Remove sensitive information from PDF downloaded from JSTOR

jstor.sh

#!/bin/bash
# Usage: ./jstor.sh <input PDF> <output PDF>

INPUT_FILE="$1"
OUTPUT_FILE="$2"

if [ $# -ne 2 ]; then
    echo "Usage: $0 <output_file> <input_file>"
    exit 1
fi

echo "[1/3] Decompressing streams..."
# -d: Decompress content streams (makes text readable for Step 2)
# -i: Compress image streams (keeps binary images compact)
mutool clean -di "$INPUT_FILE" decomp.pdf

echo "[2/3] Removing objects containing target string..."
# -0777 : Read the entire file into memory at once (slurp mode).
# -pe   : Auto-print output after executing the code.
# s/...//gs : Search and replace globally (g) treating string as single line (s).
# \Q...\E : Automatically escape special characters in the target string.
perl -0777 -pe 's/\d+\s+\d+\s+obj.*?\Q(All use subject to\E.*?endobj//gs' decomp.pdf > temp.pdf

echo "[3/3] Merging, Repairing, and removing 1st page..."
# -O garbage : Scans the file and rebuilds the XREF table, ignoring the missing objects from Step 2.
# -O compress: Re-compresses the streams.
# -O sanitize: Reads the Page's content streams to clean up graphics commands.
# 2-N        : Selects page 2 through the end (2-N).
mutool merge -O garbage,compress,sanitize -o "$OUTPUT_FILE" temp.pdf 2-N

rm decomp.pdf temp.pdf
echo "Done. Saved to $OUTPUT_FILE"