[nfl] fix extraction(closes #22245)
authorRemita Amine <remitamine@gmail.com>
Tue, 15 Dec 2020 15:06:10 +0000 (16:06 +0100)
committerRemita Amine <remitamine@gmail.com>
Tue, 15 Dec 2020 15:06:10 +0000 (16:06 +0100)
youtube_dl/extractor/extractors.py
youtube_dl/extractor/nfl.py

index d0fc56b79d8fbf454171a34c2056427f7d8a24c9..18b7168ffa420d6167b41b8cfabf5b8777a37496 100644 (file)
@@ -724,7 +724,10 @@ from .nexx import (
     NexxIE,
     NexxEmbedIE,
 )
-from .nfl import NFLIE
+from .nfl import (
+    NFLIE,
+    NFLArticleIE,
+)
 from .nhk import (
     NhkVodIE,
     NhkVodProgramIE,
index 460deb162df7994caa389b1f37c4174cec3fbf78..e234fad38b12715d34e8741a83495e0cf181f2c6 100644 (file)
@@ -4,19 +4,15 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..compat import (
-    compat_urllib_parse_urlparse,
-)
 from ..utils import (
-    ExtractorError,
-    int_or_none,
-    remove_end,
+    clean_html,
+    determine_ext,
+    get_element_by_class,
 )
 
