[Padma] Malformed Conjuncts

Nagarjuna Venna vnagarjuna at gmail.com
Tue Feb 10 12:59:33 PST 2009

Hi Golam,

I don't think the parser works as you described. It actually pushes
the vowel sign to the end (modulo other things that can go at the end)
or at least that's what I remember. I think if it did not do that,
most of the transformations would be broken.

I didn't have a chance to look at the code yet, but may be the problem
is something else?


On Sun, Feb 8, 2009 at 2:55 PM, Golam Mortuza Hossain
<gmhossain at gmail.com> wrote:
> Hi,
> While trying to add a new font (Datasoft, www.aajkaal.net)
> in Padma,  I am facing couple of issues that are breaking
> conjunct formation.
> The current handling of prefixList in Padma breaks conjunct
> formation, if the conjunct follows a matra which is in prefixList.
> For example: If the ordering in original font is say
> Vowelsn_E  + (Letter_LA + Halant + Letter_PA).   { "ে" + "ল্প" }
> As Vowelsn_E is in prefixList, Padma correctly pushes it down to
>  Letter_LA + Vowelsn_E  + Halant + Letter_PA.   { "ল্েপ" }
> However, notice that it now breaks conjunct formation. The correct
> ordering should be
>  (Letter_LA + Halant + Letter_PA)  + Vowelsn_E  .   { "ল্পে" }
> This along with few other issues can be easily handled if Padma
> API is expanded to have a post processing subroutine. This
> should be analogous to the current "preprocessMessage"  API.
> For example, I am considering following code
>     if (encoding.postprocessMessage != undefined)
>        this.text = encoding.postprocessMessage(input);
> However, I am not sure about the best place to put the
> code in Padma. Can anyone please shed some light on this?
> Thanks,
> Golam
> _______________________________________________
> Padma mailing list
> Padma at mozdev.org
> https://www.mozdev.org/mailman/listinfo/padma

More information about the Padma mailing list