bidi rules #497
bidi rules #497
Conversation
|
☔ The latest upstream changes (presumably #517) made this pull request unmergeable. Please resolve the merge conflicts. |
There was a problem hiding this comment.
I've promised to find the time to do the review here.
I've tried to read it thoroughly and I believe the changes do what you say in the comment, with some very minor nitpicking and questions.
That doesn't answer the question if this is the behaviour all the standards agree on, though. I'll find the time to try reading through the standards mentioned there later and try to come up with an answer.
| @@ -37,6 +40,13 @@ pub fn collect_tests<F: FnMut(String, TestFn)>(add_test: &mut F) { | |||
| let to_ascii = pieces.remove(0); | |||
| let nv8 = if pieces.len() > 0 { pieces.remove(0) } else { "" }; | |||
|
|
|||
| for skip in SKIP_TEST { | |||
| if original == *skip { | |||
There was a problem hiding this comment.
Can this be done with SKIP_TEST.contains(original)?
| @@ -10,6 +10,9 @@ use std::char; | |||
| use idna::uts46; | |||
| use test::TestFn; | |||
|
|
|||
| const SKIP_TEST: &'static [&'static str] = &["0A.\\u05D0", "0a.\\u05D0", "0a.xn--4db" | |||
| ,"1.걾6.𐱁\\u06D0", "1.걾6.𐱁\\u06D0","1.xn--6-945e.xn--glb1794k"]; | |||
There was a problem hiding this comment.
Probably explaining why these are skipped in a comment would be nice ‒ maybe linking to the issue?
| } | ||
| _ => { break; } | ||
| } | ||
| } |
There was a problem hiding this comment.
This loop doesn't really check if it is truly a LDH label. I think it would accept something like 3_4 as well (_ is ascii, but neither letter, digit or hyphen). Does something else check that in the existing code?
|
|
This is a pull request for issue #489
This change is