1
0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2024-11-02 17:22:31 +01:00

[ooyala] extract domain,handle errors and change related tests

This commit is contained in:
remitamine 2015-10-16 16:02:40 +01:00
parent dd414c970b
commit cce9d15d01
7 changed files with 48 additions and 31 deletions

View File

@ -14,9 +14,10 @@ class BYUtvIE(InfoExtractor):
'info_dict': {
'id': 'studio-c-season-5-episode-5',
'ext': 'mp4',
'description': 'md5:5438d33774b6bdc662f9485a340401cc',
'description': 'md5:e07269172baff037f8e8bf9956bc9747',
'title': 'Season 5 Episode 5',
'thumbnail': 're:^https?://.*\.jpg$'
'thumbnail': 're:^https?://.*\.jpg$',
'duration': 1486486,
},
'params': {
'skip_download': True,

View File

@ -335,6 +335,7 @@ class GenericIE(InfoExtractor):
'id': 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ',
'ext': 'mp4',
'title': '2cc213299525360.mov', # that's what we get
'duration': 238231,
},
'add_ie': ['Ooyala'],
},
@ -346,6 +347,7 @@ class GenericIE(InfoExtractor):
'ext': 'mp4',
'title': '"Steve Jobs: Man in the Machine" trailer',
'description': 'The first trailer for the Alex Gibney documentary "Steve Jobs: Man in the Machine."',
'duration': 135427,
},
'params': {
'skip_download': True,
@ -943,8 +945,9 @@ class GenericIE(InfoExtractor):
'info_dict': {
'id': '50YnY4czr4ms1vJ7yz3xzq0excz_pUMs',
'ext': 'mp4',
'description': 'VIDEO: Index/Match versus VLOOKUP.',
'description': 'VIDEO: INDEX/MATCH versus VLOOKUP.',
'title': 'This is what separates the Excel masters from the wannabes',
'duration': 191933,
},
'params': {
# m3u8 downloads
@ -1454,7 +1457,7 @@ def _playlist_from_matches(matches, getter=None, ie=None):
re.search(r'SBN\.VideoLinkset\.ooyala\([\'"](?P<ec>.{32})[\'"]\)', webpage) or
re.search(r'data-ooyala-video-id\s*=\s*[\'"](?P<ec>.{32})[\'"]', webpage))
if mobj is not None:
return OoyalaIE._build_url_result(mobj.group('ec'))
return OoyalaIE._build_url_result(smuggle_url(mobj.group('ec'), {'domain': url}))
# Look for multiple Ooyala embeds on SBN network websites
mobj = re.search(r'SBN\.VideoLinkset\.entryGroup\((\[.*?\])', webpage)
@ -1462,7 +1465,7 @@ def _playlist_from_matches(matches, getter=None, ie=None):
embeds = self._parse_json(mobj.group(1), video_id, fatal=False)
if embeds:
return _playlist_from_matches(
embeds, getter=lambda v: OoyalaIE._url_for_embed_code(v['provider_video_id']), ie='Ooyala')
embeds, getter=lambda v: OoyalaIE._url_for_embed_code(smuggle_url(v['provider_video_id'], {'domain': url})), ie='Ooyala')
# Look for Aparat videos
mobj = re.search(r'<iframe .*?src="(http://www\.aparat\.com/video/[^"]+)"', webpage)

View File

@ -18,6 +18,8 @@ class GrouponIE(InfoExtractor):
'id': 'tubGNycTo_9Uxg82uESj4i61EYX8nyuf',
'ext': 'mp4',
'title': 'Bikram Yoga Huntington Beach | Orange County',
'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
'duration': 44961,
},
}],
'params': {

View File

@ -16,6 +16,7 @@ class HowcastIE(InfoExtractor):
'description': 'md5:dbe792e5f6f1489027027bf2eba188a3',
'timestamp': 1276081287,
'upload_date': '20100609',
'duration': 56823,
},
'params': {
# m3u8 download

