X-Git-Url: https://jfr.im/git/yt-dlp.git/blobdiff_plain/9160a0c6a251312917ea34ff60fdc9a22e364f11..e63faa101cf7b9bf9f899cabb74ce03c7f893572:/yt_dlp/extractor/nbc.py

diff --git a/yt_dlp/extractor/nbc.py b/yt_dlp/extractor/nbc.py
index 2f25b9e7b..3de8c1508 100644
--- a/yt_dlp/extractor/nbc.py
+++ b/yt_dlp/extractor/nbc.py
@@ -1,5 +1,3 @@
-from __future__ import unicode_literals
-
 import base64
 import json
 import re
@@ -9,13 +7,20 @@
 from .adobepass import AdobePassIE
 from ..compat import compat_urllib_parse_unquote
 from ..utils import (
+    ExtractorError,
     int_or_none,
     parse_age_limit,
     parse_duration,
+    RegexNotFoundError,
     smuggle_url,
+    str_or_none,
+    traverse_obj,
     try_get,
+    unified_strdate,
     unified_timestamp,
     update_url_query,
+    url_basename,
+    variadic,
 )
 
 
@@ -85,7 +90,7 @@ class NBCIE(ThePlatformIE):
     ]
 
     def _real_extract(self, url):
-        permalink, video_id = re.match(self._VALID_URL, url).groups()
+        permalink, video_id = self._match_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={
@@ -185,6 +190,7 @@ def _real_extract(self, url):
 class NBCSportsVPlayerIE(InfoExtractor):
     _VALID_URL_BASE = r'https?://(?:vplayer\.nbcsports\.com|(?:www\.)?nbcsports\.com/vplayer)/'
     _VALID_URL = _VALID_URL_BASE + r'(?:[^/]+/)+(?P<id>[0-9a-zA-Z_]+)'
+    _EMBED_REGEX = [r'(?:iframe[^>]+|var video|div[^>]+data-(?:mpx-)?)[sS]rc\s?=\s?"(?P<url>%s[^\"]+)' % _VALID_URL_BASE]
 
     _TESTS = [{
         'url': 'https://vplayer.nbcsports.com/p/BxmELC/nbcsports_embed/select/9CsDKds0kvHI',
@@ -196,27 +202,22 @@ class NBCSportsVPlayerIE(InfoExtractor):
             'timestamp': 1426270238,
             'upload_date': '20150313',
             'uploader': 'NBCU-SPORTS',
+            'duration': 72.818,
+            'chapters': [],
+            'thumbnail': r're:^https?://.*\.jpg$'
         }
     }, {
-        'url': 'https://vplayer.nbcsports.com/p/BxmELC/nbcsports_embed/select/media/_hqLjQ95yx8Z',
+        'url': 'https://vplayer.nbcsports.com/p/BxmELC/nbcsports_embed/select/media/PEgOtlNcC_y2',
         'only_matching': True,
     }, {
         'url': 'https://www.nbcsports.com/vplayer/p/BxmELC/nbcsports/select/PHJSaFWbrTY9?form=html&autoPlay=true',
         'only_matching': True,
     }]
 
-    @staticmethod
-    def _extract_url(webpage):
-        iframe_m = re.search(
-            r'<(?:iframe[^>]+|div[^>]+data-(?:mpx-)?)src="(?P<url>%s[^"]+)"' % NBCSportsVPlayerIE._VALID_URL_BASE, 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).replace(
-            'vplayer.nbcsports.com', 'player.theplatform.com')
+        theplatform_url = self._html_search_regex(r'tp:releaseUrl="(.+?)"', webpage, 'url')
         return self.url_result(theplatform_url, 'ThePlatform')
 
 
@@ -234,6 +235,9 @@ class NBCSportsIE(InfoExtractor):
             'uploader': 'NBCU-SPORTS',
             'upload_date': '20150330',
             'timestamp': 1427726529,
+            'chapters': [],
+            'thumbnail': 'https://hdliveextra-a.akamaihd.net/HD/image_sports/NBCU_Sports_Group_-_nbcsports/253/303/izzodps.jpg',
+            'duration': 528.395,
         }
     }, {
         # data-mpx-src
@@ -304,7 +308,7 @@ def _real_extract(self, url):
         self._sort_formats(formats)
         return {
             'id': video_id,
-            'title': self._live_title(title) if is_live else title,
+            'title': title,
             'description': live_source.get('description'),
             'formats': formats,
             'is_live': is_live,
@@ -313,6 +317,7 @@ def _real_extract(self, url):
 
 class NBCNewsIE(ThePlatformIE):
     _VALID_URL = r'(?x)https?://(?:www\.)?(?:nbcnews|today|msnbc)\.com/([^/]+/)*(?:.*-)?(?P<id>[^/?]+)'
+    _EMBED_REGEX = [r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//www\.nbcnews\.com/widget/video-embed/[^"\']+)\1']
 
