On Fri, 17 Nov 2006, Vyacheslav Frolov wrote:
The few months is OK for me. If you will need volunteers for
testing that tool please let me know.
OK. It'll certainly be generally announced once written.
If you (or anyone else) feel ambitious and want to try writing one on
their own, the key is to use c-client to read the source mailbox but *NOT*
write the destination mix mailbox.
Instead, once the source mailbox is opened, do the following steps:
get the current time, and remember it as the modseq
create the destination directory
create .mixmeta in the destination directory
flock() the .mixmeta file LOCK_EX
write modseq, uidvalidity, uidlast, new message data file
number (use modseq for this), and keywords list
create .mixindex and .mixstatus
decide upon a first data file number; anything less than the
modseq is good enough. Since today's date is higher
than 0x40000000, choose something like 0x10000000.
create that data file
for each message
write the appropriate entry into the .mixindex and
.mixstatus files, then copy the message to the data
file
if data file size has exceeded the threshhold (e.g., 1MB)
start a new data file with a higher number than the
previous one. If you monotonically increase you
shouldn't bump up against the limit...
end
create new message data file using modseq as its number
close everything
The imap-2006?/docs/mixfmt.txt file should explain what the necessary
formats are. Remember that the fields are fixed format!
-- Mark --
http://staff.washington.edu/mrc
Science does not emerge from voting, party politics, or public debate.
Si vis pacem, para bellum.
_______________________________________________
Imap-uw mailing list
[email protected]
https://mailman1.u.washington.edu/mailman/listinfo/imap-uw
