On Jul 7, 2007, at 3:55 PM, Jim Syler wrote:
On Jul 7, 2007, at 12:37 AM, LuKreme wrote:
On 6-Jul-2007, at 16:23, Jim Syler wrote:
Now, if you can do womething like
s/<br><br>--<br>/<p style="color: gray; font-size:80%">/ with
confidence (that is, it always stats with two <br>'s and then -- and
then a <br>
and it ALWAYS ends with a date stamp you can anchor the </p> against,
then sed in the procmail should be fine:
:0
* tests
* go here
* to ensure right message
Speaking of tests, I'm having a problem there, too. The test
* ^Subject: [Genius/Idiot.*
is not matching
Subject: [Genius/Idiot—Current Thoughts] getting
or anything similar. Any ideas why?
{
:0fwB
| sed 's/<br><br>--<br>/<p style="color: gray; font-size:80%">/'
:0fwB
| sed 's/\d:\d\d:\d\d .M/&</p>/'
}
Why the single quotes around the sed commands? I don't see anything
about that in the sed manpage.
The first part works great. The second plays Hob somehow; I'm not
entirely sure what it's doing, but I end up with a whole bunch of
header lines in the body. I thought 'B' meant that the header wouldn't
be evaluated?
Alright, the reason this isn't working is that sed was interpreting '/'
as a delimiter and 'p' as a command in '</p>'. How do I escape '/'?
With '\'? I can't tell from the sed manpage. Also, \d isn't listed as a
synonym in grep(1). Will sed recognize it?
--
Ron Paul President 2008!
It is time for President Paul!
The only candidate who defends liberty and the
Constitution!
http://ronpaul2008.com
Visit Jim's Web Portal: http://homepage.mac.com/calion/
PGP.sig
Description: This is a digitally signed message part
____________________________________________________________
procmail mailing list Procmail homepage: http://www.procmail.org/
procmail(_at_)lists(_dot_)RWTH-Aachen(_dot_)DE
http://MailMan.RWTH-Aachen.DE/mailman/listinfo/procmail