>[...]
It turns out Eudora's filtering mechanism does see header content
beyond that first line break. I couldn't find any information on how to
match a line break with a "contains" clause, so I used "matches regexp (case
insensitive)" as follows:
" defunct domain:[^[:alpha:]]{1,3}jdoe@"
(The quotes are to delimit the content I used for readability here; they're
not part of the actual match string.) I wasn't sure if Eudora used the
line break as defined by the RFCs (CR-LF), or only one of those, so the
"1" represents the space character that will always exist between the end
of "domain:" in the case where the line is not folded (to use the RFC
terminology) and the user name, and the "3" includes the space plus the
RFC-defined line break or a single-character line break (since the second
value represents a maximum, not an absolute).
...Now to tackle this in Thunderbird.
Cheers,
Mike
P.S. The RFC in question is 5322, which obsoletes 2822, which obsoletes 822.