prepare IIS W3C log files for use by siege - concatenate cs-uri-stem and the cs-uri-query

PrepareForSiege.sh

#!/bin/bash

# make sure you always put $f in double quotes to avoid any nasty surprises i.e. "$f"
for f in $@
do

if [ ! -f $f ];
then
    echo "File ($f) not found!"
    exit 0
fi

  echo "Processing $f..."

# dos 2 unix, remove comments and empty lines
sed -i -e 's/.$//' -e 's/#.*//' -e '/^$/ d' "$f"

echo "  Extracting URI and Query String..."
cut -d' ' -f6-8 "$f" > method_uri

#strip out load balancer check
grep -v "/load/balancer/check.HTML" method_uri > urls
rm method_uri

#i only want GETs
grep "GET" urls > urls.txt
rm urls

echo "  Adding hostname..."
sed -i -e 's/GET /http:\/\/target.domain.com/' urls.txt

echo "  Fixing up uri stem..."
# clear empty cs-uri-query fields
sed -i -e '1,/^ -/s/ -$//' urls.txt

# put '?' between the cs-uri-stem and the cs-uri-query
sed -i -e 's/ /?/' urls.txt

echo "  Setting dates in cs-uri-query to `date +%Y-%m-%d`"

# update dates in cs-uri to today
sed -i -e 's/sDate=[0-9]\{4\}\-[0-9]\{2\}\-[0-9]\{2\}/sDate='`date +%Y-%m-%d`'/g' urls.txt

echo "  Splitting load into 100,000 lines..."
split -l 100000 urls.txt SiegeUrls.
rm urls.txt

echo " Ready for Siege!"

done