Whenever it's "safe," we'll need to enable the pref which is blocking off the ruby changes from non-debug builds.

(adding dependency on bug 1021952, since that's where we're creating the pref ("layout.css.ruby.enabled"))

Comment on attachment 8543692 [details] [diff] [review] patch With the expection that the only current blocker bug 1132008 will land m-c, I guess it is time to turn on this pref now.

Wow, don't you enable CSS ruby only on non-release build first for getting more feedback from our testers?

(In reply to Masayuki Nakano (:masayuki) (Mozilla Japan) from comment #4) > Wow, don't you enable CSS ruby only on non-release build first for getting > more feedback from our testers? As I understand it, that's exactly the plan here. This bug is on enabling the pref on mozilla-central / nightly, and letting that ride the trains. I don't think anyone's suggested flipping this on in official release builds right away. (Maybe I'm misunderstanding your concern, though?)

Comment on attachment 8543692 [details] [diff] [review] patch r=dbaron Please write a blog post soliciting testing and feedback. One other thing I noticed, though: many of the reftests should probably be contributed to the W3C's test suite. The easiest way to do that is to move them to layout/reftests/w3c-css/submitted/ruby/ and add the appropriate metadata. Could you do that as well?

(In reply to Daniel Holbert [:dholbert] from comment #5) > (In reply to Masayuki Nakano (:masayuki) (Mozilla Japan) from comment #4) > > Wow, don't you enable CSS ruby only on non-release build first for getting > > more feedback from our testers? > > As I understand it, that's exactly the plan here. This bug is on enabling > the pref on mozilla-central / nightly, and letting that ride the trains. I > don't think anyone's suggested flipping this on in official release builds > right away. > > (Maybe I'm misunderstanding your concern, though?) I was thinking that using #ifdef RELEASE_BUILD for a while because m-c will be merged next week. So, actually, it's not enough tested by Nightly users.

(In reply to David Baron [:dbaron] (UTC+13) (needinfo? for questions) from comment #6) > Comment on attachment 8543692 [details] [diff] [review] > patch > > r=dbaron > > Please write a blog post soliciting testing and feedback. > > One other thing I noticed, though: many of the reftests should probably be > contributed to the W3C's test suite. The easiest way to do that is to move > them to layout/reftests/w3c-css/submitted/ruby/ and add the appropriate > metadata. Could you do that as well? Yes, I'll do this in a separate bug 1134069.

(In reply to Masayuki Nakano (:masayuki) (Mozilla Japan) from comment #7) > (In reply to Daniel Holbert [:dholbert] from comment #5) > > (In reply to Masayuki Nakano (:masayuki) (Mozilla Japan) from comment #4) > > > Wow, don't you enable CSS ruby only on non-release build first for getting > > > more feedback from our testers? > > > > As I understand it, that's exactly the plan here. This bug is on enabling > > the pref on mozilla-central / nightly, and letting that ride the trains. I > > don't think anyone's suggested flipping this on in official release builds > > right away. > > > > (Maybe I'm misunderstanding your concern, though?) > > I was thinking that using #ifdef RELEASE_BUILD for a while because m-c will > be merged next week. So, actually, it's not enough tested by Nightly users. I just discussed this with dbaron, and he thought it's fine to just ship this and let it ride the train, because we can always turn off it anytime when something serious happens.

GJ! This increased our BrowserMark 2.1 score on CSS 2.1 conformity from 53% to 54%: [url]http://arewefastyet.com/#machine=30&view=single&suite=browsermark&subtest=Conformity CSS 2.1&start=1423040036&end=1424679226[/url]

Compat tables updated: https://developer.mozilla.org/en-US/docs/Web/CSS/ruby-align https://developer.mozilla.org/en-US/docs/Web/CSS/ruby-position https://developer.mozilla.org/en-US/Firefox/Releases/38#CSS