     _TESTS = [
         {
@@ -402,9 +407,7 @@ def _real_extract(self, url):
         video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
 
-        data = self._parse_json(self._search_regex(
-            r'<script[^>]+id="__NEXT_DATA__"[^>]*>({.+?})</script>',
-            webpage, 'bootstrap json'), video_id)['props']['initialState']
+        data = self._search_nextjs_data(webpage, video_id)['props']['initialState']
         video_data = try_get(data, lambda x: x['video']['current'], dict)
         if not video_data:
             video_data = data['article']['content'][0]['primaryMedia']['video']
@@ -460,7 +463,7 @@ def _real_extract(self, url):
 
 class NBCOlympicsIE(InfoExtractor):
     IE_NAME = 'nbcolympics'
-    _VALID_URL = r'https?://www\.nbcolympics\.com/video/(?P<id>[a-z-]+)'
+    _VALID_URL = r'https?://www\.nbcolympics\.com/videos?/(?P<id>[0-9a-z-]+)'
 
     _TEST = {
         # Geo-restricted to US
@@ -483,13 +486,18 @@ def _real_extract(self, url):
 
         webpage = self._download_webpage(url, display_id)
 
-        drupal_settings = self._parse_json(self._search_regex(
-            r'jQuery\.extend\(Drupal\.settings\s*,\s*({.+?})\);',
-            webpage, 'drupal settings'), display_id)
+        try:
+            drupal_settings = self._parse_json(self._search_regex(
+                r'jQuery\.extend\(Drupal\.settings\s*,\s*({.+?})\);',
+                webpage, 'drupal settings'), display_id)
 
-        iframe_url = drupal_settings['vod']['iframe_url']
-        theplatform_url = iframe_url.replace(
-            'vplayer.nbcolympics.com', 'player.theplatform.com')
+            iframe_url = drupal_settings['vod']['iframe_url']
+            theplatform_url = iframe_url.replace(
+                'vplayer.nbcolympics.com', 'player.theplatform.com')
+        except RegexNotFoundError:
+            theplatform_url = self._search_regex(
+                r"([\"'])embedUrl\1: *([\"'])(?P<embedUrl>.+)\2",
+                webpage, 'embedding URL', group="embedUrl")
 
         return {
             '_type': 'url_transparent',
@@ -502,43 +510,77 @@ def _real_extract(self, url):
 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}$',
-        },
-        'params': {
-            # m3u8 download
-            'skip_download': True,
+    _TESTS = [
+        {
+            'note': 'Tokenized m3u8 source URL',
+            'url': 'https://stream.nbcolympics.com/womens-soccer-group-round-11',
+            'info_dict': {
+                'id': '2019740',
+                'ext': 'mp4',
+                'title': r"re:Women's Group Stage - Netherlands vs\. Brazil [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$",
+            },
+            'params': {
+                'skip_download': 'm3u8',
+            },
+        }, {
+            'note': 'Plain m3u8 source URL',
+            'url': 'https://stream.nbcolympics.com/gymnastics-event-finals-mens-floor-pommel-horse-womens-vault-bars',
+            'info_dict': {
+                'id': '2021729',
+                'ext': 'mp4',
+                'title': r're:Event Finals: M Floor, W Vault, M Pommel, W Uneven Bars [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
+            },
+            'params': {
+                'skip_download': 'm3u8',
+            },
         },
-    }
-    _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'])
+            f'http://stream.nbcolympics.com/data/event_config_{pid}.json',
+            pid, 'Downloading event config')['eventConfig']
+
+        title = event_config['eventTitle']
+        is_live = {'live': True, 'replay': False}.get(event_config.get('eventStatus'))
+
         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')
+            f'https://api-leap.nbcsports.com/feeds/assets/{pid}?application=NBCOlympics&platform=desktop&format=nbc-player&env=staging',
+            pid, 'Downloading leap config'
+        )['videoSources'][0]['cdnSources']['primary'][0]['sourceUrl']
+
+        if event_config.get('cdnToken'):
+            ap_resource = self._get_mvpd_resource(
+                event_config.get('resourceId', 'NBCOlympics'),
+                re.sub(r'[^\w\d ]+', '', event_config['eventTitle']), pid,
+                event_config.get('ratingId', 'NO VALUE'))
+            media_token = self._extract_mvpd_auth(url, pid, event_config.get('requestorId', 'NBCOlympics'), ap_resource)
+
+            source_url = self._download_json(
+                'https://tokens.playmakerservices.com/', pid, 'Retrieving tokenized URL',
+                data=json.dumps({
+                    'application': 'NBCSports',
+                    'authentication-type': 'adobe-pass',
+                    'cdn': 'akamai',
+                    'pid': pid,
+                    'platform': 'desktop',
+                    'requestorId': 'NBCOlympics',
+                    'resourceId': base64.b64encode(ap_resource.encode()).decode(),
+                    'token': base64.b64encode(media_token.encode()).decode(),
+                    'url': source_url,
+                    'version': 'v1',
+                }).encode(),
+            )['akamai'][0]['tokenizedUrl']
+
+        formats = self._extract_m3u8_formats(source_url, pid, 'mp4', live=is_live)
+        for f in formats:
+            # -http_seekable requires ffmpeg 4.3+ but it doesnt seem possible to
+            # download with ffmpeg without this option
+            f['downloader_options'] = {'ffmpeg_args': ['-seekable', '0', '-http_seekable', '0', '-icy', '0']}
         self._sort_formats(formats)
 
