[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Devanagari ra + vocalic r encoding and rendering problem in yudit 2.7.2



I'm not sure whether this is the right forum for this.  I found a
reference to this list on yudit's web site and didn't find a list
specially dedicated to yudit anywhere.

I will say right off the bat that I've also tried what follows with
yudit 2.7.5 beta11 and found the same problem.  I couldn't look for
clarifications in Unicode 4.0 since I don't have access to that
version of the standard yet.  I had to use 3.0 as my source.  (I
looked in the erratums and corrigendums (ie. version 3.1 and 3.2) and
didn't find anything relevant to this discussion).

The problem is twofold:

1. The Devanagari.kmap file (used to create Devanagari.my) doesn't
   declare a proper encoding for ra + vocalic r.  The mapping is declared as:

   "rR^i=0x0930 0x0943"

   so it creates the sequence U+0930 U+0943.  I think the proper
   sequence is U+0930 U+094D U+090B.

   I base my opinion on Chapter 9 (p. 214) of the Unicode 3.0
   standard.  Figure 9-3, item 4 shows the conversion of the very case
   I'm having a problem with from the Unicode stream of characters to
   a stream of glyphs.

2. Even if I get yudit to produce U+0930 U+094D U+090B, the conjunct
   is not properly rendered.  It is rendered as two separate glyphs:
   ra + virama and an initial vocalic r instead of just one glyph
   resulting from the combination of an initial vocalic r and repha.
   In other words, what I see is what appears on the left of the
   bottom arrow in item 4 of figure 9-3.  What I expect is what
   appears on the right of the arrow.

I am using the TrueType font to render this and given the installation
I have, I'm guessing that these characters are actually taken from the
raghu.ttf font.

Note that the equivalent problems exist with long vocalic r.

        Thanks,
        ldd
--
Linux-UTF8:   i18n of Linux on all levels
Archive:      http://mail.nl.linux.org/linux-utf8/