National Geographic's Picture of the Day downloader

ng_pod.py

#!/usr/bin/env python
#
#  ng_pod.py
#
#  Copyright 2011 Costin STROIE <[email protected]>
#
#  NG_PoD is free software: you can redistribute it and/or modify
#  it under the terms of the GNU General Public License as published by
#  the Free Software Foundation, either version 3 of the License, or
#  (at your option) any later version.
#
#  NG_PoD is distributed in the hope that it will be useful,
#  but WITHOUT ANY WARRANTY; without even the implied warranty of
#  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
#  GNU General Public License for more details.
#
#  You should have received a copy of the GNU General Public License
#  along with NG_PoD.  If not, see <http://www.gnu.org/licenses/>.
#

""" Download the National Geographic Picture of the Day and 
    set the xfdesktop to use it as background """


# The Photo of the Day page
POD_ROOT = 'http://photography.nationalgeographic.com'
POD_URL = POD_ROOT + '/photography/photo-of-the-day'


# Import the required modules
import sys, urllib, os
from lxml import html as HTML
from lxml.cssselect import CSSSelector
from datetime import datetime
import subprocess


# Download directory
POD_DIRECTORY = os.path.expanduser(os.path.join('~', 'Pictures', 'PoD', 'NG'))
if not os.path.isdir(POD_DIRECTORY):
    os.makedirs(POD_DIRECTORY)


def download_image(url, recursive = False):
    """ The main method """
    # Try to open the page
    try:
        pod_fd = urllib.urlopen(url)
        pod_page = HTML.parse(pod_fd)
    except:
        raise

    # Find the previous photo
    csssel = CSSSelector('div#pod_nav>p.prev>a')
    try:
        prev_a = csssel(pod_page)[0]
    except IndexError:
        previous = None
        print 'ERROR: No previous link.'
    else:
        previous = POD_ROOT + prev_a.get('href')
        print 'INFO: Previous link: %s' % previous

    # Find the publising time
    csssel = CSSSelector('p.publication_time')
    try:
        pubtime = csssel(pod_page)[0]
    except IndexError:
        print 'INFO: No publishing time.'
        date = ''
    else:
        pubtime = str(pubtime.text)
        print 'INFO: Published %s' % pubtime
        try:
            pubdate = datetime.strptime(pubtime, '%B %d, %Y')
        except ValueError:
            print 'ERROR: Could not parse the pubtime.'
            date = ''
        else:
            date = datetime.strftime(pubdate, '%Y%m%d')

    # Find the primary photo
    csssel = CSSSelector('div.primary_photo a img')
    try:
        photo_img = csssel(pod_page)[0]
    except IndexError:
        print 'ERROR: No primary photo.'
        photo_src = None
        photo_alt = None
    else:
        # Get its src and alt
        photo_src = photo_img.get('src')
        if photo_src.startswith('//'):
            photo_src = 'http:' + photo_src
        photo_alt = photo_img.get('alt')
        # Fix the alt
        if photo_alt.startswith('Photo:'):
            photo_alt = photo_alt[6:].strip()
        print 'INFO: Photo src: %s' % photo_src
        print 'INFO: Photo alt: %s' % photo_alt

    # Find the download link
    csssel = CSSSelector('div.download_link a')
    try:
        dw_a = csssel(pod_page)[0]
    except IndexError:
        print 'INFO: No download link, using the primary photo.'
        pod_url = photo_src
        use_photo = False
    else:
        # Get the href
        pod_url = dw_a.get('href')
        print 'INFO: Photo href: %s' % pod_url
        use_photo = True

    if pod_url is not None:
        # Download the image
        filename = '%s %s.jpg' % (date, photo_alt)
        file_path = os.path.join(POD_DIRECTORY, filename)
        if os.path.exists(file_path):
            # Stop the recurence
            recursive = False
            print 'INFO: Recursive mode stop.'
        else:
            print 'INFO: Downloading "%s"' % photo_alt
            urllib.urlretrieve(pod_url, file_path)

    # Recurence
    if recursive and previous is not None:
        download_image(previous, recursive = recursive)

    if use_photo:
        return file_path


if __name__ == '__main__':
    recursive = False
    if len(sys.argv) > 1 and sys.argv[1] == '-R':
        # Recursive
        recursive = True
        print 'INFO: Recursive mode start.'
    # Download
    file_path = download_image(POD_URL, recursive = recursive)
    # Set the wallpaper in XFCE
    if file_path is not None:
        subprocess.call(['xfconf-query', '-c', 'xfce4-desktop', '-p', '/backdrop/screen0/monitor0/image-path', '-s', file_path])

# vim: set ft=python ai ts=4 sts=4 et sw=4 sta nowrap nu :