Update app.py

app.py

@@ -1,58 +1,62 @@
 import fitz  # PyMuPDF
-from pytesseract import image_to_string
 from PIL import Image
+import pytesseract
 from fpdf import FPDF
 import gradio as gr
 
-# Function to process PDF
-def process_pdf(pdf_file):
-    # Open the uploaded PDF
-    pdf_document = fitz.open(stream=pdf_file.read(), filetype="pdf")
-    processed_pages = []
-
-    for page_number in range(len(pdf_document)):
-        # Extract each page as an image
-        page = pdf_document.load_page(page_number)
-        pix = page.get_pixmap()
-        img = Image.frombytes("RGB", [pix.width, pix.height], pix.samples)
-
-        # Perform OCR on the image
-        text = image_to_string(img)
-
-        # Store extracted text and original image for later processing
-        processed_pages.append({"image": img, "text": text})
-
-    # Create a new PDF with editable text overlaying preserved shapes
-    pdf_output = FPDF()
-    pdf_output.set_auto_page_break(auto=True, margin=15)
-
-    for page_data in processed_pages:
-        pdf_output.add_page()
-        pdf_output.set_font("Arial", size=12)
-
-        # Overlay extracted text as editable content
-        pdf_output.multi_cell(0, 10, page_data["text"])
-
-    # Save the output PDF
-    output_file = "processed_output.pdf"
-    pdf_output.output(output_file)
-
-    return output_file
-
-# Gradio interface function
-def gradio_interface(pdf_file):
-    output_pdf_path = process_pdf(pdf_file)
+# Path to Tesseract executable
+pytesseract.pytesseract.tesseract_cmd = "/usr/bin/tesseract"
+
+# Step 1: Convert PDF Pages to Images
+def pdf_to_images(pdf_path):
+    pdf_document = fitz.open(pdf_path)
+    images = []
+    for page_num in range(len(pdf_document)):
+        page = pdf_document[page_num]
+        pix = page.get_pixmap()  # Render the page as an image
+        image_path = f"page_{page_num + 1}.png"
+        pix.save(image_path)
+        images.append(image_path)
+    return images
+
+# Step 2: Extract Text Using Tesseract OCR
+def extract_text(images):
+    text_pages = []
+    for image_path in images:
+        text = pytesseract.image_to_string(Image.open(image_path))  # Perform OCR
+        text_pages.append(text)
+    return text_pages
+
+# Step 3: Replace Curved Text with Editable Text in a New PDF
+def create_editable_pdf(images, text_pages, output_pdf_path):
+    pdf = FPDF()
+    pdf.set_auto_page_break(auto=True, margin=15)
+    for text in text_pages:
+        pdf.add_page()
+        pdf.set_font("Arial", size=12)
+        pdf.multi_cell(0, 10, text)
+    pdf.output(output_pdf_path)
+
+# Main Function
+def process_pdf(file):
+    input_pdf_path = file.name
+    output_pdf_path = "Editable_Output.pdf"
+    
+    # Convert PDF to images and perform OCR
+    images = pdf_to_images(input_pdf_path)
+    text_pages = extract_text(images)
+    
+    # Create a new editable PDF
+    create_editable_pdf(images, text_pages, output_pdf_path)
     return output_pdf_path
 
-# Create Gradio app
-interface = gr.Interface(
-    fn=gradio_interface,
-    inputs=gr.inputs.File(label="Upload a PDF"),
-    outputs=gr.outputs.File(label="Download Processed PDF"),
-    title="PDF Text Processing",
-    description="Upload a PDF, extract text, and download a processed version with editable text and preserved diagrams."
+# Gradio Interface
+iface = gr.Interface(
+    fn=process_pdf,
+    inputs=gr.File(label="Upload PDF"),
+    outputs=gr.File(label="Download Editable PDF"),
+    title="OCR PDF to Editable Text",
+    description="Upload a PDF to extract and replace text while preserving shapes and layout.",
 )
 
-# Launch the app
-if __name__ == "__main__":
-    interface.launch()
+iface.launch(share=True)