W3C home > Mailing lists > Public > www-style@w3.org > January 2015

Re: [css-text][css-writing-modes] Line breaking around Emoji, Gaiji, U+FFFC, and text-combine-horizontal

From: Koji Ishii <kojiishi@gmail.com>
Date: Thu, 22 Jan 2015 03:42:32 +0900
Message-ID: <CAN9ydbUqEn7AKz4HkSfPzgNQbySVThi9MkKn6KQ8NObQF+s=EQ@mail.gmail.com>
To: fantasai <fantasai.lists@inkedblade.net>
Cc: "www-style@w3.org" <www-style@w3.org>
On Thu, Jan 22, 2015 at 2:00 AM, fantasai <fantasai.lists@inkedblade.net> wrote:
> On 01/17/2015 05:53 AM, Koji Ishii wrote:
>>
>> I’m revising the proposal after discussed with Unicode experts.
>>
>> 1. Change the 5th bullet of 5.1 Line Breaking Details[1] to force a soft
>> wrap opportunity both before and after a replaced element and other atomic
>> inline for web-compatibility, but not by changing the behavior of U+FFFC nor
>> the rule orders of UAX#14.
>
> I think this is fine, although it doesn't appear to be a valid tailoring
> of UAX#14, either.

It was primarily said hard to implement using general libraries such
as ICU. I didn't hear which violates worse or better.

>> 2. Add ‘line-break-string’ property. When this property is set to a string
>> of 1 character length, the element looks as if its text is the character to
>> line breakers and justification algorithms.
>
> I don't understand this property, and do not want to add a new property.

It'll let <img> handled as if it were a single specified character to
the line breaker, just like text-combine is treated as if it were
U+FFFC. I wish to limit the effect to the minimal set of elements such
as img. So:

img.emoji { line-break-string: 'U+4E00'; }

can make the img to break just like an ID character. Does this make clearer?

>> One of the experts recommended ID (Ideographic) class[2] for inline images
>> such as Emoji. One idea is to have a boolean property to opt-in to this
>> behavior. Another idea was to for authors to specify UAX#14 class. Among the
>> three options, I think the 1 character length string gives the best
>> flexibility.
>
> Is there a reason we can't just treat as ID always?

It was pointed out that all browsers except Presto breaks
<img>&nbsp;<img> and following UAX#14 will break web-compatibility[1].
It's a bit hard to imagine for me if authors expect this sequence to
break before and after the &nbsp;, so if we want to go that route, I'm
good, but I suspect this route needs discussion?

[1] https://lists.w3.org/Archives/Public/www-style/2013Nov/0136.html

/koji
Received on Wednesday, 21 January 2015 18:43:00 UTC

This archive was generated by hypermail 2.4.0 : Friday, 25 March 2022 10:08:50 UTC