bug#69625: 30.0.50; [PATCH] rust-ts-mode doesn't fontify some enum

[Yuan Fu <casouri@gmail.com>]

> On Jun 27, 2024, at 7:40 PM, Randy Taylor <dev@rjt.dev> wrote:
> 
> On Saturday, June 22nd, 2024 at 19:17, Yuan Fu <casouri@gmail.com> wrote:
>> 
>> 
>> 
>> 
>>> On Jun 22, 2024, at 4:07 AM, Stefan Kangas stefankangas@gmail.com wrote:
>>> 
>>> Yuan Fu casouri@gmail.com writes:
>>> 
>>>>> On Mar 14, 2024, at 6:52 PM, Dmitry Gutov dmitry@gutov.dev wrote:
>>>>> 
>>>>> On 09/03/2024 05:50, Randy Taylor wrote:
>>>>> 
>>>>>> VariantA gets highlighted as a type and not a function at level 3 because that
>>>>>> level doesn't support functions, but does support types. Maybe that could be
>>>>>> considered a bug in that it shouldn't be highlighted at all for level 3?
>>>>> 
>>>>> Probably.
>>>>> 
>>>>>> I'm not sure how worth it would be to do something about that though, or how
>>>>>> easy.
>>>>> 
>>>>> Same. I haven't looked into it.
>>>>> 
>>>>>> For VariantC, our (and tree-sitter's) best guess is that it's a variable.
>>>>>> We can't really know it's a type without guessing - like assuming a capitalized
>>>>>> identifier is a type, and I don't think that's something we can assume. People
>>>>>> can have capitalized functions and variables even if that goes against Rust's
>>>>>> usual style. Perhaps as a compromise we could introduce a variable (or something)
>>>>>> that lets the user specify that all capitalized identifiers should be treated as
>>>>>> types? Maybe it even makes sense to default it to that behaviour since I believe
>>>>>> most Rust code follows that style.
>>>>> 
>>>>> We do have some rules already that are based off whether an identifier is capitalized. I.e. some for use_as_clause, and another for highlighting an identifier with font-lock-constant-face if it's ALL_CAPS. So it might be logical to carry on with that approach.
>>>>> 
>>>>> If the style is consistent enough across the ecosystem, of course.
>>>>> 
>>>>> We could add a variable too, but that'd make the rules more complex so it would be helpful to understand first whether there are users who would benefit.
>>>> 
>>>> For some reason I couldn’t see Randy’s messages. So sorry if I missed anything. I suggest we go ahead with guessing and add the variable if enough people complain. Personally speaking I believe the vast majority of Rust community wouldn’t write Rust code with capitalized variable and non-capitalized types. The Rust community is very much inclined to the standard style, AFAICT.
>>> 
>>> Yuan, did you make any progress here?
>> 
>> 
>> From what I can tell Randy isn’t very convince of this idea, so I didn’t make any changes. Randy, should we keep the status quo and close this or should we explore something else?
>> 
>> Yuan
> 
> Sorry for the late response.
> 
> I thought we were waiting to see if enough people complain, and AFAIK no one else has.
> Perhaps I misinterpreted your message and
> you meant we should go ahead with adding this and only add the
> variable to control it if people complain - apologies if so.

That’s what I proposed, yeah. You didn’t say “let’s do this” (maybe I missed it, apologies if I did), so obviously I didn’t want to jump the gun ;-)

> 
> I'm not opposed to the idea - it's the best we can do with what
> we've got. The only thing I am undecided on is if we want to bother
> with the variable to control it. If we did add it, it should be
> on by default since, as you mentioned, the vast majority of Rust
> code follows the same convention, and at that point who is actually
> going to turn it off? So it's probably not worth the hassle...
> 
> Were you thinking of adding this to the type feature or its own
> feature?

I was thinking adding this to the type feature. IMHO config options should be implemented by variables, not by splitting stuff into separate tree-sitter font-lock features. I admit the name “feature” is a bit misleading in this regard...

Yuan