Re: Towards a better testsuite

On 15/04/2016 15:08, fantasai wrote:
> On 04/15/2016 01:03 AM, Geoffrey Sneddon wrote:
>> On Wed, Apr 13, 2016 at 12:29 PM,  <ishida@w3.org> wrote:
>>> (Btw, i'm wondering about the implications of automatically suppressing
>>> blank spaces at the end of lines of content for text that is written in
>>> scripts such as Japanese or Thai, where the whitespace rules may produce
>>> unexpected effects if a space at the end of your line is removed.)
>>
>> It shouldn't, except in the face of white-space: pre, because
>> otherwise "[SP][LF]" is equivalent to "[LP]".
>
> Assuming you meant [LF] there, this is true for white-space: pre-wrap
> as well.
>    https://drafts.csswg.org/css-text/#white-space-phase-1
>
> Note that there isn't such an equivalency in any tests that make sure
> this actually happens correctly, though. ;)

not sure what you mean there, but hopefully you don't mean that these 
tests don't exist ;-)

https://www.w3.org/International/tests/repo/results/white-space


the thing i'm concerned about (and this is off-topic AND i haven't had 
time to think it through clearly) is where a  person is creating thai 
content.  Since spaces are used only for sentence/phrase separators in 
Thai, and as such are punctuation marks rather than just gaps, the 
author may put a space at the end of a line in the source code to 
separate two sentences.  If the editor removes that trailing space, the 
sentence separator is lost and the two sentences will run together when 
rendered by the browser, because the white space rules remove LFs 
between Thai characters.  So for thai content, editors that silently 
strip out spaces at the end of a line could be problematic.

Similar things can happen, of course, for Chinese, Japanese, Tibetan, 
etc., where spaces have meaning over and above white space.

ri

Received on Friday, 15 April 2016 14:31:10 UTC