View File

@ -6,13 +6,16 @@
from ..utils import (
int_or_none,
float_or_none,
ExtractorError,
unsmuggle_url,
)
from ..compat import compat_urllib_parse
class OoyalaBaseIE(InfoExtractor):
def _extract(self, player_url, video_id):
content_tree = self._download_json(player_url, video_id)['content_tree']
def _extract(self, content_tree_url, video_id, domain='example.org'):
content_tree = self._download_json(content_tree_url, video_id)['content_tree']
metadata = content_tree[list(content_tree)[0]]
embed_code = metadata['embed_code']
pcode = metadata.get('asset_pcode') or embed_code
@ -27,33 +30,36 @@ def _extract(self, player_url, video_id):
formats = []
for supported_format in ('mp4', 'm3u8', 'hds', 'rtmp'):
auth_data = self._download_json(
'http://player.ooyala.com/sas/player_api/v1/authorization/embed_code/%s/%s?domain=www.example.org&supportedFormats=%s' % (pcode, embed_code, supported_format),
'http://player.ooyala.com/sas/player_api/v1/authorization/embed_code/%s/%s?' % (pcode, embed_code) + compat_urllib_parse.urlencode({'domain': domain, 'supportedFormats': supported_format}),
video_id, 'Downloading %s JSON' % supported_format)
cur_auth_data = auth_data['authorization_data'][embed_code]
for stream in cur_auth_data['streams']:
url = base64.b64decode(stream['url']['data'].encode('ascii')).decode('utf-8')
delivery_type = stream['delivery_type']
if delivery_type == 'remote_asset':
video_info['url'] = url
return video_info
if delivery_type == 'hls':
formats.extend(self._extract_m3u8_formats(url, embed_code, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
elif delivery_type == 'hds':
formats.extend(self._extract_f4m_formats(url, embed_code, -1, 'hds', fatal=False))
else:
formats.append({
'url': url,
'ext': stream.get('delivery_type'),
'vcodec': stream.get('video_codec'),
'format_id': '%s-%s-%sp' % (stream.get('profile'), delivery_type, stream.get('height')),
'width': int_or_none(stream.get('width')),
'height': int_or_none(stream.get('height')),
'abr': int_or_none(stream.get('audio_bitrate')),
'vbr': int_or_none(stream.get('video_bitrate')),
'fps': float_or_none(stream.get('framerate')),
})
if cur_auth_data['authorized']:
for stream in cur_auth_data['streams']:
url = base64.b64decode(stream['url']['data'].encode('ascii')).decode('utf-8')
delivery_type = stream['delivery_type']
if delivery_type == 'remote_asset':
video_info['url'] = url
return video_info
if delivery_type == 'hls':
formats.extend(self._extract_m3u8_formats(url, embed_code, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
elif delivery_type == 'hds':
formats.extend(self._extract_f4m_formats(url, embed_code, -1, 'hds', fatal=False))
else:
formats.append({
'url': url,
'ext': stream.get('delivery_type'),
'vcodec': stream.get('video_codec'),
'format_id': '%s-%s-%sp' % (stream.get('profile'), delivery_type, stream.get('height')),
'width': int_or_none(stream.get('width')),
'height': int_or_none(stream.get('height')),
'abr': int_or_none(stream.get('audio_bitrate')),
'vbr': int_or_none(stream.get('video_bitrate')),
'fps': float_or_none(stream.get('framerate')),
})
else:
raise ExtractorError('%s said: %s' % (self.IE_NAME, cur_auth_data['message']), expected=True)
self._sort_formats(formats)
video_info['formats'] = formats
@ -108,9 +114,11 @@ def _build_url_result(cls, embed_code):
ie=cls.ie_key())
def _real_extract(self, url):
url, smuggled_data = unsmuggle_url(url, {})
embed_code = self._match_id(url)
domain = smuggled_data.get('domain')
content_tree_url = 'http://player.ooyala.com/player_api/v1/content_tree/embed_code/%s/%s' % (embed_code, embed_code)
return self._extract(content_tree_url, embed_code)
return self._extract(content_tree_url, embed_code, domain)
class OoyalaExternalIE(OoyalaBaseIE):

View File

@ -16,6 +16,7 @@ class TeachingChannelIE(InfoExtractor):
'ext': 'mp4',
'title': 'A History of Teaming',
'description': 'md5:2a9033db8da81f2edffa4c99888140b3',
'duration': 422255,
},
'params': {
# m3u8 download

View File

@ -15,6 +15,7 @@ class ViceIE(InfoExtractor):
'id': '43cW1mYzpia9IlestBjVpd23Yu3afAfp',
'ext': 'mp4',
'title': 'VICE_COWBOYCAPITALISTS_PART01_v1_VICE_WM_1080p.mov',
'duration': 725983,
},
'params': {
# Requires ffmpeg (m3u8 manifest)