[HarfBuzz] A problem about indic shape (for khmer)

[datao zhang]

Hi:
 
When I test some character about khmer, I find some Matra can't be decomposed. Example: I input 0x1782,0x17C4, the 0x17C4 should be decomposed as two parts and the 0x1782 should be located in the middle.
 
I use the ICU library. After call the unorm_normalize() function, no normalized value returned.  Then I also try use the glib, it is same as ICU. I don’t know if there is problem in my ICU library or the harfbuzz should handle such case.
 
I also test some case in Bengali, example : 0x09CB. I find my glib and ICU can decompose it correctly. I don’t know if others test such case. If this is a problem, maybe the harfbuzz need handle such case.
 
The following Matra can’t be decomposed:
0x17C0
0x17C4
0x17C5
0x17BE
0x17BF
 
Br,
Dean
  		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.freedesktop.org/archives/harfbuzz/attachments/20120420/d2dae463/attachment.html>