Merge remote-tracking branch 'origin/API_YT_playlists'

This commit is contained in:
Philipp Hagemeister 2013-02-26 19:58:04 +01:00
commit 6bf48bd866
3 changed files with 120 additions and 75 deletions

View file

@ -11,12 +11,18 @@ from youtube_dl.InfoExtractors import YoutubeIE, YoutubePlaylistIE
class TestAllURLsMatching(unittest.TestCase):
def test_youtube_playlist_matching(self):
self.assertTrue(YoutubePlaylistIE().suitable(u'ECUl4u3cNGP61MdtwGTqZA0MreSaDybji8'))
self.assertTrue(YoutubePlaylistIE().suitable(u'PL63F0C78739B09958'))
self.assertFalse(YoutubePlaylistIE().suitable(u'PLtS2H6bU1M'))
self.assertTrue(YoutubePlaylistIE.suitable(u'ECUl4u3cNGP61MdtwGTqZA0MreSaDybji8'))
self.assertTrue(YoutubePlaylistIE.suitable(u'UUBABnxM4Ar9ten8Mdjj1j0Q')) #585
self.assertTrue(YoutubePlaylistIE.suitable(u'PL63F0C78739B09958'))
self.assertTrue(YoutubePlaylistIE.suitable(u'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q'))
self.assertTrue(YoutubePlaylistIE.suitable(u'https://www.youtube.com/course?list=ECUl4u3cNGP61MdtwGTqZA0MreSaDybji8'))
self.assertTrue(YoutubePlaylistIE.suitable(u'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC'))
self.assertTrue(YoutubePlaylistIE.suitable(u'https://www.youtube.com/watch?v=AV6J6_AeFEQ&playnext=1&list=PL4023E734DA416012')) #668
self.assertFalse(YoutubePlaylistIE.suitable(u'PLtS2H6bU1M'))
def test_youtube_matching(self):
self.assertTrue(YoutubeIE().suitable(u'PLtS2H6bU1M'))
self.assertTrue(YoutubeIE.suitable(u'PLtS2H6bU1M'))
self.assertFalse(YoutubeIE.suitable(u'https://www.youtube.com/watch?v=AV6J6_AeFEQ&playnext=1&list=PL4023E734DA416012')) #668
def test_youtube_extract(self):
self.assertEqual(YoutubeIE()._extract_id('http://www.youtube.com/watch?&v=BaW_jenozKc'), 'BaW_jenozKc')

View file

@ -8,7 +8,7 @@ import json
import os
sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
from youtube_dl.InfoExtractors import YoutubeUserIE,YoutubePlaylistIE
from youtube_dl.InfoExtractors import YoutubeUserIE, YoutubePlaylistIE, YoutubeIE
from youtube_dl.utils import *
PARAMETERS_FILE = os.path.join(os.path.dirname(os.path.abspath(__file__)), "parameters.json")
@ -38,11 +38,20 @@ class TestYoutubeLists(unittest.TestCase):
DL = FakeDownloader()
IE = YoutubePlaylistIE(DL)
IE.extract('https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re')
self.assertEqual(DL.result, [
['http://www.youtube.com/watch?v=bV9L5Ht9LgY'],
['http://www.youtube.com/watch?v=FXxLjLQi3Fg'],
['http://www.youtube.com/watch?v=tU3Bgo5qJZE']
])
self.assertEqual(map(lambda x: YoutubeIE()._extract_id(x[0]), DL.result),
[ 'bV9L5Ht9LgY', 'FXxLjLQi3Fg', 'tU3Bgo5qJZE' ])
#661
DL = FakeDownloader()
IE = YoutubePlaylistIE(DL)
IE.extract('PLMCmkNmxw6Z9eduM7BZjSEh7HiU543Ig0')
self.assertTrue(len(DL.result) > 20)
#673
DL = FakeDownloader()
IE = YoutubePlaylistIE(DL)
IE.extract('PLBB231211A4F62143')
self.assertTrue(len(DL.result) > 40)
def test_youtube_playlist_long(self):
DL = FakeDownloader()
@ -50,14 +59,22 @@ class TestYoutubeLists(unittest.TestCase):
IE.extract('https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q')
self.assertTrue(len(DL.result) >= 799)
def test_youtube_playlist_with_deleted(self):
#651
DL = FakeDownloader()
IE = YoutubePlaylistIE(DL)
IE.extract('https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC')
self.assertFalse('pElCt5oNDuI' in map(lambda x: YoutubeIE()._extract_id(x[0]), DL.result))
self.assertFalse('KdPEApIVdWM' in map(lambda x: YoutubeIE()._extract_id(x[0]), DL.result))
def test_youtube_course(self):
DL = FakeDownloader()
IE = YoutubePlaylistIE(DL)
# TODO find a > 100 (paginating?) videos course
IE.extract('https://www.youtube.com/course?list=ECUl4u3cNGP61MdtwGTqZA0MreSaDybji8')
self.assertEqual(DL.result[0], ['http://www.youtube.com/watch?v=j9WZyLZCBzs'])
self.assertEqual(YoutubeIE()._extract_id(DL.result[0][0]), 'j9WZyLZCBzs')
self.assertEqual(len(DL.result), 25)
self.assertEqual(DL.result[-1], ['http://www.youtube.com/watch?v=rYefUsYuEp0'])
self.assertEqual(YoutubeIE()._extract_id(DL.result[-1][0]), 'rYefUsYuEp0')
def test_youtube_channel(self):
# I give up, please find a channel that does paginate and test this like test_youtube_playlist_long

