gmarks2

#!/bin/bash

# Potetialy useful for extracting the titles:
# add something like this to the script
# pdfgrep -n '' notes.pdf | grep '^[0-9]*:[ \t]*[0-9]\.' | tr -s " " > tmp


# TODO: maybe add pdfjoin, which also combines the indexes

# Example of .otl format:
#<Level> <PageNr> <Title>
#1	1	1. Basic Notions and Notation
#1	2	2. Random Variables
#1	4	3. Expectation–Integrals
#1	12	4. Lp spaces
#1	14	5. Convergence of measurable functions
#1	17	6. Independence of events and independence of random variables
#1	21	7. Classical Theorems of Probability Theory
#2	21	7.1. Law of Large Numbers.
#2	22	7.2. Central Limit Theorem.
#1	22	8. Conditional Expectation
#1	24	9. Appendix: Lebesgue measure. Caratheodory extension

# TODO ::
 - Comments in the otl format

#end TODO

pages() { pdfinfo "$1" | grep Pages | sed 's/^[^:]*:\s*//'; }

# count=0; for i in {1..7}; do gmarks2 i2o -a $count $i.info; count="$(( count + $(pages "$i.pdf") ))"; done

DEL='[\t]+'
OFFSET=0

usage() {
    echo "$0 [o2i|i2o|ext|add] ..."
    echo "     o2i  [opts] <file.otl>:   converts a .otl file to a .info file"
    echo "     i2o  [opts] <file.info>:  converts a .info file to a .otl file"
    echo "     addi [opts] <file.pdf> <file.info>: Updates the meta data of a pdf."
    echo "     addo [opts] <file.pdf> <file.otl>:  Updates the index of a pdf."
    echo "     ext  [opts] <file.pdf>:   extract the meta data from a .pdf file,"
    echo "     					                ie. a .info file"
    echo "     join [opts] <file.pdf>... :  joins pdf files,"
}

o2i_usage(){
    echo "o2i [opts] >file.otl>"
    echo "     converts a .otl file to a .info file"
    echo "OPTIONS:"
    echo "     -d DEL     use DEL instead of TAB for field delimiter."
    echo "     -a OFFSET  offset the pagenumber."
    echo "     -h prints  this message" 
}

i2o_usage(){
    echo "i2o  [opts] <file.info>"
    echo "      converts a .info file to a .otl file."
    echo "OPTIONS:"
    echo "     -h prints this message" 
}

addi_usage() {
    echo "addi [opts] <file.pdf> <file.info>"
    echo "     Updates the meta data of the pdf file."
    echo "OPTIONS:"
    echo "     -h prints this message" 
}

addo_usage() {
    echo "addo [opts] <file.pdf> <file.otl>"
    echo "      Updates the index of a pdf."
    echo "OPTIONS:"
    echo "     -d DEL   use DEL instead of TAB for field delimiter"
    echo "     -a OFFSET  offset the pagenumber."
    echo "     -h prints this message" 
}

join_usage() {
    echo "join [opts] <file.pdf>..."
    echo "      joins pdf files."
    echo "OPTIONS:"
    echo "     -d DEL   use DEL instead of TAB for field delimiter"
    echo "     -h prints this message" 
}

o2i() {
    OPTIND=0;
    while getopts ":d:a:h" o; do
        case "${o}" in
            d) DEL="${OPTARG}" ;;
            a) OFFSET="${OPTARG}" ;;
            h) o2i_usage; exit 0;;
        esac
    done
    shift $((OPTIND-1))

    [ "$#" -lt 1 ] && {
        echo "ERROR: Not enough arguments!"; o2i_usage; exit 1;
    }
    
	awk -F "$DEL" '{print "BookmarkBegin"} 
				 {print "BookmarkTitle: " $3}
                 {print "BookmarkPageNumber: " $2+'$OFFSET'}
				 {print "BookmarkLevel: " $1}' "$1"
}

i2o() {
  OPTIND=0;
  while getopts ":a:h" o; do
    case "${o}" in
      h) i2o_usage; exit 0;;
      a) OFFSET=${OPTARG} ;;
    esac
  done
  shift $((OPTIND-1))
  
