Skip to content

Weird char limit while translating using the translate_document_from_filepath endpoint #135

Description

@adityanshastry

While using the python document translation endpoint translate_document_from_filepath, there seems to be a character limit of 15368 characters. This was tested with multiple configurations of flores-200 eng_latn.devtest file along with a few private text documents, both with English as the source and target language being French.

When the character limit is exceeded, the exception just says too many errors along with some references for document ids.

Activity

  1. JanEbbing commented on Jun 30, 2025

    @JanEbbing
    Contributor

    Which document type are you using and how exactly are you testing this? The character limit should be 1 million characters, but it's not straightforward to use this as an end user, we might add more characters than are purely in the text for formatting etc depending on document type.

  2. adityanshastry commented on Jun 30, 2025

    @adityanshastry
    Author

    This is meant to be a .txt file with one source sentence per line. The API call being used is the python function deepl_client.translate_document_from_filepath.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions