Closed michael-kerscher closed 1 month ago
Attention: Patch coverage is 85.56701%
with 14 lines
in your changes are missing coverage. Please review.
Project coverage is 90.95%. Comparing base (
2b14491
) to head (bb97df5
). Report is 2 commits behind head on main.
Files | Patch % | Lines |
---|---|---|
i18n-helpers/src/lib.rs | 76.27% | 10 Missing and 4 partials :warning: |
:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.
Thanks @michael-kerscher for putting this up! Based on the test changes, it looks great.
@dyoo and @kdarkhan, I would love to have one of you look at this since you've both been working with this code recently.
The fuzz step is failing due to some changes in rust nightly. The failure seems to occur from one of the dependencies.
Until the dependency is fixed, you can temporarily pin to specific nightly version by changing this line to rustup default nightly-2024-05-10
.
The rules that determine whether the HTML is block or inline surprised me. I naively assumed that if a piece of HTML is on a separate line, then it is a block, and inline HTML otherwise. Turns out I was wrong. The rules are defined in https://spec.commonmark.org/0.31.2/#html-blocks. Some tags like p
or div
in a single line are considered to be blocks, while span
is an inline.
Here is an example which demonstrate how they differ. Switch to AST
tab to view inlines vs blocks.
I believe even with above rules, not splitting inline elements into separate groups is an improvement.
Fixes #125 by keeping inline HTML tags in the translation text group but skip block level HTML tags. This is possible with the upgrade of pulldown-cmark in #183