]> Raphaƫl G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/nbc.py
Update changelog.
[youtubedl] / youtube_dl / extractor / nbc.py
index d2a44d05dffbf7d42298a260ba728f53e43d5efd..6f3cb30034da7f5fcebb99fc6dec05f1ff3cd8e4 100644 (file)
@@ -1,23 +1,26 @@
 from __future__ import unicode_literals
 
+import base64
+import json
 import re
 
 from .common import InfoExtractor
 from .theplatform import ThePlatformIE
 from .adobepass import AdobePassIE
-from ..compat import compat_urllib_parse_urlparse
+from ..compat import compat_urllib_parse_unquote
 from ..utils import (
-    find_xpath_attr,
-    lowercase_escape,
+    int_or_none,
+    js_to_json,
+    parse_duration,
     smuggle_url,
-    unescapeHTML,
+    try_get,
+    unified_timestamp,
     update_url_query,
-    int_or_none,
 )
 
 
 class NBCIE(AdobePassIE):
-    _VALID_URL = r'https?://(?:www\.)?nbc\.com/(?:[^/]+/)+(?P<id>n?\d+)'
+    _VALID_URL = r'https?(?P<permalink>://(?:www\.)?nbc\.com/(?:classic-tv/)?[^/]+/video/[^/]+/(?P<id>n?\d+))'
 
     _TESTS = [
         {
@@ -36,16 +39,6 @@ class NBCIE(AdobePassIE):
                 'skip_download': True,
             },
         },
-        {
-            'url': 'http://www.nbc.com/the-tonight-show/episodes/176',
-            'info_dict': {
-                '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': {
@@ -63,11 +56,6 @@ class NBCIE(AdobePassIE):
             },
             '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,
-        },
         {
             # HLS streams requires the 'hdnea3' cookie
             'url': 'http://www.nbc.com/Kings/video/goliath/n1806',
@@ -84,73 +72,100 @@ class NBCIE(AdobePassIE):
                 'skip_download': True,
             },
             'skip': 'Only works from US',
+        },
+        {
+            'url': 'https://www.nbc.com/classic-tv/charles-in-charge/video/charles-in-charge-pilot/n3310',
+            'only_matching': True,
+        },
+        {
+            # Percent escaped url
+            'url': 'https://www.nbc.com/up-all-night/video/day-after-valentine%27s-day/n2189',
+            'only_matching': True,
         }
     ]
 
