Re: [Varnamproject-discuss] Libvarnam issue

[Navaneeth K N]

On Thu, May 12, 2016 at 12:52:52AM +0530, malayalee coder wrote:
> As I would be doing my GSoC project in varnam and Indic-Keyboard, for
> starters, I am trying to solve some beginner issues in both the areas.
> In,
> 
> https://github.com/varnamproject/libvarnam/issues/133
> 
> a doubt raised, whether, when accepting a sentence, how are the
> transliterated suggestions be displayed? I mean, should it be just
> word by word suggestion, one after the other in vertical blocks? Or
> the entire combination of the words which the sentence is composed of
> as a suggested sentence?
> 
> Example,
> varnamc -s ml -t ente keralam ethre sundaram
> 
> if the suggestions are like a sentence, there would be 10*9*6*6 = 3240
> lines of suggestions which is obsolete. So can anyone suggest me a
> better option.

When it is a sentence, it is very difficult to provide suggestions. Like
you have observed, suggestions are for individual words. 

What you can do is to get the first suggestion for each word and output
a single sentence by combining them.

> 
> Thanks.
> 

-- 
Thanks
Navaneeth