Closed Bug 1928480 Opened 1 year ago Closed 2 months ago

New wpt failures in /css/css-values/signs-abs-computed.html

Categories

(Core :: CSS Parsing and Computation, defect)

defect

Tracking

()

RESOLVED FIXED

People

(Reporter: wpt-sync, Unassigned)

References

(Blocks 1 open bug)

Details

(Whiteboard: [wpt])

Syncing wpt PR 48890 found new untriaged test failures in CI

Tests Affected

Firefox-only failures

  • /css/css-values/signs-abs-computed.html [wpt.fyi]
    • sign(10px - 1em) should be used-value-equivalent to 0; fontSize=10px: FAIL
    • sign(10px - 2em) should be used-value-equivalent to -1; fontSize=10px: FAIL
    • calc(2.5 - sign(41px - 2em) / 2) should be used-value-equivalent to 2: FAIL
    • calc(2.5 - sign(40px - 2em) / 2) should be used-value-equivalent to 2.5: FAIL
    • calc(2.5 - sign(39px - 2em) / 2) should be used-value-equivalent to 3: FAIL
    • calc(3 + sign(42px - 2em)) should be used-value-equivalent to 4: FAIL
    • calc(3 + sign(40px - 2em)) should be used-value-equivalent to 3: FAIL
    • calc(3 + sign(38px - 2em)) should be used-value-equivalent to 2: FAIL
    • calc(90deg + 30deg * sign(42px - 2em)) should be used-value-equivalent to 120deg: FAIL
    • calc(90deg + 30deg * sign(40px - 2em)) should be used-value-equivalent to 90deg: FAIL
    • calc(90deg + 30deg * sign(38px - 2em)) should be used-value-equivalent to 60deg: FAIL
    • calc(5s + 15s * sign(42px - 2em)) should be used-value-equivalent to 20s: FAIL
    • calc(5s + 15s * sign(40px - 2em)) should be used-value-equivalent to 5s: FAIL
    • calc(5s + 15s * sign(38px - 2em)) should be used-value-equivalent to -10s: FAIL
    • calc(2.5 - sign(33px - 2rem) / 2) should be used-value-equivalent to 2: FAIL
    • calc(2.5 - sign(32px - 2rem) / 2) should be used-value-equivalent to 2.5: FAIL
    • calc(2.5 - sign(31px - 2rem) / 2) should be used-value-equivalent to 3: FAIL

CI Results

Gecko CI (Treeherder)
GitHub PR Head

Notes

These updates will be on mozilla-central once bug 1928147 lands.

Note: this bug is for tracking fixing the issues and is not
owned by the wpt sync bot.

This bug is linked to the relevant tests by an annotation in
https://github.com/web-platform-tests/wpt-metadata. These annotations
can be edited using the wpt interop dashboard
https://jgraham.github.io/wptdash/

If this bug is split into multiple bugs, please also update the
annotations, otherwise we are unable to track which wpt issues are
already triaged. Resolving as duplicate or closing this issue should
be cause the bot to automatically update or remove the annotation.

The failures seem to all be cases where the test is using something like sign(10px - 1em) as part of a value for a number-valued-property like z-index.

Without having looked, I'm guessing we reject that because it looks like a type mismatch when we parse the <length> components, even though in fact the <length> units don't end up mattering since sign() just checks whether the value is positive or negative.

This is probably related to bug 1827404, though probably not exactly the same.

Blocks: css-values-4
Severity: -- → S3
Depends on: 1827404

This was actually resolved by Bug 2035572 so I'm going to associate it with that and close this out. The issue with almost all of these sign() constructions in a lot of WPT suites is that they used computed-value-time units like em or vh. But before Bug 2035572, Firefox always tried to eagerly resolve calc expressions at parse time for all types other than <length-percentage>, so it would never make it to the computed value stage.

Status: NEW → RESOLVED
Closed: 2 months ago
Depends on: 2035572
No longer depends on: 1827404
Resolution: --- → FIXED
You need to log in before you can comment on or make changes to this bug.