Update app.py

app.py

@@ -4,42 +4,42 @@ from pdf2image import convert_from_path
 from openai import OpenAI
 import base64
 import asyncio
+from datetime import datetime
 import gradio as gr
 
-# Get the API key from environment variables
+# We'll use an environment variable for the API key in Spaces
 OPENAI_API_KEY = os.environ.get("OPENAI_API_KEY")
 
-class PDFFormExtractor:
+class PDFTextExtractor:
     def __init__(self, api_key):
         self.client = OpenAI(api_key=api_key)
-    
-    async def extract_first_form(self, pdf_path):
+
+    async def extract_text_from_pdf(self, pdf_path):
         try:
             if not os.path.exists(pdf_path):
                 raise FileNotFoundError(f"PDF file not found: {pdf_path}")
             
             print(f"Processing PDF: {pdf_path}")
             
-            # Convert only the first page to an image
-            images = convert_from_path(pdf_path, first_page=1, last_page=1)
-            
-            if not images:
-                return "No pages found in the PDF."
-            
-            image = images[0]
-            print("Processing first page...")
+            images = convert_from_path(pdf_path)
             
-            img_buffer = io.BytesIO()
-            image.save(img_buffer, format='PNG')
-            img_base64 = base64.b64encode(img_buffer.getvalue()).decode('utf-8')
-            
-            # Custom prompt specifically for extracting the Patient Admission Form
-            prompt = """
-You are an expert in document processing and OCR with deep knowledge of Markdown formatting. Your task is to extract the structured content from a PDF page image and convert it into a clean Markdown format.
-
-Extract the Patient Admission Form (KMHIPF002V3) from this image and format it as follows:
+            extracted_texts = []
+            for i, image in enumerate(images):
+                print(f"Processing page {i+1}...")
+                
+                img_buffer = io.BytesIO()
+                image.save(img_buffer, format='PNG')
+                img_base64 = base64.b64encode(img_buffer.getvalue()).decode('utf-8')
+                
+                response = self.client.chat.completions.create(
+                model="gpt-4o",
+                messages=[
+                    {
+                        "role": "system",
+                        "content": """You are an expert in document processing and OCR with deep knowledge of Markdown formatting. 
+                        Extract the Patient Admission Form (KMHIPF002V3) from this image and format it in clean markdown.
+                        Follow this exact markdown structure:
 
-```markdown
 # PATIENT ADMISSION FORM
 ## DR.KAMAKSHI MEMORIAL HOSPITAL, PALLIKARANAI, CHENNAI.
 
@@ -123,54 +123,55 @@ Contact No.: ________________
 *\\* Subject to change during the course of diseases*
 
 ---
-*Form No: KMHIPF002V3* 
-"""
-
-            response = self.client.chat.completions.create(
-            model="gpt-4o",
-            messages=[
-                {"role": "system", "content": prompt},
+*Form No: KMHIPF002V3*
+            """
+        },
+        {
+            "role": "user",
+            "content": [
+                {
+                    "type": "text",
+                    "text": "Extract and format the Patient Admission Form from this image according to the specified markdown format. Preserve all form fields and checkboxes (as □)."
+                },
                 {
-                    "role": "user",
-                    "content": [
-                        {
-                            "type": "text",
-                            "text": "Extract and format the Patient Admission Form from this image according to the specified markdown format."
-                        },
-                        {
-                            "type": "image_url",
-                            "image_url": {"url": f"data:image/png;base64,{img_base64}"}
-                        }
-                    ]
+                    "type": "image_url",
+                    "image_url": {
+                        "url": f"data:image/png;base64,{img_base64}"
+                    }
                 }
-            ],
-            max_tokens=4096
-        )
-
-        result = response.choices[0].message["content"]
-        return result
-
-    except Exception as e:
-        print(f"Error in form extraction: {str(e)}")
-        return f"Error: {str(e)}"
+            ]
+        }
+    ],
+    max_tokens=4096
+)
 
+return response.choices[0].message.content
 
-def extract_first_form(pdf_file):
+def extract_text(pdf_file):
     if OPENAI_API_KEY is None:
         return "Error: OpenAI API key not found. Please set the OPENAI_API_KEY environment variable."
 
-        extractor = PDFFormExtractor(OPENAI_API_KEY)
+    extractor = PDFTextExtractor(OPENAI_API_KEY)
+    
+    pdf_path = pdf_file.name
+    extracted_texts = asyncio.run(extractor.extract_text_from_pdf(pdf_path))
+    
+    if extracted_texts:
+        output = ""
+        for page in extracted_texts:
+            output += f"\n\n=== Page {page['page']} ===\n\n"
+            output += page['text']
         
-        pdf_path = pdf_file.name
-        result = asyncio.run(extractor.extract_first_form(pdf_path))
-        return result
-
+        return output
+    else:
+        return "Failed to extract text from PDF"
 
 iface = gr.Interface(
-fn=extract_first_form,
-inputs=gr.File(label="Upload PDF with Patient Admission Form"),
-outputs=gr.Textbox(label="Extracted Form in Markdown", lines=30),
-title="Patient Admission Form Extractor",
-description="Upload a PDF file to extract the Patient Admission Form (first form) in markdown format."
+    fn=extract_text,
+    inputs=gr.File(label="Upload PDF"),
+    outputs="text",
+    title="PDF Text Extractor",
+    description="Upload a PDF file to extract all text using OpenAI's GPT-4 Vision."
 )
-iface.launch()
+
+iface.launch()