#!/bin/bash

# find compression ratio of source files
# pipe this through 'sort -n' to find files with a lot of redundancy

ZIP=gzip
#ZIP=bzip2
for FILE in `find -name '*.[ch]' | grep -v _darcs`; do
	SIZE=`ls -l $FILE | awk '{ print $5 }'`
	ZSIZE=`$ZIP $FILE -c | wc -c`
	SCORE=`dc -e "$SIZE 100 * $ZSIZE / n"`
	echo $SCORE $FILE
done