Greetings,
First let me apologize for posting this to the developer list when it
should probably be posted to the bug tracking system; for some reason I
don't seem to be able to post a message there (maybe Mantis is more
strict about this than Bugzilla?).
Anyway, I would like to point out this wiki note:
http://www.dbmail.org/dokuwiki/doku.php?id=unicode_postgresql_database
to whomever is working on bug 218:
http://www.dbmail.org/mantis/view.php?id=218
In response to the last note there, I would say that this issue *will*
persist, because basically dbmail is trying to submit arbitrary octet
sequences as values for columns of type "text", and Postgres will always
try to interpret the input according to the defined encoding for that
database. I have patches to fix this if there is interest. The patches
change the column type to "bytea", which tells Postgres to not try to
decode the text, but just store it.
Thanks,
Robert
