imgur album downloader

imgur.py

#!/usr/bin/env python
# -*- coding: utf-8 -*-
#
# author: kyle isom <[email protected]>
# license: isc / public domain dual-license
#     see (http://www.brokenlcd.net/standard-license.txt)
#
"""
This is a module (with a main() function) to grab a list of image links in
an imgur album. The main() function demonstrates how to use that to
download an album to a directory.
"""

# PEP8 status:
# a warning still exist, namely
# imgur.py:61:36: W601 .has_key() is deprecated, use 'in'
#     this is annotated in ImgurAlbum.__image_generator but the reason
#     is that the BeautifulSoup nodes don't support the in keyword.


# pylint status:
# several warnings still exist, namely
# imgur.py:48: [E, ImgurAlbum.__init__] Instance of 'Request' has no 'content'
#     member (but some types could not be inferred)
#     pylint doesn't realise the Request object has a content method
# imgur.py:24: [R, ImgurAlbum] Too few public methods (0/2)
#     the ImgurAlbum has as many methods as are applicable


import argparse
import os

# external dependencies
import BeautifulSoup
import requests


class ImgurAlbum:
    """
    An abstraction of an imgur album. Provides access to the album's title
    and the list of image links. It is initialised with the url to library:

        album = ImgurAlbum('http://www.imgur.com/a/0ruA9')

    The list of links is grabbed from album.image_generator, which is a
    generator object. The generator can be reset by:
        album.image_generator = album.__image_generator()

    """
    url = None
    title = None
    req = None
    soup = None

    def __init__(self, target):
        """
        The album must be initialised with the url to the album. The
        class will handle the rest.
        """
        open('/tmp/imgur.txt', 'a').write('received target %s\n' % (
            target))
        self.url = target
        self.req = requests.get(target)
        self.soup = BeautifulSoup.BeautifulSoup(self.req.content)
        self.title = self.soup.find('title').contents[0]
        self.title = self.title.strip().strip(' - Imgur')
        self.image_generator = self.__image_generator()

    def __image_generator(self):
        """
        This is an internal function to initialise the link generator.
        """

        # has_key is deprecated, but the nodes don't act like proper
        # dictionaries - 'class' in img won't work, for example.
        return (img['data-src'] for img in self.soup.findAll('img') if
                                img.has_key('class') and
                                img.has_key('data-src') and
                                'unloaded' == img['class'])


def main(target, verbose=True):
    """
    Main code that takes a url to an imgur album and downloads it
    to a directory (named by the album's title).
    """

    album = ImgurAlbum(target)
    if not os.access(album.title, os.F_OK):
        os.makedirs(album.title)

    for link in album.image_generator:
        filename = os.path.join(album.title, os.path.basename(link))
        if verbose:
            print 'processing: %s' % (filename, )
        image = requests.get(link)
        open(filename, 'w').write(image.content)


if '__main__' == __name__:
    DESC = 'script to download an imgur album'
    PARSER = argparse.ArgumentParser(description=DESC)
    PARSER.add_argument('-d', '--directory', action='store',
                        help='top level directory to store album in')
    PARSER.add_argument('target', help='album to download')

    ARGS = PARSER.parse_args()

    if ARGS.directory:
        PUSHDIR = os.getcwd()
        os.chdir(ARGS.directory)

    main(ARGS.target)

    if ARGS.directory:
        os.chdir(PUSHDIR)

Glad to hear!