Script parse Squid logs

[Boaz60]

Hi, this is my first post as I'm just starting to learn about grep and all it's powers since my company purchased Linux based software for our proxy server.

Anyways, I'm trying to write a script that I can use to parse the squid logs.

The logs look like this:

2007.12.30 4:24:26 - xx.xx.xx.xx http://url.url.com/
2008.1.3 8:42:15 - xx.xx.xx.xx http://url.url.com/

It would be great if I could have a way to enter a cetain date and ip and parse that out into a file.

Right now I only know how to get the lines matching an IP, but I'd like to match an IP and Date. The time is irrelevant to me.

grep 'xx\.xx\.xx\.xx' file >> todaysdate.log

Thats what I'm doing now. How can I improve on this or even have a script that I can update the IP and Date I want and have it create a new log file naming it the current date.

Thanks in advance for any help.

[Franklin52]

If awk is allowed:

#!/bin/sh

awk -v date="$1" -v ip="$2" '
$1 == date && $2 == ip {print}
' logfile > todaysdate.log

Usage: <scriptname> <date> <IP>

Regards

[Boaz60]

Hey fantastic! Thanks a bunch, it'll be a big help.

One thing I had hoped, is there a way to add into that, instead of always naming the output the same thing, to have it auto create the filename to the days actual current date?

[anomie]

[fugu ~]$ outfile=some-file-$(date +%F)
[fugu ~]$ echo ${outfile}
some-file-2008-01-06

You get the idea. Initialize 'outfile' in your script, and then redirect to ${outfile}.