> SZEDER Gábor <szeder.dev@xxxxxxxxx> writes:
>
> > There is a minor behaviour change here, though. This
> >
> > prune,*)
> > ;;
> >
> > case arm ensured that we don't list refs for 'git notes prune <TAB>',
> > because it doesn't accept them (and then we take our usual fallback and
> > let Bash complete filenames; yeah, 'git notes prune' doesn't accept
> > filenames either, but, as I said, that's our usual fallback when we
> > can't offer anything for completion).
> >
> > This patch removes that case arm, and refs will be offered for 'git
> > notes prune <TAB>'.
> >
> >> + *,--*)
> >> + __gitcomp_builtin notes_$subcommand
> >> ;;
> >> *)
> >> case "$prev" in
>
> I have this tentatively queued on the topic. Can we wrap the topic
> up and move it forward, instead of leaving it (and other topics)
> hanging around and causing conflicts with other topics in flight,
> please?
>
> Thanks.
>
>
> Subject: [PATCH] SQUASH???
>
> By Szeder <CAM0VKjmmF2t=B2s3rJoKmO3j6Bk5d6PEfCV==ONRGMcdUMLFpQ@xxxxxxxxxxxxxx>
> ---
> contrib/completion/git-completion.bash | 5 +++++
> 1 file changed, 5 insertions(+)
>
> diff --git a/contrib/completion/git-completion.bash b/contrib/completion/git-completion.bash
> index ab80f4e6e8..5f7495cda3 100644
> --- a/contrib/completion/git-completion.bash
> +++ b/contrib/completion/git-completion.bash
> @@ -1836,6 +1836,11 @@ _git_notes ()
> add,--reedit-message=*|append,--reedit-message=*)
> __git_complete_refs --cur="${cur#*=}"
> ;;
> + prune,--*)
> + __gitcomp_builtin notes_prune
> + ;;
> + prune,*)
> + ;;
> *,--*)
> __gitcomp_builtin notes_$subcommand
> ;;
> --
> 2.16.2-345-g7e31236f65
>
That works fine, but this would work just as well and has one less
case arm:
diff --git a/contrib/completion/git-completion.bash b/contrib/completion/git-completion.bash
index ab80f4e6e8..038af63c1a 100644
--- a/contrib/completion/git-completion.bash
+++ b/contrib/completion/git-completion.bash
@@ -1839,6 +1839,8 @@ _git_notes ()
*,--*)
__gitcomp_builtin notes_$subcommand
;;
+ prune,*)
+ ;;
*)
case "$prev" in
-m|-F)
