X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fnba.py;h=9d26030d3a43e11b0430aecbc5506ab10a1e5973;hb=ac25992bc74dec638cb33e8da088c325a03036a5;hp=a0cc58c126857693a3e75a0473f6aed8b682b372;hpb=6a11bb77baf9f70da76f2595b74061b31223d4ff;p=youtube-dl diff --git a/youtube_dl/extractor/nba.py b/youtube_dl/extractor/nba.py index a0cc58c12..9d26030d3 100644 --- a/youtube_dl/extractor/nba.py +++ b/youtube_dl/extractor/nba.py @@ -6,6 +6,8 @@ from .common import InfoExtractor from ..utils import ( parse_duration, int_or_none, + xpath_text, + xpath_attr, ) @@ -26,7 +28,7 @@ class NBAIE(InfoExtractor): }, { 'url': 'http://www.nba.com/video/games/hornets/2014/12/05/0021400276-nyk-cha-play5.nba/', 'only_matching': True, - },{ + }, { 'url': 'http://watch.nba.com/video/channels/playoffs/2015/05/20/0041400301-cle-atl-recap.nba', 'md5': 'b2b39b81cf28615ae0c3360a3f9668c4', 'info_dict': { @@ -42,12 +44,14 @@ class NBAIE(InfoExtractor): def _real_extract(self, url): path, video_id = re.match(self._VALID_URL, url).groups() + if path.startswith('nba/'): + path = path[3:] video_info = self._download_xml('http://www.nba.com/%s.xml' % path, video_id) - video_id = video_info.find('slug').text - title = video_info.find('headline').text - description = video_info.find('description').text - duration = parse_duration(video_info.find('length').text) - timestamp = int_or_none(video_info.find('dateCreated').attrib.get('uts')) + video_id = xpath_text(video_info, 'slug') + title = xpath_text(video_info, 'headline') + description = xpath_text(video_info, 'description') + duration = parse_duration(xpath_text(video_info, 'length')) + timestamp = int_or_none(xpath_attr(video_info, 'dateCreated', 'uts')) thumbnails = [] for image in video_info.find('images'): @@ -59,24 +63,28 @@ class NBAIE(InfoExtractor): }) formats = [] - for video_file in video_info.find('files').iter('file'): + for video_file in video_info.findall('.//file'): video_url = video_file.text if video_url.startswith('/'): continue if video_url.endswith('.m3u8'): - formats.extend(self._extract_m3u8_formats(video_url, video_id, m3u8_id='hls')) + formats.extend(self._extract_m3u8_formats(video_url, video_id, m3u8_id='hls', fatal=False)) elif video_url.endswith('.f4m'): - formats.extend(self._extract_f4m_formats(video_url + '?hdcore=3.4.1.1', video_id, f4m_id='hds')) + formats.extend(self._extract_f4m_formats(video_url + '?hdcore=3.4.1.1', video_id, f4m_id='hds', fatal=False)) else: key = video_file.attrib.get('bitrate') - width, height, bitrate = re.search(r'(\d+)x(\d+)(?:_(\d+))?', key).groups() - formats.append({ + format_info = { 'format_id': key, 'url': video_url, - 'width': int_or_none(width), - 'height': int_or_none(height), - 'tbr': int_or_none(bitrate), - }) + } + mobj = re.search(r'(\d+)x(\d+)(?:_(\d+))?', key) + if mobj: + format_info.update({ + 'width': int(mobj.group(1)), + 'height': int(mobj.group(2)), + 'tbr': int_or_none(mobj.group(3)), + }) + formats.append(format_info) self._sort_formats(formats) return {