Behdad/Jonathan, Thank you for the awesome work you guys are doing.
I just realized that you both are having fun at Hackfest. I wanted to share one lingering category of issues for Telugu (my mother tongue). Fixing this should make Telugu failure rate reduce considerably. Shaping U+0C15,U+0C4D,U+0C30,U+0C3E should appear as క్రా but it appears as in attached file. If I understand anything here, looks like matra needs to be reordered ( https://www.microsoft.com/typography/otfntdev/teluguot/shaping.aspx). Also, can you share failure rate for Telugu? How can I access test framework you guys are using? Sorry, If you guys already sent info on this as I am not able to find it. Thanks, Srikar
<<attachment: output.png>>
_______________________________________________ HarfBuzz mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/harfbuzz
