Cannot translate a file with azure translator, try to upload on blob storage but "File not Found" error

Hi @Dipankar Bhattacharyya ,

Thank you for using the Microsoft Q&A.

As per your query regarding other approaches without blob storage, I recommend you place the source and target files in your local system itself and then hit the Microsoft Azure Translator API (URL) to translate a source text file from one language to other.

Below is the code, that translates simple English word to French. Kindly enhance it as per your requirements.

import requests
import json

# Set up the URL, headers, and body
url = "https://api.cognitive.microsofttranslator.com/translate?api-version=3.0&from=en&to=fr"
headers = {
    "Ocp-Apim-Subscription-Key": "Subscription_Key",
    "Content-Type": "application/json"
}

# Read content from the local text file
with open("path_to_local_text_file.txt", "r", encoding="utf-8") as file:
    source_text_content = file.read()
    print(f"Source Document Content: {source_text_content}")

# Set up the request body
body = [
    {
        "text": source_text_content
    }
]

# Send the request and get the response
response = requests.post(url, headers=headers, json=body)
response_json = json.loads(response.content.decode('utf-8'))

# Get the translated text
translated_text = response_json[0]["translations"][0]["text"]
print(f"Translated Document Content: {translated_text}")

# Write the translated text to a file
with open("translated.txt", "w", encoding="utf-8") as file:
    file.write(translated_text)

Hope this helps. Thank you!

Hi @Dipankar Bhattacharyya ,

Thank you for posting the detailed error.

Regarding the error, please check if the source and target container URLs are correct and accessible. Also try to access the container using the Azure Storage Explorer or Azure Portal to verify if the container exists and is accessible.

As part of debugging the code you provided, I made few changes. Below is the perfectly working version at my end using BLOB containers for source and target :

import os
from datetime import datetime, timedelta
from azure.core.credentials import AzureKeyCredential
from azure.ai.translation.document import DocumentTranslationClient
from azure.storage.blob import generate_account_sas, ResourceTypes, AccountSasPermissions

def translate_text(source_text, source_language, target_language, key):
    endpoint = "https://api.cognitive.microsofttranslator.com/translate"
    
    headers = {
        "Ocp-Apim-Subscription-Key": key,
        "Content-Type": "application/json",
    }

    params = {
        "api-version": "3.0",
        "from": source_language,
        "to": target_language,
    }

    body = [{"text": source_text}]

    response = requests.post(endpoint, headers=headers, params=params, json=body)
    response.raise_for_status()

    translated_text = response.json()[0]["translations"][0]["text"]
    return translated_text

def sample_translation():
    key = "key"
    source_container_url = "FULL_SRC_CONTAINER_WITH_FILENAME_SAS"
    target_container_url = "FULL_TARGET_CONTAINER_WITH_FILENAME_SAS"
    
    source_response = requests.get(source_container_url)
    source_text = source_response.content.decode("utf-8")

    translated_text = translate_text(source_text, "el", "fr", key)

    # Upload the translated text to the target blob container
    target_response = requests.put(
        target_container_url,
        data=translated_text.encode("utf-8"),
        headers={
            "x-ms-blob-type": "BlockBlob",
            "Content-Type": "text/plain",
        },
        params={"timeout": 30},
    )

    # Print the response status code
    print(target_response.status_code)

if __name__ == "__main__":
    sample_translation()

Note: As part of debugging, manually formed the source and target URLs in above code( as 1-liners). You please try to add more debug statements / PRINT statements / Try-catch block to debug more in the code you provided at the beginning.

Output of above code :

If you think that expired SAS tokens are causing an issue, please regenerate yours and use them in the code.

Hope this helps.

If this answers your query, do click Accept Answer and Yes for was this answer helpful. And, if you have any further query do let us know.

Hi @Dipankar Bhattacharyya ,

We haven’t heard from you on the last response and was just checking back to see if you have a resolution yet or you got a chance to try my above suggestion.

Thanks!

I don't want to get text, want to preserve the format. Do you guys join on zoom, then can show you the issue? It is working without SAS when I just give the source and target folder, it takes everything from source and translate and put to target folder.

My requirement is to put the file in source container and then translate it to target container. I am getting the source file SAS with User delegation key as "Read" and getting the target container (not the file, because the file is not translated yet), SAS with User delegation key as "Write" and "List", here is the code below

The return status is 202, but I cannot see the output in target folder

response status code: 202

response status: Accepted

response headers: {'Content-Length': '0', 'X-RequestId': '12379c86-7fe6-41d6-81e6-108f92e51456', 'Operation-Location': 'https://translatordiptest2devmain.cognitiveservices.azure.com/translator/text/batch/v1.1/batches/f1aefcc8-0469-4ee2-ae38-5fab8b8f0cd2', 'Strict-Transport-Security': 'max-age=31536000; includeSubDomains; preload', 'apim-request-id': '12379c86-7fe6-41d6-81e6-108f92e51456', 'x-content-type-options': 'nosniff', 'x-ms-region': 'West US 2', 'Date': 'Mon, 11 Dec 2023 23:45:42 GMT'}

import requests

endpoint = "https://api.cognitive.microsofttranslator.com"
key = "3fbcf97xxxxxxxxx"

path = '/batches'
constructed_url = "https://translatordiptest2devmain.cognitiveservices.azure.com/translator/text/batch/v1.1/batches"

payload= {
    "inputs": [
        {
            "storageType": "File",
            "source": {
                "sourceUrl": "https://diptestXXXXX.blob.core.windows.net/diptest2devmainconta/TMFiles.pdf?sp=r&st=2023-12-11T23:28:25Z&se=2023-12-13T07:28:25Z&skoid=69ee639b-29f3-487f-ad71-2f477e5c18c2&sktid=70e732cf-2ee8-4ddd-8c95-23ae8158469e&skt=2023-12-11T23:28:25Z&ske=2023-12-13T07:28:25Z&sks=b&skv=2022-11-02&spr=https&sv=2022-11-02&sr=b&sig=BS3LMbH%2FUNORW67BNZmO6t%2Fk%2BeJ%2FTK9ho7qphz5YNX4%3D",
                "storageSource": "AzureBlob",
            },
            "targets": [
                {
                    "targetUrl": "https://diptestXXXXX.blob.core.windows.net/diptest2devmaincontb?sp=wl&st=2023-12-11T23:32:59Z&se=2023-12-13T07:32:59Z&skoid=69ee639b-29f3-487f-ad71-2f477e5c18c2&sktid=70e732cf-2ee8-4ddd-8c95-23ae8158469e&skt=2023-12-11T23:32:59Z&ske=2023-12-13T07:32:59Z&sks=b&skv=2022-11-02&spr=https&sv=2022-11-02&sr=c&sig=sor1oaj105F511wxgXqtBnRDfzWKU8qU5pAdx3XL1Dw%3D",
                    "language": "fr"
                }
            ]
        }
    ]
}
headers = {
  'Ocp-Apim-Subscription-Key': key,
  'Content-Type': 'application/json'
}

response = requests.post(constructed_url, headers=headers, json=payload)

print(f'response status code: {response.status_code}\nresponse status: {response.reason}\nresponse headers: {response.headers}')