app.py

import gradio as gr
import pdfplumber
import pandas as pd

# Function to process PDF and classify transactions
def process_pdf(file):
    if file is None:
        return "No file uploaded."
    
    # Extract text from the uploaded PDF
    with pdfplumber.open(file.name) as pdf:
        text = "\n".join([page.extract_text() for page in pdf.pages if page.extract_text()])

    # Extract transactions (Modify based on statement format)
    lines = text.split("\n")
    transactions = [line for line in lines if any(char.isdigit() for char in line)]  

    # Convert to DataFrame
    df = pd.DataFrame([line.split()[:3] for line in transactions], columns=["Date", "Description", "Amount"])

    # Classification function (Modify as needed)
    def classify_transaction(description):
        categories = {
            "Grocery": ["Walmart", "Kroger", "Whole Foods"],
            "Dining": ["McDonald's", "Starbucks", "Chipotle"],
            "Bills": ["Verizon", "AT&T", "Con Edison"],
            "Entertainment": ["Netflix", "Spotify", "Amazon Prime"],
            "Transport": ["Uber", "Lyft", "MetroCard"],
        }
        for category, keywords in categories.items():
            if any(keyword in description for keyword in keywords):
                return category
        return "Other"

    # Apply classification
    df["Category"] = df["Description"].apply(classify_transaction)

    return df  # Display the table

# Gradio Interface
app = gr.Interface(fn=process_pdf, inputs=gr.File(type="file"), outputs="dataframe", title="Bank Statement Classifier")
app.launch()