Stop excluding base::return() and base::function from completions #768
Conversation
| @@ -53,7 +53,7 @@ fn completions_from_search_path( | |||
| let mut completions = vec![]; | |||
|
|
|||
| const R_CONTROL_FLOW_KEYWORDS: &[&str] = &[ | |||
| "if", "else", "for", "in", "while", "repeat", "break", "next", "return", "function", | |||
| "if", "else", "for", "in", "while", "repeat", "break", "next", | |||
There was a problem hiding this comment.
Do we need to exclude any of these actually? The shorter ones won't be triggered if they are below the character limit for completions but it presumably wouldn't hurt to include them.
There was a problem hiding this comment.
As we discussed in person, I suspect it's not good that we exclude these at all. I will open a new issue shortly with a systematic look at how reserved words currently fare in positron's R completions. And that will lead to some new proposals.
|
Does Ark provide the appropriate escaping so that the completion from |
|
@kevinushey No. That is already on my radar and will appear soon in its own issue or added to an existing issue. I definitely bumped into this while playing around for this PR. |
I'm opening Completion Month with a real banger! Let's stop explicitly excluding
base::return()andbase::functionfrom completions.Relates to posit-dev/positron#4842
I've
git blamein arkand I can't find any reason to not just let
base::return()andbase::functionshow up in our completions. They will both be contributed by our search path completions, coming frombase.Both are represented by snippets, which I have some mixed feelings about. So I plan to discuss some additional improvements relating to the snippet treatment of
functionandreturn(). We could also discuss posit-dev/positron#1850 at the same time. But perhaps this simple fix can happen first, especially forreturn().