drott@, could you help triage this, to give some idea of how close the existing implementation is to other browsers and the spec? If this is a simple unprefixing with no changes needed, then it'd be great to just ship it.

The reason Kent found that this wasn't trivial in  bug 614955  was that we don't support multiple CSS property aliases and there's already -epub-text-emphasis --

But hypothetically, if we did support multiple aliases, then -epub-text-emphasis and -webkit-text-emphasis could probably be made an alias for webkit-text-emphasis, right?

Based on the (not 100% comprehensive) tests in
//third_party/WebKit/LayoutTests/fast/text/, it looks like they might be the same as the unprefixed versions. When experimenting with this later, we could try importing csswg-test/vendor-imports/mozilla/mozilla-central-reftests/text-decor-3/, which has a more comprehensive test for text-emphasis.

"we don't support multiple CSS property aliases" -> I'd guess you just need to update the enum assignment logic in css_properties.py and isPropertyAlias()/resolveCSSPropertyID().

Removing -epub-text-emphasis would also be an option, if this is the only case of multiple aliases. There are only 67 hits in httparchive, and at a glance it doesn't look like there's any real usage.

I think the implementation is working well, I have no objections unprefixing it. I am not sure I agree with what the tests are doing, though. I was just looking at the test import, and we're placing the emphasis marks centered on top of graphemes, while the test seems to expect them left aligned with the left grapheme boundary.

I take it this isn't observable to scripts in any way? Would you describe the differences as... trivial? If so it certainly seems like shipping now and fixing later is the better trade-off.

One caveat perhaps: I am currently not sure to which degree we support text-emphasis-position.

Test enabling in  issue 666657 .

So a quick check with http://roettsch.es/emphasistest.html shows that we do support over and under, and these values are effective in vertical as well, and do sort of the right thing, but we don't support left/right - which would not be terribly complicated to fix. This should probably be done before shipping it, but other than that I believe the emphasis mark implementation works quite well. (I haven't verified it's interoperability with ruby text.)

I doubt we'll manage to pass the css decor the tests as they are now, they are doing some CSS absolute positioning acrobatics to try and place the emphasis mark above where the emphasis code would place it, this is quite prone to subpixel errors.

We must support the new syntax.

[ over | under ] && [ right | left ]

https://bugs.chromium.org/p/chromium/issues/detail?id=522882

Looks like we'd need to support multiple aliases per property to fix this. While that's not super-easy, it's doable. Please reach out to me or shend@ if you want to work on this.

This issue has been Available for over a year. If it's no longer important or seems unlikely to be fixed, please consider closing it out. If it is important, please re-triage the issue.

Sorry for the inconvenience if the bug really should have been left as Available.

For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot