Enforce escape functions to operate over UTF-8 #423
Conversation
Mingun
added
the
encoding
| @@ -901,11 +897,13 @@ impl<'a> BytesCData<'a> { | |||
| /// | `&` | `&` | |||
| /// | `'` | `'` | |||
| /// | `"` | `"` | |||
| pub fn escape(self) -> BytesText<'a> { | |||
| BytesText::from_escaped(match escape(&self.content) { | |||
| pub fn escape(self, decoder: Decoder) -> Result<BytesText<'a>> { | |||
There was a problem hiding this comment.
Because BytesText::escape and BytesText::partial_escape introduced only in new 0.24 release, the changes in their signatures are not reflected in the changelog. Moreover, the next PRs will change that signature back
Codecov Report
@@ Coverage Diff @@
## master #423 +/- ##
==========================================
- Coverage 49.53% 49.50% -0.04%
==========================================
Files 22 22
Lines 13856 13861 +5
==========================================
- Hits 6864 6862 -2
- Misses 6992 6999 +7
Flags with carried forward coverage won't be shown. Click here to find out more.
Continue to review full report at Codecov.
|
|
FYI, pretty sure that 1a1f4bf doesn't work, as equality comparison on Cow ignores the variant. You probably need to use |
|
Yes, you're right. As far as I remember, there are a few more places in tests where we rely on such behavior |
They are always operate over UTF-8 input, but that was not reflected in types.
This set of changes is more isolated that one in #422, and it also does not enforce owning of data in some cases where that is not required.