derma/bin/scrape

#!/bin/bash

readonly BIN_DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" >/dev/null 2>&1 && pwd )"
readonly RUN_DIR=$(pwd)

function usage {
  cat <<- EOF
  ./scrape [options] <page>

  Scrape facebook event pages.

  page                                Facebook page id. Scrape all events of a
                                      specific facebook page.
  OPTIONS:
    -h --help -?                      print usage
    --events                          File in JSON format that contains an array
                                      of prevously parsed events.
    -o --output                       Output events into this path instead of
                                      stdout.
    --skip-upcoming-events            Default: The scraper will automatically
                                      scrape upcoming events, with this option
                                      enabled they will be skipped.
    --past-events (experimental)      Default: The scraper will not scrape past
                                      events by default. Enabling this option
                                      makes the scraper include past events.
                                      Please note that this might take a while
                                      depending on the number of past events.
    --no-headless                     The scraper will not run in headless mode.

  NOTE:
  Events and pages needs to be public. Private events or pages are not yet
  supported.

  EXAMPLES:
  # Select files with options
  ./scrape -o events.json --events=events.json livesentralen

  # You can redirect standard output into a file
  ./scrape tyventrondheim > events.json
EOF
}

function parse_args {
  if [ "$1" = "" ]; then
    usage
    exit 0
  fi
}

function install_node_dependencies {
  if ! [ -d node_modules ]; then
    yarn
  fi
}

function check_dependencies {
  local missing;
  missing=false;

  if [ ! $(command -v node) ]; then
    echo "Dependency missing. Please install node.js and make it available to the path as 'node'."
    missing=true
  fi

  if [ ! $(command -v yarn) ]; then
    echo "Dependency missing. Please install yarn and make it available to the path as 'yarn'."
    missing=true
  fi

  install_node_dependencies

  if [ "${missing}" != "false" ]; then
    exit 1;
  fi

  return 0;
}

function scrape {
  node src/scrape.js "$@" || usage
}

cd "$BIN_DIR"/..

check_dependencies \
  && parse_args "$@" \
  && scrape "$@"

cd "$RUN_DIR"