app.py

import gradio as gr
from PIL import Image
import pytesseract
import cv2
import numpy as np
import re
import csv
import tempfile
import os

# Synonyms for features
feature_synonyms = {
    "Caller Number Display": ["Caller Number Display", "Caller ID", "Calling Number"],
    "Incoming Calls": ["Incoming Calls", "Inbound Calls"],
    "Incoming SMS": ["Incoming SMS", "Received SMS", "Inbound SMS"],
}

def find_feature(text, synonyms_list):
    for feature_name in synonyms_list:
        pattern = rf"({feature_name})\s*(FREE|Included|Yes)?"
        match = re.search(pattern, text, re.IGNORECASE)
        if match:
            return match.group(0).strip()
    return ""

def parse_text(text):
    data = {
        "Mobile Number": "",
        "Receipt Date": "",
        "Total Amount Paid": "",
        "Local Data": "",
        "Call Time": "",
        "Number of SMS": "",
        "Roaming Details": "",
        "Caller Number Display": "",
        "Incoming Calls": "",
        "Incoming SMS": ""
    }

    mobile_match = re.search(r"Mobile\s*No\.?\s*[:\-]?\s*(\d+)", text, re.IGNORECASE)
    if mobile_match:
        data["Mobile Number"] = mobile_match.group(1)

    date_match = re.search(r"Receipt Date\s*[:\-]?\s*([\d]{1,2}[-/][A-Za-z]{3}[-/][\d]{4})", text, re.IGNORECASE)
    if date_match:
        data["Receipt Date"] = date_match.group(1)

    total_match = re.search(r"Total Amount Paid\s*\$?([\d\.,]+)", text, re.IGNORECASE)
    if total_match:
        data["Total Amount Paid"] = total_match.group(1)

    local_data_match = re.search(r"(\d+GB\s*plan.*?)(?:\n|$)", text, re.IGNORECASE)
    if local_data_match:
        data["Local Data"] = local_data_match.group(1).strip()

    call_time_match = re.search(r"(\d+\s*mins.*?)(?:\n|$)", text, re.IGNORECASE)
    if call_time_match:
        data["Call Time"] = call_time_match.group(1).strip()

    sms_match = re.search(r"(\d+\s*SMS.*?)(?:\n|$)", text, re.IGNORECASE)
    if sms_match:
        data["Number of SMS"] = sms_match.group(1).strip()

    roaming_match = re.search(r"(\d+GB.*?Roaming.*?)(?:\n|$)", text, re.IGNORECASE)
    if roaming_match:
        data["Roaming Details"] = roaming_match.group(1).strip()

    data["Caller Number Display"] = find_feature(text, feature_synonyms["Caller Number Display"])
    data["Incoming Calls"] = find_feature(text, feature_synonyms["Incoming Calls"])
    data["Incoming SMS"] = find_feature(text, feature_synonyms["Incoming SMS"])

    return data

def ocr_extract(image):
    if image is None:
        return "Please upload an image.", None

    # Convert to OpenCV format
    image_np = np.array(image)
    image_cv = cv2.cvtColor(image_np, cv2.COLOR_RGB2BGR)

    # Preprocess image for OCR
    gray = cv2.cvtColor(image_cv, cv2.COLOR_BGR2GRAY)
    _, thresh = cv2.threshold(gray, 150, 255, cv2.THRESH_BINARY)

    # Extract text using pytesseract
    text = pytesseract.image_to_string(thresh)

    # Parse the extracted text
    parsed_data = parse_text(text)

    # Create a temp file to save CSV
    tmpfile = tempfile.NamedTemporaryFile(delete=False, suffix=".csv", mode="w", encoding="utf-8")
    writer = csv.DictWriter(tmpfile, fieldnames=parsed_data.keys())
    writer.writeheader()
    writer.writerow(parsed_data)
    tmpfile_path = tmpfile.name
    tmpfile.close()

    # Prepare display text as before
    display_text = "\n".join(f"{k}: {v}" for k, v in parsed_data.items())

    return display_text, tmpfile_path  # Return path string, not BytesIO

iface = gr.Interface(
    fn=ocr_extract,
    inputs=gr.Image(type="pil"),
    outputs=[gr.Textbox(label="Extracted Info"), gr.File(label="Download CSV")],
    title="Bill OCR Scanner with CSV Export",
    description="Upload a bill image, extract key fields, and download the data as CSV."
)

if __name__ == "__main__":
    iface.launch(server_name="0.0.0.0", server_port=7860)