[DOC-FIX] Implement HTTP 301 redirect for MLflow 1.28.0 change of mlflow.tracking.MlflowClient to mlflow.client.MlflowClient to preserve existing documentation hyperlinks

Willingness to contribute

No. I cannot contribute a documentation fix at this time.

URL(s) with the issue

https://mlflow.org/docs/latest/python_api/mlflow.client.html

Description of proposal (what needs changing)

For example, the link https://mlflow.org/docs/latest/python_api/mlflow.tracking.html#mlflow.tracking.MlflowClient.get_latest_versions returns an HTTP 403 (not 404) status code

<Error>
<Code>AccessDenied</Code>
<Message>Access Denied</Message>
<RequestId>7NX9Q00FWYZJQ86X</RequestId>
<HostId>9yRMs47uE5KOSp37A57DkQUhf0Iz2eYui709kgXm04Nf+5er68hldsGD+7hHor9sAx3vUX8c2ZE=</HostId>
</Error>

The new link now is: https://mlflow.org/docs/latest/python_api/mlflow.client.html#mlflow.tracking.MlflowClient.get_latest_versions

Solution is to implement an HTTP 301 Moved Permanently redirect mechanism to preserve existing hyperlinks.

[amesar]

@WeichenXu123 But that still requires me (and all others) to edit every document or web page and make change. Why can’t our webserver do a 301 redirect? Pretty standard procedure to prevent broken links. 😉

[harupy]

@WeichenXu123 The issue is https://mlflow.org/docs/latest/python_api/mlflow.tracking.html#mlflow.tracking.MlflowClient.get_latest_versions used to be available but it’s not now because we changed the module structure.