Phil Pennock <exim-users@???> (Di 14 Jul 2009 04:55:19 CEST):
> On 2009-07-14 at 00:35 +0200, Heiko Schlittermann wrote:
> > I'm still at my version - instead of cutting away the tail, I'm
> > selecting the head of the logical header line:
> >
> > ${lc:${sg {$message_headers_raw}{\N(?m)(^\S+(?=\s*):)?.*?\n\N}{\$1}}}
>
> The \S+(?=\s*): part doesn't do what I think you think it does.
>
> (?=foo) is a zero-width positive lookahead assertion. It matches if and
> only if followed by foo, but does *not* advance the "current position"
> past foo.
>
> So X(?=\s*): will match if, after matching X, it can match zero or more
> spaces and then, immediately after X, match a colon. In the degenerate
> case of zero spaces, this works. But it won't match when there is
> space.
>
> ${lc:${sg {$message_headers_raw}{\N(?m)(?:(^\S+)\s*(:))?.*?\n\N}{\$1\$2}}}
>
> I suggest taking a mail header for your -bem test file and inserting
> some whitespace for testing purposes.
ohohoho, I just wanted to show you that it does what I thing it should
do, using my mail file (according to your suggestion yesterday) - and -
voila - it doesn't do what I think it should do. I'm missing the
"To : fred" header :-( and I don't know why I didn't see it yesterday.
So, exploring regex goes on ;)
Best regards from Dresden/Germany
Heiko Schlittermann
--
SCHLITTERMANN.de ---------------------------- internet & unix support -
Heiko Schlittermann HS12-RIPE -----------------------------------------
gnupg encrypted messages are welcome - key ID: 48D0359B ---------------
gnupg fingerprint: 3061 CFBF 2D88 F034 E8D2 7E92 EE4E AC98 48D0 359B -
