R-yan/l1ving_youtube-dl
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

[espn] fix info extraction

Browse Source
This commit is contained in:
remitamine 2015-10-09 15:13:02 +01:00
parent 9d5fb3b58d
commit aeba993411
1 changed files with 57 additions and 19 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

76
youtube_dl/extractor/espn.py
View File

@ -1,18 +1,26 @@
from __future__ import unicode_literals from __future__ import unicode_literals
import re
from .common import InfoExtractor from .common import InfoExtractor
from ..utils import (
parse_iso8601,
ExtractorError,
)
class ESPNIE(InfoExtractor): class ESPNIE(InfoExtractor):
_VALID_URL = r'https?://espn\.go\.com/(?:[^/]+/)*(?P<id>[^/]+)' _VALID_URL = r'https?://espn\.go\.com/(?:[^/]+/)*(?P<id>[^/]+)'
_WORKING = False
_TESTS = [{ _TESTS = [{
'url': 'http://espn.go.com/video/clip?id=10365079', 'url': 'http://espn.go.com/video/clip?id=10365079',
'info_dict': { 'info_dict': {
'id': 'FkYWtmazr6Ed8xmvILvKLWjd4QvYZpzG', 'id': '10365079',
'ext': 'mp4', 'ext': 'mp4',
'title': 'dm_140128_30for30Shorts___JudgingJewellv2', 'title': '30 for 30 Shorts: Judging Jewell',
'description': '', 'description': 'On July 27, 1996, a terrorist\'s bomb exploded in a crowded Centennial Olympic Park during the Atlanta Olympic Games. The death toll might have been far higher if not for security guard Richard Jewell, who hours after his heroism was called a murderer.',
'duration': 1302,
'timestamp': 1390936111,
'upload_date': '20140128',
}, },
'params': { 'params': {
# m3u8 download # m3u8 download
@ -35,21 +43,51 @@ class ESPNIE(InfoExtractor):
'only_matching': True, 'only_matching': True,
}] }]
def _extract_video_info(self, video_data):
video_id = str(video_data['id'])
formats = []
for source in video_data['links']['source'].values():
if isinstance(source, dict) and source.get('href'):
source_url = source['href']
if '.m3u8' in source_url:
formats.extend(self._extract_m3u8_formats(source_url, video_id, m3u8_id='hls'))
elif '.f4m' in source_url:
formats.extend(self._extract_f4m_formats(source_url + '?hdcore=2.10.3', video_id, f4m_id='hds'))
elif '.smil' in source_url:
formats.extend(self._extract_smil_formats(source_url, video_id))
else:
formats.append({
'url': source_url,
})
self._sort_formats(formats)
return {
'id': video_id,
'title': video_data['headline'],
'description': video_data.get('description') or video_data.get('caption'),
'duration': video_data.get('duration'),
'thumbnail': video_data.get('thumbnail'),
'timestamp': parse_iso8601(video_data.get('originalPublishDate')),
'formats': formats,
}
def _real_extract(self, url): def _real_extract(self, url):
video_id = self._match_id(url) video_id = self._match_id(url)
webpage = self._download_webpage(url, video_id) webpage = self._download_webpage(url + ('&xhr=1' if '?' in url else '?xhr=1'), video_id)
json_data = self._parse_json(webpage, video_id, fatal=False)
video_id = self._search_regex( if json_data:
r'class="video-play-button"[^>]+data-id="(\d+)', page_type = json_data['meta']['type']
webpage, 'video id') if page_type == 'video':
return self._extract_video_info(json_data['content'])
player = self._download_webpage( else:
'https://espn.go.com/video/iframe/twitter/?id=%s' % video_id, video_id) videos_data = json_data['content'].get('video')
if videos_data:
pcode = self._search_regex( entries = []
r'["\']pcode=([^"\']+)["\']', player, 'pcode') for video_data in videos_data:
entries.append(self._extract_video_info(video_data))
return self.url_result( return self.playlist_result(entries, str(json_data['uid']), json_data['content']['title'], json_data['content']['description'])
'ooyalaexternal:espn:%s:%s' % (video_id, pcode), else:
'OoyalaExternal') raise ExtractorError('No videos in the webpage', expected=True)
else:
return self.url_result(self._search_regex(r'mobileLink\s*=\s*"([^"]+)";', webpage, 'mobile link'), 'ESPN')