Skip to content

fix: force completions/generations to be in appropriate list form

What does this merge request do and why?

fix: force completions/generations to be in appropriate list form

Requesting completion(v2/code/completions) when using the AI gateway with a AIGW_USE_FAKE_MODELS set to true yields a 500 error. This was occurring because the mock response was returning a single entry instead of a list.

In a similar but opposite fashion, requesting generations (v2/code/generations) was also yielding a 500 error as it was expecting a single entry but was sometimes receiving a list.

This MR aims to fix these issues.

related issue: #497 (closed)

How to set up and validate locally

Numbered steps to set up and validate the change are strongly suggested.

  1. Check out to this merge request's branch.
  2. Set AIGW_USE_FAKE_MODELS to true in your .env file
  3. Head to http://localhost:5052/docs
  4. Using the following prompt:
{
    "prompt_version": 1,
    "telemetry":
    [],
    "current_file":
    {
        "file_name": "/test.rb",
        "content_above_cursor": "def set_name(whitespace_name)\\\\n    this.name = whitespace_name.",
        "content_below_cursor": "\\n\\n\\n\\n\\n",
        "language_identifier": "ruby"
    },
    "intent": "completion"
}

execute the v2/code/generations and v2/code/completions endpoints and ensure you receive an OK response.

Merge request checklist

  • Tests added for new functionality. If not, please raise an issue to follow up.
  • Documentation added/updated, if needed.

Closes #497 (closed)

Edited by Jeff Park

Merge request reports

Loading