In the unicode book, page 220, Rule 5, the diagram shows that for
conformance with the ISCII standard:
RRa + Virama -> EyelashRa
The question is, is this to be taken as an unconditional.
I mean, in an ISCII implementation, does RRa Virama -> EylashRa, when final
in a word?
If not then the diagram can not be taken as literally as I did.
As far as I can see it the rules for Ra and RRa are:
Ra Virama -> Ra_Virama (when final)
Ra Virama X -> RX_Virama
Ra Virama ZWNJ -> Ra_Virama X
Ra Virama ZWJ -> EyelashRa
*RRa Virama -> RRa_Virama (when final)*
*RRa Virama X -> EyelashRa_X*
RRa Virama ZWNJ -> RRa_Virama
RRa Virama ZWJ -> EyelashRa
(X is any consonant)
Does in fact everyone agree with this?
Abdul.
----- Original Message -----
From: "James Kass" <jameskass@worldnet.att.net>
To: <opentype@list.sirius.com>
Sent: Friday, May 19, 2000 3:49 AM
Subject: Re: Eyelash Repha
>Abdul Malik wrote:
>"...I must learn not to take the diagrams shown in the book quite so
>literally
>The diagram shows RRa + Virama -> EyelashRRa
>(perhaps the words: 'when followed by another consonant', could have been
>added, for people like me) ..."
>
>I don't see a problem with taking the diagram literally, suppose you
wanted
>to encode the diagram in Unicode? How would you do it? If RRa + Virama
>only equals ElashRa when followed by another consonant, then how to encode
>ElashRa for illustrative purposes ?
>
>Or, suppose that a Unicode file contained multilingual material in
>Devanagari. One language in the document uses the Eyelash Ra form and
>another language doesn't, but uses a rra with virama. How would each
>section be properly encoded ? (Does any language use rra with virama? For
>transliterations?)
>
>Unicode 3.0 apparently matches ISCII in all save some special Vedic
letters
>and the backwards compatibility for Eyelash Ra from Unicode 2.0, Ra +
>Virama + ZWJ (which is still included as valid Unicode in Unicode 3.0, see
>p. 218, R5a).
>
>So, If I get this right...
>Ra + Virama + ZWNJ = ra with virama
>Ra + Virama + ZWJ = eyelash ra
>Rra + Virama = eyelash ra
>
>How would one encode rra with virama,
>RRa + Virama + ZWNJ ?
>
>If the font developer doesn't include a glyph for eyelash ra along with a
>lookup table, the default will be displayed, rra with virama (or, ra with
>virama). But, if the font includes a glyph for eyelash ra, then both
>lookups should be present in the font's GSUB table.
>
>Anyone encoding the eyelash ra with an eye towards converting to/from
ISCII
>would use the Rra + Virama, as this offers a one-to-one correlation
between
>Unicode and ISCII .
>
>
>
________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com
This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:21:02 EDT