split-pdf.sh

#!/usr/bin/env bash
# Copyright 2018 mrl5
# Distributed under the terms of the GNU General Public License v3

# Creates new pdf from part of another pdf
#
# $1: output file
# $2: input file
# $3: first range
# $4: second range
# $n-2: nth range
#
# Example: 'sh split-pdf.sh /path/to/output.pdf /path/to/input.pdf 1-1 6-9 18-21'
#  creates output.pdf from input.pdf pages: 1, 6, 7, 8, 9, 18, 19, 20, 21

# checks if enough arguments were provided
if [ $# -lt 3 ]; then
        echo "Expected minimum 3 arguments. Please specify output file, input files and ranges."
        exit 1
else
	TMPNAME="temp-pdf-snapshot"
	TMPDIR="/tmp/temp-pdf-snapshots-$(tr -dc A-Za-z0-9_ < /dev/urandom | head -c 16 | xargs)"
	# stores directory of the script
	# source: https://stackoverflow.com/a/246128
	SCRIPT_DIR="$( cd "$( dirname "$0" )" >/dev/null && pwd -P)"
	# get directory of /path/to/outputfile.pdf file
	OUTPUTDIR=${2%/*}
	OUTPUT_FILE="$1"
	INPUT_FILE="$2"
	PAGES="${@:3}"
fi

if [ $OUTPUTDIR == $OUTPUT_FILE ]
then
    OUTPUTDIR="."
fi

# split input.pdf
split() {
	mkdir $TMPDIR
	INDEX=1
	# start loop from 3rd argument
	for RANGE in $PAGES
	do
	    START=$(echo $RANGE | cut -d'-' -f1)
	    END=$(echo $RANGE | cut -d'-' -f2)
	
	    # create new pdf from range $START-$END
	    #'$(printf "%04g" $INDEX)' = padding $INDEX with leading zeros
	    gs -sDEVICE=pdfwrite -dNOPAUSE -dBATCH -dSAFER -dFirstPage=$START -dLastPage=$END -sOutputFile=$TMPDIR/$TMPNAME-$(printf "%04g" $INDEX).pdf "$INPUT_FILE"
	
	    ((INDEX++))
	done
}

# merge all 'temp-pdf-snapshot' files into one output.pdf
merge() {
	sh $SCRIPT_DIR/merge-pdf.sh $OUTPUT_FILE $TMPDIR/$TMPNAME*.pdf
}

# main()
split
if [ $? -eq 0 ]; then merge; fi
rm -rf $TMPDIR