1 from __future__
import unicode_literals
7 from .common
import InfoExtractor
22 class NBAIE(InfoExtractor
):
23 _VALID_URL
= r
'https?://(?:watch\.|www\.)?nba\.com/(?P<path>(?:[^/]+/)+(?P<id>[^?]*?))/?(?:/index\.html)?(?:\?.*)?$'
25 'url': 'http://www.nba.com/video/games/nets/2012/12/04/0021200253-okc-bkn-recap.nba/index.html',
26 'md5': '9e7729d3010a9c71506fd1248f74e4f4',
28 'id': '0021200253-okc-bkn-recap',
30 'title': 'Thunder vs. Nets',
31 'description': 'Kevin Durant scores 32 points and dishes out six assists as the Thunder beat the Nets in Brooklyn.',
33 'timestamp': 1354638466,
34 'upload_date': '20121204',
38 'skip_download': True,
41 'url': 'http://www.nba.com/video/games/hornets/2014/12/05/0021400276-nyk-cha-play5.nba/',
42 'only_matching': True,
44 'url': 'http://watch.nba.com/video/channels/playoffs/2015/05/20/0041400301-cle-atl-recap.nba',
45 'md5': 'b2b39b81cf28615ae0c3360a3f9668c4',
47 'id': '0041400301-cle-atl-recap',
49 'title': 'Hawks vs. Cavaliers Game 1',
50 'description': 'md5:8094c3498d35a9bd6b1a8c396a071b4d',
52 'timestamp': 1432134543,
53 'upload_date': '20150520',
56 'url': 'http://www.nba.com/clippers/news/doc-rivers-were-not-trading-blake',
58 'id': '1455672027478-Doc_Feb16_720',
60 'title': 'Practice: Doc Rivers - 2/16/16',
61 'description': 'Head Coach Doc Rivers addresses the media following practice.',
62 'upload_date': '20160217',
63 'timestamp': 1455672000,
67 'skip_download': True,
70 'url': 'http://www.nba.com/timberwolves/wiggins-shootaround#',
73 'title': 'Shootaround Access - Dec. 12 | Andrew Wiggins',
77 # Download the whole playlist takes too long time
78 'playlist_items': '1-30',
81 'url': 'http://www.nba.com/timberwolves/wiggins-shootaround#',
85 'title': 'Shootaround Access - Dec. 12 | Andrew Wiggins',
86 'description': 'Wolves rookie Andrew Wiggins addresses the media after Friday\'s shootaround.',
87 'upload_date': '20141212',
88 'timestamp': 1418418600,
93 'skip_download': True,
99 def _fetch_page(self
, team
, video_id
, page
):
100 search_url
= 'http://searchapp2.nba.com/nba-search/query.jsp?' + compat_urllib_parse
.urlencode({
102 'start': page
* self
._PAGE
_SIZE
+ 1,
103 'npp': (page
+ 1) * self
._PAGE
_SIZE
+ 1,
108 results
= self
._download
_json
(
109 search_url
, video_id
, note
='Download page %d of playlist data' % page
)['results'][0]
111 yield self
.url_result(compat_urlparse
.urljoin('http://www.nba.com/', item
['url']))
113 def _extract_playlist(self
, orig_path
, video_id
, webpage
):
114 team
= orig_path
.split('/')[0]
116 if self
._downloader
.params
.get('noplaylist'):
117 self
.to_screen('Downloading just video because of --no-playlist')
118 video_path
= self
._search
_regex
(
119 r
'nbaVideoCore\.firstVideo\s*=\s*\'([^
\']+)\';', webpage, 'video path
')
120 video_url = 'http
://www
.nba
.com
/%s/video
/%s' % (team, video_path)
121 return self.url_result(video_url)
123 self.to_screen('Downloading playlist
- add
--no
-playlist to just download video
')
124 playlist_title = self._og_search_title(webpage, fatal=False)
125 entries = OnDemandPagedList(
126 functools.partial(self._fetch_page, team, video_id),
127 self._PAGE_SIZE, use_cache=True)
129 return self.playlist_result(entries, team, playlist_title)
131 def _real_extract(self, url):
132 path, video_id = re.match(self._VALID_URL, url).groups()
134 if path.startswith('nba
/'):
137 if 'video
/' not in path:
138 webpage = self._download_webpage(url, video_id)
139 path = remove_start(self._search_regex(r'data
-videoid
="([^"]+)"', webpage, 'video id'), '/')
142 return self._extract_playlist(orig_path, video_id, webpage)
144 # See prepareContentId() of pkgCvp.js
145 if path.startswith('video/teams'):
146 path = 'video/channels/proxy/' + path[6:]
148 video_info = self._download_xml('http://www.nba.com/%s.xml' % path, video_id)
149 video_id = os.path.splitext(xpath_text(video_info, 'slug'))[0]
150 title = xpath_text(video_info, 'headline')
151 description = xpath_text(video_info, 'description')
152 duration = parse_duration(xpath_text(video_info, 'length'))
153 timestamp = int_or_none(xpath_attr(video_info, 'dateCreated', 'uts'))
156 for image in video_info.find('images'):
158 'id': image.attrib.get('cut'),
160 'width': int_or_none(image.attrib.get('width')),
161 'height': int_or_none(image.attrib.get('height')),
165 for video_file in video_info.findall('.//file'):
166 video_url = video_file.text
167 if video_url.startswith('/'):
169 if video_url.endswith('.m3u8'):
170 formats.extend(self._extract_m3u8_formats(video_url, video_id, ext='mp4', m3u8_id='hls', fatal=False))
171 elif video_url.endswith('.f4m'):
172 formats.extend(self._extract_f4m_formats(video_url + '?hdcore=3.4.1.1', video_id, f4m_id='hds', fatal=False))
174 key = video_file.attrib.get('bitrate')
179 mobj = re.search(r'(\d+)x(\d+)(?:_(\d+))?', key)
182 'width': int(mobj.group(1)),
183 'height': int(mobj.group(2)),
184 'tbr': int_or_none(mobj.group(3)),
186 formats.append(format_info)
187 self._sort_formats(formats)
192 'description': description,
193 'duration': duration,
194 'timestamp': timestamp,
195 'thumbnails': thumbnails,