getmsg.cgi used a "\n^From " pattern to find the end of

one email message and the start of another.  this hueristic
	failed on lines like "\n^From Ushant to Scilly is 35
	leagues".

	the new pattern relies upon our From line ending in the
	four digits of the year.  which is also an hueristic that
	will from fail from time to time.  For example "\n^From
	2000 substract 1 to get 1999$".

PR:	docs/21762
This commit is contained in:
Jonathan M. Bresler 2000-10-29 16:29:45 +00:00
parent 293a1c98a2
commit aecc5899ee
Notes: svn2git 2020-12-08 03:00:23 +00:00
svn path=/www/; revision=8233

View file

@ -6,7 +6,7 @@
# by John Fieber
# February 26, 1998
#
# $FreeBSD: www/en/cgi/getmsg.cgi,v 1.23 1999/12/17 14:23:52 phantom Exp $
# $FreeBSD: www/en/cgi/getmsg.cgi,v 1.24 2000/01/05 15:47:44 phantom Exp $
#
require "./cgi-lib.pl";
@ -74,7 +74,7 @@ sub Fetch
# Unknown length, guess the end of the E-Mail
my($newline) = 0;
while(<DATA>) {
last if ($newline && /^From /);
last if ($newline && /^From .* \d{4}/);
if (/^$/) { $newline = 1 } else { $newline = 0; }
$message .= $_;
}