rename-pdf-to-title.sh

#!/bin/bash

# Copyright Maeve Kennedy 2019
# Under MIT License

set -eu

if ! [ "${1:-""}" ] ; then
  echo "Usage: $0 filename..." >&2
  echo "   converts filename... to <pdf title>.pdf" >&2
  exit 1
fi

if ! which pdftotext > /dev/null ;
  echo "pdftotext is required, please install or add to path" >&2
  exit 1
fi

mkdir -p backups

for file in "$@"; do
  # File exist Guard
  if ! [ -f "$file" ] ; then echo "$file" not found >&2 ; continue ; fi

  txt=`pdftotext -htmlmeta "${file}" /dev/stdout | head -n 3`
  [[ "$txt" =~ \<title\>(.*)\</title\> ]]
  newname="${BASH_REMATCH[1]:-""}"

  # Guard
  if ! [ "${newname}" ] ; then echo "$file" has no pdf title >&2 ; continue ; fi

  newname="${newname}.pdf"

  # Replace non alpha words, html encoding
  if which recode > /dev/null & which sed > /dev/null ; then
    newname=`echo "${newname}" | recode html`
    newname=`echo "${newname}" | sed 's/<.*>//g'`
  fi

  # Replace / 's
  newname="${newname//\//-}"
  # Replace spaces with dashes
  newname="${newname//[[:space:]]/-}"

  echo "$file -> $newname"
  cp "${file}" backups
  mv "${file}" "$newname"
done