 
-class NFLIE(InfoExtractor):
-    IE_NAME = 'nfl.com'
-    _VALID_URL = r'''(?x)
+class NFLBaseIE(InfoExtractor):
+    _VALID_URL_BASE = r'''(?x)
                     https?://
                         (?P<host>
                             (?:www\.)?
@@ -34,15 +30,15 @@ class NFLIE(InfoExtractor):
                                     houstontexans|
                                     colts|
                                     jaguars|
-                                    titansonline|
+                                    (?:titansonline|tennesseetitans)|
                                     denverbroncos|
-                                    kcchiefs|
+                                    (?:kc)?chiefs|
                                     raiders|
                                     chargers|
                                     dallascowboys|
                                     giants|
                                     philadelphiaeagles|
-                                    redskins|
+                                    (?:redskins|washingtonfootball)|
                                     chicagobears|
                                     detroitlions|
                                     packers|
@@ -52,180 +48,112 @@ class NFLIE(InfoExtractor):
                                     neworleanssaints|
                                     buccaneers|
                                     azcardinals|
-                                    stlouisrams|
+                                    (?:stlouis|the)rams|
                                     49ers|
                                     seahawks
                                 )\.com|
                                 .+?\.clubs\.nfl\.com
                             )
                         )/
-                        (?:.+?/)*
-                        (?P<id>[^/#?&]+)
                     '''
+    _VIDEO_CONFIG_REGEX = r'<script[^>]+id="[^"]*video-config-[0-9a-f]{8}-(?:[0-9a-f]{4}-){3}[0-9a-f]{12}[^"]*"[^>]*>\s*({.+})'
+
+    def _parse_video_config(self, video_config, display_id):
+        video_config = self._parse_json(video_config, display_id)
+        item = video_config['playlist'][0]
+        mcp_id = item.get('mcpID')
+        if mcp_id:
+            info = self.url_result(
+                'anvato:GXvEgwyJeWem8KCYXfeoHWknwP48Mboj:' + mcp_id,
+                'Anvato', mcp_id)
+        else:
+            media_id = item.get('id') or item['entityId']
+            title = item['title']
+            item_url = item['url']
+            info = {'id': media_id}
+            ext = determine_ext(item_url)
+            if ext == 'm3u8':
+                info['formats'] = self._extract_m3u8_formats(item_url, media_id, 'mp4')
+                self._sort_formats(info['formats'])
+            else:
+                info['url'] = item_url
+                if item.get('audio') is True:
+                    info['vcodec'] = 'none'
+            is_live = video_config.get('live') is True
+            thumbnails = None
+            image_url = item.get(item.get('imageSrc')) or item.get(item.get('posterImage'))
+            if image_url:
+                thumbnails = [{
+                    'url': image_url,
+                    'ext': determine_ext(image_url, 'jpg'),
+                }]
+            info.update({
+                'title': self._live_title(title) if is_live else title,
+                'is_live': is_live,
+                'description': clean_html(item.get('description')),
+                'thumbnails': thumbnails,
+            })
+        return info
+
+
+class NFLIE(NFLBaseIE):
+    IE_NAME = 'nfl.com'
+    _VALID_URL = NFLBaseIE._VALID_URL_BASE + r'(?:videos?|listen|audio)/(?P<id>[^/#?&]+)'
     _TESTS = [{
-        'url': 'http://www.nfl.com/videos/nfl-game-highlights/0ap3000000398478/Week-3-Redskins-vs-Eagles-highlights',
-        'md5': '394ef771ddcd1354f665b471d78ec4c6',
+        'url': 'https://www.nfl.com/videos/baker-mayfield-s-game-changing-plays-from-3-td-game-week-14',
         'info_dict': {
-            'id': '0ap3000000398478',
+            'id': '899441',
             'ext': 'mp4',
-            'title': 'Week 3: Redskins vs. Eagles highlights',
-            'description': 'md5:56323bfb0ac4ee5ab24bd05fdf3bf478',
-            'upload_date': '20140921',
-            'timestamp': 1411337580,
+            'title': "Baker Mayfield's game-changing plays from 3-TD game Week 14",
+            'description': 'md5:85e05a3cc163f8c344340f220521136d',
+            'upload_date': '20201215',
+            'timestamp': 1608009755,
             'thumbnail': r're:^https?://.*\.jpg$',
+            'uploader': 'NFL',
         }
     }, {
-        'url': 'http://prod.www.steelers.clubs.nfl.com/video-and-audio/videos/LIVE_Post_Game_vs_Browns/9d72f26a-9e2b-4718-84d3-09fb4046c266',
-        'md5': 'cf85bdb4bc49f6e9d3816d130c78279c',
+        'url': 'https://www.chiefs.com/listen/patrick-mahomes-travis-kelce-react-to-win-over-dolphins-the-breakdown',
+        'md5': '6886b32c24b463038c760ceb55a34566',
         'info_dict': {
-            'id': '9d72f26a-9e2b-4718-84d3-09fb4046c266',
-            'ext': 'mp4',
-            'title': 'LIVE: Post Game vs. Browns',
-            'description': 'md5:6a97f7e5ebeb4c0e69a418a89e0636e8',
-            'upload_date': '20131229',
-            'timestamp': 1388354455,
-            'thumbnail': r're:^https?://.*\.jpg$',
+            'id': 'd87e8790-3e14-11eb-8ceb-ff05c2867f99',
+            'ext': 'mp3',
+            'title': 'Patrick Mahomes, Travis Kelce React to Win Over Dolphins | The Breakdown',
+            'description': 'md5:12ada8ee70e6762658c30e223e095075',
         }
     }, {
-        'url': 'http://www.nfl.com/news/story/0ap3000000467586/article/patriots-seahawks-involved-in-lategame-skirmish',
-        'info_dict': {
-            'id': '0ap3000000467607',
-            'ext': 'mp4',
-            'title': 'Frustrations flare on the field',
-            'description': 'Emotions ran high at the end of the Super Bowl on both sides of the ball after a dramatic finish.',
-            'timestamp': 1422850320,
-            'upload_date': '20150202',
-        },
-    }, {
-        'url': 'http://www.patriots.com/video/2015/09/18/10-days-gillette',
-        'md5': '4c319e2f625ffd0b481b4382c6fc124c',
-        'info_dict': {
-            'id': 'n-238346',
-            'ext': 'mp4',
-            'title': '10 Days at Gillette',
-            'description': 'md5:8cd9cd48fac16de596eadc0b24add951',
-            'timestamp': 1442618809,
-            'upload_date': '20150918',
-        },
-    }, {
-        # lowercase data-contentid
-        'url': 'http://www.steelers.com/news/article-1/Tomlin-on-Ben-getting-Vick-ready/56399c96-4160-48cf-a7ad-1d17d4a3aef7',
-        'info_dict': {
-            'id': '12693586-6ea9-4743-9c1c-02c59e4a5ef2',
-            'ext': 'mp4',
-            'title': 'Tomlin looks ahead to Ravens on a short week',
-            'description': 'md5:32f3f7b139f43913181d5cbb24ecad75',
-            'timestamp': 1443459651,
-            'upload_date': '20150928',
-        },
-        'params': {
-            'skip_download': True,
-        },
-    }, {
-        'url': 'http://www.nfl.com/videos/nfl-network-top-ten/09000d5d810a6bd4/Top-10-Gutsiest-Performances-Jack-Youngblood',
+        'url': 'https://www.buffalobills.com/video/buffalo-bills-military-recognition-week-14',
         'only_matching': True,
     }, {
-        'url': 'http://www.buffalobills.com/video/videos/Rex_Ryan_Show_World_Wide_Rex/b1dcfab2-3190-4bb1-bfc0-d6e603d6601a',
+        'url': 'https://www.raiders.com/audio/instant-reactions-raiders-week-14-loss-to-indianapolis-colts-espn-jason-fitz',
         'only_matching': True,
     }]
 
