LGTM3

On 9/13/23 5:46 PM, Daniel Bratell wrote:

LGTM2

/Daniel

On 2023-09-13 17:14, Philip Jägenstedt wrote:
LGTM1

If there is feedback on the TAG review or Mozilla issue while this feature is on its way to stable, can you loop back to this thread?

On Thu, Sep 7, 2023 at 4:01 PM Philip Jägenstedt <foo...@chromium.org> wrote:

    Thanks for investigating and fixing the failures, Koji!

    On the UA defined rules, if other vendors are happy with the
    examples used, then that's what matters in practice. If you do
    get pushback on specific examples I hope there are others that
    can be used that are a common ground.

    I think everything looks good here, but since the TAG review and
    Mozilla issue were filed recently, I'd like to give those a bit
    more time.

    On Thu, Sep 7, 2023 at 11:14 AM Koji Ishii <ko...@chromium.org>
    wrote:

        I've got two updates on the questions:

        On Thu, Sep 7, 2023 at 2:46 AM Koji Ishii
        <ko...@chromium.org> wrote:

            Good point, thanks for asking.

            Technically speaking, we can't write any tests because
            the language-specific content analysis is UA defined.
            Tests use common and easy words that most engines would
            analyze the same way, but you're right that we may need
            to modify tests if any engine analyzes them differently.

            Tests contain that in comments, like this
            
<https://github.com/web-platform-tests/wpt/blob/master/css/css-text/word-break/auto-phrase/word-break-auto-phrase-001.html>
            or this
            
<https://github.com/web-platform-tests/wpt/blob/master/css/css-text/word-break/auto-phrase/word-break-auto-phrase-wbr-nobr-001.html>.
            I'm also talking to WebKit about tests, since there are
            currently two known engines; one in ICU and one in
            macOS/iOS. We'll work together to ensure words in tests
            are common and easy enough for both engines.


        In case this helps, the situation is the same as hyphenation
        tests. In the past, wpt tests needed updates when one
        implementation hyphenates differently from other browsers. I
        remember this happened at least once before.

            Four tests failing is strange, thanks for pointing them
            out too, they all pass in Chromium bots. I'll check them
            and make sure they're all green before shipping.


        I've figured them out, all fixes landed, I'll watch when
        wpt.fyi will be updated. One was a recent spec change I
        wasn't aware of, three were differences between wpt bots and
        chromium bots (#41851
        <https://github.com/web-platform-tests/wpt/issues/41851>.)

--
You received this message because you are subscribed to the Google Groups "blink-dev" group. To unsubscribe from this group and stop receiving emails from it, send an email to blink-dev+unsubscr...@chromium.org. To view this discussion on the web visit https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAARdPYfPQhus6RhhrdbBuhzPQaQraqSLOhHfxAyQ5eMgoy1sOA%40mail.gmail.com <https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAARdPYfPQhus6RhhrdbBuhzPQaQraqSLOhHfxAyQ5eMgoy1sOA%40mail.gmail.com?utm_medium=email&utm_source=footer>.

--
You received this message because you are subscribed to the Google Groups 
"blink-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to blink-dev+unsubscr...@chromium.org.
To view this discussion on the web visit 
https://groups.google.com/a/chromium.org/d/msgid/blink-dev/ce7a4dba-0160-427b-905c-ad537021da2d%40chromium.org.

Reply via email to