     def _real_extract(self, url):
-        video_id = self._match_id(url)
-        webpage = self._download_webpage(url, video_id)
-        info = {
+        permalink, video_id = re.match(self._VALID_URL, url).groups()
+        permalink = 'http' + compat_urllib_parse_unquote(permalink)
+        video_data = self._download_json(
+            'https://friendship.nbc.co/v2/graphql', video_id, query={
+                'query': '''query bonanzaPage(
+  $app: NBCUBrands! = nbc
+  $name: String!
+  $oneApp: Boolean
+  $platform: SupportedPlatforms! = web
+  $type: EntityPageType! = VIDEO
+  $userId: String!
+) {
+  bonanzaPage(
+    app: $app
+    name: $name
+    oneApp: $oneApp
+    platform: $platform
+    type: $type
+    userId: $userId
+  ) {
+    metadata {
+      ... on VideoPageData {
+        description
+        episodeNumber
+        keywords
+        locked
+        mpxAccountId
+        mpxGuid
+        rating
+        resourceId
+        seasonNumber
+        secondaryTitle
+        seriesShortTitle
+      }
+    }
+  }
+}''',
+                'variables': json.dumps({
+                    'name': permalink,
+                    'oneApp': True,
+                    'userId': '0',
+                }),
+            })['data']['bonanzaPage']['metadata']
+        query = {
+            'mbr': 'true',
+            'manifest': 'm3u',
+        }
+        video_id = video_data['mpxGuid']
+        title = video_data['secondaryTitle']
+        if video_data.get('locked'):
+            resource = self._get_mvpd_resource(
+                video_data.get('resourceId') or 'nbcentertainment',
+                title, video_id, video_data.get('rating'))
+            query['auth'] = self._extract_mvpd_auth(
+                url, video_id, 'nbcentertainment', resource)
+        theplatform_url = smuggle_url(update_url_query(
+            'http://link.theplatform.com/s/NnzsPC/media/guid/%s/%s' % (video_data.get('mpxAccountId') or '2410887629', video_id),
+            query), {'force_smil_url': True})
+        return {
             '_type': 'url_transparent',
-            'ie_key': 'ThePlatform',
             'id': video_id,
+            'title': title,
+            'url': theplatform_url,
+            'description': video_data.get('description'),
+            'tags': video_data.get('keywords'),
+            'season_number': int_or_none(video_data.get('seasonNumber')),
+            'episode_number': int_or_none(video_data.get('episodeNumber')),
+            'episode': title,
+            'series': video_data.get('seriesShortTitle'),
+            'ie_key': 'ThePlatform',
         }
-        video_data = None
-        preload = self._search_regex(
-            r'PRELOAD\s*=\s*({.+})', webpage, 'preload data', default=None)
-        if preload:
-            preload_data = self._parse_json(preload, video_id)
-            path = compat_urllib_parse_urlparse(url).path.rstrip('/')
-            entity_id = preload_data.get('xref', {}).get(path)
-            video_data = preload_data.get('entities', {}).get(entity_id)
-        if video_data:
-            query = {
-                'mbr': 'true',
-                'manifest': 'm3u',
-            }
-            video_id = video_data['guid']
-            title = video_data['title']
-            if video_data.get('entitlement') == 'auth':
-                resource = self._get_mvpd_resource(
-                    'nbcentertainment', title, video_id,
-                    video_data.get('vChipRating'))
-                query['auth'] = self._extract_mvpd_auth(
-                    url, video_id, 'nbcentertainment', resource)
-            theplatform_url = smuggle_url(update_url_query(
-                'http://link.theplatform.com/s/NnzsPC/media/guid/2410887629/' + video_id,
-                query), {'force_smil_url': True})
-            info.update({
-                'id': video_id,
-                'title': title,
-                'url': theplatform_url,
-                'description': video_data.get('description'),
-                'keywords': video_data.get('keywords'),
-                'season_number': int_or_none(video_data.get('seasonNumber')),
-                'episode_number': int_or_none(video_data.get('episodeNumber')),
-                'series': video_data.get('showName'),
-            })
-        else:
-            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
-            info['url'] = smuggle_url(theplatform_url, {'source_url': url})
-        return info
 
 
 class NBCSportsVPlayerIE(InfoExtractor):
     _VALID_URL = r'https?://vplayer\.nbcsports\.com/(?:[^/]+/)+(?P<id>[0-9a-zA-Z_]+)'
 
     _TESTS = [{
-        'url': 'https://vplayer.nbcsports.com/p/BxmELC/nbcsports_share/select/9CsDKds0kvHI',
+        'url': 'https://vplayer.nbcsports.com/p/BxmELC/nbcsports_embed/select/9CsDKds0kvHI',
         'info_dict': {
             'id': '9CsDKds0kvHI',
-            'ext': 'flv',
+            'ext': 'mp4',
             'description': 'md5:df390f70a9ba7c95ff1daace988f0d8d',
             'title': 'Tyler Kalinoski hits buzzer-beater to lift Davidson',
             'timestamp': 1426270238,
@@ -158,7 +173,7 @@ class NBCSportsVPlayerIE(InfoExtractor):
             'uploader': 'NBCU-SPORTS',
         }
     }, {
-        'url': 'http://vplayer.nbcsports.com/p/BxmELC/nbc_embedshare/select/_hqLjQ95yx8Z',
+        'url': 'https://vplayer.nbcsports.com/p/BxmELC/nbcsports_embed/select/media/_hqLjQ95yx8Z',
         'only_matching': True,
     }]
 
@@ -172,7 +187,8 @@ class NBCSportsVPlayerIE(InfoExtractor):
     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)
+        theplatform_url = self._og_search_video_url(webpage).replace(
+            'vplayer.nbcsports.com', 'player.theplatform.com')
         return self.url_result(theplatform_url, 'ThePlatform')
 
 
@@ -200,6 +216,65 @@ class NBCSportsIE(InfoExtractor):
             NBCSportsVPlayerIE._extract_url(webpage), 'NBCSportsVPlayer')
 
