gutenberg-dl - download text books from Project Gutenberg

gutenberg-dl.bash

#! /usr/bin/env bash

# All my gist code is licensed under the MIT license.

# gutenberg-dl <search-term>
# - Download books from Project Gutenberg in plain text format
# - Actually uses "gutendex" to search for the download link. It seems to be
#   the current best third-party Gutemberg API
#   - Source: https://github.com/garethbjohnson/gutendex

shopt -s xpg_echo

bold='\x1b[1m'
green='\x1b[32m'
blue='\x1b[34m'
reset='\x1b[0m'

function usage () {
  echo "${bold}Usage:${reset} gutenberg-dl [OPTIONS] <search-term>"
  echo "  - Download books from Project Gutenberg in plain text format"
  echo "  - If muliple matching books are found, the first one is downloaded."
  echo "    Books are sorted by popularity, so this should be ok in most cases."
  echo "  - The output file will be named '<title>_<author>.txt', with spaces"
  echo "    replaced by underscores."
  echo ""
  echo "${bold}Options:${reset}"
  echo "  -h, --help        Display this help message"
  echo '  -d, --output-dir DIR  Output directory (default: current directory)'
  exit 0
}

# print usage on -h/--help or no arguments
if [[ "$#" -eq 1 && ("$1" == "-h" || "$1" == "--help") || "$#" -eq 0 ]]; then
  usage
fi

# parse options
output_dir="."
while [[ "$#" -gt 0 ]]; do
  case "$1" in
    -d|--output-dir)
      output_dir="$2"
      shift
      ;;
    *)
      break
      ;;
  esac
  shift
done

url_encode () {
  # yes, this is rudiementary, please keep search terms to alphanumeric
  # characters
  echo -n "$1" | sed 's/ /%20/g'
}

endpoint="http://gutendex.com/books?search="
search_term=$(url_encode "$*")
url="$endpoint$search_term"
response=$(curl --location --header 'Accept: application/json' --silent "$url")

# on error, print message and bail out
if [[ "$?" -ne 0 ]]; then
  echo "Error: Could not fetch data from Gutendex"
  echo "  - URL: $url"
  exit 1
fi

# use jq to extract id, title and authors
data=$(echo $response | jq -r '.results[0] | "\(.id)|\(.title)|\(.authors[0].name)"')
IFS='|' read -r id title author <<< "$data"

# print message about found book
echo "Found book: ${bold}${blue}$title${reset} by ${bold}${blue}$author${reset}"

# create filename by joining title and author, removing non alphanum
# characters, and replacing spaces with underscores
# make all lowercase
filename=$(echo "${title}_${author}" | sed -E -e 's/[^a-zA-Z0-9_ ]//g' -e 's/ /_/g' | tr '[:upper:]' '[:lower:]').txt


# download book
book_url="http://gutenberg.org/ebooks/$id.txt.utf-8"
output_file="$output_dir/$filename"
echo "Downloading book to $output_file"
echo "  - From: $book_url"
curl --location --silent --output "$output_file" "$book_url"

if [[ "$?" -ne 0 ]]; then
  echo "Error: Could not download book"
  echo "  - URL: $book_url"
  exit 1
fi

echo "${green}Done!${reset}"