[BRLTTY] RTL languages and BRLTTY (was Re: BRLTTY in GRUB)

[Vladimir 'φ-coder/phcoder' Serbinenko]

On 16.03.2012 20:53, Dave Mielke wrote:
>> >I've just checked with linux terminal, gnome-terminal and mlterm and
>> >only mlterm got Hebrew (on screen) right, all others have written it
> >from left-to-right. Perhaps I just miss a software component. So this
>> >doesn't give much info as to how it should be done.
>> >According to wikipediahttp://en.wikipedia.org/wiki/Hebrew_Braille
>> >braille in Hebrew is read from left-to-right. This creates a weird
>> >situation with screen-reading solution that you have to get it wrong
>> >on screen to get it right on braille.
>> >Similar situation with Arabic:http://libbraille.org/arabic_alphabet.php
>> >For GRUB it won't be any problem as I can easily make GRUB to do bidi
>> >on visual terminal and don't reorder it on braille.
> Or just render it properly and leave it to us to find a solution. We;ll clearly
> have to, anyway, for other environments.
Problem is that there are additionally several unicode bidi control 
characters which make it impossible to restore original reliably. If you 
see an alef left to beth then it might be either Hebrew word ba' (come) 
or a beginning of alphabet used in some generally left-to-right text 
with overrides.
Another example: Let's say we have some table with title line
2ELTIT 1ELTIT
         2         1
This is how a table would look like. But without knowing the element 
boundary and without control characters you would reverse the first line:
TITLE1 TITLE2
But then second line is purely numerical so it stays the way it is:
          2         1
Now our table is obviously screwed.
Ideally we need some way to reliably get not-reordered version while 
allowing the visual screen to be reordered.
Another issue with Hebrew is that braille character depends on both 
dagesh and sin dot. Both are usually omitted in normal text. There are 2 
types of dagesh: lenae and forte. Former depends only on consonant in 
question and preceding vocal. Unfortunately vocals aren't written 
either.  But in Hebrew both dagesh forte and vocals depend on 
grammatical form and not only marginally on root. Most of the roots in 
Hebrew are three-letter ones. So a possible approach is to have the list 
of forms which infer dagesh forte or lene on one of its letters with 
root consonants replaced by wildcard which can match any Hebrew consonant.
This doesn't work for loanwords though where the choice between putting 
dagesh or not is usually decided based on original writing or pronunciation.
Similar applies for the two uses of yud: as a vowel or a consonant.
Shin with a sin dot is another story, it's a complete letter which can't 
be inferred from grammar but is a possible root letter.  Fortunately 
it's a rare letter. An example I can come up with offhand is "Israel". 
In Hebrew it's usually written ישראל. Full vocalisation would be יִשְׂרָאֵל, 
where you can clearly see the sin dot. So we need a rule for Israel. I 
see no exception file for Hebrew. Could someone render "Israel" from 
vocalised version I pasted and put it as a rule for non-vocalised one?
In Yiddish, which is Germanic language which uses Hebrew script grammar 
doesn't help to disambiguate between those letters but vocalisation on 
ambiguous letters is more common but isn't universal.




-- 
Regards
Vladimir 'φ-coder/phcoder' Serbinenko