Script to download The dilbert comic strips

dilbert.py

#!/usr/bin/env python
"""
Simple script to download the Dilbert comic strips in a defined period of time

If no arguments are passed to the script, it will download all the Dilbert comic
strips in the current folder (It may take a while).

Acknowledgments
---------------
This script is strongly based in the work from:
https://community.spiceworks.com/scripts/show/982-download-all-dilbert-comics

2019-05-06: added downloading by default only missing dates.
"""

from __future__ import print_function

import datetime
import os
import re
import sys
import time
import argparse
from dateutil import rrule, parser
from os import walk
import glob

# for backwards compatibility
if sys.version_info[0] > 2:
	import urllib.request as ul
else:
	import urllib as ul


def main():
	args = parse_input_arguments()

	# If a dump folder has been defiled, create if (if does not already exists)
	# and move to it
	try:
		if args.output != '.' and not(os.path.isdir(args.output)):
			os.makedirs(args.output)
	except:
		args.output = '.'
	os.chdir(args.output)

	download_strips(args.start_date, args.end_date)

def get_last_file_date():
	f = glob.glob('./??????????.jpg')
	f.sort(reverse = True);
	last_file_date = datetime.datetime.strptime(f[0][2:12], '%Y-%m-%d')
	newstartdate = last_file_date + datetime.timedelta(days=1) 
	return newstartdate.strftime('%Y-%m-%d')

def parse_input_arguments():
	argp = argparse.ArgumentParser(description='Dilbert strips download script augmented.')
	
	argp.add_argument("-s", "--start",
						dest="start_date",
						help="start date (1989-04-17, 1st published strip, defaults to last file found in current directory).",
						default=get_last_file_date())
	argp.add_argument("-e", "--end",
						dest="end_date",
						help="End date (default, today)",
						default=None)
	argp.add_argument("-o", "--output",
						dest="output",
						help="Comics dump folder",
						default='.')

	args = argp.parse_args()
	if args.end_date is None:
		args.end_date = datetime.datetime.now().date()
	else:
		args.end_date = parser.parse(args.end_date)
	args.start_date = parser.parse(args.start_date)
	
	return args
	
def download_strips(start_date, end_date):
	for date in list(rrule.rrule(rrule.DAILY, dtstart=start_date, until=end_date)):
		comic_date = '%04d-%02d-%02d' % (date.year, date.month, date.day)
		url  = 'http://dilbert.com/strip/' + comic_date
		comic_name = comic_date + '.jpg'
		print('getting comic from', comic_date)
		ul.urlretrieve(get_true_comic_url(url), comic_name)
		#time.sleep(0.01)


def get_true_comic_url(comic_url, comic_name='comic'):
	"""
	get the true comic strip url from http://dilbert.com/strip/<date>

	It looks like Scott Adams has protected himself against pointy haired
	pirates by hiding him comic strips within the assets.amuniversal domain.
	This function digs into the comic strip web-page, finds (and returns)
	the URL where the original image lives.
	"""

	html=str(ul.urlopen(comic_url).read())
	comic_strip_pattern = 'http://assets\.amuniversal\.com/[a-zA-Z\d]+'
	return re.search(comic_strip_pattern, html).group()


if __name__ == '__main__':
	main()