Re: archiving UTF-8 encoded messages

[Earl Hood <ehood@hydra.acs.uci.edu>]

On December 19, 2000 at 10:04, Eric Marsden wrote:

> I see that this subject has been touched on in the list archives. It
> would be nice if MHonarc were able correctly to archive UTF-8 encoded
> messages. Currently a false http-equiv "charset=ISO-8859-1" is used.

MHonArc does not set http-equiv be default.  If the meta tag
is showing up, then it is in whatever resource file you are using.

> Assuming that common browsers can handle this, would it be difficult
> (and sufficient?) to use a "charset=UTF-8" http-equiv instead? Perhaps
> by augmenting %readmail::MIMECharSetConverters?

Unicode support has not been integrated into MHonArc.  Charset
converters need to be created and then some mechanism for a filter
to specify what the charset is for the entire message page written.
Since that is nothing prohibiting a raw message from have parts
with different charsets, then a routine is needed that convert
from various charsets into Unicode/UTF-8.  I have not checked
in awhile if there are any pure Perl implementations of character
set translation modules.  I also have not checked to see what
the level of UTF-8 support there is in Perl 5.6.

Any contributions are welcome,

--ewh