-    @staticmethod
-    def prepend_host(host, url):
-        if not url.startswith('http'):
-            if not url.startswith('/'):
-                url = '/%s' % url
-            url = 'http://{0:}{1:}'.format(host, url)
-        return url
-
-    @staticmethod
-    def format_from_stream(stream, protocol, host, path_prefix='',
-                           preference=0, note=None):
-        url = '{protocol:}://{host:}/{prefix:}{path:}'.format(
-            protocol=protocol,
-            host=host,
-            prefix=path_prefix,
-            path=stream.get('path'),
-        )
-        return {
-            'url': url,
-            'vbr': int_or_none(stream.get('rate', 0), 1000),
-            'preference': preference,
-            'format_note': note,
-        }
-
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id, host = mobj.group('id'), mobj.group('host')
-
-        webpage = self._download_webpage(url, video_id)
-
-        config_url = NFLIE.prepend_host(host, self._search_regex(
-            r'(?:(?:config|configURL)\s*:\s*|<nflcs:avplayer[^>]+data-config\s*=\s*)(["\'])(?P<config>.+?)\1',
-            webpage, 'config URL', default='static/content/static/config/video/config.json',
-            group='config'))
-        # For articles, the id in the url is not the video id
-        video_id = self._search_regex(
-            r'(?:<nflcs:avplayer[^>]+data-content[Ii]d\s*=\s*|content[Ii]d\s*:\s*)(["\'])(?P<id>(?:(?!\1).)+)\1',
-            webpage, 'video id', default=video_id, group='id')
-        config = self._download_json(config_url, video_id, 'Downloading player config')
-        url_template = NFLIE.prepend_host(
-            host, '{contentURLTemplate:}'.format(**config))
-        video_data = self._download_json(
-            url_template.format(id=video_id), video_id)
-
-        formats = []
-        cdn_data = video_data.get('cdnData', {})
-        streams = cdn_data.get('bitrateInfo', [])
-        if cdn_data.get('format') == 'EXTERNAL_HTTP_STREAM':
-            parts = compat_urllib_parse_urlparse(cdn_data.get('uri'))
-            protocol, host = parts.scheme, parts.netloc
-            for stream in streams:
-                formats.append(
-                    NFLIE.format_from_stream(stream, protocol, host))
-        else:
-            cdns = config.get('cdns')
-            if not cdns:
-                raise ExtractorError('Failed to get CDN data', expected=True)
-
-            for name, cdn in cdns.items():
-                # LimeLight streams don't seem to work
-                if cdn.get('name') == 'LIMELIGHT':
-                    continue
+        display_id = self._match_id(url)
+        webpage = self._download_webpage(url, display_id)
+        return self._parse_video_config(self._search_regex(
+            self._VIDEO_CONFIG_REGEX, webpage, 'video config'), display_id)
 
-                protocol = cdn.get('protocol')
-                host = remove_end(cdn.get('host', ''), '/')
-                if not (protocol and host):
-                    continue
 
-                prefix = cdn.get('pathprefix', '')
-                if prefix and not prefix.endswith('/'):
-                    prefix = '%s/' % prefix
-
-                preference = 0
-                if protocol == 'rtmp':
-                    preference = -2
-                elif 'prog' in name.lower():
-                    preference = 1
-
-                for stream in streams:
-                    formats.append(
-                        NFLIE.format_from_stream(stream, protocol, host,
-                                                 prefix, preference, name))
-
-        self._sort_formats(formats)
-
-        thumbnail = None
-        for q in ('xl', 'l', 'm', 's', 'xs'):
-            thumbnail = video_data.get('imagePaths', {}).get(q)
-            if thumbnail:
-                break
+class NFLArticleIE(NFLBaseIE):
+    IE_NAME = 'nfl.com:article'
+    _VALID_URL = NFLBaseIE._VALID_URL_BASE + r'news/(?P<id>[^/#?&]+)'
+    _TEST = {
+        'url': 'https://www.buffalobills.com/news/the-only-thing-we-ve-earned-is-the-noise-bills-coaches-discuss-handling-rising-e',
+        'info_dict': {
+            'id': 'the-only-thing-we-ve-earned-is-the-noise-bills-coaches-discuss-handling-rising-e',
+            'title': "'The only thing we've earned is the noise' | Bills coaches discuss handling rising expectations",
+        },
+        'playlist_count': 4,
+    }
 
-        return {
-            'id': video_id,
-            'title': video_data.get('headline'),
-            'formats': formats,
-            'description': video_data.get('caption'),
-            'duration': video_data.get('duration'),
-            'thumbnail': thumbnail,
-            'timestamp': int_or_none(video_data.get('posted'), 1000),
-        }
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+        webpage = self._download_webpage(url, display_id)
+        entries = []
+        for video_config in re.findall(self._VIDEO_CONFIG_REGEX, webpage):
+            entries.append(self._parse_video_config(video_config, display_id))
+        title = clean_html(get_element_by_class(
+            'nfl-c-article__title', webpage)) or self._html_search_meta(
+            ['og:title', 'twitter:title'], webpage)
+        return self.playlist_result(entries, display_id, title)