app.py

import gradio as gr
import pdfplumber
from pdf2image import convert_from_path
import pytesseract
import shutil

def extract_text_debug(file_path):
    logs = []
    text = ""

    # Debug paths
    logs.append(f"pdftoppm path: {shutil.which('pdftoppm')}")
    logs.append(f"pdftocairo path: {shutil.which('pdftocairo')}")

    # Try pdfplumber
    try:
        with pdfplumber.open(file_path) as pdf:
            for page in pdf.pages:
                t = page.extract_text() or ""
                text += t
        if text.strip():
            logs.append("✅ Extracted text using pdfplumber")
            return text[:800] + "\n\n---\n" + "\n".join(logs)
        else:
            logs.append("⚠️ pdfplumber gave empty text, trying OCR…")
    except Exception as e:
        logs.append(f"❌ pdfplumber failed: {e}")

    # OCR fallback
    try:
        images = convert_from_path(file_path, dpi=200, poppler_path="/usr/bin")
        ocr_text = [pytesseract.image_to_string(img) for img in images[:2]]
        text = "\n".join(ocr_text)
        if text.strip():
            logs.append("✅ OCR worked via pdf2image + Tesseract")
        else:
            logs.append("⚠️ OCR returned empty text")
    except Exception as e:
        logs.append(f"❌ OCR fallback failed: {e}")

    return (text[:800] if text.strip() else "❌ No text extracted") + "\n\n---\n" + "\n".join(logs)

with gr.Blocks() as demo:
    gr.Markdown("# 📄 PDF Extractor Debug")
    inp = gr.File(file_types=[".pdf"], type="filepath")
    out = gr.Textbox(lines=20, label="Text + Debug Logs")
    inp.change(extract_text_debug, inputs=inp, outputs=out)

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