         return {
@@ -546,5 +588,171 @@ def _real_extract(self, url):
             'display_id': display_id,
             'title': title,
             'formats': formats,
-            'is_live': True,
+            'is_live': is_live,
+        }
+
+
+class NBCStationsIE(InfoExtractor):
+    _DOMAIN_RE = '|'.join(map(re.escape, (
+        'nbcbayarea', 'nbcboston', 'nbcchicago', 'nbcconnecticut', 'nbcdfw', 'nbclosangeles',
+        'nbcmiami', 'nbcnewyork', 'nbcphiladelphia', 'nbcsandiego', 'nbcwashington',
+        'necn', 'telemundo52', 'telemundoarizona', 'telemundochicago', 'telemundonuevainglaterra',
+    )))
+    _VALID_URL = rf'https?://(?:www\.)?(?P<site>{_DOMAIN_RE})\.com/(?:[^/?#]+/)*(?P<id>[^/?#]+)/?(?:$|[#?])'
+
+    _TESTS = [{
+        'url': 'https://www.nbclosangeles.com/news/local/large-structure-fire-in-downtown-la-prompts-smoke-odor-advisory/2968618/',
+        'md5': '462041d91bd762ef5a38b7d85d6dc18f',
+        'info_dict': {
+            'id': '2968618',
+            'ext': 'mp4',
+            'title': 'Large Structure Fire in Downtown LA Prompts Smoke Odor Advisory',
+            'description': None,
+            'timestamp': 1661135892,
+            'upload_date': '20220821',
+            'uploader': 'NBC 4',
+            'uploader_id': 'KNBC',
+            'channel': 'nbclosangeles',
+        },
+    }, {
+        'url': 'https://www.telemundoarizona.com/responde/huracan-complica-reembolso-para-televidente-de-tucson/2247002/',
+        'md5': '0917dcf7885be1023a9220630d415f67',
+        'info_dict': {
+            'id': '2247002',
+            'ext': 'mp4',
+            'title': 'Huracán complica que televidente de Tucson reciba reembolso',
+            'description': 'md5:af298dc73aab74d4fca6abfb12acb6cf',
+            'timestamp': 1660886507,
+            'upload_date': '20220819',
+            'uploader': 'Telemundo Arizona',
+            'uploader_id': 'KTAZ',
+            'channel': 'telemundoarizona',
+        },
+    }]
+
+    _RESOLUTIONS = {
+        '1080': '1920',
+        '720': '1280',
+        '540': '960',
+        '360': '640',
+        '234': '416',
+    }
+
+    def _real_extract(self, url):
+        channel, video_id = self._match_valid_url(url).group('site', 'id')
+        webpage = self._download_webpage(url, video_id)
+
+        nbc_data = self._search_json(
+            r'<script>var\s*nbc\s*=', webpage, 'NBC JSON data', video_id)
+        pdk_acct = nbc_data.get('pdkAcct') or 'Yh1nAC'
+        fw_ssid = traverse_obj(nbc_data, ('video', 'fwSSID'))
+        fw_network_id = traverse_obj(nbc_data, ('video', 'fwNetworkID'), default='382114')
+
+        video_data = self._parse_json(self._html_search_regex(
+            r'data-videos="([^"]*)"', webpage, 'video data', default='{}'), video_id)
+        video_data = variadic(video_data)[0]
+        video_data.update(self._parse_json(self._html_search_regex(
+            r'data-meta="([^"]*)"', webpage, 'metadata', default='{}'), video_id))
+
+        formats = []
+
+        if video_data.get('mpx_is_livestream') == '1':
+            live = True
+            player_id = traverse_obj(
+                video_data, 'mpx_m3upid', ('video', 'meta', 'mpx_m3upid'), 'mpx_pid',
+                ('video', 'meta', 'mpx_pid'), 'pid_streaming_web_medium')
+            query = {
+                'mbr': 'true',
+                'assetTypes': 'LegacyRelease',
+                'fwsitesection': fw_ssid,
+                'fwNetworkID': fw_network_id,
+                'pprofile': 'ots_desktop_html',
+                'sensitive': 'false',
+                'w': '1920',
+                'h': '1080',
+                'rnd': '1660303',
