X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/a070911bf98c751e4425008f1335cc57c86d30fc..139d14b198add5b26a11ed5dca022feaa6581d98:/youtube_dl/extractor/nbc.py diff --git a/youtube_dl/extractor/nbc.py b/youtube_dl/extractor/nbc.py index 690c46b..dc2091b 100644 --- a/youtube_dl/extractor/nbc.py +++ b/youtube_dl/extractor/nbc.py @@ -1,30 +1,32 @@ from __future__ import unicode_literals import re -import json from .common import InfoExtractor from ..compat import ( compat_str, + compat_HTTPError, ) from ..utils import ( ExtractorError, find_xpath_attr, + lowercase_escape, + unescapeHTML, ) class NBCIE(InfoExtractor): - _VALID_URL = r'http://www\.nbc\.com/(?:[^/]+/)+(?Pn?\d+)' + _VALID_URL = r'https?://www\.nbc\.com/(?:[^/]+/)+(?Pn?\d+)' _TESTS = [ { - 'url': 'http://www.nbc.com/chicago-fire/video/i-am-a-firefighter/2734188', + 'url': 'http://www.nbc.com/the-tonight-show/segments/112966', # md5 checksum is not stable 'info_dict': { - 'id': 'bTmnLCvIbaaH', + 'id': 'c9xnCo0YPOPH', 'ext': 'flv', - 'title': 'I Am a Firefighter', - 'description': 'An emergency puts Dawson\'sf irefighter skills to the ultimate test in this four-part digital series.', + 'title': 'Jimmy Fallon Surprises Fans at Ben & Jerry\'s', + 'description': 'Jimmy gives out free scoops of his new "Tonight Dough" ice cream flavor by surprising customers at the Ben & Jerry\'s scoop shop.', }, }, { @@ -37,23 +39,92 @@ class NBCIE(InfoExtractor): }, 'skip': 'Only works from US', }, + { + 'url': 'http://www.nbc.com/saturday-night-live/video/star-wars-teaser/2832821', + 'info_dict': { + 'id': '8iUuyzWDdYUZ', + 'ext': 'flv', + 'title': 'Star Wars Teaser', + 'description': 'md5:0b40f9cbde5b671a7ff62fceccc4f442', + }, + 'skip': 'Only works from US', + }, + { + # This video has expired but with an escaped embedURL + 'url': 'http://www.nbc.com/parenthood/episode-guide/season-5/just-like-at-home/515', + 'skip': 'Expired' + } ] def _real_extract(self, url): video_id = self._match_id(url) webpage = self._download_webpage(url, video_id) - theplatform_url = self._search_regex( - '(?:class="video-player video-player-full" data-mpx-url|class="player" src)="(.*?)"', - webpage, 'theplatform url').replace('_no_endcard', '') + theplatform_url = unescapeHTML(lowercase_escape(self._html_search_regex( + [ + r'(?:class="video-player video-player-full" data-mpx-url|class="player" src)="(.*?)"', + r'"embedURL"\s*:\s*"([^"]+)"' + ], + webpage, 'theplatform url').replace('_no_endcard', '').replace('\\/', '/'))) if theplatform_url.startswith('//'): theplatform_url = 'http:' + theplatform_url return self.url_result(theplatform_url) +class NBCSportsVPlayerIE(InfoExtractor): + _VALID_URL = r'https?://vplayer\.nbcsports\.com/(?:[^/]+/)+(?P[0-9a-zA-Z_]+)' + + _TESTS = [{ + 'url': 'https://vplayer.nbcsports.com/p/BxmELC/nbcsports_share/select/9CsDKds0kvHI', + 'info_dict': { + 'id': '9CsDKds0kvHI', + 'ext': 'flv', + 'description': 'md5:df390f70a9ba7c95ff1daace988f0d8d', + 'title': 'Tyler Kalinoski hits buzzer-beater to lift Davidson', + } + }, { + 'url': 'http://vplayer.nbcsports.com/p/BxmELC/nbc_embedshare/select/_hqLjQ95yx8Z', + 'only_matching': True, + }] + + @staticmethod + def _extract_url(webpage): + iframe_m = re.search( + r']+src="(?Phttps?://vplayer\.nbcsports\.com/[^"]+)"', webpage) + if iframe_m: + return iframe_m.group('url') + + def _real_extract(self, url): + video_id = self._match_id(url) + webpage = self._download_webpage(url, video_id) + theplatform_url = self._og_search_video_url(webpage) + return self.url_result(theplatform_url, 'ThePlatform') + + +class NBCSportsIE(InfoExtractor): + # Does not include https becuase its certificate is invalid + _VALID_URL = r'http://www\.nbcsports\.com//?