_ Git - youtube-dl/blob - youtube_dl/extractor/espn.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from .once import OnceIE
   7 from ..compat import compat_str
   8 from ..utils import (
   9     determine_ext,
  10     int_or_none,
  11     unified_timestamp,
  12 )
  13
  14
  15 class ESPNIE(OnceIE):
  16     _VALID_URL = r'''(?x)
  17                     https?://
  18                         (?:
  19                             (?:
  20                                 (?:
  21                                     (?:(?:\w+\.)+)?espn\.go|
  22                                     (?:www\.)?espn
  23                                 )\.com/
  24                                 (?:
  25                                     (?:
  26                                         video/(?:clip|iframe/twitter)|
  27                                         watch/player
  28                                     )
  29                                     (?:
  30                                         .*?\?.*?\bid=|
  31                                         /_/id/
  32                                     )
  33                                 )
  34                             )|
  35                             (?:www\.)espnfc\.(?:com|us)/(?:video/)?[^/]+/\d+/video/
  36                         )
  37                         (?P<id>\d+)
  38                     '''
  39
  40     _TESTS = [{
  41         'url': 'http://espn.go.com/video/clip?id=10365079',
  42         'info_dict': {
  43             'id': '10365079',
  44             'ext': 'mp4',
  45             'title': '30 for 30 Shorts: Judging Jewell',
  46             'description': 'md5:39370c2e016cb4ecf498ffe75bef7f0f',
  47             'timestamp': 1390936111,
  48             'upload_date': '20140128',
  49         },
  50         'params': {
  51             'skip_download': True,
  52         },
  53     }, {
  54         'url': 'https://broadband.espn.go.com/video/clip?id=18910086',
  55         'info_dict': {
  56             'id': '18910086',
  57             'ext': 'mp4',
  58             'title': 'Kyrie spins around defender for two',
  59             'description': 'md5:2b0f5bae9616d26fba8808350f0d2b9b',
  60             'timestamp': 1489539155,
  61             'upload_date': '20170315',
  62         },
  63         'params': {
  64             'skip_download': True,
  65         },
  66         'expected_warnings': ['Unable to download f4m manifest'],
  67     }, {
  68         'url': 'http://nonredline.sports.espn.go.com/video/clip?id=19744672',
  69         'only_matching': True,
  70     }, {
  71         'url': 'https://cdn.espn.go.com/video/clip/_/id/19771774',
  72         'only_matching': True,
  73     }, {
  74         'url': 'http://www.espn.com/watch/player?id=19141491',
  75         'only_matching': True,
  76     }, {
  77         'url': 'http://www.espn.com/watch/player?bucketId=257&id=19505875',
  78         'only_matching': True,
  79     }, {
  80         'url': 'http://www.espn.com/watch/player/_/id/19141491',
  81         'only_matching': True,
  82     }, {
  83         'url': 'http://www.espn.com/video/clip?id=10365079',
  84         'only_matching': True,
  85     }, {
  86         'url': 'http://www.espn.com/video/clip/_/id/17989860',
  87         'only_matching': True,
  88     }, {
  89         'url': 'https://espn.go.com/video/iframe/twitter/?cms=espn&id=10365079',
  90         'only_matching': True,
  91     }, {
  92         'url': 'http://www.espnfc.us/video/espn-fc-tv/86/video/3319154/nashville-unveiled-as-the-newest-club-in-mls',
  93         'only_matching': True,
  94     }, {
  95         'url': 'http://www.espnfc.com/english-premier-league/23/video/3324163/premier-league-in-90-seconds-golden-tweets',
  96         'only_matching': True,
  97     }]
  98
  99     def _real_extract(self, url):
 100         video_id = self._match_id(url)
 101
 102         clip = self._download_json(
 103             'http://api-app.espn.com/v1/video/clips/%s' % video_id,
 104             video_id)['videos'][0]
 105
 106         title = clip['headline']
 107
 108         format_urls = set()
 109         formats = []
 110
 111         def traverse_source(source, base_source_id=None):
 112             for source_id, source in source.items():
 113                 if source_id == 'alert':
 114                     continue
 115                 elif isinstance(source, compat_str):
 116                     extract_source(source, base_source_id)
 117                 elif isinstance(source, dict):
 118                     traverse_source(
 119                         source,
 120                         '%s-%s' % (base_source_id, source_id)
 121                         if base_source_id else source_id)
 122
 123         def extract_source(source_url, source_id=None):
 124             if source_url in format_urls:
 125                 return
 126             format_urls.add(source_url)
