105 lines
3.2 KiB
Python
105 lines
3.2 KiB
Python
from flask import Flask, request
|
|
from ocr_runner import pdf_to_json, ocr_pdf
|
|
import requests
|
|
import os
|
|
import tempfile
|
|
from pathlib import Path
|
|
import logging
|
|
import threading
|
|
|
|
|
|
# Set up logging
|
|
logging.basicConfig(level=logging.INFO)
|
|
logger = logging.getLogger(__name__)
|
|
|
|
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")
|
|
COORDINATOR_URL = os.getenv("COORDINATOR_URL", "http://localhost:5050")
|
|
|
|
|
|
def convert_pdf_async(temp_path, pitchbook_id):
|
|
try:
|
|
logger.info("Starting OCR process...")
|
|
|
|
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_file.seek(0)
|
|
result = pdf_to_json(ocr_file)
|
|
|
|
|
|
payload = {
|
|
"id": int(pitchbook_id),
|
|
"extracted_text_per_page": result["pages"]
|
|
}
|
|
|
|
logger.info("Sending payload to EXXETA and SPACY services")
|
|
|
|
requests.post(COORDINATOR_URL + "/api/progress", json={"id": pitchbook_id, "progress": 35})
|
|
try:
|
|
exxeta_response = requests.post(EXXETA_URL, json=payload, timeout=600)
|
|
logger.info(f"EXXETA response: {exxeta_response.status_code}")
|
|
except Exception as e:
|
|
logger.error(f"Error calling EXXETA: {e}")
|
|
|
|
try:
|
|
spacy_response = requests.post(SPACY_URL, json=payload, timeout=600)
|
|
logger.info(f"SPACY response: {spacy_response.status_code}")
|
|
except Exception as e:
|
|
logger.error(f"Error calling SPACY: {e}")
|
|
|
|
files=[
|
|
('file',('',open(ocr_path,'rb'),'application/pdf'))
|
|
]
|
|
headers = {}
|
|
|
|
try:
|
|
|
|
requests.put(f"{COORDINATOR_URL}/api/pitch_book/{pitchbook_id}", files=files, timeout=600, headers=headers)
|
|
logger.info("COORDINATOR response: Progress + File updated")
|
|
except Exception as e:
|
|
logger.error(f"Error calling COORDINATOR: {e}")
|
|
|
|
ocr_path.unlink()
|
|
temp_path.unlink()
|
|
except Exception as e:
|
|
logger.error(f"Exception in OCR processing: {str(e)}", exc_info=True)
|
|
|
|
|
|
@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")
|
|
|
|
logger.info(f"Processing file for pitchbook_id: {pitchbook_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)
|
|
|
|
thread = threading.Thread(target=convert_pdf_async, args=(temp_path, pitchbook_id))
|
|
thread.start()
|
|
|
|
return {
|
|
"status": "sent",
|
|
"message": "PDF successfully OCR'd and processed"
|
|
}, 200
|
|
|
|
|
|
if __name__ == "__main__":
|
|
logger.info("Starting OCR service on port 5000")
|
|
app.run(host="0.0.0.0", port=5051, debug=True)
|