Fwd: Status of support for Indian languages

[Linda W]

For some reason google kicked back vim_multibyte as not
existing... Anyway...see below...

-------- Original Message --------

Subject:	Re: Status of support for Indian languages
Date:	Mon, 12 Aug 2013 17:00:05 -0700
To:	vim_mu...@googlegroups.com
CC:	vim-mu...@vim.org

Arnab Bhattacharya wrote:

Till now, Vim has very poor support for Indian languages.

Is there as workable solution yet?

Arnab

-----
What were you looking for?  I.e. it may be your OS version of Vim.


The above has Devangari and a few other characters.

But it's displaying using X11 on Windows displaying from linux.

Using a Windows-text window:



Microsoft is poor about keeping up with the Unicode standard....
Most of their stuff is back at Unicode V1.0 (when UCS-2 held
everything).  Ever since unicode has been growing MS has been
getting more behind.

[Tony Mechelynck]

On 13/08/13 02:06, Linda W wrote:
> For some reason google kicked back vim_multibyte as not
> existing... Anyway...see below...
>
> -------- Original Message --------
> Subject: Re: Status of support for Indian languages
> Date: Mon, 12 Aug 2013 17:00:05 -0700
> To: vim_mu...@googlegroups.com
> CC: vim-mu...@vim.org
>
>
> Arnab Bhattacharya wrote:
>> Till now, Vim has very poor support for Indian languages.
>> Is there as workable solution yet?
>> Arnab
> -----
> What were you looking for? I.e. it may be your OS version of Vim.
>
>

> The above has Devangari and a few other characters.
>
> But it's displaying using X11 on Windows displaying from linux.
>
> Using a Windows-text window:
>
>
>

> Microsoft is poor about keeping up with the Unicode standard....
> Most of their stuff is back at Unicode V1.0 (when UCS-2 held
> everything). Ever since unicode has been growing MS has been
> getting more behind.

Even in the first image, you can see that there is poor (if any) support
of the way nāgarī letters change when joined. The successive glyphs in a
word shouldn't just be placed next to each other unchanged, see
http://en.wikipedia.org/wiki/Devanagari_script#Biconsonantal_conjuncts
(and some of the info elsewhere on that page too, in particular the
consonant-vowel combinations and the photograph of an Australian tram
with "incorrect" devanāgarī lettering in an advert).

Of course, if the (Indian, or sanskriptologist, or…) user of Vim can see
different glyphs and understand what is being meant, it's better than
nothing, even if the rendering is less than perfect.


Best regards,
Tony.
--
"The Puritan through Life's sweet garden goes
To pluck the thorn and cast away the rose."
[Kenneth Hare]

[Ven Tadipatri]

On Tue, Aug 13, 2013 at 8:37 AM, Tony Mechelynck <antoine.m...@gmail.com> wrote:

Even in the first image, you can see that there is poor (if any) support of the way nāgarī letters change when joined. The successive glyphs in a word

It seems this issue occurs in other Indian languages as well. On a CentOS 5 machine with Kannada Unicode installed, typing u+0C95 followed by u+0CBF produces a weird consonant-vowel combination. But I was surprised to see that the same problem does not occur in Eclipse. When the second unicode character is typed, it is correctly appended to the first one (after a bit of a delay). The same poor rendering of glyphs observed in VI occurs in a standard bash shell (Gnome terminal). So there is support for Indic scripts in some programs, but it doesn't appear to be there in a standard shell.

   It seems like this isn't so much a problem with VI as the underlying shell that's invoking it right? The way text is rendered on the shell prompt matches exactly with the way text is rendered in VI.

 

shouldn't just be placed next to each other unchanged, see http://en.wikipedia.org/wiki/Devanagari_script#Biconsonantal_conjuncts (and some of the info elsewhere on that page too, in particular the consonant-vowel combinations and the photograph of an Australian tram with "incorrect" devanāgarī lettering in an advert).

I was staring at that advertisement for a while, but I don't quite see what you mean. There are no vowels in the middle of the word and it seems to be rendered ok. Where's the incorrect lettering?

 

Of course, if the (Indian, or sanskriptologist, or…) user of Vim can see different glyphs and understand what is being meant, it's better than nothing, even if the rendering is less than perfect.

I think rendering for Indic scripts has a long way to go. Even trying gmail's input tools with the same version of Firefox on two different OS's, I noticed that glyphs were not combined correctly on a Mac, but they seemed fine on a Windows machine. Perhaps the fonts were not correctly installed on the Mac, or perhaps there just isn't enough interest out there for someone to fix these bugs/standardize the way Indic fonts are rendered.

 

Best regards,
Tony.
--

Thanks,
Ven