reject macros with empty repetitions #36721
Conversation
|
r? @nrc (rust_highfive has picked a reviewer for you, use r? to override) |
| @@ -358,6 +358,10 @@ pub fn compile<'cx>(cx: &'cx mut ExtCtxt, | |||
| valid &= check_rhs(cx, rhs); | |||
| } | |||
|
|
|||
| for lhs in &lhses { | |||
| valid &= check_lhs_no_empty_seq(cx, &[lhs.clone()]) | |||
| } | |||
There was a problem hiding this comment.
could use lhses.iter().all() instead of the for loop.
There was a problem hiding this comment.
The for loop was mainly used to use the same style as the loop of the rhses directly above.
Also .iter().all() would abort early, I think, which means that not all lhses would be checked, ie. not all errors we could report would be reported.
There was a problem hiding this comment.
ok, good point on not reporting all errors. Could you add a comment to explain that please?
|
r=me with the |
TimNN
force-pushed
the
infinite-emptiness
branch
from
2bbef65
to
d06a9ca
Compare
|
@bors: r+ |
|
📌 Commit d06a9ca has been approved by |
|
☔ The latest upstream changes (presumably #36616) made this pull request unmergeable. Please resolve the merge conflicts. |
TimNN
force-pushed
the
infinite-emptiness
branch
from
d06a9ca
to
51ea050
Compare
|
Rebased. |
|
@bors: r+ |
|
📌 Commit 51ea050 has been approved by |
reject macros with empty repetitions Fixes rust-lang#5067 by checking the lhs of `macro_rules!` for repetitions which could match an empty token tree.
reject macros with empty repetitions Fixes rust-lang#5067 by checking the lhs of `macro_rules!` for repetitions which could match an empty token tree.
Fixes #5067 by checking the lhs of
macro_rules!for repetitions which could match an empty token tree.