app.py

import gradio as gr
from docx import Document
from docx.shared import RGBColor
from difflib import get_close_matches
from pdf_parser import extract_text_from_pdf, parse_data_blocks

def is_red_color(run):
    color = run.font.color
    if not color or not color.rgb:
        return False
    r, g, b = color.rgb[0], color.rgb[1], color.rgb[2]
    return r >= 200 and g <= 100 and b <= 100  # red-dominant

def replace_red_text_with_data(doc_path, data_dict):
    doc = Document(doc_path)

    for para in doc.paragraphs:
        for run in para.runs:
            if is_red_color(run):
                original_text = run.text.strip()
                # Try exact or close match
                match = get_close_matches(original_text.lower(), [k.lower() for k in data_dict.keys()], n=1, cutoff=0.6)
                if match:
                    for key in data_dict:
                        if key.lower() == match[0]:
                            run.text = data_dict[key]
                            break
    return doc

def process_files(pdf_file, word_template):
    # Extract data from PDF
    raw_text = extract_text_from_pdf(pdf_file)
    data_dict = parse_data_blocks(raw_text)

    # Replace red text in Word
    final_doc = replace_red_text_with_data(word_template, data_dict)

    # Save and return output
    output_path = "filled_output.docx"
    final_doc.save(output_path)
    return output_path

demo = gr.Interface(
    fn=process_files,
    inputs=[
        gr.File(label="Upload PDF Report", file_types=[".pdf"]),
        gr.File(label="Upload Word Template (.docx)", file_types=[".docx"])
    ],
    outputs=gr.File(label="Download Updated Word (.docx)"),
    title="Audit Report Auto-Filler",
    description="Replaces outdated red text in Word using updated values from a PDF report."
)

if __name__ == "__main__":
    demo.launch()