import re
from .common import InfoExtractor
-from ..compat import (
- compat_str,
- compat_HTTPError,
-)
+from .theplatform import ThePlatformIE
from ..utils import (
- ExtractorError,
find_xpath_attr,
+ lowercase_escape,
+ smuggle_url,
+ unescapeHTML,
+ update_url_query,
+ int_or_none,
+ HEADRequest,
+ parse_iso8601,
)
_TESTS = [
{
'url': 'http://www.nbc.com/the-tonight-show/segments/112966',
- # md5 checksum is not stable
'info_dict': {
- 'id': 'c9xnCo0YPOPH',
- 'ext': 'flv',
+ 'id': '112966',
+ 'ext': 'mp4',
'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.',
},
+ 'params': {
+ # m3u8 download
+ 'skip_download': True,
+ },
},
{
'url': 'http://www.nbc.com/the-tonight-show/episodes/176',
'info_dict': {
- 'id': 'XwU9KZkp98TH',
+ 'id': '176',
'ext': 'flv',
'title': 'Ricky Gervais, Steven Van Zandt, ILoveMakonnen',
'description': 'A brand new episode of The Tonight Show welcomes Ricky Gervais, Steven Van Zandt and ILoveMakonnen.',
},
+ 'skip': '404 Not Found',
+ },
+ {
+ 'url': 'http://www.nbc.com/saturday-night-live/video/star-wars-teaser/2832821',
+ 'info_dict': {
+ 'id': '2832821',
+ 'ext': 'mp4',
+ 'title': 'Star Wars Teaser',
+ 'description': 'md5:0b40f9cbde5b671a7ff62fceccc4f442',
+ },
+ 'params': {
+ # m3u8 download
+ 'skip_download': True,
+ },
'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',
+ 'only_matching': True,
+ }
]
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'<iframe[^>]+src="((?:https?:)?//player\.theplatform\.com/[^"]+)"',
+ 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)
+ return {
+ '_type': 'url_transparent',
+ 'url': smuggle_url(theplatform_url, {'source_url': url}),
+ 'id': video_id,
+ }
class NBCSportsVPlayerIE(InfoExtractor):
class NBCSportsIE(InfoExtractor):
- # Does not include https becuase its certificate is invalid
- _VALID_URL = r'http://www\.nbcsports\.com//?(?:[^/]+/)+(?P<id>[0-9a-z-]+)'
+ # Does not include https because its certificate is invalid
+ _VALID_URL = r'https?://www\.nbcsports\.com//?(?:[^/]+/)+(?P<id>[0-9a-z-]+)'
_TEST = {
'url': 'http://www.nbcsports.com//college-basketball/ncaab/tom-izzo-michigan-st-has-so-much-respect-duke',
NBCSportsVPlayerIE._extract_url(webpage), 'NBCSportsVPlayer')
-class NBCNewsIE(InfoExtractor):
+class NBCNewsIE(ThePlatformIE):
_VALID_URL = r'''(?x)https?://(?:www\.)?nbcnews\.com/
(?:video/.+?/(?P<id>\d+)|
- (?:feature|nightly-news)/[^/]+/(?P<title>.+))
+ ([^/]+/)*(?P<display_id>[^/?]+))
'''
_TESTS = [
},
},
{
- 'url': 'http://www.nbcnews.com/feature/edward-snowden-interview/how-twitter-reacted-snowden-interview-n117236',
- 'md5': 'b2421750c9f260783721d898f4c42063',
+ 'url': 'http://www.nbcnews.com/watch/nbcnews-com/how-twitter-reacted-to-the-snowden-interview-269389891880',
+ 'md5': 'af1adfa51312291a017720403826bb64',
'info_dict': {
- 'id': 'I1wpAI_zmhsQ',
+ 'id': '269389891880',
'ext': 'mp4',
'title': 'How Twitter Reacted To The Snowden Interview',
'description': 'md5:65a0bd5d76fe114f3c2727aa3a81fe64',
},
- 'add_ie': ['ThePlatform'],
},
{
