label names handled by the shortcut command should be configurable

[JohnTitor]

Currently the label names handled by the shortcut command are hardcoded:

triagebot/src/handlers/shortcut.rs

Lines 29 to 31 in a2ae093

	let waiting_on_review = "S-waiting-on-review";
	let waiting_on_author = "S-waiting-on-author";
	let blocked = "S-blocked";

That would enforce other repos to use the exact same names, but that restriction shouldn't be necessary and we should be able to change them as we want.

[Mark-Simulacrum]

Happy to see a PR changing this, so long as we preserve the defaults (at least until rust-lang/rust updates its configuration with the right labels).

[Saplyn]

Hi, new to contributing to rust, I'd like to have a try at this issue.

That would enforce other repos to use the exact same names, but that restriction shouldn't be necessary and we should be able to change them as we want.

So how should the name be specified? Are there any usage examples or implementation examples that I could use as reference to get an idea of how it is should look like?

... so long as we preserve the defaults (at least until rust-lang/rust updates its configuration with the right labels).

Is this (wait for rust-lang/rust's update) still relevant today (after 3 years)? Is the one hard-coded in the code the "defaults"?

---

What and where should I check out before start working on this? Besides README and Forge - Triagebot?

Also, where is this function (handle_command()) being called? My rust-analyzer says "No results found for lsp_references", it is because it's being called by some external hook or something?

And... should I claim it?

[Kobzol]

So how should the name be specified? Are there any usage examples or implementation examples that I could use as reference to get an idea of how it is should look like?

You can take a look at how we specify other configuration options in the TOML, e.g. like this:

triagebot/src/config.rs

Line 731 in 4fbd33a

[backport.teamRed]

So it could be something like

[shortcut]
waiting-on-review-label = "S-waiting-on-review"
waiting-on-author-label = "S-waiting-on-author"
blocked-label = "S-blocked"

added to ShortcutConfig.

Is this (wait for rust-lang/rust's update) still relevant today (after 3 years)? Is the one hard-coded in the code the "defaults"?

Yup, still the same.

Also, where is this function (handle_command()) being called? My rust-analyzer says "No results found for lsp_references", it is because it's being called by some external hook or something?

It's called through a macro:

triagebot/src/handlers.rs

Line 377 in 614aef7

command_handlers! {

You don't really need to worry about that though.

And... should I claim it?

You don't have to, but if you want to, feel free to :)

[Saplyn]

You can take a look at how we specify other configuration options in the TOML ...

Acknowledged, will check it out then

Yup, still the same.

Please correct me if I'm wrong. The default value is the one hard-coded there, and we should fallback to default value if there's no shortcut specified. Right?

It's called through a macro ... You don't really need to worry about that though.

Okay then... but I guess it's good to know that it is being called :3

You don't have to, but if you want to, feel free to :)

@rustbot claim

---

Is it normal that struct ShortcutConfig is empty? Am I expected implement it (both the struct and the config parsing) or should I keep it as is (leave it empty)?

Also, will it be fine to do the "hard work" tomorrow (it's about bedtime here in Australia)? I wasn't expecting such immediate reply TBH.

Thanks!

[Kobzol]

Well, it's currently empty only because it has no configuration options :) Once we add some (the labels), then it won't be empty anymore.

Please correct me if I'm wrong. The default value is the one hard-coded there, and we should fallback to default value if there's no shortcut specified. Right?

Yeah.

[apiraino]

Wouldn't this feature request be satisfied by #2172?

[Kobzol]

Sorry, I didn't realize that. Yeah, if we have fully generics handlers, then that would be even more general than what this issue proposes.

[Saplyn]

(So should I continue to work on this? Or should I release this issue?)

[apiraino]

you can release the assignment. I will close it once #2172 is done

[Saplyn]

(sad 😿)

@rustbot release-assignment