 
+class NBCSportsStreamIE(AdobePassIE):
+    _VALID_URL = r'https?://stream\.nbcsports\.com/.+?\bpid=(?P<id>\d+)'
+    _TEST = {
+        'url': 'http://stream.nbcsports.com/nbcsn/generic?pid=206559',
+        'info_dict': {
+            'id': '206559',
+            'ext': 'mp4',
+            'title': 'Amgen Tour of California Women\'s Recap',
+            'description': 'md5:66520066b3b5281ada7698d0ea2aa894',
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
+        'skip': 'Requires Adobe Pass Authentication',
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        live_source = self._download_json(
+            'http://stream.nbcsports.com/data/live_sources_%s.json' % video_id,
+            video_id)
+        video_source = live_source['videoSources'][0]
+        title = video_source['title']
+        source_url = None
+        for k in ('source', 'msl4source', 'iossource', 'hlsv4'):
+            sk = k + 'Url'
+            source_url = video_source.get(sk) or video_source.get(sk + 'Alt')
+            if source_url:
+                break
+        else:
+            source_url = video_source['ottStreamUrl']
+        is_live = video_source.get('type') == 'live' or video_source.get('status') == 'Live'
+        resource = self._get_mvpd_resource('nbcsports', title, video_id, '')
+        token = self._extract_mvpd_auth(url, video_id, 'nbcsports', resource)
+        tokenized_url = self._download_json(
+            'https://token.playmakerservices.com/cdn',
+            video_id, data=json.dumps({
+                'requestorId': 'nbcsports',
+                'pid': video_id,
+                'application': 'NBCSports',
+                'version': 'v1',
+                'platform': 'desktop',
+                'cdn': 'akamai',
+                'url': video_source['sourceUrl'],
+                'token': base64.b64encode(token.encode()).decode(),
+                'resourceId': base64.b64encode(resource.encode()).decode(),
+            }).encode())['tokenizedUrl']
+        formats = self._extract_m3u8_formats(tokenized_url, video_id, 'mp4')
+        self._sort_formats(formats)
+        return {
+            'id': video_id,
+            'title': self._live_title(title) if is_live else title,
+            'description': live_source.get('description'),
+            'formats': formats,
+            'is_live': is_live,
+        }
+
+
 class CSNNEIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?csnne\.com/video/(?P<id>[0-9a-z-]+)'
 
@@ -228,31 +303,17 @@ class CSNNEIE(InfoExtractor):
 
 
 class NBCNewsIE(ThePlatformIE):
-    _VALID_URL = r'''(?x)https?://(?:www\.)?(?:nbcnews|today|msnbc)\.com/
-        (?:video/.+?/(?P<id>\d+)|
-        ([^/]+/)*(?:.*-)?(?P<mpx_id>[^/?]+))
-        '''
+    _VALID_URL = r'(?x)https?://(?:www\.)?(?:nbcnews|today|msnbc)\.com/([^/]+/)*(?:.*-)?(?P<id>[^/?]+)'
 
