from flask import Flask, request, jsonify
from ocr_runner import pdf_to_json, ocr_pdf
import requests
import os
import tempfile
import base64
from pathlib import Path

app = Flask(__name__)

EXXETA_URL = os.getenv("EXXETA_SERVICE_URL", "http://localhost:5053/extract")
SPACY_URL = os.getenv("SPACY_SERVICE_URL", "http://localhost:5052/extract")

@app.route('/ocr', methods=['POST'])
def convert_extract_text_from_pdf():
    if "file" not in request.files:
        return {"error": "No file"}, 400

    file = request.files["file"]
    pitchbook_id = request.form.get("id")

    if not pitchbook_id:
        return {"error": "No ID"}, 400

    with tempfile.NamedTemporaryFile(delete=False, suffix='.pdf') as temp_file:
        file.seek(0)
        temp_file.write(file.read())
        temp_path = Path(temp_file.name)
    ocr_path = ocr_pdf(temp_path)

    if not ocr_path or not ocr_path.exists():
        temp_path.unlink()  # cleanup
        return {"error": "OCR processing failed - all PDFs must be OCR'd"}, 500


    with open(ocr_path, 'rb') as ocr_file:
        ocr_pdf_data = ocr_file.read()
        ocr_pdf_base64 = base64.b64encode(ocr_pdf_data).decode('utf-8')

        ocr_file.seek(0)
        result = pdf_to_json(ocr_file)

    ocr_path.unlink()
    temp_path.unlink()

    payload = {
        "id": int(pitchbook_id),
        "extracted_text_per_page": result["pages"]
    }

    requests.post(EXXETA_URL, json=payload, timeout=600)
    requests.post(SPACY_URL, json=payload, timeout=600)

    return {
        "status": "sent",
        "ocr_pdf": ocr_pdf_base64,
        "message": "PDF successfully OCR'd and processed"
    }, 200

if __name__ == "__main__":
    app.run(host="0.0.0.0", port=5051)