'url': 'http://www.nbcnews.com/feature/dateline-full-episodes/full-episode-family-business-n285156',
'title': 'FULL EPISODE: Family Business',
'description': 'md5:757988edbaae9d7be1d585eb5d55cc04',
},
+ 'skip': 'This page is unavailable.',
},
{
'url': 'http://www.nbcnews.com/nightly-news/video/nightly-news-with-brian-williams-full-broadcast-february-4-394064451844',
- 'md5': 'b5dda8cddd8650baa0dcb616dd2cf60d',
+ 'md5': '73135a2e0ef819107bbb55a5a9b2a802',
'info_dict': {
- 'id': 'sekXqyTVnmN3',
+ 'id': '394064451844',
'ext': 'mp4',
'title': 'Nightly News with Brian Williams Full Broadcast (February 4)',
'description': 'md5:1c10c1eccbe84a26e5debb4381e2d3c5',
},
},
+ {
+ 'url': 'http://www.nbcnews.com/business/autos/volkswagen-11-million-vehicles-could-have-suspect-software-emissions-scandal-n431456',
+ 'md5': 'a49e173825e5fcd15c13fc297fced39d',
+ 'info_dict': {
+ 'id': '529953347624',
+ 'ext': 'mp4',
+ 'title': 'Volkswagen U.S. Chief: We \'Totally Screwed Up\'',
+ 'description': 'md5:d22d1281a24f22ea0880741bb4dd6301',
+ },
+ 'expected_warnings': ['http-6000 is not available']
+ },
+ {
+ 'url': 'http://www.nbcnews.com/watch/dateline/full-episode--deadly-betrayal-386250819952',
+ 'only_matching': True,
+ },
]
def _real_extract(self, url):
'title': info.find('headline').text,
'ext': 'flv',
'url': find_xpath_attr(info, 'media', 'type', 'flashVideo').text,
- 'description': compat_str(info.find('caption').text),
+ 'description': info.find('caption').text,
'thumbnail': find_xpath_attr(info, 'media', 'type', 'thumbnail').text,
}
else:
# "feature" and "nightly-news" pages use theplatform.com
- title = mobj.group('title')
- webpage = self._download_webpage(url, title)
+ display_id = mobj.group('display_id')
+ webpage = self._download_webpage(url, display_id)
+ info = None
bootstrap_json = self._search_regex(
- r'var\s+(?:bootstrapJson|playlistData)\s*=\s*({.+});?\s*$',
- webpage, 'bootstrap json', flags=re.MULTILINE)
- bootstrap = self._parse_json(bootstrap_json, video_id)
- info = bootstrap['results'][0]['video']
- mpxid = info['mpxId']
-
- base_urls = [
- info['fallbackPlaylistUrl'],
- info['associatedPlaylistUrl'],
- ]
-
- for base_url in base_urls:
- if not base_url:
- continue
- playlist_url = base_url + '?form=MPXNBCNewsAPI'
+ r'(?m)var\s+(?:bootstrapJson|playlistData)\s*=\s*({.+});?\s*$',
+ webpage, 'bootstrap json', default=None)
+ if bootstrap_json:
+ bootstrap = self._parse_json(bootstrap_json, display_id)
+ info = bootstrap['results'][0]['video']
+ else:
+ player_instance_json = self._search_regex(
+ r'videoObj\s*:\s*({.+})', webpage, 'player instance')
+ info = self._parse_json(player_instance_json, display_id)
+ video_id = info['mpxId']
+ title = info['title']
- try:
- all_videos = self._download_json(playlist_url, title)
- except ExtractorError as ee:
- if isinstance(ee.cause, compat_HTTPError):
- continue
- raise
+ subtitles = {}
+ caption_links = info.get('captionLinks')
+ if caption_links:
+ for (sub_key, sub_ext) in (('smpte-tt', 'ttml'), ('web-vtt', 'vtt'), ('srt', 'srt')):
+ sub_url = caption_links.get(sub_key)
+ if sub_url:
