Content-Type: text/plain; charset=us-ascii
Content-Description: message body text
Content-Transfer-Encoding: 7bit
For those of you on the other mailing lists (and Hovik), the message that
prompted this reply is attached at the end of this message so you don't miss
any of the details.
>> FE (U2019) Right Single Quotation Mark FF (U0027) Apostrofe
Pablo> In fact 0xFE is "Armenian Apostrophe" (0x055A) (both 0x2019 and
Pablo> 0x0027 are absurd mappings at those positions: 0x2019 because the
Pablo> paired opening quotation mark is missing, so why having only the
Pablo> right side ? 0x0027 because it is already at 0x27 ! ).
Ahh. A regretable error. I think it slipped into my table as a result of
some unclear pre-1997 standardization efforts that I was following.
Pablo> 0xFF is undefined in armscii-8; I put there the "armenian section
Pablo> sign" (0x00A7) which was at 0xA2 on previous version of armscii-8;
Pablo> so losless conversion can be done.
For various reasons, the section mark was replaced with the EV ligature. I
agree with Pablo that for conversion between the older and newer standards
without loss, 0xFF should be used for the paragraph mark.
Pablo> Armscii-8 defines a char "Armenian eternity sign" wich looks a bit
Pablo> like the indian wheel in the Indian flag (or like a daisy flower).
Pablo> However, it isn't on unicode... I've mapped it to the value
Pablo> 0x0530, as that one is on the Armenian section of unicode and
Pablo> empty.
Because there is no real effort being made to add an Armenian eternity symbol
to Unicode that I know of, my opinion is that this should be mapped to U+E530
for round trip conversion. Using U+0530 may cause errors in strictly
conforming programs.
I will be adding AST34.001 tables to the CSets distribution by Monday, which
should be used in place of the current AST166 tables.
http://crl.nmsu.edu/~mleisher/csets.html
Hovik, if you want to make these tables up and donate them, I would be glad to
distribute them. Otherwise, I'll just put them together from the info on your
web site. Of course you are welcome to donate these mapping tables directly
to the Unicode Consortium as well.
-----------------------------------------------------------------------------
Mark Leisher
Computing Research Lab Cinema, radio, television, magazines are a
New Mexico State University school of inattention: people look without
Box 30001, Dept. 3CRL seeing, listen without hearing.
Las Cruces, NM 88003 -- Robert Bresson
From: Pablo Saratxaga <[EMAIL PROTECTED]>
Sender: [EMAIL PROTECTED]
To: [EMAIL PROTECTED]
Subject: Re: [I18n] Some Armenian signs.
Date: Fri, 10 Nov 2000 16:20:57 +0100
Kaixo!
On Fri, Nov 10, 2000 at 08:09:52PM +0700, Ivan Pascal wrote:
>
> I found some dismatches in Armenian codes between keysym->char mapping
> and unicode tables.
> When I tried to find any information in Internet I found that ...
>
> http://crl.nmsu.edu/~mleisher/csets/AST166-8.TXT
> A2 (U00A7) Section Sign
that is the old armscii-8; it has been updated some time ago and now 0xa2
holds the "ew" ligature (which is to armenian script what the "&" ("et"
ligature) is to latin script)
> FE (U2019) Right Single Quotation Mark
> FF (U0027) Apostrofe
In fact 0xFE is "Armenian Apostrophe" (0x055A) (both 0x2019 and 0x0027
are absurd mappings at those positions: 0x2019 because the paired opening
quotation mark is missing, so why having only the right side ?
0x0027 because it is already at 0x27 ! ).
0xFF is undefined in armscii-8; I put there the "armenian section sign"
(0x00A7) which was at 0xA2 on previous version of armscii-8; so losless
conversion can be done.
> http://www.freenet.am/armscii/armcs-006.html
> (It seems to me that it must be more autoritative source but it incomplete.)
> A2 (U0587) armew (Armenian ligature "ew")
> FE (U055A) Armenian apostrofe
> FF - none
It is the right one.
> 'Armenian section sign' mentioned but without any codes.
I put it at the 0xFF
> http://www.linuxi18n.org/locales/armenian/ARMSCII-8.gz
> A2 (U0587) Armenian ligature "ew"
> FE (U02BC) Armenian apostrofe
Oops, typo of mine; should be:
FE (U055A) Armenian apostrofe
> So it is clear that Bruno use AST166-8.TXT as source for converter.
> Puzzle is - what is source for armscii-8.enc and Pablo's ARMSCII-8.gz?
The freenet page; but with typos of my own :-\
> And main question - what mapping is right ?!
The one of freenet.am page.
> (And what must be changed - lcUniConv/armscii_8.h or armscii-8.enc?)
Both ?
There is also another pending point:
Armscii-8 defines a char "Armenian eternity sign" wich looks a bit like
the indian wheel in the Indian flag (or like a daisy flower).
However, it isn't on unicode...
I've mapped it to the value 0x0530, as that one is on the Armenian section
of unicode and empty.
--
Ki �a vos v�ye b�n,
Pablo Saratxaga
http://www.srtxg.easynet.be/ PGP Key available, key ID: 0x8F0E4975
_______________________________________________
I18n mailing list
[EMAIL PROTECTED]
http://XFree86.Org/mailman/listinfo/i18n
