From: Terry Howerton (terry@scouter.com)
Date: Tue Dec 07 1999 - 22:12:35 CST
Below is a message from several months ago about converting Listserv(tm)
style mailboxes to UNIX Mbox format. I am trying to run the scrip on a large
mailbox with about 1,500 messages, but to no avail.
The command line on Win32 that I am using is:
ls2mail.pl inbox.txt outbox.mbox
That cycles through all the messages, but does not write anything to the
outbox file and does not modify the inbox file.
I have two questions:
1) Do current versions of Hypermail work on the listserv style mail?
2) Any idea what might be wrong with the script below, or is there a better
way to convert this mail?
TERRY HOWERTON
>Below please find a Perl script I helped to write for archiving the
>ADV-HTML mailing list with Hypermail. I originally wrote the script for
>Patrick Douglas Crispen <crispen@netsquirrel.com>, but he said it would
>be okay to include the script with Hypermail (if so desired).
>
>This script appears to function similarly to the n2folder Perl script
>that Peter Murray <pem@po.cwru.edu> recently sent, and they may even
>operate on the same type of listserv archive.
>
>One thing unique feature that ls2mail may have is that it generates a
>"Message-ID" for each message when converting the listserv archive to an
>mbox archive. While this won't help when linking message threads, it
>did help Hypermail (possibly required by the old 1.x versions?) to
>create its output.
>
>Dave
------- Forwarded Message
#!/usr/local/bin/perl
#
# ls2mail -- converts listserv formated digests to UNIX "mail" format
#
# Usage:
# ./ls2mail < infile > outfile
# cat infile | ./ls2mail > outfile
#
# Written by David Kilzer <ddkilzer@ti.com>
# Tue, Mar 24, 1998
#
use strict;
my $first_time = 1; # marks first time through script
my $line; # stores one input line
my $header; # stores mail message header lines
my $from_address; # stores "From:" address
my @date; # stores "Date:" information
my $message_id; # stores new message ID info
while ($line = <>) # use '<>' operator so we act like a UNIX filter
{
chomp ($line); # remove extra newlines
if ($line !~ m/^={73}$/) # Separator line?
{
print $line, "\n"; # Not separator, just print
}
else # Found separator line, process
{
$header = ""; # clear variable
$from_address = ""; # clear variable
@date = (); # clear variable
$message_id = ""; # clear variable
# Read in email header lines
while ($line = <>)
{
last if ($line =~ m/^\s*$/); # message header ends with "blank" line
$header .= $line; # add $line to $header
}
$header =~ s/^Sender:\s/To: /mi; # change "Sender:" to "To:"
$header =~ s/^([^\s:]+:)\s+/$1 /mg; # remove extra space from all lines
$header =~ s/\n\s+/\n /mg; # continued lines used 8 spaces
# Find "From" address to use
if ($header =~ m/^Reply-To:\s.*<([^>]+)>/mi)
{
$from_address = $1;
}
elsif ($header =~ m/^Reply-To:\s.*\n\s.*<([^>]+)>/mi)
{
$from_address = $1;
}
$header =~ m/^Date:\s(.*)$/mi; # find "Date" header
@date = split (' ', $1); # split date into an array
$date[0] =~ tr/,//d; # remove commas from first date element
$date[1] = " " . $date[1] if (length($date[1]) == 1);
# add space to single days
$message_id = uc (join ('.', @date, $from_address)); # create message ID
$message_id =~ tr/[A-Z][0-9].@//cd; # remove bad characters
# Print new UNIX mail header
print "\n" if (! $first_time);
$first_time &&= 0;
print "From $from_address $date[0] $date[2] $date[1] $date[4]
$date[3]\n";
print "Message-Id: <$message_id>\n";
print $header, "\n";
}
}
exit 0;
__END__
------- End of Forwarded Message