如何使用 JSON 文件作为使用 Python 的 MS Azure 文本分析的输入？

Question

I've looked through many responses to variants of this problem but still not able to get my code working.

I am trying to use the MS Azure text analytics service and when I paste the example code (including 2/3 sample sentences) it works as you might expect. However, my use case requires the same analysis to be performed on hundreds of free text survey responses so rather than pasting in each and every sentence, I would like to use a JSON file containing these responses as an input, pass that to Azure for analysis and receive back a JSON output.

The code I am using and the response it yields is shown below (note that the last bit of ID 2 response has been chopped off before the error message).

        key = "xxxxxxxxxxx"
endpoint = "https://blablabla.cognitiveservices.azure.com/"

import json
with open(r'example.json', encoding='Latin-1') as f:
  data = json.load(f)

print (data) 

import os
from azure.cognitiveservices.language.textanalytics import TextAnalyticsClient
from msrest.authentication import CognitiveServicesCredentials

def authenticateClient():
    credentials = CognitiveServicesCredentials(key)
    text_analytics_client = TextAnalyticsClient(
        endpoint=endpoint, credentials=credentials)
    return text_analytics_client

import requests
# pprint is used to format the JSON response
from pprint import pprint

import os

subscription_key = "xxxxxxxxxxxxx"
endpoint = "https://blablabla.cognitiveservices.azure.com/"

entities_url = "https://blablabla.cognitiveservices.azure.com/text/analytics/v2.1/entities/"

documents = data 

headers = {"Ocp-Apim-Subscription-Key": subscription_key}
response = requests.post(entities_url, headers=headers, json=documents)
entities = response.json()
pprint(entities)

[{'ID': 1, 'text': 'dog ate my homework', {'ID': 2, 'text': 'cat sat on the

{'code': 'BadRequest', 'innerError': {'code': 'InvalidRequestBodyFormat', 'message': 'Request body format is wrong. Make sure the json ' 'request is serialized correctly and there are no ' 'null members.'}, 'message': 'Invalid request'}

Answer 1

According to my research, when we call Azure text analytics rest API to Identify Entities, the request body should be like

{
  "documents": [
    {
      "id": "1",
      "text": "."
    },
    {

      "id": "2",
      "text": ""
    }
  ]
}

For example

My json file

[{
        "id": "1",
        "text": "dog ate my homework"
    }, {
        "id": "2",
        "text": "cat sat on the sofa"
    }
]

My code

key = ''
endpoint = "https://<>.cognitiveservices.azure.com/"

import requests
from pprint import pprint
import os

import json
with open(r'd:\data.json', encoding='Latin-1') as f:
  data = json.load(f)
pprint(data)
entities_url = endpoint + "/text/analytics/v2.1/entities?showStats=true"
headers = {"Ocp-Apim-Subscription-Key": key}
documents=data 
response = requests.post(entities_url, headers=headers, json=documents)
entities = response.json()
pprint(entities)
pprint("--------------------------------")
documents ={}
documents["documents"]=data
response = requests.post(entities_url, headers=headers, json=documents)
entities = response.json()
pprint(entities)