>>>>> "Michael" == Michael Monnerie <michael.monne...@is.it-management.at> >>>>> writes:
Michael> I would have thought convert_from, possibly combined with Michael> convert, would do it. And this should really do the magic, Michael> but I'm on 8.1 today, will test on 8.3 soon. Indeed, convert_from() can be used to coerce the messageblk to text, but only if dbmail knows how the blk is encoded. This query in psql does give me the header of a message sent to this group: ,---- | select physmessage_id from dbmail_messageblks where is_header = 1 | and convert_from(messageblk,'SQL_ASCII') ~ 'dbmail-dev' limit 1; `---- It is, however, as one might expect quite slow. If one blindly uses 'SQL_ASCII', octets other than those from in /[\n\t -~]/ aka /\n\t\x20-\x7E/ seem to be output as C-style backslash escapes. That may or may not be useful for imap search. There is also the question of whether imap search is supposed to search the over-the-wire format of the mail or the as-viewed-in-an-MUA format. Ie, how imap search and mime are supposed to interact. This query would get all of the mail which have 'dmail-dev' in their bodies, ordered by physmessage_id. But it is wildly disk intensive, as psql has to read through every messageblk in dbmail_messageblks. ,---- | select convert_from(messageblk,'SQL_ASCII') from dbmail_messageblks | where physmessage_id in ( select physmessage_id | from dbmail_messageblks | where is_header=0 and | convert_from(messageblk,'SQL_ASCII') ~ 'dbmail-dev' ) | order by physmessage_id, is_header desc; `---- Generating an index of the output of convert_from(messageblk,'SQL_ASCII') would be painful at best. -JimC -- James Cloos <cl...@jhcloos.com> OpenPGP: 1024D/ED7DAEA6 _______________________________________________ DBmail mailing list DBmail@dbmail.org http://mailman.fastxs.nl/cgi-bin/mailman/listinfo/dbmail
