-
Notifications
You must be signed in to change notification settings - Fork 379
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
_split_longopt rename is breaking downstream (3rd party) completion scripts #1135
Comments
The function was removed in d659507 |
Any functions and variables whose name contains
For the semver, we might have kept the old names. I actually had a doubt about removing the old names even if they become a part of private APIs. But I haven't tried to keep it because a reasoning was given in the commit message, and also |
I've checked the
I think the |
I thought about submitting a PR to the upstream Instead, there seem to be many alternative tools [1]. For example, the latest version of ripgrep ( |
JFYI/aside rg 3rd party loader in #1148 |
Describe the bug
The removal of the function
_split_longopt
in 19a3798 (released in 2.12.0) is causing completions for the-silver-searcher (ag
) to failTo reproduce
(both the above installed via homebrew)
ag SEARCHQUERY app<TAB>
get:
Expected behavior
I'm not the author of ag's completion script so I'm not entirely sure what they expect that function to do. But as the consumer, of course, I expect ag's completion script to work. :D I opened this as a bug against ag's completion as well, but wanted to get guidance here first. Is this function not intended to be public? If not, I'm not certain how to remedy this in ag's completion. If it is then I suspect there needs to be some migration period where the function is aliased to allow updating of downstream completion scripts?
Versions (please complete the following information)
echo "$BASH_VERSION"
: 5.2.26(1)-release(IFS=.; echo "${BASH_COMPLETION_VERSINFO[*]}")
: 2.12.0Additional context
I recognize that it's possible (probable even?) that this function should have been considered an internal private function and thus was never intended for completion scripts to use.
However, other completion tooling is also being affected by this change (in addition to silver-searcher):
FWIW, i manually edited ag's completion script to use
_comp__split_longopt
instead, and all is well. So i think the real question is a matter of "public api management" and "semver". (in scare quotes because of the complexity of package version management with shell utilities like this)Debug trace
I suspect tracing isn't necessary in this case because the root cause is known? (The function rename.)
The text was updated successfully, but these errors were encountered: