A bash script to parse my nginx access logs.

analytics.sh

#!/bin/bash

rm /home/james/logs/*

log_files=/home/james/logs/

# Copy log files to temporary directory

cp /var/log/nginx/access*.gz $log_files

for f in $log_files/*;
do
	gunzip -d "$f"
done

# Manually copy plain-text access files

cp /var/log/nginx/access.log $log_files
cp /var/log/nginx/access.log.1 $log_files

page_view_counter=`cat $log_files* | wc -l`
feed_counter=`cat $log_files* | grep "feed.xml" | wc -l`

# Show top five performing pages

echo "Top Five Pages"

awk '$7 !~ "/assets/*|robots.txt|favicon.ico|400|/2020/*|wp-*|vendor" {print $7}' $log_files/* | \
sort | \
uniq -c | \
sort -r | \
head -n 5

echo
echo "Top Five Posts"

awk '$7 ~ "/2020/*" {print $7}' $log_files/* | \
sort | \
uniq -c | \
sort -r | \
head -n 5

echo
echo "Status Codes"

awk '{print $9}' $log_files/* | \
sort | \
uniq -c | \
sort -r | \
head -n 5

echo
echo "Total Pageviews: $page_view_counter"
echo "Total Feed Hits: $feed_counter"