Per https://phabricator.services.mozilla.com/D33233#986429 , I suspect emilio may actually have been asking for a "testharness.js"-based test here, not a reftest, to test the parsing behavior. (This will be sort of redundant in our own test suite, since we've got the `property_database.js`-based mochitests that should exercise this pretty well. But I suppose it's still good to add as a validation that our expectations about parsing matches what other browsers do. And it lets us express that e.g. `from-font` is supposed to be supported but is not yet, which we can do by adding a failure annotation for just that one part of the test, if we hypothetically had a web-platform-test that tests all the values.) You could probably start with this testcase as an example to use: https://searchfox.org/mozilla-central/source/testing/web-platform/tests/css/css-text-decor/text-decoration-skip-ink.html Note that we fail that test right now, and we have the specific failures annotated here: https://searchfox.org/mozilla-central/source/testing/web-platform/meta/css/css-text-decor/text-decoration-skip-ink.html.ini (Note the paths are almost the same, except for "meta" instead of "tests" and the ".ini" suffix. This is how our test harness discovers known/expected test failures.) So you could, for example, make a copy of that testcase and just swap out the property & values for `text-underline-offset` and its values in your copy.
Bug 1556041 Comment 1 Edit History
Note: The actual edited comment in the bug view page will always show the original commenter’s name and original timestamp.
Per https://phabricator.services.mozilla.com/D33233#986429 , I suspect emilio may actually have been asking for a "testharness.js"-based test here, not a reftest, to test the parsing behavior (the part that we'll support [aside from `from-font`] when your patch there lands). (This will be sort of redundant in our own test suite, since we've got the `property_database.js`-based mochitests that should exercise this pretty well. But I suppose it's still good to add as a validation that our expectations about parsing matches what other browsers do. And it lets us express that e.g. `from-font` is supposed to be supported but is not yet, which we can do by adding a failure annotation for just that one part of the test, if we hypothetically had a web-platform-test that tests all the values.) You could probably start with this testcase as an example to use: https://searchfox.org/mozilla-central/source/testing/web-platform/tests/css/css-text-decor/text-decoration-skip-ink.html Note that we fail that test right now, and we have the specific failures annotated here: https://searchfox.org/mozilla-central/source/testing/web-platform/meta/css/css-text-decor/text-decoration-skip-ink.html.ini (Note the paths are almost the same, except for "meta" instead of "tests" and the ".ini" suffix. This is how our test harness discovers known/expected test failures.) So you could, for example, make a copy of that testcase and just swap out the property & values for `text-underline-offset` and its values in your copy.