_pages)
+            images_to_process, total_pages = process_pdf(file_path, int(num_pages))
             if len(images_to_process) == 0:
                 yield "Error: Could not extract pages from PDF.", "", "", None
                 return
@@ -82,7 +82,7 @@ def process_input(file_input, temperature, num_pages):
             yield f"Error opening image: {str(e)}", "", "", None
             return
     
-    content = [{"type": "text", "text": ""}]
+    content = [{"type": "text", "text": "Extract the text from this image."}]
     
     for img in images_to_process:
         try:
@@ -97,12 +97,7 @@ def process_input(file_input, temperature, num_pages):
     
     payload = {
         "model": MODEL,
-        "messages": [
-            {
-                "role": "user",
-                "content": content
-            }
-        ],
+        "messages": [{"role": "user", "content": content}],
         "temperature": temperature,
         "stream": True
     }
@@ -149,17 +144,17 @@ def process_input(file_input, temperature, num_pages):
 
 
 with gr.Blocks(title="📖 Image/PDF OCR", theme=gr.themes.Soft()) as demo:
-    gr.Markdown(
-        """
-        # 📖 Image/PDF to Text Extraction
-        **💡 How to use:**
-        1. Upload an image or PDF
-        2. For PDFs: choose how many pages to process (1-5, default is 1)
-        3. Adjust temperature if needed
-        4. Click "Extract Text"
-        Note: The Markdown rendering for tables is not always correct, check the raw output for complex tables!
-        """
-    )
+    gr.Markdown("""
+# 📖 Image/PDF to Text Extraction
+
+**💡 How to use:**
+1. Upload an image or PDF
+2. For PDFs: choose how many pages to process (1-5, default is 1)
+3. Adjust temperature if needed
+4. Click "Extract Text"
+
+**Note:** The Markdown rendering for tables is not always correct, check the raw output for complex tables!
+""")
     
     with gr.Row():
         with gr.Column(scale=1):
@@ -200,8 +195,7 @@ with gr.Blocks(title="📖 Image/PDF OCR", theme=gr.themes.Soft()) as demo:
         with gr.Column(scale=2):
             output_text = gr.Markdown(
                 label="📄 Extracted Text (Rendered)",
-                value="<div style='min-height: 600px; padding: 10px; border: 1px solid #e0e0e0; border-radius: 4px; background-color: #f9f9f9;'><em>Extracted text will appear here...</em></div>",
-                height=600
+                value="*Extracted text will appear here...*"
             )
     
     with gr.Row():
@@ -221,7 +215,7 @@ with gr.Blocks(title="📖 Image/PDF OCR", theme=gr.themes.Soft()) as demo:
     )
     
     clear_btn.click(
-        fn=lambda: (None, "", "", "", None, 1),
+        fn=lambda: (None, "*Extracted text will appear here...*", "", "", None, 1),
         outputs=[file_input, output_text, raw_output, page_info, rendered_image, num_pages]
     )