[extractor/nbc] Add NBCStations extractor (#5077)

Closes #4571
Authored by: bashonly

yt_dlp/extractor/_extractors.py

@@ -1079,6 +1079,7 @@ from .nbc import (
     NBCSportsIE,
     NBCSportsStreamIE,
     NBCSportsVPlayerIE,
+    NBCStationsIE,
 )
 from .ndr import (
     NDRIE,

yt_dlp/extractor/nbc.py

@@ -7,14 +7,20 @@ from .theplatform import ThePlatformIE
 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,
 )
 
 
@@ -584,3 +590,169 @@ class NBCOlympicsStreamIE(AdobePassIE):
             'formats': formats,
             '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*=\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,
+        }