Azure Function Ouput to ADF variable

Question

Hello folks,

I am testing Azure Function app in ADF pipeline. Based on input I am receiving to Azure python function, need to do some processings and write the output to variable, which will be input to next step.

Here is my test code:

@app.route(route="http_trigger_2")
def http_trigger_2(req: func.HttpRequest) -> func.HttpResponse:
    logging.info('Python HTTP trigger function processed a request.')

    name = req.params.get('name')
    if not name:
        try:
            req_body = req.get_json()
        except ValueError:
            pass
        else:
            name = req_body.get('name')

    if name:
        text =req.params.get('name')
        d = p.DataFrame({'Data': [text], 'Size': [len(text)]})
        res = d.to_json(orient='records')
        print(res)
        return func.HttpResponse(res, mimetype='application/json')
    else:
        return func.HttpResponse(
             "This HTTP triggered function executed successfully. Pass a name in the query string or in the request body for a personalized response.",
             status_code=200
        )

When I trigger the function using url in browser passing value for name, it executes fine and give output in browser as below

[{"Data":"Testing","Size":7}]

But while POST this function in ADF pipeline with body, getting error as below:

Operation on target Azure Function1 failed: Call to provided Azure function 'http_trigger_2' failed with status-'InternalServerError' while invoking 'POST' on 'https://func-app.azurewebsites.net' and message - 'Invoking Azure function failed with HttpStatusCode - InternalServerError.'.

I believe it might be related to way output is returned.

Any comments or suggestions are much appreciated.

Answer 1

@Jatinder Luthra Thanks for your patience on this.

Post making the below changes to the above shared code, I am able to execute the function from portal and from the data factory pipelines without any issues.

Removed this line text =req.params.get('name') and updated this line : d = p.DataFrame({'Data': [name], 'Size': [len(name)]})

Here is the modified code:

import logging
import azure.functions as func
import pandas as p 

from azure.functions import HttpRequest, HttpResponse


def main(req: HttpRequest) -> HttpResponse:
    logging.info('Python HTTP trigger function processed a request.')

    
    name = req.params.get('name')
    if not name:
        try:
            req_body = req.get_json()
        except ValueError:
            pass
        else:
            name = req_body.get('name')

    if name:
        d = p.DataFrame({'Data': [name], 'Size': [len(name)]})
        res = d.to_json(orient='records')
        print(res)
        return func.HttpResponse(res, mimetype='application/json')
    else:
        return func.HttpResponse(
             "This HTTP triggered function executed successfully. Pass a name in the query string or in the request body for a personalized response.",
             status_code=200
        )

Here are the sample outputs for your reference.

1. Output from Code+Test in portal.

2. Output from ADF pipeline Azure Function activity:

Feel free to reach back to me if you have any further questions on this.