From 6324fd1d7494449c168805db8e3f9fa45396367b Mon Sep 17 00:00:00 2001 From: Filippo Valsorda Date: Tue, 26 Feb 2013 10:39:26 +0100 Subject: [PATCH 1/2] Switch YTPlaylistIE to API (relevant: #586); fixes #651; fixes #673; fixes #661 --- test/test_youtube_lists.py | 20 ++++--- youtube_dl/InfoExtractors.py | 103 ++++++++++++++++++++--------------- 2 files changed, 71 insertions(+), 52 deletions(-) diff --git a/test/test_youtube_lists.py b/test/test_youtube_lists.py index 3044e0852..69b0f4447 100644 --- a/test/test_youtube_lists.py +++ b/test/test_youtube_lists.py @@ -8,7 +8,7 @@ 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 +38,8 @@ def test_youtube_playlist(self): 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' ]) def test_youtube_playlist_long(self): DL = FakeDownloader() @@ -50,14 +47,21 @@ def test_youtube_playlist_long(self): IE.extract('https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q') self.assertTrue(len(DL.result) >= 799) + def test_youtube_playlist_with_deleted(self): + 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 diff --git a/youtube_dl/InfoExtractors.py b/youtube_dl/InfoExtractors.py index d661d517d..021579ce0 100755 --- a/youtube_dl/InfoExtractors.py +++ b/youtube_dl/InfoExtractors.py @@ -15,6 +15,7 @@ import xml.etree.ElementTree import random import math +import operator from .utils import * @@ -1662,22 +1663,40 @@ def _download_n_results(self, query, n): 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) + \? .*? (p|a|list)= + | user/.*?/user/ + | p/ + | user/.*?#[pg]/c/ + ) + (?:PL|EC)? + |PL|EC) + ([0-9A-Za-z-_]{10,}) + (?:/.*?/([0-9A-Za-z_-]+))? + .*""" + _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) + def suitable(self, url): + """Receives a URL and returns True if suitable for this IE.""" + return re.match(self._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 @@ -1687,55 +1706,51 @@ def _real_extract(self, url): 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' + # Download playlist videos from API playlist_id = mobj.group(2) - video_ids = [] - pagenum = 1 + 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: + videos += [(entry['yt$position']['$t'], entry['content']['src']) for entry in response['feed']['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 @@ -3605,9 +3620,9 @@ def _real_extract(self, url): 'upload_date': upload_date } return [info] - + class SteamIE(InfoExtractor): - _VALID_URL = r"""http://store.steampowered.com/ + _VALID_URL = r"""http://store.steampowered.com/ (?Pvideo|app)/ #If the page is only for videos or for a game (?P\d+)/? (?P\d*)(?P\??) #For urltype == video we sometimes get the videoID @@ -3707,7 +3722,7 @@ def _real_extract(self, url): class YouPornIE(InfoExtractor): """Information extractor for youporn.com.""" _VALID_URL = r'^(?:https?://)?(?:\w+\.)?youporn\.com/watch/(?P[0-9]+)/(?P[^/]+)' - + def _print_formats(self, formats): """Print all available formats""" print(u'Available formats:') @@ -3769,8 +3784,8 @@ def _real_extract(self, url): links = re.findall(LINK_RE, download_list_html) if(len(links) == 0): raise ExtractorError(u'ERROR: no known formats available for video') - - self._downloader.to_screen(u'[youporn] Links found: %d' % len(links)) + + self._downloader.to_screen(u'[youporn] Links found: %d' % len(links)) formats = [] for link in links: @@ -3821,7 +3836,7 @@ def _real_extract(self, url): return return [format] - + class PornotubeIE(InfoExtractor): """Information extractor for pornotube.com.""" @@ -3893,7 +3908,7 @@ def _real_extract(self, url): embed_page_url = result.group(0).strip() video_id = result.group('videoid') - + webpage = self._download_webpage(embed_page_url, video_id) # Get the video URL @@ -4053,7 +4068,7 @@ def _talk_info(self, url, video_id=0): class MySpassIE(InfoExtractor): _VALID_URL = r'http://www.myspass.de/.