X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/c512650955de0b16d37e7fa7fb29ea0985e415bb..7df4343e553a9942a290e5e1dffe931261ab844e:/youtube_dl/extractor/bbccouk.py diff --git a/youtube_dl/extractor/bbccouk.py b/youtube_dl/extractor/bbccouk.py index 75e608f..249bc6b 100644 --- a/youtube_dl/extractor/bbccouk.py +++ b/youtube_dl/extractor/bbccouk.py @@ -1,15 +1,19 @@ from __future__ import unicode_literals -import re +import xml.etree.ElementTree -from .subtitles import SubtitlesInfoExtractor -from ..utils import ExtractorError +from .common import InfoExtractor +from ..utils import ( + ExtractorError, + int_or_none, +) +from ..compat import compat_HTTPError -class BBCCoUkIE(SubtitlesInfoExtractor): +class BBCCoUkIE(InfoExtractor): IE_NAME = 'bbc.co.uk' IE_DESC = 'BBC iPlayer' - _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:programmes|iplayer/episode)/(?P[\da-z]{8})' + _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:(?:(?:programmes|iplayer(?:/[^/]+)?/(?:episode|playlist))/)|music/clips[/#])(?P[\da-z]{8})' _TESTS = [ { @@ -17,8 +21,8 @@ class BBCCoUkIE(SubtitlesInfoExtractor): 'info_dict': { 'id': 'b039d07m', 'ext': 'flv', - 'title': 'Kaleidoscope: Leonard Cohen', - 'description': 'md5:db4755d7a665ae72343779f7dacb402c', + 'title': 'Kaleidoscope, Leonard Cohen', + 'description': 'The Canadian poet and songwriter reflects on his musical career.', 'duration': 1740, }, 'params': { @@ -55,6 +59,85 @@ class BBCCoUkIE(SubtitlesInfoExtractor): 'skip_download': True, }, 'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only', + }, + { + 'url': 'http://www.bbc.co.uk/iplayer/episode/p026c7jt/tomorrows-worlds-the-unearthly-history-of-science-fiction-2-invasion', + 'info_dict': { + 'id': 'b03k3pb7', + 'ext': 'flv', + 'title': "Tomorrow's Worlds: The Unearthly History of Science Fiction", + 'description': '2. Invasion', + 'duration': 3600, + }, + 'params': { + # rtmp download + 'skip_download': True, + }, + 'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only', + }, { + 'url': 'http://www.bbc.co.uk/programmes/b04v20dw', + 'info_dict': { + 'id': 'b04v209v', + 'ext': 'flv', + 'title': 'Pete Tong, The Essential New Tune Special', + 'description': "Pete has a very special mix - all of 2014's Essential New Tunes!", + 'duration': 10800, + }, + 'params': { + # rtmp download + 'skip_download': True, + } + }, { + 'url': 'http://www.bbc.co.uk/music/clips/p02frcc3', + 'note': 'Audio', + 'info_dict': { + 'id': 'p02frcch', + 'ext': 'flv', + 'title': 'Pete Tong, Past, Present and Future Special, Madeon - After Hours mix', + 'description': 'French house superstar Madeon takes us out of the club and onto the after party.', + 'duration': 3507, + }, + 'params': { + # rtmp download + 'skip_download': True, + } + }, { + 'url': 'http://www.bbc.co.uk/music/clips/p025c0zz', + 'note': 'Video', + 'info_dict': { + 'id': 'p025c103', + 'ext': 'flv', + 'title': 'Reading and Leeds Festival, 2014, Rae Morris - Closer (Live on BBC Three)', + 'description': 'Rae Morris performs Closer for BBC Three at Reading 2014', + 'duration': 226, + }, + 'params': { + # rtmp download + 'skip_download': True, + } + }, { + 'url': 'http://www.bbc.co.uk/iplayer/episode/b054fn09/ad/natural-world-20152016-2-super-powered-owls', + 'info_dict': { + 'id': 'p02n76xf', + 'ext': 'flv', + 'title': 'Natural World, 2015-2016: 2. Super Powered Owls', + 'description': 'md5:e4db5c937d0e95a7c6b5e654d429183d', + 'duration': 3540, + }, + 'params': { + # rtmp download + 'skip_download': True, + }, + 'skip': 'geolocation', + }, { + 'url': 'http://www.bbc.co.uk/iplayer/playlist/p01dvks4', + 'only_matching': True, + }, { + 'url': 