(?:[^/]+/)+(?P[0-9a-z-]+)' + + _TEST = { + 'url': 'http://www.nbcsports.com//college-basketball/ncaab/tom-izzo-michigan-st-has-so-much-respect-duke', + 'info_dict': { + 'id': 'PHJSaFWbrTY9', + 'ext': 'flv', + 'title': 'Tom Izzo, Michigan St. has \'so much respect\' for Duke', + 'description': 'md5:ecb459c9d59e0766ac9c7d5d0eda8113', + } + } + + def _real_extract(self, url): + video_id = self._match_id(url) + webpage = self._download_webpage(url, video_id) + return self.url_result( + NBCSportsVPlayerIE._extract_url(webpage), 'NBCSportsVPlayer') + + class NBCNewsIE(InfoExtractor): - _VALID_URL = r'''(?x)https?://www\.nbcnews\.com/ - ((video/.+?/(?P\d+))| - (feature/[^/]+/(?P.+))) + _VALID_URL = r'''(?x)https?://(?:www\.)?nbcnews\.com/ + (?:video/.+?/(?P<id>\d+)| + (?:feature|nightly-news)/[^/]+/(?P<title>.+)) ''' _TESTS = [ @@ -78,6 +149,26 @@ class NBCNewsIE(InfoExtractor): }, 'add_ie': ['ThePlatform'], }, + { + 'url': 'http://www.nbcnews.com/feature/dateline-full-episodes/full-episode-family-business-n285156', + 'md5': 'fdbf39ab73a72df5896b6234ff98518a', + 'info_dict': { + 'id': 'Wjf9EDR3A_60', + 'ext': 'mp4', + 'title': 'FULL EPISODE: Family Business', + 'description': 'md5:757988edbaae9d7be1d585eb5d55cc04', + }, + }, + { + 'url': 'http://www.nbcnews.com/nightly-news/video/nightly-news-with-brian-williams-full-broadcast-february-4-394064451844', + 'md5': 'b5dda8cddd8650baa0dcb616dd2cf60d', + 'info_dict': { + 'id': 'sekXqyTVnmN3', + 'ext': 'mp4', + 'title': 'Nightly News with Brian Williams Full Broadcast (February 4)', + 'description': 'md5:1c10c1eccbe84a26e5debb4381e2d3c5', + }, + }, ] def _real_extract(self, url): @@ -96,13 +187,13 @@ class NBCNewsIE(InfoExtractor): 'thumbnail': find_xpath_attr(info, 'media', 'type', 'thumbnail').text, } else: - # "feature" pages use theplatform.com + # "feature" and "nightly-news" pages use theplatform.com title = mobj.group('title') webpage = self._download_webpage(url, title) bootstrap_json = self._search_regex( - r'var bootstrapJson = ({.+})\s*$', webpage, 'bootstrap json', - flags=re.MULTILINE) - bootstrap = json.loads(bootstrap_json) + r'var\s+(?:bootstrapJson|playlistData)\s*=\s*({.+});?\s*$', + webpage, 'bootstrap json', flags=re.MULTILINE) + bootstrap = self._parse_json(bootstrap_json, video_id) info = bootstrap['results'][0]['video'] mpxid = info['mpxId'] @@ -115,10 +206,19 @@ class NBCNewsIE(InfoExtractor): if not base_url: continue playlist_url = base_url + '?form=MPXNBCNewsAPI' - all_videos = self._download_json(playlist_url, title)['videos'] try: - info = next(v for v in all_videos if v['mpxId'] == mpxid) + all_videos = self._download_json(playlist_url, title) + except ExtractorError as ee: + if isinstance(ee.cause, compat_HTTPError): + continue + raise + + if not all_videos or 'videos' not in all_videos: + continue + + try: + info = next(v for v in all_videos['videos'] if v['mpxId'] == mpxid) break except StopIteration: continue