Uploading text file yields dorky encoding in the UI.

[erip]

I'm trying to pretranslate a list of raw strings. The easiest way to do this using the API seems to be uploading a document with pretranslation options enabled. Unfortunately it seems like submitting .txt files causes issues from the Python client.

My code below:

configuration = lilt.Configuration(
 host = host,
 api_key = {
 'key': api_key
 },
)
with lilt.ApiClient(configuration) as api_client:
 api_instance = lilt.DocumentsApi(api_client)
 name = 'dummy.txt' 
 project_id = 4 
 s = BytesIO("Hello, world!".encode())
 body = s.read() .
 pretranslate = 'tm+mt' 

 try:
 api_response = api_instance.upload_document(name, project_id, body, pretranslate=pretranslate)
 pprint(api_response)
 except ApiException as e:
 print("Exception when calling TranslateApi->translate_segment: %s\n" % e)

The corresponding uploaded document in UI:

Perhaps I'm misunderstanding the API as I see the same behavior with cURL:

curl -k -X POST "$host/documents/files?key=$key" \
 --header "LILT-API: {\"name\": \"test2.txt\",\"pretranslate\": \"tm+mt\",\"project_id\": 4}" \
 --header "Content-Type: application/octet-stream" \
 --data-binary @test.txt

[erip]

Indeed, this doesn't work through the UI either. ☹️