View file

@ -15,6 +15,7 @@ import email.utils
import xml.etree.ElementTree
import random
import math
import operator
from .utils import *
@ -73,13 +74,15 @@ class InfoExtractor(object):
self._ready = False
self.set_downloader(downloader)
def suitable(self, url):
@classmethod
def suitable(cls, url):
"""Receives a URL and returns True if suitable for this IE."""
return re.match(self._VALID_URL, url) is not None
return re.match(cls._VALID_URL, url) is not None
def working(self):
@classmethod
def working(cls):
"""Getter method for _WORKING."""
return self._WORKING
return cls._WORKING
def initialize(self):
"""Initializes an instance (authentication, etc)."""
@ -136,7 +139,6 @@ class YoutubeIE(InfoExtractor):
(?:youtu\.be/|(?:\w+\.)?youtube(?:-nocookie)?\.com/|
tube\.majestyc\.net/) # the various hostnames, with wildcard subdomains
(?:.*?\#/)? # handle anchor (#/) redirect urls
(?!view_play_list|my_playlists|artist|playlist) # ignore playlist URLs
(?: # the various things that can precede the ID:
(?:(?:v|embed|e)/) # v/ or embed/ or e/
|(?: # or the v= param in all its forms
@ -188,9 +190,11 @@ class YoutubeIE(InfoExtractor):
}
IE_NAME = u'youtube'
def suitable(self, url):
@classmethod
def suitable(cls, url):
"""Receives a URL and returns True if suitable for this IE."""
return re.match(self._VALID_URL, url, re.VERBOSE) is not None
if YoutubePlaylistIE.suitable(url): return False
return re.match(cls._VALID_URL, url, re.VERBOSE) is not None
def report_lang(self):
"""Report attempt to set language."""
@ -1666,80 +1670,95 @@ class YahooSearchIE(InfoExtractor):
class YoutubePlaylistIE(InfoExtractor):
"""Information Extractor for YouTube playlists."""
_VALID_URL = r'(?:(?:https?://)?(?:\w+\.)?youtube\.com/(?:(?:course|view_play_list|my_playlists|artist|playlist)\?.*?(p|a|list)=|user/.*?/user/|p/|user/.*?#[pg]/c/)(?:PL|EC)?|PL|EC)([0-9A-Za-z-_]{10,})(?:/.*?/([0-9A-Za-z_-]+))?.*'
_TEMPLATE_URL = 'http://www.youtube.com/%s?%s=%s&page=%s&gl=US&hl=en'
_VIDEO_INDICATOR_TEMPLATE = r'/watch\?v=(.+?)&([^&"]+&)*list=.*?%s'
_MORE_PAGES_INDICATOR = u"Next \N{RIGHT-POINTING DOUBLE ANGLE QUOTATION MARK}"
_VALID_URL = r"""(?:
(?:https?://)?
(?:\w+\.)?
youtube\.com/
(?:
(?:course|view_play_list|my_playlists|artist|playlist|watch)
\? (?:.*?&)*? (?:p|a|list)=
| user/.*?/user/
| p/
| user/.*?#[pg]/c/
)
((?:PL|EC|UU)?[0-9A-Za-z-_]{10,})
.*
|
((?:PL|EC|UU)[0-9A-Za-z-_]{10,})
)"""
_TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/playlists/%s?max-results=%i&start-index=%i&v=2&alt=json'
_MAX_RESULTS = 50
IE_NAME = u'youtube:playlist'
def __init__(self, downloader=None):
InfoExtractor.__init__(self, downloader)
@classmethod
def suitable(cls, url):
"""Receives a URL and returns True if suitable for this IE."""
return re.match(cls._VALID_URL, url, re.VERBOSE) is not None
def report_download_page(self, playlist_id, pagenum):
"""Report attempt to download playlist page with given number."""
self._downloader.to_screen(u'[youtube] PL %s: Downloading page #%s' % (playlist_id, pagenum))
def _real_extract(self, url):
# Extract playlist id
mobj = re.match(self._VALID_URL, url)
mobj = re.match(self._VALID_URL, url, re.VERBOSE)
if mobj is None:
self._downloader.trouble(u'ERROR: invalid url: %s' % url)
return
# Single video case
if mobj.group(3) is not None:
self._downloader.download([mobj.group(3)])
return
# Download playlist pages
# prefix is 'p' as default for playlists but there are other types that need extra care
playlist_prefix = mobj.group(1)
if playlist_prefix == 'a':
playlist_access = 'artist'
else:
playlist_prefix = 'p'
playlist_access = 'view_play_list'
playlist_id = mobj.group(2)
video_ids = []
pagenum = 1
# Download playlist videos from API
playlist_id = mobj.group(1) or mobj.group(2)
page_num = 1
