[HarfBuzz] Lemongrass HarfBuzz Hackfest, end of day 4

[Rajeesh K Nambiar]

On Tue, Jul 24, 2012 at 9:05 AM, Shriramana Sharma <samjnaa at gmail.com> wrote:
> On Mon, Jul 23, 2012 at 3:27 PM, Rajeesh K Nambiar
> <rajeeshknambiar at gmail.com> wrote:
>>
>> From the in-tree test file, test cases 15, 19, 20, 53 and 54 all
>> involve 'Ra'. Case 5 (കാര്‍ക്കോടകന്‍) involves 'RA+ZWJ' in the middle
>> of the word. At present all are failing (interestingly, case 5 was
>> fine during the crash ;-)).
>
> Hello -- running your sequence through UniView
> (http://rishida.net/scripts/uniview/) shows me that the sequence you
> have used is actually RA + VIRAMA + ZWJ (and not RA + ZWJ). However,

Right, it is RA + VIRAMA + ZWJ.

> ever since the encoding of the atomic chillu-s, the above sequence is
> no longer recommended for producing the chillu-RA because it would
> cause distinct encoded sequences to be rendered the same.

An unfortunate decision by Unicode consortium, yet we are left with
supporting thousands of articles using those encodings.

>
> As such I think it is not advisable for HB to support the old sequence
> RA + VIRAMA + ZWJ. Behdad might like to ask about this on the Unicode
> (or Unicore, if he has access) before implementing support for those
> old sequences.
>
>> Also noticed that post-base 'LA' is not correctly rendered for 'സ്പ്ലേ'.
>
> Can you clarify what you mean by *post*-base LA? You mean *sub*-base
> LA? Because AFAIK only YA and VA have post-base forms, but perhaps I
> am mistaken...

Right again, sub-base LA. My mistake.

>
> --
> Shriramana Sharma


-- 
Cheers,
Rajeesh