     _TESTS = [
-        {
-            'url': 'http://www.nbcnews.com/video/nbc-news/52753292',
-            'md5': '47abaac93c6eaf9ad37ee6c4463a5179',
-            'info_dict': {
-                'id': '52753292',
-                'ext': 'flv',
-                'title': 'Crew emerges after four-month Mars food study',
-                'description': 'md5:24e632ffac72b35f8b67a12d1b6ddfc1',
-            },
-        },
         {
             'url': 'http://www.nbcnews.com/watch/nbcnews-com/how-twitter-reacted-to-the-snowden-interview-269389891880',
-            'md5': 'af1adfa51312291a017720403826bb64',
+            'md5': 'cf4bc9e6ce0130f00f545d80ecedd4bf',
             'info_dict': {
-                'id': 'p_tweet_snow_140529',
+                'id': '269389891880',
                 'ext': 'mp4',
                 'title': 'How Twitter Reacted To The Snowden Interview',
                 'description': 'md5:65a0bd5d76fe114f3c2727aa3a81fe64',
-                'uploader': 'NBCU-NEWS',
                 'timestamp': 1401363060,
                 'upload_date': '20140529',
             },
@@ -270,55 +331,51 @@ class NBCNewsIE(ThePlatformIE):
         },
         {
             'url': 'http://www.nbcnews.com/nightly-news/video/nightly-news-with-brian-williams-full-broadcast-february-4-394064451844',
-            'md5': '73135a2e0ef819107bbb55a5a9b2a802',
+            'md5': '8eb831eca25bfa7d25ddd83e85946548',
             'info_dict': {
-                'id': 'nn_netcast_150204',
+                'id': '394064451844',
                 'ext': 'mp4',
                 'title': 'Nightly News with Brian Williams Full Broadcast (February 4)',
                 'description': 'md5:1c10c1eccbe84a26e5debb4381e2d3c5',
                 'timestamp': 1423104900,
-                'uploader': 'NBCU-NEWS',
                 'upload_date': '20150205',
             },
         },
         {
             'url': 'http://www.nbcnews.com/business/autos/volkswagen-11-million-vehicles-could-have-suspect-software-emissions-scandal-n431456',
-            'md5': 'a49e173825e5fcd15c13fc297fced39d',
+            'md5': '4a8c4cec9e1ded51060bdda36ff0a5c0',
             'info_dict': {
-                'id': 'x_lon_vwhorn_150922',
+                'id': 'n431456',
                 'ext': 'mp4',
-                'title': 'Volkswagen U.S. Chief:\xa0 We Have Totally Screwed Up',
-                'description': 'md5:c8be487b2d80ff0594c005add88d8351',
+                'title': "Volkswagen U.S. Chief:  We 'Totally Screwed Up'",
+                'description': 'md5:d22d1281a24f22ea0880741bb4dd6301',
                 'upload_date': '20150922',
                 'timestamp': 1442917800,
-                'uploader': 'NBCU-NEWS',
             },
         },
         {
             'url': 'http://www.today.com/video/see-the-aurora-borealis-from-space-in-stunning-new-nasa-video-669831235788',
             'md5': '118d7ca3f0bea6534f119c68ef539f71',
             'info_dict': {
-                'id': 'tdy_al_space_160420',
+                'id': '669831235788',
                 'ext': 'mp4',
                 'title': 'See the aurora borealis from space in stunning new NASA video',
                 'description': 'md5:74752b7358afb99939c5f8bb2d1d04b1',
                 'upload_date': '20160420',
                 'timestamp': 1461152093,
-                'uploader': 'NBCU-NEWS',
             },
         },
         {
             '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',
+                'id': '314487875924',
                 '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': r're:^https?://.*\.jpg$',
                 'timestamp': 1406937606,
                 'upload_date': '20140802',
-                'uploader': 'NBCU-NEWS',
             },
         },
         {
@@ -333,63 +390,67 @@ class NBCNewsIE(ThePlatformIE):
     ]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-        if video_id is not None:
-            all_info = self._download_xml('http://www.nbcnews.com/id/%s/displaymode/1219' % video_id, video_id)
-            info = all_info.find('video')
-
-            return {
-                'id': video_id,
-                'title': info.find('headline').text,
-                'ext': 'flv',
-                'url': find_xpath_attr(info, 'media', 'type', 'flashVideo').text,
-                'description': info.find('caption').text,
-                'thumbnail': find_xpath_attr(info, 'media', 'type', 'thumbnail').text,
-            }
-        else:
-            # "feature" and "nightly-news" pages use theplatform.com
-            video_id = mobj.group('mpx_id')
-            webpage = self._download_webpage(url, video_id)
-
-            filter_param = 'byId'
-            bootstrap_json = self._search_regex(
-                [r'(?m)(?:var\s+(?:bootstrapJson|playlistData)|NEWS\.videoObj)\s*=\s*({.+});?\s*$',
-                 r'videoObj\s*:\s*({.+})', r'data-video="([^"]+)"',
-                 r'jQuery\.extend\(Drupal\.settings\s*,\s*({.+?})\);'],
-                webpage, 'bootstrap json', default=None)
-            if bootstrap_json:
-                bootstrap = self._parse_json(
-                    bootstrap_json, video_id, transform_source=unescapeHTML)
-
-                info = None
-                if 'results' in bootstrap:
-                    info = bootstrap['results'][0]['video']
-                elif 'video' in bootstrap:
-                    info = bootstrap['video']
-                elif 'msnbcVideoInfo' in bootstrap:
-                    info = bootstrap['msnbcVideoInfo']['meta']
-                elif 'msnbcThePlatform' in bootstrap:
-                    info = bootstrap['msnbcThePlatform']['videoPlayer']['video']
-                else:
-                    info = bootstrap
-
-                if 'guid' in info:
-                    video_id = info['guid']
-                    filter_param = 'byGuid'
-                elif 'mpxId' in info:
-                    video_id = info['mpxId']
-
-            return {
-                '_type': 'url_transparent',
-                'id': video_id,
-                # http://feed.theplatform.com/f/2E2eJC/nbcnews also works
-                'url': update_url_query('http://feed.theplatform.com/f/2E2eJC/nnd_NBCNews', {filter_param: video_id}),
-                'ie_key': 'ThePlatformFeed',
-            }
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+
+        data = self._parse_json(self._search_regex(
+            r'window\.__data\s*=\s*({.+});', webpage,
+            'bootstrap json'), video_id, js_to_json)
+        video_data = try_get(data, lambda x: x['video']['current'], dict)
+        if not video_data:
+            video_data = data['article']['content'][0]['primaryMedia']['video']
+        title = video_data['headline']['primary']
+
+        formats = []
+        for va in video_data.get('videoAssets', []):
+            public_url = va.get('publicUrl')
+            if not public_url:
+                continue
+            if '://link.theplatform.com/' in public_url:
+                public_url = update_url_query(public_url, {'format': 'redirect'})
+            format_id = va.get('format')
+            if format_id == 'M3U':
+                formats.extend(self._extract_m3u8_formats(
+                    public_url, video_id, 'mp4', 'm3u8_native',
+                    m3u8_id=format_id, fatal=False))
+                continue
+            tbr = int_or_none(va.get('bitrate'), 1000)
+            if tbr:
+                format_id += '-%d' % tbr
+            formats.append({
+                'format_id': format_id,
+                'url': public_url,
+                'width': int_or_none(va.get('width')),
+                'height': int_or_none(va.get('height')),
+                'tbr': tbr,
+                'ext': 'mp4',
+            })
+        self._sort_formats(formats)
+
+        subtitles = {}
+        closed_captioning = video_data.get('closedCaptioning')
+        if closed_captioning:
+            for cc_url in closed_captioning.values():
+                if not cc_url:
+                    continue
+                subtitles.setdefault('en', []).append({
+                    'url': cc_url,
+                })
+
+        return {
+            'id': video_id,
+            'title': title,
+            'description': try_get(video_data, lambda x: x['description']['primary']),
+            'thumbnail': try_get(video_data, lambda x: x['primaryImage']['url']['primary']),
+            'duration': parse_duration(video_data.get('duration')),
+            'timestamp': unified_timestamp(video_data.get('datePublished')),
+            'formats': formats,
+            'subtitles': subtitles,
+        }
 
