RE: Characters changing (was Collation TR)

From: stephen_holmes@lionbridge.com
Date: Tue Jan 12 1999 - 03:21:29 EST


It's not the cc:Mail client software that's the problem, but the server-side
apps. When you install the server software and create the initial post office,
the installation requests a preferred codepage for managing this post office.
You may want to take a look at this. You also can't change the post office code
page without recreating the post office, so you'd probably want to export your
mailing lists and reimport them into the new schema.

Although I'm not sure whether it affects Thai, it certainly affects ANSI Latin1
(i.e., cp850 at the server, cp1252 at the client).

Hope this helps.

Cheers
Steve.

-----Original Message-----
From: Mark Davis <marked@best.com>
Sent: 11 January 1999 09:29
To: Unicode List <unicode@unicode.org>
Cc: Unicode List <unicode@unicode.org>
Subject: Re: Characters changing (was Collation TR)

You will have to get fairly recent versions of the software. You'll have to
upgrade your Netscape to 4.X--I don't know about ccMail. Older software will
misinterpret the character codes.

Also, I have noticed that Internet Explorer (Windows) sometimes stops reading
http://www.macchiato.com/mark/unicode/UTF8List.html around Armenian. Hitting
the refresh button loads a few more lines (sometimes), but it doesn't go much
farther.

Any MS folk know what is going on?

Mark

Chris White wrote:

>
>
> Help! I am confused about when various pieces of software under
> various operating environments may quietly transform characters.
>
> To take bit one example:-
> On
> http://www.macchiato.com/mark/unicode/UTF8List.html
> the UTF-8 string for the Thai character cho ching is happily rendered
> by my browser (Netscape Navigator 3.0) as:
> a-grave cedilla per_mille_sign
> (and it comes out on the printer like that as well)
> which by my reckoning is the correct UTF-8 string for U+0E09.
>
> When I read the email from Mark Davis 8/1/99 14:19, that same string
> is rendered by my email (Lotus cc:Mail 6.3) as:
> a-grave comma percent_sign
> which is a mal-formed UTF-8 string anyway.
>
> Both the Netscape and the email are running under Windows 3.11.
>
> I am endeavouring to assess how much trouble this sort of quiet
> character transformation (or bit pattern transformation if you like)
> is going to give us in the BL as we move towards keeping our huge
> quantity of bibliographic data in Unicode.
>
>
> Chris White
> Systems Analyst
> The British Library
>
>

--
business: medavis2@us.ibm.com, mark@unicode.org
personal: mark@macchiato.com, http://www.macchiato.com
--



This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:20:44 EDT