Empty string response after request to import a Conversational Language Understanding project

Question

Greetings,

I am following the the following Azure Guide to import a project in Azure AI Language. https://learn.microsoft.com/en-us/azure/ai-services/language-service/conversational-language-understanding/how-to/fail-over#import-to-a-new-project

The document states "Upon a successful request, the API response will contain an operation-location header with a URL you can use to check the status of the import job." However, as I receive response code 202, the response content is ''. And from Language Studio, the expected project is not seen.

My code is pasted here

os.environ["LS_CONVERSATIONS_ENDPOINT"] = "{my-endpoint}"
os.environ['LS_CONVERSATIONS_KEY'] = "{my-key}"

# Import project
azure_endpoint = os.environ.get("LS_CONVERSATIONS_ENDPOINT")
azure_key = os.environ.get("LS_CONVERSATIONS_KEY")
project_name = 'training-assistant'
API_VERSION = "2023-04-01"
LS_PROJECT_URL =  f"{azure_endpoint}/language/authoring/analyze-conversations/projects/{project_name}/:import?api-version={API_VERSION}"

header = {
    "Ocp-Apim-Subscription-Key": azure_key,
}

file_path = "{path to the following json file}\\training-assistant-project.json"

with open(file_path, 'r') as file:
    data = json.load(file)

response = requests.post(LS_PROJECT_URL, headers=header, json=data)
response.content

file_path is pointing to the following file.

{
  "projectFileVersion":"2023-04-01",
  "stringIndexType":"Utf16CodeUnit",
  "metadata":{
     "projectKind":"Conversation",
     "settings":{
        "confidenceThreshold":0.7
     },
     "projectName":"training-assistant",
     "multilingual":true,
     "description":"Trying out CLU",
     "language":"en-us"
  },
  "assets":{
     "projectKind":"Conversation",
     "intents":[
        {
           "category":"CalculateTime"
        }
     ],
     "entities":[
		{           
			"category":"LastXActivity"
        }
     ],
     "utterances":[
        {"text": "What is the longest time did I run at one go?","dataset": "Train","intent": "CalculateTime","entities": []},
        {"text": "How long did I run for the last run?","dataset": "Train","intent": "CalculateTime","entities": [{"category": "LastXActivity","offset": 28,"length": 8}]}
        
     ]
  }
}

Answer 1

Hi @Zhaoyi Zhu,

Thank you for reaching out to Microsoft Q&A forum!

I tried to repro the scenario with the following Python code and JSON file.

import os
import json
import requests

os.environ["LS_CONVERSATIONS_ENDPOINT"] = "ENDPOINT>"
os.environ['LS_CONVERSATIONS_KEY'] = "<KEY>"

# Import project
azure_endpoint = os.environ.get("LS_CONVERSATIONS_ENDPOINT")
azure_key = os.environ.get("LS_CONVERSATIONS_KEY")
project_name = 'training-assistant'
API_VERSION = "2023-04-01"
LS_PROJECT_URL = f"{azure_endpoint}/language/authoring/analyze-conversations/projects/{project_name}/:import?api-version={API_VERSION}"

header = {
    "Ocp-Apim-Subscription-Key": azure_key,
}

file_path = r"C:\Users\XXXXXXX\lang.json" # Provide your file path

with open(file_path, 'r') as file:
    data = json.load(file)

# Debug: Print the data being sent
print(json.dumps(data, indent=2))

response = requests.post(LS_PROJECT_URL, headers=header, json=data)

# Check if the request was accepted
if response.status_code == 202:
    print("Import request accepted. Checking status...")

    # Debug: Print response headers
    print("Response Headers:", response.headers)

    # Use the correct header to get the operation location
    operation_location = response.headers.get("operation-location")
    if not operation_location:
        print("No operation-location header found in the response.")
    else:
        while True:
            # Check the operation status
            status_response = requests.get(operation_location, headers=header)
            status_code = status_response.status_code

            if status_code == 200:
                status_data = status_response.json()
                print("Operation Status:", status_data)
                break
            elif status_code == 202:
                print("Still processing... waiting for 5 seconds.")
                time.sleep(5)  # Wait before checking again
            else:
                print("Error checking status:", status_response.text)
                break
else:
    print("Error:", response.text)

JSON file:

{
  "projectFileVersion": "2023-04-01",
  "stringIndexType": "Utf16CodeUnit",
  "metadata": {
    "projectKind": "Conversation",
    "settings": {
      "confidenceThreshold": 0.7
    },
    "projectName": "training-assistant",
    "multilingual": true,
    "description": "Trying out CLU",
    "language": "en-us"
  },
  "assets": {
    "projectKind": "Conversation",
    "intents": [
      {
        "category": "CalculateTime"
      }
    ],
    "entities": [
      {
        "category": "LastXActivity"
      }
    ],
    "utterances": [
      {
        "text": "What is the longest time did I run at one go?",
        "dataset": "Train",
        "intent": "CalculateTime",
        "entities": []
      },
      {
        "text": "How long did I run for the last run?",
        "dataset": "Train",
        "intent": "CalculateTime",
        "entities": [
          {
            "category": "LastXActivity",
            "offset": 28,
            "length": 8
          }
        ]
      }
    ]
  }
}

Project reflected in the Language Studio:

I hope this helps. Still if you face any errors, do let us know will try to figure out the issue or escalate this to the concerned team.

---

If this answers your query, do click Accept Answer and Yes for was this answer helpful.