procmail
[Top] [All Lists]

RE: html bouncer (was no subject)

2001-03-21 10:33:00

On your corrections, is there a missing ) or perhaps you type one extra?

Yes, and there was a missing }

I meant it to be like this:

(BTW you don't need

* ! ^majordomo(_at_)bus\(_dot_)ucf\(_dot_)edu

because majordomo is included in ^FROM_DAEMON)


:0
* ! ^FROM_DAEMON
{
         :0 H
         * cba(stf|fac)?@(atlantis\.)?bus\.ucf\.edu
         {
                 :0 B
                 * ()<HTML>
                 * ()</HTML>
                 {
                         :0 c
                         /tmp/$file_name.txt

                         :0
                         | /usr/local/bin/html_bouncer.pl $file_name.txt
                 }

        }
}


# I should know better than to forget to indent properly to make sure I don't
leave a stray { or } hanging


What I need is an exact matching of the To: addresses, on the first part,
i.e.,

cba(_at_)atlantis(_dot_)bus(_dot_)ucf(_dot_)edu
cba(_at_)bus(_dot_)ucf(_dot_)edu
^    ^
|    |- disregard hostnames, match bus.ucf.edu.
|
- exact match for each list (cba, cbafac, cbastf, etc)

So you want to match cba/cbafac/cbastf exactly,
and you want to match 'bus.ucf.edu' with or without further subdomains?

cba(stf|fac)?                           #matches "usernames" cba, or cbafac, or 
cbastf
@
(.*\.)?\<bus\.ucf\.edu\>                #matches (I think) bus.ucf.edu or any 
subdomain above
it

I'm still new to the use of \< and \> so their placement may be off.... well,
I'm pretty sure that \> is right, but I'm less sure about where to put \<

Put it all together,
        and instead of :0 H why not just use

:0
* ^(To|Cc):.*cba(stf|fac)?@(.*\.)?\<bus\.ucf\.edu\>

Or you could use ^TO_ or ^TO also

Content-type are too many... I tried and I could not make it work at all. If
someone has a solution for that approach... Any new feedback?

Too many content-types for HTMLs?

I just did a quick 'egrep -i "^content-type.*html|<html>"' on my inbox and found
that in every case I found a

^Content-Type: text/html

for each <HTML>

But if you search the headers for

        *   ^Mime-Version: 1.0
        * ! ^Content-Type:[     ]Text/Plain

*then* you could search the body for 'Content-Type: text/html' (rather than
searching every message body for <HTML> & </HTML>)

In your current method, someone could send you a text message that says

"Here, copy and paste this in to make a webpage

<html><head><title>hello world</title></head>
<body>
this is a test
</body>
</html>

"

and your recipe would bounce a text message that happened to have those HTML
tags in it.

Just some thoughts from a really basic procmail user

TjL

ps -- personally I prefer using $HOME/.tmp to /tmp but that's my own idiomatic
thing...


_______________________________________________
procmail mailing list
procmail(_at_)lists(_dot_)RWTH-Aachen(_dot_)DE
http://MailMan.RWTH-Aachen.DE/mailman/listinfo/procmail

<Prev in Thread] Current Thread [Next in Thread>