dups.bash

#!/usr/bin/env bash

##
# Duplicate detector: a general-purpose line-oriented key tracker.
#
# The input lines use this format:
#
#     <key> <note>
#
# The output lines are the duplicates use this format:
#
#     <key> <note of first hit> <note of current hit>
#
# The key is any identifier you want to compare, such as a checksum.
# The key must be a command line token, i.e. no spaces, no quotes, etc.
#
# The note is anything you want in the printout, such as a file name.
# The note is only for printout; the note is not used for comparison.
#
# The output prints the note from the first matching line first,
# then the note from the second matching linee.
#
# Example file demo.txt:
#
#     A Alice
#     B Bob
#     C Carol
#     A Anna
#
# Example command:
#
#     $ cat demo.txt | dups.bash
#     A Alice Anna
#
# To detect duplicate files, we can use a checksum:
#
#     find . -type f -exec sha512sum '{}' \; | dups.bash
#
# This script requires the bash shell.
# If you use zsh instead of bash,
# see our repo for a zsh version.
#
# Contact: Joel Parker Henderson (joel@joelparkerhenderson.com)
# License: GPL
# Updated: 2015-01-25
##
set -euf -o pipefail

unset track
declare -A track
while read -r key val; do
 if [[ ${track[$key]} ]]; then
   printf "%s %s %s\n" "$key" "${track[$key]}" "$val"
 else
   track[$key]="$val"
 fi
done