W3C home > Mailing lists > Public > www-international@w3.org > January to March 2015

Re: New hyphens tests

From: Richard Ishida <ishida@w3.org>
Date: Thu, 05 Mar 2015 12:45:45 +0000
Message-ID: <54F84FF9.3050507@w3.org>
To: John Daggett <jdaggett@mozilla.com>
CC: www International <www-international@w3.org>
thanks for the comments.

On 05/03/2015 11:23, John Daggett wrote:
> Richard Ishida wrote:
>
>  > i just posted a set of tests for CSS Text Level 3, section 6.1
>  > Hyphenation Control: the hyphens property, with results for major
>  > browsers, at
>  >
>  > http://www.w3.org/International/tests/repo/results/hyphens
>
> Prior to submission it might be better to structure the tests so that
> they pick up the prefixed hyphens property that is already widely
> implemented. The "results" here are basically bogus since for all
> current browsers since the unprefixed 'hyphens' property is not supported.

i think the tests do usefully show that the unprefixed versions are not 
yet available. They don't, however, as you say, reflect the work that 
has gone into the prefixed implementations. I was indeed already 
thinking about how to extend the tests to check the prefixed 
implementations, so hopefully that will appear shortly.  I want to keep 
a clear distinction between prefixed and non-prefixed.

>
> MDN 'hyphens' page:
>
> https://developer.mozilla.org/en/docs/Web/CSS/hyphens
>
> Not sure I believe the language columns for non-Firefox browsers...
>
> I think it might be more useful to have tests that pass/fail per
> language, since this is a unique issue for this property (without
> underlying dictionary support, using the hyphens property has no effect
> on text display). And you should be aware that the "support" will vary
> by platform version in the case of Safari and IE which are relying on
> underlying platform dictionary resources, at least as I understand it.

It would indeed be nice to have per language tests, although with 
limited resources for test devt it would still be necessary to guess at 
what would and would not be supported with the auto setting, given the 
points you rightly make about the potentially variable nature of the 
dictionaries used.  I don't have a lot of time for test development, but 
I thought that these tests would at least be better than no tests.

cheers,
ri
Received on Thursday, 5 March 2015 12:45:54 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 21 September 2016 22:37:38 UTC