Re: [Hamlib-developer] The use of LLM generated code in Hamlib (long)

[Greg T. <gd...@le...>]

Nate Bargmann <n0...@n0...> writes:

> As another data point we have PR 2008:
>
> https://github.com/Hamlib/Hamlib/pull/2008
>
> Mike clearly notes the use of Code QL to diagnose and offer a fix for
> using the return value of sscanf() correctly in the Codan backend.  A
> disclosure  regarding the use of AI is clearly stated in the commit
> message.
>
> As I see it, this is a useful use of LLM where our code base is examined
> and fixes offered in terms of C language best practices.

It was a useful fix, but the PR is full of LLM text that should have
been replaced by what a human would have written.  I have commented in
the PR.

(and +1 for getting off github, but I know it's hard; haven't done it
yet for the the project I maintain)