 
 class NBCOlympicsIE(InfoExtractor):
+    IE_NAME = 'nbcolympics'
     _VALID_URL = r'https?://www\.nbcolympics\.com/video/(?P<id>[a-z-]+)'
 
     _TEST = {
@@ -427,3 +488,54 @@ class NBCOlympicsIE(InfoExtractor):
             'ie_key': ThePlatformIE.ie_key(),
             'display_id': display_id,
         }
+
+
+class NBCOlympicsStreamIE(AdobePassIE):
+    IE_NAME = 'nbcolympics:stream'
+    _VALID_URL = r'https?://stream\.nbcolympics\.com/(?P<id>[0-9a-z-]+)'
+    _TEST = {
+        'url': 'http://stream.nbcolympics.com/2018-winter-olympics-nbcsn-evening-feb-8',
+        'info_dict': {
+            'id': '203493',
+            'ext': 'mp4',
+            'title': 're:Curling, Alpine, Luge [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
+    }
+    _DATA_URL_TEMPLATE = 'http://stream.nbcolympics.com/data/%s_%s.json'
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+        webpage = self._download_webpage(url, display_id)
+        pid = self._search_regex(r'pid\s*=\s*(\d+);', webpage, 'pid')
+        resource = self._search_regex(
+            r"resource\s*=\s*'(.+)';", webpage,
+            'resource').replace("' + pid + '", pid)
+        event_config = self._download_json(
+            self._DATA_URL_TEMPLATE % ('event_config', pid),
+            pid)['eventConfig']
+        title = self._live_title(event_config['eventTitle'])
+        source_url = self._download_json(
+            self._DATA_URL_TEMPLATE % ('live_sources', pid),
+            pid)['videoSources'][0]['sourceUrl']
+        media_token = self._extract_mvpd_auth(
+            url, pid, event_config.get('requestorId', 'NBCOlympics'), resource)
+        formats = self._extract_m3u8_formats(self._download_webpage(
+            'http://sp.auth.adobe.com/tvs/v1/sign', pid, query={
+                'cdn': 'akamai',
+                'mediaToken': base64.b64encode(media_token.encode()),
+                'resource': base64.b64encode(resource.encode()),
+                'url': source_url,
+            }), pid, 'mp4')
+        self._sort_formats(formats)
+
+        return {
+            'id': pid,
+            'display_id': display_id,
+            'title': title,
+            'formats': formats,
+            'is_live': True,
+        }