*' - + def _real_extract(self, url): META_DATA_URL_TEMPLATE = 'http://www.myspass.de/myspass/includes/apps/video/getvideometadataxml.php?id=%s' @@ -4063,12 +4078,12 @@ def _real_extract(self, url): url_parent_path, video_id = os.path.split(url_path) if not video_id: _, video_id = os.path.split(url_parent_path) - + # get metadata metadata_url = META_DATA_URL_TEMPLATE % video_id metadata_text = self._download_webpage(metadata_url, video_id) metadata = xml.etree.ElementTree.fromstring(metadata_text.encode('utf-8')) - + # extract values from metadata url_flv_el = metadata.find('url_flv') if url_flv_el is None: From 89de9eb125abd273d50c9c5cc2d2ce80fa8922f3 Mon Sep 17 00:00:00 2001 From: Filippo Valsorda <filippo.valsorda@gmail.com> Date: Tue, 26 Feb 2013 19:02:31 +0100 Subject: [PATCH 2/2] Modified Youtube video/playlist matching; fixes #668; fixes #585 --- test/test_all_urls.py | 14 +++++--- test/test_youtube_lists.py | 13 ++++++++ youtube_dl/InfoExtractors.py | 65 ++++++++++++++++++++---------------- 3 files changed, 59 insertions(+), 33 deletions(-) diff --git a/test/test_all_urls.py b/test/test_all_urls.py index 06de8e7b8..69717b3fc 100644 --- a/test/test_all_urls.py +++ b/test/test_all_urls.py @@ -11,12 +11,18 @@ 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') diff --git a/test/test_youtube_lists.py b/test/test_youtube_lists.py index 69b0f4447..7055d5a7a 100644 --- a/test/test_youtube_lists.py +++ b/test/test_youtube_lists.py @@ -41,6 +41,18 @@ def test_youtube_playlist(self): 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() IE = YoutubePlaylistIE(DL) @@ -48,6 +60,7 @@ def test_youtube_playlist_long(self): 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') diff --git a/youtube_dl/InfoExtractors.py b/youtube_dl/InfoExtractors.py index 021579ce0..490f30491 100755 --- a/youtube_dl/InfoExtractors.py +++ b/youtube_dl/InfoExtractors.py @@ -74,13 +74,15 @@ def __init__(self, downloader=None): 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).""" @@ -137,7 +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 @@ -189,9 +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.""" @@ -1668,17 +1671,17 @@ class YoutubePlaylistIE(InfoExtractor): (?:\w+\.)? youtube\.com/ (?: - (?:course|view_play_list|my_playlists|artist|playlist) - \? .*? (p|a|list)= + (?:course|view_play_list|my_playlists|artist|playlist|watch) + \? (?:.*?&)*? (?:p|a|list)= | user/.*?/user/ | p/ | user/.*?#[pg]/c/ ) - (?:PL|EC)? - |PL|EC) - ([0-9A-Za-z-_]{10,}) - (?:/.*?/([0-9A-Za-z_-]+))? - .*""" + ((?: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' @@ -1686,9 +1689,10 @@ class YoutubePlaylistIE(InfoExtractor): def __init__(self, downloader=None): InfoExtractor.__init__(self, 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, re.VERBOSE) is not None + 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.""" @@ -1701,13 +1705,8 @@ def _real_extract(self, url): 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 videos from API - playlist_id = mobj.group(2) + playlist_id = mobj.group(1) or mobj.group(2) page_num = 1 videos = [] @@ -1727,7 +1726,12 @@ def _real_extract(self, url): self._downloader.trouble(u'ERROR: Invalid JSON in API response: ' + compat_str(err)) return - videos += [(entry['yt$position']['$t'], entry['content']['src']) for entry in response['feed']['entry']] + 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 @@ -2313,9 +2317,10 @@ 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) @@ -3628,9 +3633,10 @@ 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) @@ -4004,9 +4010,10 @@ 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)