https://issues.apache.org/SpamAssassin/show_bug.cgi?id=6759
--- Comment #4 from Kevin A. McGrail <[email protected]> 2012-02-20 14:27:17 UTC --- (In reply to comment #3) > As far as the addition of the semicolon to the end of the regex goes, why do > we > need a real-world example if such is perfectly legal given the ABNF syntax in > RFC 5321? If the "id" and "for" clauses are omitted, the semicolon will > indeed > trail the "with" clause's parameter. However, as a semicolon followed by a > date is always required per the syntax (cf. RFC 5322), the "$" to terminate > the > string should NEVER occur in a properly formatted message. Even if we're > testing a wrapped header without unwrapping it, the semicolon will appear > BEFORE the newline+tab wrapping the date on the next line. I'd *prefer* real-world but will accept synthesized. Overall though, if you add failure and success cases it will help avoid issues where things get forgotten because according to Alex, this bug was caused by an update that overrode an older bug. Test cases would have solved that issue and they would prove the regexs work as intended both for success and failure. Regards, KAM -- Configure bugmail: https://issues.apache.org/SpamAssassin/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are the assignee for the bug.
