Azure function to run OCR using Azure AI vision.

function_app.py

import azure.functions as func
import json
import logging
import os
import time

import requests

app = func.FunctionApp(http_auth_level=func.AuthLevel.FUNCTION)

AZURE_AI_VISION_ENDPOINT = os.getenv("AZURE_AI_VISION_ENDPOINT")
AZURE_AI_VISION_KEY = os.getenv("AZURE_AI_VISION_KEY")


@app.function_name(name="HttpTrigger1")
@app.route(route="run_ocr")
def run_ocr(req: func.HttpRequest) -> func.HttpResponse:
    logging.info("Python HTTP trigger function processed a request.")

    binary_data = req.get_body()
    text = process_document(binary_data)

    return func.HttpResponse(body=json.dumps({"text": text}), status_code=200)


def process_document(binary_data) -> str:
    response = requests.post(
        f"{AZURE_AI_VISION_ENDPOINT}/vision/v3.2/read/analyze",
        headers={
            "Content-Type": "application/pdf",
            "Ocp-Apim-Subscription-Key": AZURE_AI_VISION_KEY,
        },
        data=binary_data,
    )

    operation_location = response.headers["Operation-Location"]

    while True:
        result_response = requests.get(
            operation_location,
            headers={"Ocp-Apim-Subscription-Key": AZURE_AI_VISION_KEY},
        )
        result = result_response.json()

        status = result.get("status")

        if status in ["succeeded", "failed"]:
            break
        time.sleep(1)

    lines = []
    for page in result["analyzeResult"]["readResults"]:
        for line in page["lines"]:
            lines.append(line["text"])
    text = "\n".join(lines)

    return text