Package: spamassassin Version: 3.0.4-2 Severity: minor Tags: upstream Yo!
Occasionally, I see in my log: == Parsing of undecoded UTF-8 will give garbage when decoding entities at /usr/share/perl5/Mail/SpamAssassin/HTML.pm line 182, <GEN5218> line 105. == (1) this is probably some minor bug in HTML handling. AFAICT spamassassin runs in unicode locale (LC_CTYPE=en_US.UTF-8 is the default, so I guess it applies to daemons in /etc/init.d as well.) (2) wishlist: Could this kind of apparently malformed UTF-8 in mail be used for Spamassassin hitpoints? (No, I didn't track it down to see if it's really malformed mail.) thanks & greetings -- vbi -- Pzat!
pgpyCWnZfOnBf.pgp
Description: PGP signature
