etc/yt-dlp
0
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2024-12-30 02:01:05 +00:00
Code Issues Projects Releases Wiki Activity

Switched to use media detail XML to extract video URL

Browse source
This commit is contained in:
Charles Chen 2014-07-15 13:55:23 -07:00
parent b1b01841af
commit 172240c0a4
1 changed files with 32 additions and 25 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

57
youtube_dl/extractor/mlb.py
View file

@ -28,37 +28,44 @@ def _real_extract(self, url):
title = self._og_search_title(webpage, default=video_id) title = self._og_search_title(webpage, default=video_id)
description = self._html_search_regex(r'<meta name="description" (?:content|value)="(.*?)"/>', webpage, 'description', fatal=False) description = self._html_search_regex(r'<meta name="description" (?:content|value)="(.*?)"/>', webpage, 'description', fatal=False)
thumbnail = self._html_search_regex(r'<meta itemprop="image" (?:content|value)="(.*?)" />', webpage, 'image', fatal=False) thumbnail = self._html_search_regex(r'<meta itemprop="image" (?:content|value)="(.*?)" />', webpage, 'image', fatal=False)
# use the video_id to find the Media detail XML
id_len = len(video_id)
_mediadetail_url = 'http://m.mlb.com/gen/multimedia/detail/'+video_id[id_len-3]+'/'+video_id[id_len-2]+'/'+video_id[id_len-1]+'/'+video_id+'.xml'
# use the thumbnail URL to find the folder that contains the videos mediadetails = self._download_xml(_mediadetail_url, video_id, "Downloading media detail...")
_image_url = r'http://mediadownloads.mlb.com/mlbam/(?P<_date>n?.+)/images/.*$' has1500K = 0
bobj = re.match(_image_url, thumbnail) has1200K = 0
datestr = bobj.group('_date') has600K = 0
base_url = 'http://mediadownloads.mlb.com/mlbam/' + datestr # loop through the list of url's and only get the highest quality MP4 content
filespage = self._download_webpage(base_url, video_id) for element in mediadetails.findall('url'):
scenario = element.attrib['playback_scenario']
# Try 1800K, 1500K, 1200K, 600K, then 300K videos if scenario.startswith(u'FLASH'):
video = self._html_search_regex(r'<li><a href="(.*?)_'+video_id+'_1800K.mp4"', filespage, '1800K', fatal=False) if scenario.startswith(u'FLASH_1800K'):
if video is not None: video_url = element.text
video_url = base_url+'/'+video+'_'+video_id+'_1800K.mp4' # 1800K is the current highest quality video on MLB.com
else: break
video = self._html_search_regex(r'<li><a href="(.*?)_'+video_id+'_1500K.mp4"', filespage, '1500K', fatal=False)
if video is not None:
video_url = base_url+'/'+video+'_'+video_id+'_1500K.mp4'
else:
video = self._html_search_regex(r'<li><a href="(.*?)_'+video_id+'_600K.mp4"', filespage, '600K', fatal=False)
if video is not None:
video_url = base_url+'/'+video+'_'+video_id+'_600K.mp4'
else: else:
video = self._html_search_regex(r'<li><a href="(.*?)_'+video_id+'_300K.mp4"', filespage, 'MLB', fatal=False) if scenario.startswith(u'FLASH_1500K'):
if video is not None: video_url = element.text
video_url = base_url+'/'+video+'_'+video_id+'_300K.mp4' has1500K = 1
else: else:
# nothing valuable to return if (scenario.startswith(u'FLASH_1200K') and not has1500K):
return None video_url = element.text
has1200K = 1
else:
if (scenario.startswith(u'FLASH_600K') and not has1200K):
video_url = element.text
has600K = 1
else:
if (scenario.startswith(u'FLASH_300K') and not has600K):
video_url = element.text
return { return {
'id': video_id, 'id': video_id,
'url': video_url, 'url': video_url,
'extractor': 'mlb',
'webpage_url': url,
'title': title, 'title': title,
'ext': 'mp4', 'ext': 'mp4',
'format': 'mp4', 'format': 'mp4',