On 9/27/2013 11:41 PM, John Hardin wrote:
On Fri, 27 Sep 2013, Kevin A. McGrail wrote:
or should I be using this:
header __KAM_CREDIT8 Subject =~ /Your
(\xd0\xa1|C)ompl(\d1\d96|i)mentary/i
I had the best results using this format:
header __KAM_CREDIT8 Subject =~ /Your
(?:[\xd0][\xa1]|C)ompl(?:[\xd1][\x96]|i)mentary/i
I found that without making each of the individual bytes a separate
character class, I'd get unpredictable UTF-8 warnings during lint.
Thanks. And you had the best luck with the bytecode because otherwise
it wasn't clear when something was 2 bytes which made things uncertain
as well I'm guessing?
Regards,
KAM