'http://www.bbc.co.uk/music/clips#p02frcc3', + 'only_matching': True, + }, { + 'url': 'http://www.bbc.co.uk/iplayer/cbeebies/episode/b0480276/bing-14-atchoo', + 'only_matching': True, } ] @@ -102,6 +185,10 @@ class BBCCoUkIE(SubtitlesInfoExtractor): return playlist.findall('./{http://bbc.co.uk/2008/emp/playlist}item') def _extract_medias(self, media_selection): + error = media_selection.find('./{http://bbc.co.uk/2008/mp/mediaselection}error') + if error is not None: + raise ExtractorError( + '%s returned error: %s' % (self.IE_NAME, error.get('id')), expected=True) return media_selection.findall('./{http://bbc.co.uk/2008/mp/mediaselection}media') def _extract_connections(self, media): @@ -145,71 +232,140 @@ class BBCCoUkIE(SubtitlesInfoExtractor): formats.extend(conn_formats) return formats - def _extract_captions(self, media, programme_id): + def _get_subtitles(self, media, programme_id): subtitles = {} for connection in self._extract_connections(media): captions = self._download_xml(connection.get('href'), programme_id, 'Downloading captions') lang = captions.get('{http://www.w3.org/XML/1998/namespace}lang', 'en') ps = captions.findall('./{0}body/{0}div/{0}p'.format('{http://www.w3.org/2006/10/ttaf1}')) srt = '' + + def _extract_text(p): + if p.text is not None: + stripped_text = p.text.strip() + if stripped_text: + return stripped_text + return ' '.join(span.text.strip() for span in p.findall('{http://www.w3.org/2006/10/ttaf1}span')) for pos, p in enumerate(ps): - srt += '%s\r\n%s --> %s\r\n%s\r\n\r\n' % (str(pos), p.get('begin'), p.get('end'), - p.text.strip() if p.text is not None else '') - subtitles[lang] = srt + srt += '%s\r\n%s --> %s\r\n%s\r\n\r\n' % (str(pos), p.get('begin'), p.get('end'), _extract_text(p)) + subtitles[lang] = [ + { + 'url': connection.get('href'), + 'ext': 'ttml', + }, + { + 'data': srt, + 'ext': 'srt', + }, + ] return subtitles - def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - group_id = mobj.group('id') + def _download_media_selector(self, programme_id): + try: + media_selection = self._download_xml( + 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/pc/vpid/%s' % programme_id, + programme_id, 'Downloading media selection XML') + except ExtractorError as ee: + if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403: + media_selection = xml.etree.ElementTree.fromstring(ee.cause.read().encode('utf-8')) + else: + raise - webpage = self._download_webpage(url, group_id, 'Downloading video page') - if re.search(r'id="emp-error" class="notinuk">', webpage): - raise ExtractorError('Currently BBC iPlayer TV programmes are available to play in the UK only', - expected=True) + formats = [] + subtitles = None + + for media in self._extract_medias(media_selection): + kind = media.get('kind') + if kind == 'audio': + formats.extend(self._extract_audio(media, programme_id)) + elif kind == 'video': + formats.extend(self._extract_video(media, programme_id)) + elif kind == 'captions': + subtitles = self.extract_subtitles(media, programme_id) + + return formats, subtitles - playlist = self._download_xml('http://www.bbc.co.uk/iplayer/playlist/%s' % group_id, group_id, - 'Downloading playlist XML') + def _download_playlist(self, playlist_id): + try: + playlist = self._download_json( + 'http://www.bbc.co.uk/programmes/%s/playlist.json' % playlist_id, + playlist_id, 'Downloading playlist JSON') + + version = playlist.get('defaultAvailableVersion') + if version: + smp_config = version['smpConfig'] + title = smp_config['title'] + description = smp_config['summary'] + for item in smp_config['items']: + kind = item['kind'] + if kind != 'programme' and kind != 'radioProgramme': + continue + programme_id = item.get('vpid') + duration = int(item.get('duration')) + formats, subtitles = self._download_media_selector(programme_id) + return programme_id, title, description, duration, formats, subtitles + except ExtractorError as ee: + if not (isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 404): + raise + + # fallback to legacy playlist + playlist = self._download_xml( + 'http://www.bbc.co.uk/iplayer/playlist/%s' % playlist_id, + playlist_id, 'Downloading legacy playlist XML') no_items = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}noItems') if no_items is not None: reason = no_items.get('reason') if reason == 'preAvailability': - msg = 'Episode %s is not yet available' % group_id + msg = 'Episode %s is not yet available' % playlist_id elif reason == 'postAvailability': - msg = 'Episode %s is no longer available' % group_id + msg = 'Episode %s is no longer available' % playlist_id + elif reason == 'noMedia': + msg = 'Episode %s is not currently available' % playlist_id else: - msg = 'Episode %s is not available: %s' % (group_id, reason) + msg = 'Episode %s is not available: %s' % (playlist_id, reason) raise ExtractorError(msg, expected=True) - formats = [] - subtitles = None - for item in self._extract_items(playlist): kind = item.get('kind') if kind != 'programme' and kind != 'radioProgramme': continue title = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}title').text description = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}summary').text - programme_id = item.get('identifier') duration = int(item.get('duration')) + formats, subtitles = self._download_media_selector(programme_id) - media_selection = self._download_xml( - 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/pc/vpid/%s' % programme_id, - programme_id, 'Downloading media selection XML') + return programme_id, title, description, duration, formats, subtitles - for media in self._extract_medias(media_selection): - kind = media.get('kind') - if kind == 'audio': - formats.extend(self._extract_audio(media, programme_id)) - elif kind == 'video': - formats.extend(self._extract_video(media, programme_id)) - elif kind == 'captions': - subtitles = self._extract_captions(media, programme_id) + def _real_extract(self, url): + group_id = self._match_id(url) + + webpage = self._download_webpage(url, group_id, 'Downloading video page') + + programme_id = None - if self._downloader.params.get('listsubtitles', False): - self._list_available_subtitles(programme_id, subtitles) - return + tviplayer = self._search_regex( + r'mediator\.bind\(({.+?})\s*,\s*document\.getElementById', + webpage, 'player', default=None) + + if tviplayer: + player = self._parse_json(tviplayer, group_id).get('player', {}) + duration = int_or_none(player.get('duration')) + programme_id = player.get('vpid') + + if not programme_id: + programme_id = self._search_regex( + r'"vpid"\s*:\s*"([\da-z]{8})"', webpage, 'vpid', fatal=False, default=None) + + if programme_id: + formats, subtitles = self._download_media_selector(programme_id) + title = self._og_search_title(webpage) + description = self._search_regex( + r'

([^<]+)

', + webpage, 'description', fatal=False) + else: + programme_id, title, description, duration, formats, subtitles = self._download_playlist(group_id) self._sort_formats(formats) @@ -217,7 +373,8 @@ class BBCCoUkIE(SubtitlesInfoExtractor): 'id': programme_id, 'title': title, 'description': description, + 'thumbnail': self._og_search_thumbnail(webpage, default=None), 'duration': duration, 'formats': formats, 'subtitles': subtitles, - } \ No newline at end of file + }