PDF concat with section rewrites

README.md

Concat PDFs

This script concatenates PDFs and preseverses their sections (like pdftk normally does)

Additionally it writes the current pdf's title (or file name if it doesn't have a title) as a section on the start page of the pdf in the merged pdf.

All other sections are then moved down a level to preserve the section hierarchy within the document

concat_pdf.bash

#!/bin/bash
PDFS=$1
TMP="/tmp/pdfcat"
rm -r "$TMP"
mkdir "$TMP"

for i in $(ls -v "$1"*.pdf); do
        echo $i
        pdftk $i data_dump output "$TMP/$i.txt"
        TITLE=$(grep "InfoKey: Title" "$TMP/$i.txt" -A 1 | grep InfoValue | cut -d " " -f "2-")
        TITLE=$(echo $TITLE | cut -d "-" -f "2-")
        TITLE=${TITLE:=$i}
        TITLE=${TITLE:="No Title"}
        FIRST_BOOKMARK=$(grep BookmarkBegin "$TMP/$i.txt" -n | head -n 1 | cut -d: -f1)
        grep BookmarkBegin "$TMP/$i.txt" -n | head -n 1 | cut -d: -f1
        head -n $(expr $FIRST_BOOKMARK - 1) "$TMP/$i.txt" >> $(echo "$TMP/""$i""2.txt")
        head -n $(expr $FIRST_BOOKMARK - 1) "$TMP/$i.txt" 
        echo "BookmarkBegin" >> $(echo "$TMP/""$i""2.txt")
        echo "BookmarkTitle: $TITLE" >> $(echo "$TMP/""$i""2.txt")
        echo "BookmarkLevel: 1" >> $(echo "$TMP/""$i""2.txt")
        echo "BookmarkPageNumber: 1" >> $(echo "$TMP/""$i""2.txt")
        # tail -n +22 "$TMP/$i.txt" |  >> "$TMP/$($i)2.txt"
        tail -n +$FIRST_BOOKMARK "$TMP/$i.txt" | awk '{if ( $1 == "BookmarkLevel:" ) print $1 " " $2 + 1; else print $0 }' >> $(echo "$TMP/""$i""2.txt")
        pdftk $i update_info $(echo "$TMP/""$i""2.txt") output "$TMP/$i"
done
pdftk $(ls -v "$TMP"/*.pdf) cat output $2