jcfr/git_list_largest_file_from_history.sh

## git_list_largest_file_from_history.sh
#!/bin/bash

set -eo pipefail

#
# This script will list the N largest files found in the history of a Git project.
#
# References:
# * https://docs.acquia.com/article/removing-large-files-git-without-losing-history
# * https://stackoverflow.com/questions/10622179/how-to-find-identify-large-files-commits-in-git-history
# * https://git-scm.com/book/en/v2/Git-Internals-Packfiles
#

PROG=$(basename $0)

#-------------------------------------------------------------------------------
err() { echo -e >&2 ERROR: $@\\n; }
die() { err $@; exit 1; }
help() {
  cat >&2 <<ENDHELP
Usage: $PROG N [options]

List the N largest files found in the history of a Git project.

Options:
  -h, --human-readable       print human readable sizes(e.g., 1K 234M 2G)
  -t, --table                pretty print results as a table instead of comma separated lines
ENDHELP
}

#-------------------------------------------------------------------------------
if [ ! -d .git ]; then
  err "Execute the script at the root of a git-versioned project"
fi

table=0
human_readable=0

while [[ $# != 0 ]]; do
    case $1 in
        --human-readable|-h)
            human_readable=1
            shift 1
            ;;
        --table|-t)
            table=1
            shift 1
            ;;
        -*)
            err Unknown option \"$1\"
            help
            exit 1
            ;;
        *)
            break
            ;;
    esac
done

N=$1
if [ "$N" == "" ]; then
  err Missing N option
  help
  exit 1
fi

format_size(){
  if [ $human_readable == 1 ]; then
    echo $1 | numfmt --to=iec-i --suffix=B --padding=12
  else
    echo $1
  fi
}

display_line(){
  if [ $table == 1 ]; then
    printf "%12s %12s %-40s %s\n" $@
  else
    printf "%s,%s,%s,%s\n" $@
  fi
}


all_objects=`git rev-list --all --objects`

display_line "size" "pack_size" "sha" "location"

git verify-pack -v .git/objects/pack/pack-*.idx | sort -k 3 -nr | head -n $N | while read line
do
  sha=$(echo $line | cut -f1 -d" ");
  size=$(format_size $(echo $line | cut -f3 -d" "));
  compressed_size=$(format_size $(echo $line | cut -f4 -d" "));
  sha_and_location=$(echo "${all_objects}" | grep $sha 2>&1)
  sha=$(echo ${sha_and_location} | sed "r/ +//" | cut -d" " -f1)
  location=$(echo ${sha_and_location} | sed "r/ +//" | cut -d" " -f2)
  display_line $size $compressed_size $sha $location
done
	#!/bin/bash

	set -eo pipefail

	#
	# This script will list the N largest files found in the history of a Git project.
	#
	# References:
	# * https://docs.acquia.com/article/removing-large-files-git-without-losing-history
	# * https://stackoverflow.com/questions/10622179/how-to-find-identify-large-files-commits-in-git-history
	# * https://git-scm.com/book/en/v2/Git-Internals-Packfiles
	#

	PROG=$(basename $0)

	#-------------------------------------------------------------------------------
	err() { echo -e >&2 ERROR: $@\\n; }
	die() { err $@; exit 1; }
	help() {
	cat >&2 <<ENDHELP
	Usage: $PROG N [options]

	List the N largest files found in the history of a Git project.

	Options:
	-h, --human-readable print human readable sizes(e.g., 1K 234M 2G)
	-t, --table pretty print results as a table instead of comma separated lines
	ENDHELP
	}

	#-------------------------------------------------------------------------------
	if [ ! -d .git ]; then
	err "Execute the script at the root of a git-versioned project"
	fi

	table=0
	human_readable=0

	while [[ $# != 0 ]]; do
	case $1 in
	--human-readable\|-h)
	human_readable=1
	shift 1
	;;
	--table\|-t)
	table=1
	shift 1
	;;
	-*)
	err Unknown option \"$1\"
	help
	exit 1
	;;
	*)
	break
	;;
	esac
	done

	N=$1
	if [ "$N" == "" ]; then
	err Missing N option
	help
	exit 1
	fi

	format_size(){
	if [ $human_readable == 1 ]; then
	echo $1 \| numfmt --to=iec-i --suffix=B --padding=12
	else
	echo $1
	fi
	}

	display_line(){
	if [ $table == 1 ]; then
	printf "%12s %12s %-40s %s\n" $@
	else
	printf "%s,%s,%s,%s\n" $@
	fi
	}


	all_objects=`git rev-list --all --objects`

	display_line "size" "pack_size" "sha" "location"

	git verify-pack -v .git/objects/pack/pack-*.idx \| sort -k 3 -nr \| head -n $N \| while read line
	do
	sha=$(echo $line \| cut -f1 -d" ");
	size=$(format_size $(echo $line \| cut -f3 -d" "));
	compressed_size=$(format_size $(echo $line \| cut -f4 -d" "));
	sha_and_location=$(echo "${all_objects}" \| grep $sha 2>&1)
	sha=$(echo ${sha_and_location} \| sed "r/ +//" \| cut -d" " -f1)
	location=$(echo ${sha_and_location} \| sed "r/ +//" \| cut -d" " -f2)
	display_line $size $compressed_size $sha $location
	done