Detect more cases of unused_parens around types #142237
Conversation
|
r? @fee1-dead rustbot has assigned @fee1-dead. Use |
rustbot
added
PG-exploit-mitigations
|
Some changes occurred in tests/ui/sanitizer cc @rcvalle |
| .span_to_snippet(poly_trait_ref.span) | ||
| .map(|snip| snip.starts_with('(') && snip.ends_with(')')) |
There was a problem hiding this comment.
There has to be a better way to test this.
There was a problem hiding this comment.
Could we pass it so that the AST stores it somehow from parsing?
Maybe the unused parens lint should also be emitted ad-hoc when parsing these things
There was a problem hiding this comment.
My gut feeling is that adding this to the parser is not a worth the added complexity. However, I could see adding another Span to PolyTraitRef which excludes the parens. That would be cleaner than what I have currently.
Given my limited experience working on rustc, I'll defer to your judgement on which way I should go. (Of course, you can change your mind later. 🙂)
There was a problem hiding this comment.
Yes, please explore whether we can add a parenthesized: bool (or : Option<Span> that gives the inner span) to PolyTraitRef, we shouldn't resort to doing stringy operations here.
|
@rustbot author |
rustbot
added
S-waiting-on-author
|
Reminder, once the PR becomes ready for a review, use |
|
☔ The latest upstream changes (presumably #142878) made this pull request unmergeable. Please resolve the merge conflicts. |
With this change, more unused parentheses around bounds and types nested within bounds are detected.