On Tue, Dec 22, 2015 at 05:27:01PM +0100, sb wrote:
> Let test.eml be a complete e-mail, and
> let test-body.eml be the body of test.eml only.
>
> The following matches any e-mail body with a single link
> surrounded by spaces and newlines:
Postfix does not apply regular expressions to the entire body of a
message. As documented matching happens one line at a time.
> The difference between patterns is dictated by postfix's
> adaptation of pcre and its defaults: despite the specification,
> postfix is ignoring \A, \z and /A.
No. It matches one line at a time, so "\A" == "^" and "\z" == "$".
> postfix-3.0.3/README_FILES/PCRE_README:
>
> > When Postfix searches a pcre: or regexp: lookup table,
> > each pattern is applied to the entire input string.
The input string for body_checks is a single message body line.
--
Viktor.