  [ "$#" -lt 1 ] && {
    echo "ERROR: Not enough arguments!"; i2o_usage; exit 1;
  }

  IFS=':'
  grep Bookmark "$1" | {
    # level=0; pnr=0; title=""
    prt() {
        [ -z "$level" ] || printf "$level\t$(( $pnr + $OFFSET ))\t$title\n"
    }

    while read line; do
      key="$(echo "$line" | cut -d':' -f1)"
      value="$(echo "$line" | sed "s/^[^:]*: *//" )"

      case $key in
      	BookmarkBegin)       prt             ;;
      	BookmarkLevel)       level="$value"  ;;
      	BookmarkTitle)       title="$value"  ;;
      	BookmarkPageNumber)  pnr="$value"    ;;
      esac
    done 
    prt
  }
}

addi(){
    OPTIND=0;
    while getopts ":h" o; do
        case "${o}" in
            h) addi_usage; exit 0;;
        esac
    done
    shift $((OPTIND-1))

    [ "$#" -lt 2 ] && {
        echo "ERROR: Not enough arguments!"; addi_usage; exit 1;
    }

    pdftk "$2" update_info_utf8 "$3" output "$2-new.pdf"
}

addo(){
    OUTFILE=""
    OPTIND=0;
    while getopts ":d:a:o:h" o; do
        case "${o}" in
            d) DEL="${OPTARG}" ;;
            a) OFFSET="${OPTARG}" ;;
            o) OUTFILE="${OPTARG}" ;;
            h) addo_usage; exit 0;;
        esac
    done
    shift $((OPTIND-1))

    [ "$#" -lt 2 ] && {
        echo "ERROR: Not enough arguments!"; addo_usage; exit 1;
    }

    [ -z "$OUTFILE" ] && OUTFILE="$1-new.pdf"

    echo $@

    tempfile="$(mktemp)"
    o2i "$2" > "$tempfile";
    cat $tempfile
    
    pdftk "$1" update_info_utf8 "$tempfile" output "$OUTFILE"
    rm $tempfile
}


join() {
    OPTIND=0;
    while getopts ":h" o; do
        case "${o}" in
            h) join_usage; exit 0;;
        esac
    done
    shift $((OPTIND-1))
    
    [ "$#" -lt 1 ] && {
        echo "ERROR: Not enough arguments!"; join_usage; exit 1;
    }

	echo "Combining $@"
	echo "Prosessing metadata"
	for file in "$@"; do 
		pdftk $file dump_data_utf8 > /tmp/$file.info;
	done
    # TODO: Use jq
	python -c '
import sys; from pdf_dump import *
off=0; 
for fn in sys.argv[1:]:
    root=read_info("/tmp/%s.info" % fn); info2otl(root,pgn_off=off)
    off+=root["NumberOfPages"]
	' "$@" > /tmp/pdfjoin.otl
	o2i /tmp/pdfjoin.otl > pdfjoin.info
	echo "Combining pdfs"
	pdfjoin --outfile /tmp/pdfjoin.pdf -- "$@"
	echo "Adding metadata"
	pdftk /tmp/pdfjoin.pdf update_info_utf8 pdfjoin.info \
				output pdfjoin.pdf

	echo "Removing tmp-files"
	for file in "$@"; do rm /tmp/$file.info; done
	rm /tmp/pdfjoin.otl; rm /tmp/pdfjoin.pdf
}

main() {
    case $1 in
    		o2i) o2i "${@:2}" ;;
    		i2o) i2o "${@:2}" ;;
    		ext) pdftk "$2" dump_data_utf8  ;;
    		addi) addi "${@:2}" ;;
            addo) addo "${@:2}" ;;
    		join) join "${@:2}" ;;
            -h|-help)  usage ;;
            *) echo "ERROR: Unknown keyword '$1'."; usage ;;
    esac
}
main "$@"