Open LukasKalbertodt opened 3 years ago
I haven't had a chance to test this out yet, but I think having TryFrom
impls could be quite convenient. It would avoid the match
with a single type expected with the rest panicking (.expect
could be used or some other error handling).
I assume you are talking about TryFrom<X> for Y
where Y
is litrs::StringLit
, litrs::IntegerLit
and so on? That's a great idea! What could X be then? proc_macro[2]::Literal
for sure. But also String
and &str
? Probably all four options, right?
EDIT: no, I wasn't thinking straight. The &str
and String
cases are already covered by parse
, of course. But yeah, for the proc_macro[2]::Literal
types it makes perfect sense!
I was imagining proc_macro::Literal
and proc_macro::TokenTree
(along with proc_macro2
equivalents).
Yeah TokenTree
also makes sense. Sweet, thanks.
@jhpratt I just published 0.2.0 which contains a large number of From
and TryFrom
impls. It was actually more work than expected, but I think they really improve the library and make lots of real world uses cases easier. Thanks for the suggestion again!
@LukasKalbertodt I stumbled across this crate and really love it. Thank you so much for doing this!
I'm working on my own programming language https://github.com/xiyuzhai-husky-lang/husky/. I shall work out the literal part of my language based on this project. Amazing work!!!
Do you have any opinions, complaints, suggestions, ideas, ... about
litrs
? Let me know! This issue is less formal and more relaxed than "normal issues", so feel free to just dump your thoughts here.