[XeTeX] Sanskrit hyphenation

[Jonathan Kew]

On 5 Apr 2005, at 4:56 pm, Yves Codet wrote:

>
> Le 5 avr. 05, à 17:06, Jonathan Kew a écrit :
>
>>
>> That's odd, I would have expected it to read UTF-16 OK. You could try 
>> copying and pasting some text, just to test the font.
>
> WorldText seemed to read bytes, not characters.

If there's no BOM at the start of the file, you might need to 
explicitly select the encoding rather than allowing it to guess. You 
can do this if you use the Open dialog, I think.

>  If I paste it's alright, but there's no reordering of "two-part 
> dependent vowel signs" (U+0D4A to 0D4C); the rest looks fine. What is 
> weird is that characters U+0D46 to 0D48 (which behave like Nagari  ि  ) 
> are reordered.

Sure sounds like something is missing from the AAT tables in that font, 
then. In which case, I don't expect it'll work right in XeTeX either. 
(Unlike your OpenType font, which doesn't work in the editors that 
don't have OT support, but does work in XeTeX provided the right 
shaping engine [script] is specified.)

Welcome to the wonderful world of multiple font and rendering 
technologies!

JK