[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: [Lohit-devel-list] Kannada GPOS issue

Actually no Kannada fonts can handle a case like this in which a cluster of four consonant (0ccd) and three halanths are there! The word you have mentioned is a transliterated word from English. So in order to get good readability, it is better to use a zero-width non joiner (u0ccd) after the second consonant.

Also, in comparison with Gubbi and Navilu, Lohit doesn't handle the cluster that good. This has been already reported by Shriramana Sharma: https://bugzilla.redhat.com/show_bug.cgi?id=825081

By the way, in the image attached, you have added a space after the second 0ccd!


On Fri, Feb 1, 2013 at 1:31 PM, Pravin Satpute <psatpute redhat com> wrote:
On 02/01/2013 12:31 AM, Michael Lekman wrote:
> Hello,
> I have encountered a GPOS bug in Lohit Kannada font. The following
> text is not rendered correct.
> "\u0CB8\u0CBE\u0CAB\u0CCD\u0C9F\u0CCD\u0020\u0CB8\u0CCD\u0CA8\u0CBE\u0CAA\u0CCD"
> I did verify it both with 2.5.2 and 2.5.3.
> Is his a known issue?

Nope, first time got to know this.

While testing i found from given In the given string
\u0cab\u0ccd\u0c9f\u0ccd \u0cb8\u0ccd\u0ca8\u0cbe (ಫ್ಟ್ಸ್ನಾ ) is the
actually syllable causing problem.

This is very complex and rare combination, it involves CHCHCHCM (4
consonant and 3, C = consonant H = Halant and M = Matra). As far as i
know this is max allowed syllable for Brahmi based script though Open
Type rules not specifies it.

Still i am ok with fixing it, let me know how it should render exactly.
Anyone has any more information about such other combination?

Pravin Satpute

Lohit-devel-list mailing list
Lohit-devel-list redhat com

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]