#3855 illustrates a PR reviewed by two maintainers, but not approved due to
(objective) reasons. I am puzzled by the reaction advocating *"maintainers'
contributions *must* be merged"*.

> reject [...] a single helper without a compelling use case.

+1

> I’d like to suggest [...]:
> 1. Gary continues with the public API changes in PR #3855 by:
>    adding `getPath()` consistently across all file manager [...]
> 2. I take care of the internal migration to `Path` [...]

+1

I must admit that the 2nd option makes me very concerned that rejecting a
maintainers' contribution, which helps with his employer's use case at the
cost of downsides elaborated in the PR comments, ends up being assigned as
a feature implementation task to the reviewer.

Reply via email to