With TAG_NO_TAGFUNC remaining just for :help completion
closes 12039
https://github.com/vim/vim/pull/21316
(2 files)
—
Reply to this email directly, view it on GitHub, or unsubscribe.
Triage notifications, keep track of coding agent tasks and review pull requests on the go with GitHub Mobile for iOS and Android. Download it today!
You are receiving this because you are subscribed to this thread.![]()
Hm, I am slightly worried that a badly written tagfunc could cause some harms. But I couldn't make it crash using e.g. bw or :new or other such strange things. So let me include this now, but I fix the wrong order of assert_equal() again :) and add a few more tests for some of those nasty things.
—
Reply to this email directly, view it on GitHub, or unsubscribe.
Triage notifications, keep track of coding agent tasks and review pull requests on the go with GitHub Mobile for iOS and Android. Download it today!
You are receiving this because you are subscribed to this thread.![]()
Hm, I am slightly worried that a badly written tagfunc could cause some harms. But I couldn't make it crash using e.g.
bwor:newor other such strange things. So let me include this now, but I fix the wrong order ofassert_equal()again :) and add a few more tests for some of those nasty things.
Yes, I suppose I assumed that if calling tagfunc worked in ins-completion, it'd be ok in cmdline-completion, but I don't think that's a fair assumption. Maybe if the completion func does something with setcmdpos()/setcmdline() ? I've put in some more tests and that looks find though - #21334
(thanks for the assert_equal() order fixes - shall remember next time)
—
Reply to this email directly, view it on GitHub, or unsubscribe.
Triage notifications, keep track of coding agent tasks and review pull requests on the go with GitHub Mobile for iOS and Android. Download it today!
You are receiving this because you are subscribed to this thread.![]()