- From: Martin Duerst <duerst@w3.org>
- Date: Fri, 08 Oct 2004 09:14:34 +0900
- To: "L. David Baron" <dbaron@dbaron.org>, GEO <public-i18n-geo@w3.org>, www-international@w3.org
At 15:52 04/10/07 -0700, L. David Baron wrote:
>This seems (I think, although I'm not sure where the letter boundaries
>are) to work correctly for me in Mozilla, Opera, and Konqueror, but
>that's probably because my windowing system (GNOME 2.6 on Fedora Core 2)
>handles bidirectional text and Arabic shaping.
>
>It seems like this should really be described as a test of the windowing
>system rather than a test of the browser.
I think you have a valid point here: The test tests bidirectionality
handling for titles, but it is dependent on the windowing system.
One way to handle this would be to have two tests:
1) precondition test: e.g. a single Arabic word, to test basic
rendering, shaping, and bidi handling
2) feature test: a series of Arabic and Latin words that should get
rendered differently with different values of the dir attribute.
The table for the interpretation of the results would then be:
Result Result Interpretation
of 1) of 2)
fail don't even try 2),
your window system isn't up to it.
okay fail browser doesn't correctly pass bidi
info to windowing system
okay okay browser handles dir for title correctly
Regards, Martin.
Received on Friday, 8 October 2004 02:52:24 UTC