On 21/05/14 18:19, Devin Asay wrote:
On May 21, 2014, at 1:06 AM, Richmond <richmondmathew...@gmail.com>
wrote:
On 21/05/14 05:24, Alejandro Tejada wrote:
Hi Richmond,
Could you create any kind of custom text editor
Yes: my PISMO program is just a (much simplified) version of
my Devawriter program using a different writing set.
The main difference lies in the fact that both Cyrillic and Glagolitic (the
writing systems used in
my PISMO program) are written from left-to-right in a straight-forward way.
Devanagari (like many Indian writing systems) is written from left-to-right,
but there is a problem
to do with the positioning of a post-vocalic R, a short I (which is written
before a consonant that
it is sounded after, and the fact that a consonantal cluster aggregates into a
new written form called
a 'conjunct consonant'. So Devawriter contains something in the order of 5,000
rules to make
sure that when end-users tap in individual glyphs (the Devanagari script is an
abugida, not an
alphabet) the 'magic' happens [mind you, programming the thing didn't feel very
magical at
times].
Richmond,
Have you tried out the prerelease of LiveCode 7? I'm really curious how your
projects will work with it. My understanding is that all of the arcane,
language-specific glyph munging rules are handled transparently (or will be).
I've fiddled with it a bit with Cyrillic, and one of my students and I
experimented with Arabic. The results with Cyrillic were perfect (although just
typing Russian has never been problematic in LC, moving text around and looking
at chunks of text required lots of hoop-jumping. Arabic and other RTLS still
have some problems, but they're light years ahead of what we've had before.
These are things we ought to hammer on during the Global Jam Richard announced
last week. LC 7 I think will be a huge step forward for those of us who use
Unicode.
Regards,
Devin
Devin,
No, I haven't for 2 reasons:
1. Time (and by time I mean that I will have to do a search and replace
for thousands of numToChar statements).
2. Money (by that I mean that, currently, there is no incentive for me
to change as my programs
work as they are [version 4.5] and the money I am making from my
programs currently comes to about $150 per year).
However; "just for fun" (which, in my books at least, is the best reason
to do this sort of thing<0, if
I can find the time this coming weekend, I will try with my PISMO
programs, and then, if that works,
at a lter date with my Devawriter.
However; as I understand things the problems associated with languages
that do not write in a
straightforward RTL or LTR way will not be sorted out; my Devawriter
will still need all its
sub-routines to arrange letters:
Consider the Sanskrit word "Karma"; I type in k - a - r - m - a and the
output has to be:
'kamra' (which is read and pronounced 'karma').
Similarly with the word "hindi": I type h - i - n - d - i and the output
has to do this:
i - h - i - (nd) where '(nd)' is a "funny letter" (conjunct consonant)
formed when a 'n' knocks up
against an 'h'.
I have not found the Unicode system in versions of RR/LC from 2 to 6.7
a source of difficulty;
in fact, far from it, they have been a source of great joy as I have
been able to successfully
been able to carry my Sanskrit thing "off".
-------------------------------------------------------------------
"glyph munging rules" (what a whacked-out phrase; lovely) may be there;
but certainly not for
Sanskrit unless one uses exactly the half-cocked halant system that
others use that does not
do the writing system any favours; and certainly not for a writing
system that has to access thousands of additional characters in the
Personal Private Use areas of the Unicode system that are specific to
my specialist fonts.
-------------------------------------------------------------------
Best, Richmond.
_______________________________________________
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode
