From 8fc226ef994a82f7b1050cdb72ec38922d3ab9cf Mon Sep 17 00:00:00 2001 From: remitamine Date: Fri, 2 Oct 2015 17:24:30 +0100 Subject: [PATCH] [nba] extract all video formats and extract more info --- youtube_dl/extractor/__init__.py | 5 +- youtube_dl/extractor/nba.py | 102 +++++++++++++++++++++---------- 2 files changed, 74 insertions(+), 33 deletions(-) diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py index a73a1317eb..78478b38be 100644 --- a/youtube_dl/extractor/__init__.py +++ b/youtube_dl/extractor/__init__.py @@ -351,7 +351,10 @@ from .myvideo import MyVideoIE from .myvidster import MyVidsterIE from .nationalgeographic import NationalGeographicIE from .naver import NaverIE -from .nba import NBAIE +from .nba import ( + NBAIE, + NBAWatchIE, +) from .nbc import ( NBCIE, NBCNewsIE, diff --git a/youtube_dl/extractor/nba.py b/youtube_dl/extractor/nba.py index 944096e1ca..36ece5b64b 100644 --- a/youtube_dl/extractor/nba.py +++ b/youtube_dl/extractor/nba.py @@ -2,62 +2,100 @@ from __future__ import unicode_literals from .common import InfoExtractor from ..utils import ( - remove_end, parse_duration, + parse_iso8601, + int_or_none, ) -class NBAIE(InfoExtractor): - _VALID_URL = r'https?://(?:watch\.|www\.)?nba\.com/(?:nba/)?video(?P/[^?]*?)/?(?:/index\.html)?(?:\?.*)?$' +class NBABaseIE(InfoExtractor): + def _get_formats(self, video_id): + base_url = 'http://nba.cdn.turner.com/nba/big%s' % video_id + return [{ + 'url': base_url + '_nba_android_high.mp4', + 'width': 480, + 'height': 320, + 'format_id': '320p', + },{ + 'url': base_url + '_640x360_664b.mp4', + 'width': 640, + 'height': 360, + 'format_id': '360p', + },{ + 'url': base_url + '_768x432_1404.mp4', + 'width': 768, + 'height': 432, + 'format_id': '432p', + },{ + 'url': base_url + '_1280x720.mp4', + 'width': 1280, + 'height': 720, + 'format_id': '720p', + }] + + def _real_extract(self, url): + video_id = self._match_id(url) + webpage = self._download_webpage(url, video_id) + ret = self._extract_metadata(webpage, video_id) + ret['id'] = video_id.rpartition('/')[2] + ret['formats'] = self._get_formats(video_id) + return ret + + +class NBAIE(NBABaseIE): + IE_NAME = 'nba' + _VALID_URL = r'https?://(?:www\.)?nba\.com/(?:nba/)?video(?P/[^?]*?)/?(?:/index\.html)?(?:\?.*)?$' _TESTS = [{ 'url': 'http://www.nba.com/video/games/nets/2012/12/04/0021200253-okc-bkn-recap.nba/index.html', - 'md5': 'c0edcfc37607344e2ff8f13c378c88a4', + 'md5': '9d902940d2a127af3f7f9d2f3dc79c96', 'info_dict': { 'id': '0021200253-okc-bkn-recap.nba', 'ext': 'mp4', 'title': 'Thunder vs. Nets', 'description': 'Kevin Durant scores 32 points and dishes out six assists as the Thunder beat the Nets in Brooklyn.', 'duration': 181, + 'timestamp': 1354680189, + 'upload_date': '20121205', }, }, { 'url': 'http://www.nba.com/video/games/hornets/2014/12/05/0021400276-nyk-cha-play5.nba/', 'only_matching': True, - }, { + }] + + def _extract_metadata(self, webpage, video_id): + return { + 'title': self._html_search_meta('name', webpage), + 'description': self._html_search_meta('description', webpage), + 'duration': parse_duration(self._html_search_meta('duration', webpage)), + 'thumbnail': self._html_search_meta('thumbnailUrl', webpage), + 'timestamp': parse_iso8601(self._html_search_meta('uploadDate', webpage)) + } + +class NBAWatchIE(NBABaseIE): + IE_NAME = 'nba:watch' + _VALID_URL = r'https?://watch.nba\.com/(?:nba/)?video(?P/[^?]*?)/?(?:/index\.html)?(?:\?.*)?$' + _TESTS = [{ 'url': 'http://watch.nba.com/nba/video/channels/playoffs/2015/05/20/0041400301-cle-atl-recap.nba', + 'md5': 'b2b39b81cf28615ae0c3360a3f9668c4', 'info_dict': { 'id': '0041400301-cle-atl-recap.nba', 'ext': 'mp4', - 'title': 'NBA GAME TIME | Video: Hawks vs. Cavaliers Game 1', + 'title': 'Hawks vs. Cavaliers Game 1', 'description': 'md5:8094c3498d35a9bd6b1a8c396a071b4d', 'duration': 228, - }, - 'params': { - 'skip_download': True, + 'timestamp': 1432094400, + 'upload_date': '20150520', } }] - def _real_extract(self, url): - video_id = self._match_id(url) - webpage = self._download_webpage(url, video_id) - - video_url = 'http://ht-mobile.cdn.turner.com/nba/big' + video_id + '_nba_1280x720.mp4' - - shortened_video_id = video_id.rpartition('/')[2] - title = remove_end( - self._og_search_title(webpage, default=shortened_video_id), ' : NBA.com') - - description = self._og_search_description(webpage) - duration_str = self._html_search_meta( - 'duration', webpage, 'duration', default=None) - if not duration_str: - duration_str = self._html_search_regex( - r'Duration:\s*(\d+:\d+)', webpage, 'duration', fatal=False) - duration = parse_duration(duration_str) - + def _extract_metadata(self, webpage, video_id): + program_id = self._search_regex(r'var\s+programId\s*=\s*(\d+);', webpage, 'program id') + metadata = self._download_json( + 'http://smbsolr.cdnak.neulion.com/solr_nbav6/nba/nba/mlt/?wt=json&fl=name,description,image,runtime,releaseDate&q=sequence%3A' + program_id, video_id)['match']['docs'][0] return { - 'id': shortened_video_id, - 'url': video_url, - 'title': title, - 'description': description, - 'duration': duration, + 'title': metadata['name'], + 'description': metadata.get('description'), + 'duration': int_or_none(metadata.get('runtime')), + 'thumbnail': metadata.get('image'), + 'timestamp': parse_iso8601(metadata.get('releaseDate')) }