def dump_albums(data):
path = create_folder(BACKUP_DIR, 'facebook_app', 'albums')
photo_queue = Queue.Queue(1500)
workers = WorkerPool(Worker, photo_queue)
workers.start()
for album_id, album in data['albums'].iteritems():
album_path = create_folder(path, album['id'])
downloaded_photos = os.listdir(album_path)
for photo_id, photo in album['photos'].iteritems():
source_url = urlparse.urlparse(photo['source'])
ext = re.match("^.*/([^/]+?(\.[^/]+)?)$", source_url.path).groups()[1]
if not ext:
ext = ''
if photo['id'] not in downloaded_photos:
photo_path = create_folder(album_path, photo['id'])
photo_queue.put((dump_photo_set, (photo_path, photo, ext,), {}))
else:
# either thumbnail or photo is missing. redownloading set
if not photo_exists(album_path, photo):
photo_path = create_folder(album_path, photo['id'])
photo_queue.put((dump_photo_set, (photo_path, photo, ext,), {}))
else:
pass # photo is already in the filesystem
workers.stop()
def dump_photos(data):
path = create_folder(BACKUP_DIR, 'facebook_app', 'photos')
photo_queue = Queue.Queue(1500)
workers = WorkerPool(Worker, photo_queue)
workers.start()
downloaded_photos = os.listdir(path)
for photo in data['photos']:
photo_path = create_folder(path, photo['id'])
if not photo['id'] in downloaded_photos:
photo_queue.put((dump_photo_set, (photo_path, photo, ".jpg",), {}))
else:
if not photo_exists(path, photo):
photo_queue.put((dump_photo_set, (photo_path, photo, ".jpg",), {}))
else:
pass # photo is alread in the filesystem
workers.stop()
def dump_albums(data):
path = create_folder(BACKUP_DIR, 'picasaweb_app', 'albums')
photo_queue = Queue.Queue(1500)
workers = WorkerPool(Worker, photo_queue)
workers.start()
for album_id, album in data['albums'].iteritems():
album_path = create_folder(path, album['id'])
downloaded_photos = os.listdir(album_path)
for photo_id, photo in album['photos'].iteritems():
if photo_id not in downloaded_photos:
photo_path = create_folder(album_path, photo_id)
photo_queue.put((dump_photo_set, (photo_path, photo,),{}))
else:
if not photo_exists(album_path, photo):
photo_path = create_folder(album_path, photo['id'])
photo_queue.put((dump_photo_set, (photo_path, photo,), {}))
else:
pass
workers.stop()
photo_album = {'id': album.gphoto_id.text}
photo_album['title'] = album.title.text
photo_album['name'] = album.name.text
photo_album['timestamp'] = album.timestamp.text
photo_album['updated'] = album.updated.text
photo_album['published'] = album.published.text
photo_album['count'] = int(album.numphotos.text)
logger.debug('album: %r', photo_album)
data['albums'][album.gphoto_id.text] = photo_album
picasa_job_queue.put((func, (photo_album['id'], data, photos, album,), {}))
progress_monitor.update(10)
workers.stop()
dump_albums(data)
dump_json(data, progress_monitor.start_time)
filename = "%s.json" % progress_monitor.start_time.strftime("%Y%m%d%H%M%S")
add_to_archive("picasaweb_app", filename, progress_monitor.start_time)
progress_monitor.complete()
logger.debug("picasaweb_app backup done")
return progress_monitor.start_time
def dump_json(data, backup_time):
path = create_folder(BACKUP_DIR, 'picasaweb_app')
def run_backup(preference):
logger.debug("running backup routine for facebook_app")
logger.debug("preference: %r", preference)
progress_monitor = get_progress_monitor("facebook_app")
graph = facebook.GraphAPI(preference['tokens'])
data = {'profile': graph.get_object('me')}
facebook_job_queue = Queue.Queue(1500)
logger.debug("starting workers")
workers = WorkerPool(Worker, facebook_job_queue)
workers.start()
for k, v in preference['settings'].iteritems():
progress_monitor.update(10)
logger.debug("processing %s", k)
if v:
if k == 'albums':
data[k] = {}
def func(hash_key, result_dict, client, album):
logger.debug("fetching data for album %s", album['id'])
photos = {}
album_photos = [photo for photo in client.get_connections(album['id'], 'photos', limit=1000000)['data']]
album_photos = sorted(album_photos, key=itemgetter('created_time'), reverse=True)
for photo in album_photos:
logger.debug("fetching comments for photo %s in album %s",
photo['id'], album['id'])
comments = [comment for comment in client.get_connections(photo['id'], 'comments', limit=1000000)['data']]
photo['comments'] = sorted(comments, key=itemgetter('created_time'))
photos[photo['id']] = photo
album['photos'] = photos
result_dict[hash_key][album['id']] = album
for obj in graph.get_connections('me', 'albums', limit=1000000)['data']:
logger.debug("queueing album %s", obj['id'])
facebook_job_queue.put((func, (k, data, graph, obj,), {}))
elif k == 'photos':
data[k] = []
logger.debug("fetching data for tagged photos")
photos = graph.get_connections('me', k, limit=1000000)['data']
def func(hash_key, result_dict, client, obj):
logger.debug("fetching comments for tagged photo %s", obj['id'])
comments = [comment for comment in client.get_connections(obj['id'], 'comments', limit=1000000)['data']]
obj['comments'] = comments
result_dict[hash_key].append(obj)
for photo in photos:
logger.debug("queueing comments for tagged photo %s", photo['id'])
facebook_job_queue.put((func, (k, data, graph, photo), {}))
else:
data[k] = []
def func(hash_key, result_dict, client, id):
logger.debug("fetching data for %s %s", hash_key, id)
obj = client.get_object(id)
if hash_key in ['statuses', 'events']:
logger.debug("fetching comments for %s %s", hash_key, id)
comments = [comment for comment in client.get_connections(id, 'comments', limit=1000000)['data']]
obj['comment'] = sorted(comments, key=itemgetter('created_time'))
result_dict[hash_key].append(obj)
for obj in graph.get_connections('me', k, limit=1000000)['data']:
logger.debug("queueing %s %s", k, obj['id'])
facebook_job_queue.put((func, (k, data, graph, obj['id'],), {}))
progress_monitor.update(10)
workers.stop()
if 'albums' in data:
dump_albums(data)
if 'photos' in data:
dump_photos(data)
dump_profile_pics(data)
dump_json(data, progress_monitor.start_time)
filename = "%s.json" % progress_monitor.start_time.strftime("%Y%m%d%H%M%S")
add_to_archive("facebook_app", filename, progress_monitor.start_time)
progress_monitor.complete()
logger.debug("facebook_app backup done")
return progress_monitor.start_time