+ subtitles.setdefault('en', []).append({
+ 'url': sub_url,
+ 'ext': sub_ext,
+ })
- if not all_videos or 'videos' not in all_videos:
+ formats = []
+ for video_asset in info['videoAssets']:
+ video_url = video_asset.get('publicUrl')
+ if not video_url:
continue
-
- try:
- info = next(v for v in all_videos['videos'] if v['mpxId'] == mpxid)
- break
- except StopIteration:
+ container = video_asset.get('format')
+ asset_type = video_asset.get('assetType') or ''
+ if container == 'ISM' or asset_type == 'FireTV-Once':
continue
-
- if info is None:
- raise ExtractorError('Could not find video in playlists')
+ elif asset_type == 'OnceURL':
+ tp_formats, tp_subtitles = self._extract_theplatform_smil(
+ video_url, video_id)
+ formats.extend(tp_formats)
+ subtitles = self._merge_subtitles(subtitles, tp_subtitles)
+ else:
+ tbr = int_or_none(video_asset.get('bitRate'), 1000)
+ format_id = 'http%s' % ('-%d' % tbr if tbr else '')
+ video_url = update_url_query(
+ video_url, {'format': 'redirect'})
+ # resolve the url so that we can check availability and detect the correct extension
+ head = self._request_webpage(
+ HEADRequest(video_url), video_id,
+ 'Checking %s url' % format_id,
+ '%s is not available' % format_id,
+ fatal=False)
+ if head:
+ video_url = head.geturl()
+ formats.append({
+ 'format_id': format_id,
+ 'url': video_url,
+ 'width': int_or_none(video_asset.get('width')),
+ 'height': int_or_none(video_asset.get('height')),
+ 'tbr': tbr,
+ 'container': video_asset.get('format'),
+ })
+ self._sort_formats(formats)
return {
- '_type': 'url',
- # We get the best quality video
- 'url': info['videoAssets'][-1]['publicUrl'],
- 'ie_key': 'ThePlatform',
+ 'id': video_id,
+ 'title': title,
+ 'description': info.get('description'),
+ 'thumbnail': info.get('description'),
+ 'thumbnail': info.get('thumbnail'),
+ 'duration': int_or_none(info.get('duration')),
+ 'timestamp': parse_iso8601(info.get('pubDate')),
+ 'formats': formats,
+ 'subtitles': subtitles,
}
+
+
+class MSNBCIE(InfoExtractor):
+ # https URLs redirect to corresponding http ones
+ _VALID_URL = r'https?://www\.msnbc\.com/[^/]+/watch/(?P<id>[^/]+)'
+ _TEST = {
+ 'url': 'http://www.msnbc.com/all-in-with-chris-hayes/watch/the-chaotic-gop-immigration-vote-314487875924',
+ 'md5': '6d236bf4f3dddc226633ce6e2c3f814d',
+ 'info_dict': {
+ 'id': 'n_hayes_Aimm_140801_272214',
+ 'ext': 'mp4',
+ 'title': 'The chaotic GOP immigration vote',
+ 'description': 'The Republican House votes on a border bill that has no chance of getting through the Senate or signed by the President and is drawing criticism from all sides.',
+ 'thumbnail': 're:^https?://.*\.jpg$',
+ 'timestamp': 1406937606,
+ 'upload_date': '20140802',
+ 'categories': ['MSNBC/Topics/Franchise/Best of last night', 'MSNBC/Topics/General/Congress'],
+ },
+ }
+
+ def _real_extract(self, url):
+ video_id = self._match_id(url)
+ webpage = self._download_webpage(url, video_id)
+ embed_url = self._html_search_meta('embedURL', webpage)
+ return self.url_result(embed_url)
projects / youtube-dl / blobdiff