Searches for (and remove) large files from git repository

ls-large-files.sh

#!/bin/bash

#Lists all large files in git repostory for the given branch.
#Needs to be run from repository subdirecory,
#Assumes git and perl are availblse
#
#http://blog.jessitron.com/2013/08/finding-and-removing-large-files-in-git.html
# and http://git-scm.com/book/ch6-4.html for repo history rewrite exlanation
#
#To get argregated statistics on output use:
#$ cat large_files_dev.txt | awk '{t+=$1} END {print t/1024/1024 "Mb"}'
#
#To actually re-write the history use: (see https://help.github.com/articles/remove-sensitive-data)
#$ git filter-branch -d ~/tmp --index-filter \
#   'git rm --cached --ignore-unmatch \
#   `cat /Users/alex/Documents/_NFLabs/peloton-master/large_files_all_branches.txt | cut -d " " -f 2` ' \
#    --prune-empty --tag-name-filter cat -- --all
#
#Pro-tip: mount ~/tmp to tmpfs in RAM
#
#Make sure we filtered all refs
# for file in `cat large_files_master.txt | cut -d " " -f 2`; do git log --name-only --follow --all -- $file; done
#
#To cleanup the repository use:
# either clean clone from it
# or
#$ rm -rf .git/refs/original/
#$ git reflog expire --expire=now --all
#$ git gc --prune=now
#$ git gc --aggressive --prune=now


usage(){
	echo "Usage: $0 <brach-name>"
	echo "Lists all largre files for the given brach-name"
	echo "Pro-tip: use --all as a branch name"
	exit 1
}

# invoke  usage
# call usage() function if branch name not supplied
[[ $# -eq 0 ]] && usage 

#work in masted by default
BRANCH="master"
if [ ! -z "$1" ]
then
    BRANCH=$1
fi
OUTPUT_FILE="large_files_${BRANCH}.txt"


echo "Listing large files in brach \"${BRANCH}\" ..."
git rev-list ${BRANCH} | while read rev; do git ls-tree -lr $rev | cut -c54- | grep -v '^ '; done | sort -u | perl -e '
  while (<>) {
    chomp;
    @stuff=split("\t");
    $sums{$stuff[1]} += $stuff[0];
  }
  print "$sums{$_} $_\n" for (keys %sums);
' | sort -rn >> "${OUTPUT_FILE}"

echo "Done. See ${OUTPUT_FILE} for results."