Modified dups to allow selecting files for removal

This commit is contained in:
KlzXS 2020-10-21 17:03:03 +00:00
parent d38bed58b2
commit 782df96080
No known key found for this signature in database
GPG Key ID: 6F2A29F874A14243
1 changed files with 31 additions and 2 deletions

View File

@ -6,10 +6,39 @@
#
# Dependencies: find md5sum sort uniq xargs
#
# Shell: POSIX compliant
# Note: bash compatible required for mktemp
#
# Shell: bash
# Authors: syssyphus, KlzXS
find . -size +0 -type f -printf "%s %p\n" | sort -rn | sed -n 'N; /^\([0-9]*\) .*\n\1.*$/p;$d;D' | awk '{printf("%s\0", substr($0, index($0, $2)))}' | xargs -0 md5sum | sort | uniq -w32 --all-repeated=separate
# If the size of a file has more that $size_digits digits the file will be misplaced
# 12 digits fit files up to 931GiB
EDITOR="${EDITOR:-vi}"
TMPDIR="${TMPDIR:-/tmp}"
size_digits=12
tmpfile=$(mktemp "$TMPDIR/.nnnXXXXXX")
# shellcheck disable=SC2016
find . -size +0 -type f -printf "%${size_digits}s %p\n" | sort -rn | uniq -w"${size_digits}" -D | tr '\n' '\0' | xargs -0 -n1 sh -c 'printf "%s %s\n" "$(md5sum $@)" "d$0"' | sort | { uniq -w32 --all-repeated=separate; echo; } | sed -nE '
h
s/^(.{32}).* d([0-9]*)$/md5sum: \1 size: \2 bytes/p
g
:loop
N
/.*\n$/!b loop
p' | sed -E 's/^.{32} (.*) d[0-9]*$/\1/' > "$tmpfile"
"$EDITOR" "$tmpfile"
cat "$tmpfile"
# shellcheck disable=SC2016
sed -e 's/md5sum.*//' "$tmpfile" | tr '\n' '\0' | xargs -0 sh -c 'rm -i $0 $@ < /dev/tty'
rm "$tmpfile"
printf "Press any key to exit"
read -r _