http://bugzilla.spamassassin.org/show_bug.cgi?id=3427
------- Additional Comments From [EMAIL PROTECTED] 2004-09-28 10:57 -------
-1 this patch disables a lot of code which was specifically put in to deal
with what browsers support.
I don't mind skipping the non-schemed URLs, and agree with the logic in the
comment there.
'http:www.spamassassin.org' and 'http:/www.spamassassin.org' work in browsers!
(both work in IE, the
first works in Safari (not the second one though), both work in Firefox, etc.)
in an href, both are links
in Outlook Web (off of Exchange), and then it's up to the browser to goto the
right place which as I
stated above, will work!
I really don't like the fact that the patch changes the redirector behavior.
We want the URI list to have
both raw + cooked data on it. The original:
if ($nuri =~ m{^https?://.+?(https?:/{0,2}.+)$}i) {
push(@uris, $1);
would allow for http://yahoo.com/?http:/example.com/ to be turned into:
http://yahoo.com/?http:/example.com/
http:/example.com/
http://example.com/
then the uri rules can catch any of that. The patched version wouldn't include
the http:/example.com/
one, which means we won't be able to catch that if we wanted to.
I'd be ok with a patch which just disables the "schemeless" http guessing code.
Everything else ought
to stay.
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.