 127             ext = determine_ext(source_url)
 128             if OnceIE.suitable(source_url):
 129                 formats.extend(self._extract_once_formats(source_url))
 130             elif ext == 'smil':
 131                 formats.extend(self._extract_smil_formats(
 132                     source_url, video_id, fatal=False))
 133             elif ext == 'f4m':
 134                 formats.extend(self._extract_f4m_formats(
 135                     source_url, video_id, f4m_id=source_id, fatal=False))
 136             elif ext == 'm3u8':
 137                 formats.extend(self._extract_m3u8_formats(
 138                     source_url, video_id, 'mp4', entry_protocol='m3u8_native',
 139                     m3u8_id=source_id, fatal=False))
 140             else:
 141                 f = {
 142                     'url': source_url,
 143                     'format_id': source_id,
 144                 }
 145                 mobj = re.search(r'(\d+)p(\d+)_(\d+)k\.', source_url)
 146                 if mobj:
 147                     f.update({
 148                         'height': int(mobj.group(1)),
 149                         'fps': int(mobj.group(2)),
 150                         'tbr': int(mobj.group(3)),
 151                     })
 152                 if source_id == 'mezzanine':
 153                     f['preference'] = 1
 154                 formats.append(f)
 155
 156         links = clip.get('links', {})
 157         traverse_source(links.get('source', {}))
 158         traverse_source(links.get('mobile', {}))
 159         self._sort_formats(formats)
 160
 161         description = clip.get('caption') or clip.get('description')
 162         thumbnail = clip.get('thumbnail')
 163         duration = int_or_none(clip.get('duration'))
 164         timestamp = unified_timestamp(clip.get('originalPublishDate'))
 165
 166         return {
 167             'id': video_id,
 168             'title': title,
 169             'description': description,
 170             'thumbnail': thumbnail,
 171             'timestamp': timestamp,
 172             'duration': duration,
 173             'formats': formats,
 174         }
 175
 176
 177 class ESPNArticleIE(InfoExtractor):
 178     _VALID_URL = r'https?://(?:espn\.go|(?:www\.)?espn)\.com/(?:[^/]+/)*(?P<id>[^/]+)'
 179     _TESTS = [{
 180         'url': 'http://espn.go.com/nba/recap?gameId=400793786',
 181         'only_matching': True,
 182     }, {
 183         'url': 'http://espn.go.com/blog/golden-state-warriors/post/_/id/593/how-warriors-rapidly-regained-a-winning-edge',
 184         'only_matching': True,
 185     }, {
 186         'url': 'http://espn.go.com/sports/endurance/story/_/id/12893522/dzhokhar-tsarnaev-sentenced-role-boston-marathon-bombings',
 187         'only_matching': True,
 188     }, {
 189         'url': 'http://espn.go.com/nba/playoffs/2015/story/_/id/12887571/john-wall-washington-wizards-no-swelling-left-hand-wrist-game-5-return',
 190         'only_matching': True,
 191     }]
 192
 193     @classmethod
 194     def suitable(cls, url):
 195         return False if ESPNIE.suitable(url) else super(ESPNArticleIE, cls).suitable(url)
 196
 197     def _real_extract(self, url):
 198         video_id = self._match_id(url)
 199
 200         webpage = self._download_webpage(url, video_id)
 201
 202         video_id = self._search_regex(
 203             r'class=(["\']).*?video-play-button.*?\1[^>]+data-id=["\'](?P<id>\d+)',
 204             webpage, 'video id', group='id')
 205
 206         return self.url_result(
 207             'http://espn.go.com/video/clip?id=%s' % video_id, ESPNIE.ie_key())
 208
 209
 210 class FiveThirtyEightIE(InfoExtractor):
 211     _VALID_URL = r'https?://(?:www\.)?fivethirtyeight\.com/features/(?P<id>[^/?#]+)'
 212     _TEST = {
 213         'url': 'http://fivethirtyeight.com/features/how-the-6-8-raiders-can-still-make-the-playoffs/',
 214         'info_dict': {
 215             'id': '21846851',
 216             'ext': 'mp4',
 217             'title': 'FiveThirtyEight: The Raiders can still make the playoffs',
 218             'description': 'Neil Paine breaks down the simplest scenario that will put the Raiders into the playoffs at 8-8.',
 219             'timestamp': 1513960621,
 220             'upload_date': '20171222',
 221         },
 222         'params': {
 223             'skip_download': True,
 224         },
 225         'expected_warnings': ['Unable to download f4m manifest'],
 226     }
 227
 228     def _real_extract(self, url):
 229         video_id = self._match_id(url)
 230
 231         webpage = self._download_webpage(url, video_id)
 232
 233         video_id = self._search_regex(
 234             r'data-video-id=["\'](?P<id>\d+)',
 235             webpage, 'video id', group='id')
 236
 237         return self.url_result(
 238             'http://espn.go.com/video/clip?id=%s' % video_id, ESPNIE.ie_key())