videos = []
while True:
self.report_download_page(playlist_id, pagenum)
url = self._TEMPLATE_URL % (playlist_access, playlist_prefix, playlist_id, pagenum)
request = compat_urllib_request.Request(url)
self.report_download_page(playlist_id, page_num)
url = self._TEMPLATE_URL % (playlist_id, self._MAX_RESULTS, self._MAX_RESULTS * (page_num - 1) + 1)
try:
page = compat_urllib_request.urlopen(request).read().decode('utf-8')
page = compat_urllib_request.urlopen(url).read().decode('utf8')
except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
self._downloader.trouble(u'ERROR: unable to download webpage: %s' % compat_str(err))
return
# Extract video identifiers
ids_in_page = []
for mobj in re.finditer(self._VIDEO_INDICATOR_TEMPLATE % playlist_id, page):
if mobj.group(1) not in ids_in_page:
ids_in_page.append(mobj.group(1))
video_ids.extend(ids_in_page)
try:
response = json.loads(page)
except ValueError as err:
self._downloader.trouble(u'ERROR: Invalid JSON in API response: ' + compat_str(err))
return
if self._MORE_PAGES_INDICATOR not in page:
if not 'feed' in response or not 'entry' in response['feed']:
self._downloader.trouble(u'ERROR: Got a malformed response from YouTube API')
return
videos += [ (entry['yt$position']['$t'], entry['content']['src'])
for entry in response['feed']['entry']
if 'content' in entry ]
if len(response['feed']['entry']) < self._MAX_RESULTS:
break
pagenum = pagenum + 1
page_num += 1
total = len(video_ids)
videos = map(operator.itemgetter(1), sorted(videos))
total = len(videos)
playliststart = self._downloader.params.get('playliststart', 1) - 1
playlistend = self._downloader.params.get('playlistend', -1)
if playlistend == -1:
video_ids = video_ids[playliststart:]
videos = videos[playliststart:]
else:
video_ids = video_ids[playliststart:playlistend]
videos = videos[playliststart:playlistend]
if len(video_ids) == total:
if len(videos) == total:
self._downloader.to_screen(u'[youtube] PL %s: Found %i videos' % (playlist_id, total))
else:
self._downloader.to_screen(u'[youtube] PL %s: Found %i videos, downloading %i' % (playlist_id, total, len(video_ids)))
self._downloader.to_screen(u'[youtube] PL %s: Found %i videos, downloading %i' % (playlist_id, total, len(videos)))
for id in video_ids:
self._downloader.download(['http://www.youtube.com/watch?v=%s' % id])
for video in videos:
self._downloader.download([video])
return
@ -2302,9 +2321,10 @@ class ComedyCentralIE(InfoExtractor):
'400': '384x216',
}
def suitable(self, url):
@classmethod
def suitable(cls, url):
"""Receives a URL and returns True if suitable for this IE."""
return re.match(self._VALID_URL, url, re.VERBOSE) is not None
return re.match(cls._VALID_URL, url, re.VERBOSE) is not None
def report_extraction(self, episode_id):
self._downloader.to_screen(u'[comedycentral] %s: Extracting information' % episode_id)
@ -3617,9 +3637,10 @@ class SteamIE(InfoExtractor):
(?P<videoID>\d*)(?P<extra>\??) #For urltype == video we sometimes get the videoID
"""
def suitable(self, url):
@classmethod
def suitable(cls, url):
"""Receives a URL and returns True if suitable for this IE."""
return re.match(self._VALID_URL, url, re.VERBOSE) is not None
return re.match(cls._VALID_URL, url, re.VERBOSE) is not None
def _real_extract(self, url):
m = re.match(self._VALID_URL, url, re.VERBOSE)
@ -3993,9 +4014,10 @@ class TEDIE(InfoExtractor):
/(?P<name>\w+) # Here goes the name and then ".html"
'''
def suitable(self, url):
@classmethod
def suitable(cls, url):
"""Receives a URL and returns True if suitable for this IE."""
return re.match(self._VALID_URL, url, re.VERBOSE) is not None
return re.match(cls._VALID_URL, url, re.VERBOSE) is not None
def _real_extract(self, url):
m=re.match(self._VALID_URL, url, re.VERBOSE)