+                'mode': 'LIVE',
+                'format': 'SMIL',
+                'tracking': 'true',
+                'formats': 'M3U+none,MPEG-DASH+none,MPEG4,MP3',
+                'vpaid': 'script',
+                'schema': '2.0',
+                'SDK': 'PDK+6.1.3',
+            }
+            info = {
+                'title': f'{channel} livestream',
+            }
+
+        else:
+            live = False
+            player_id = traverse_obj(
+                video_data, ('video', 'meta', 'pid_streaming_web_high'), 'pid_streaming_web_high',
+                ('video', 'meta', 'mpx_pid'), 'mpx_pid')
+
+            date_string = traverse_obj(video_data, 'date_string', 'date_gmt')
+            if date_string:
+                date_string = self._search_regex(
+                    r'datetime="([^"]+)"', date_string, 'date string', fatal=False)
+            else:
+                date_string = traverse_obj(
+                    nbc_data, ('dataLayer', 'adobe', 'prop70'), ('dataLayer', 'adobe', 'eVar70'),
+                    ('dataLayer', 'adobe', 'eVar59'))
+
+            video_url = traverse_obj(video_data, ('video', 'meta', 'mp4_url'), 'mp4_url')
+            if video_url:
+                height = url_basename(video_url).split('-')[1].split('p')[0]
+                formats.append({
+                    'url': video_url,
+                    'ext': 'mp4',
+                    'width': int_or_none(self._RESOLUTIONS.get(height)),
+                    'height': int_or_none(height),
+                    'format_id': f'http-{height}',
+                })
+
+            query = {
+                'mbr': 'true',
+                'assetTypes': 'LegacyRelease',
+                'fwsitesection': fw_ssid,
+                'fwNetworkID': fw_network_id,
+                'format': 'redirect',
+                'manifest': 'm3u',
+                'Tracking': 'true',
+                'Embedded': 'true',
+                'formats': 'MPEG4',
+            }
+            info = {
+                'title': video_data.get('title') or traverse_obj(
+                    nbc_data, ('dataLayer', 'contenttitle'), ('dataLayer', 'title'),
+                    ('dataLayer', 'adobe', 'prop22'), ('dataLayer', 'id')),
+                'description': traverse_obj(video_data, 'summary', 'excerpt', 'video_hero_text'),
+                'upload_date': str_or_none(unified_strdate(date_string)),
+                'timestamp': int_or_none(unified_timestamp(date_string)),
+            }
+
+        if not player_id:
+            raise ExtractorError(
+                'No video player ID or livestream player ID found in webpage', expected=True)
+
+        headers = {'Origin': f'https://www.{channel}.com'}
+        manifest, urlh = self._download_webpage_handle(
+            f'https://link.theplatform.com/s/{pdk_acct}/{player_id}', video_id,
+            headers=headers, query=query, note='Downloading manifest')
+        if live:
+            manifest_url = self._search_regex(r'<video src="([^"]*)', manifest, 'manifest URL')
+        else:
+            manifest_url = urlh.geturl()
+
+        formats.extend(self._extract_m3u8_formats(
+            manifest_url, video_id, 'mp4', headers=headers, m3u8_id='hls',
+            fatal=live, live=live, errnote='No HLS formats found'))
+        self._sort_formats(formats)
+
+        return {
+            'id': str_or_none(video_id),
+            'channel': channel,
+            'uploader': str_or_none(nbc_data.get('on_air_name')),
+            'uploader_id': str_or_none(nbc_data.get('callLetters')),
+            'formats': formats,
+            'is_live': live,
+            **info,
         }