Max Caines wrote:
> I am trying to construct a system filter expression to match a value in
> dollars, to look like this:
>
> if $h_Subject: matches "\\$[0-9]+"
>
> I have tried it with numbers of backslashes before the '$' ranging from
> zero to 4, and (in desperation) making it a character class (e.g.
> "[\\$][0-9]+"). Either it won't parse; or it does, but doesn't match a
> header like 'Make $150 now'. I can't work out from the documentation what
> the expression _should_ look like. Can anyone tell me?
This is really horrible. Here's how it works.
1. Inside "..." \ needs to be backslashed to inhibit escape interpretation.
2. When string expansion may happen, $ must be backslashed to stop it,
and so must \ (because unbackslashed \ is used for inhibiting expansion).
3. In regexps, $ needs to be backslashed to stop it meaning EOL.
So: We want to match $. By (3) we need the regexp \$. By (2)
we need that to look like \\\$ (that's a backslashed \ plus a
backslashed $). By (1) all those \s need to be doubled, giving
\\\\\\$. Ouch.
On the other hand, note that
| If the text contains no white space then it can be typed
| verbatim. However, if it is part of a condition, it must
| also be free of round brackets [...]
So you can avoid both 1 and 2 by saying
if $h_Subject: matches \$[0-9]+
but this won't work if you need spaces in your matching string.
--
Gareth McCaughan Dept. of Pure Mathematics & Mathematical Statistics,
gjm11@??? Cambridge University, England.
--
*** Exim information can be found at
http://www.exim.org/ ***
