Date: Tue, 27 May 2003 08:10:18 -0500 (CDT) From: Adam Maloney <adamm@sihope.com> To: "Dave [Hawk-Systems]" <dave@hawk-systems.com> Cc: freebsd-isp@freebsd.org Subject: Re: shell/awk scripting help - parsing directories to gainuser/file information for commands Message-ID: <Pine.BSI.4.05L.10305270751531.16072-100000@unix1.sihope.com> In-Reply-To: <DBEIKNMKGOBGNDHAAKGNEEBFPEAB.dave@hawk-systems.com>
next in thread | previous in thread | raw e-mail | index | archive | help
You could pull the data for the FOR loop out of the apache config,
something like:
# Assuming your config is something like:
<VirtualHost 192.168.1.1>
DocumentRoot /www/user1
ServerName www.blah.com
CustomLog .../blah.com.access_log
...
</VirtualHost>
for DOMAIN in `grep ^ServerName httpd.conf | awk '{ print $2 }'`; do
# DocRoot is one line above ServerName
OUTPUT_DIR=`grep -B 1 $DOMAIN httpd.conf | grep ^DocumentRoot | awk '{ print $2 }'`
OUTPUT_DIR=`echo "$OUTPUT_DIR/stats"`
# CustomLog is 1 line after ServerName
ACCESS_LOG=`grep -A 1 ^CustomLog httpd.conf | awk '{ print $2 }'`
webalizer -o $OUTPUT_DIR $ACCESS_LOG
done
Obviously needs some tweaking, and there are TONS of other ways of doing
this. I have perl code somewhere that actually does this right - it
treats everything between the VirtualHost lines as a record, and can pull
all of the per-site config from it.
For most things like this in Awk, you need to know 3 things:
awk '{ print $3 }' will print field 3 (whitespace seperated)
awk '{ printf("Hello %s\n", $3 }' you can use C-style printf for formatting
echo "1X2X3X4" | awk -F X '{ print $2 }' field seperator is "X"
You can get more fancy:
Input is:
"print this line"
"not this line"
# Only print the line with the word "print" in it:
awk '/print/ { print $0 }'
$0 represents the entire line, so awk searches (/) for "print" and for
each matching line, prints it.
Get much more complicated than this and just "use Perl;" :)
Good luck.
> # ls /www/*/logs/*.access_log
>
> generates...
>
> /www/user1/.logs/user1domain1.com.access_log
> /www/user1/.logs/user1domain2.com.access_log
> /www/user2/.logs/user2domain1.com.access_log
> /www/user2/.logs/user2domain2.com.access_log
> /www/user3/.logs/user3domain1.com.access_log
> /www/user3/.logs/user3domain2.com.access_log
>
> what I am trying to script is something that does;
>
> <pseudo code>
> for i in /www/*/logs/*.access_log;
> ereg (user)(domain_name) from $i;
> do webalizer -n $domain_name -o /www/$user/stats/;
> done
> </pseudo code>
>
> ...as this would eliminate human error in maintaining a file which contains the
> appropriate lines to handle this via cron or something every night.
>
> This is one example, there are a slew of other similar applications that I would
> use this for. have played with awk as well, just can't wrap my head around this
> (not a shell scripting person by trade).
>
> Any guidance or insight would be appreciated.
>
> Dave
>
>
> _______________________________________________
> freebsd-isp@freebsd.org mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-isp
> To unsubscribe, send any mail to "freebsd-isp-unsubscribe@freebsd.org"
>
Adam Maloney
Systems Administrator
Sihope Communications
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSI.4.05L.10305270751531.16072-100000>