Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Remove Line-Number-Based Matching for Analysis Issues and Rely Solely on Message + File #516

Open
fabianvf opened this issue Dec 6, 2024 · 1 comment
Assignees

Comments

@fabianvf
Copy link
Contributor

fabianvf commented Dec 6, 2024

WE currently rely on line numbers to differentiate multiple identical violations within a single file. The line numbers probably do not help the LLM fix the correct issue and add unnecessary complexity. Instead, we should simplify the approach so that each unique task is keyed off the message and file alone, rather than line numbers. Removing line-number-based matching would also eliminate the need for similarity checks.

We should:

  • Remove line number from the equality check for at least analyzer violations, and maybe other task types as well.
  • Use only the file and the violation message (and any unique variables it contains) to determine uniqueness.
  • Eliminate the complexity introduced by comparing multiple identical violations within the same file at different line numbers.
  • Confirm that this approach does not reduce the correctness of the LLM-generated solutions.
@fabianvf fabianvf self-assigned this Dec 6, 2024
@shawn-hurley
Copy link
Contributor

@fabianvf I think we fixed the issues we had with the equality check, can we close this or am I missing something?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants