Developers of Sympa

[Chris Hastie <address@concealed>]

Hi

A good while ago I contributed a patch to provide plain text digests,
principally the PlainDigest.pm module. I've been looking at this again,
partly because I've been working on another project that needed a
similar approach, and partly because that other project is going to have
an impact on how the PlainDigest stuff works on my system - it's a pre
filter which will force messages to UTF-8, and PlainDigest.pm never
coped well with anything outside the iso-8859-1, windows-1252 group of
character sets.

So I'm thinking I should revisit PlainDigest.pm and look carefully at
character set handling. The idea would be to normalise everything to
UTF-8. Is this a worthwhile exercise? If I go forward with it what else
do I need to do to ensure the actual digests are sent with appropriate
headers for character set and Content-Transfer-Encoding? Would a small
change to the template suffice? Also, what CTE should be used? 8bit
seems to be the default, but I notice some of my digests seem to arrive
with base64. I'd actually prefer qp (7bit safe, but largely legible, in
western european languages at least).

Chris