You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

158 lines
6.8 KiB

  1. from __future__ import unicode_literals
  2. import re
  3. import base64
  4. from .common import InfoExtractor
  5. from ..utils import (
  6. int_or_none,
  7. float_or_none,
  8. ExtractorError,
  9. unsmuggle_url,
  10. )
  11. from ..compat import compat_urllib_parse
  12. class OoyalaBaseIE(InfoExtractor):
  13. def _extract(self, content_tree_url, video_id, domain='example.org'):
  14. content_tree = self._download_json(content_tree_url, video_id)['content_tree']
  15. metadata = content_tree[list(content_tree)[0]]
  16. embed_code = metadata['embed_code']
  17. pcode = metadata.get('asset_pcode') or embed_code
  18. video_info = {
  19. 'id': embed_code,
  20. 'title': metadata['title'],
  21. 'description': metadata.get('description'),
  22. 'thumbnail': metadata.get('thumbnail_image') or metadata.get('promo_image'),
  23. 'duration': float_or_none(metadata.get('duration'), 1000),
  24. }
  25. formats = []
  26. for supported_format in ('mp4', 'm3u8', 'hds', 'rtmp'):
  27. auth_data = self._download_json(
  28. '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}),
  29. video_id, 'Downloading %s JSON' % supported_format)
  30. cur_auth_data = auth_data['authorization_data'][embed_code]
  31. if cur_auth_data['authorized']:
  32. for stream in cur_auth_data['streams']:
  33. url = base64.b64decode(stream['url']['data'].encode('ascii')).decode('utf-8')
  34. delivery_type = stream['delivery_type']
  35. if delivery_type == 'remote_asset':
  36. video_info['url'] = url
  37. return video_info
  38. if delivery_type == 'hls':
  39. formats.extend(self._extract_m3u8_formats(url, embed_code, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
  40. elif delivery_type == 'hds':
  41. formats.extend(self._extract_f4m_formats(url, embed_code, -1, 'hds', fatal=False))
  42. else:
  43. formats.append({
  44. 'url': url,
  45. 'ext': stream.get('delivery_type'),
  46. 'vcodec': stream.get('video_codec'),
  47. 'format_id': '%s-%s-%sp' % (stream.get('profile'), delivery_type, stream.get('height')),
  48. 'width': int_or_none(stream.get('width')),
  49. 'height': int_or_none(stream.get('height')),
  50. 'abr': int_or_none(stream.get('audio_bitrate')),
  51. 'vbr': int_or_none(stream.get('video_bitrate')),
  52. 'fps': float_or_none(stream.get('framerate')),
  53. })
  54. else:
  55. raise ExtractorError('%s said: %s' % (self.IE_NAME, cur_auth_data['message']), expected=True)
  56. self._sort_formats(formats)
  57. video_info['formats'] = formats
  58. return video_info
  59. class OoyalaIE(OoyalaBaseIE):
  60. _VALID_URL = r'(?:ooyala:|https?://.+?\.ooyala\.com/.*?(?:embedCode|ec)=)(?P<id>.+?)(&|$)'
  61. _TESTS = [
  62. {
  63. # From http://it.slashdot.org/story/13/04/25/178216/recovering-data-from-broken-hard-drives-and-ssds-video
  64. 'url': 'http://player.ooyala.com/player.js?embedCode=pxczE2YjpfHfn1f3M-ykG_AmJRRn0PD8',
  65. 'info_dict': {
  66. 'id': 'pxczE2YjpfHfn1f3M-ykG_AmJRRn0PD8',
  67. 'ext': 'mp4',
  68. 'title': 'Explaining Data Recovery from Hard Drives and SSDs',
  69. 'description': 'How badly damaged does a drive have to be to defeat Russell and his crew? Apparently, smashed to bits.',
  70. 'duration': 853.386,
  71. },
  72. }, {
  73. # Only available for ipad
  74. 'url': 'http://player.ooyala.com/player.js?embedCode=x1b3lqZDq9y_7kMyC2Op5qo-p077tXD0',
  75. 'info_dict': {
  76. 'id': 'x1b3lqZDq9y_7kMyC2Op5qo-p077tXD0',
  77. 'ext': 'mp4',
  78. 'title': 'Simulation Overview - Levels of Simulation',
  79. 'duration': 194.948,
  80. },
  81. },
  82. {
  83. # Information available only through SAS api
  84. # From http://community.plm.automation.siemens.com/t5/News-NX-Manufacturing/Tool-Path-Divide/ba-p/4187
  85. 'url': 'http://player.ooyala.com/player.js?embedCode=FiOG81ZTrvckcchQxmalf4aQj590qTEx',
  86. 'md5': 'a84001441b35ea492bc03736e59e7935',
  87. 'info_dict': {
  88. 'id': 'FiOG81ZTrvckcchQxmalf4aQj590qTEx',
  89. 'ext': 'mp4',
  90. 'title': 'Divide Tool Path.mp4',
  91. 'duration': 204.405,
  92. }
  93. }
  94. ]
  95. @staticmethod
  96. def _url_for_embed_code(embed_code):
  97. return 'http://player.ooyala.com/player.js?embedCode=%s' % embed_code
  98. @classmethod
  99. def _build_url_result(cls, embed_code):
  100. return cls.url_result(cls._url_for_embed_code(embed_code),
  101. ie=cls.ie_key())
  102. def _real_extract(self, url):
  103. url, smuggled_data = unsmuggle_url(url, {})
  104. embed_code = self._match_id(url)
  105. domain = smuggled_data.get('domain')
  106. content_tree_url = 'http://player.ooyala.com/player_api/v1/content_tree/embed_code/%s/%s' % (embed_code, embed_code)
  107. return self._extract(content_tree_url, embed_code, domain)
  108. class OoyalaExternalIE(OoyalaBaseIE):
  109. _VALID_URL = r'''(?x)
  110. (?:
  111. ooyalaexternal:|
  112. https?://.+?\.ooyala\.com/.*?\bexternalId=
  113. )
  114. (?P<partner_id>[^:]+)
  115. :
  116. (?P<id>.+)
  117. (?:
  118. :|
  119. .*?&pcode=
  120. )
  121. (?P<pcode>.+?)
  122. (?:&|$)
  123. '''
  124. _TEST = {
  125. 'url': 'https://player.ooyala.com/player.js?externalId=espn:10365079&pcode=1kNG061cgaoolOncv54OAO1ceO-I&adSetCode=91cDU6NuXTGKz3OdjOxFdAgJVtQcKJnI&callback=handleEvents&hasModuleParams=1&height=968&playerBrandingId=7af3bd04449c444c964f347f11873075&targetReplaceId=videoPlayer&width=1656&wmode=opaque&allowScriptAccess=always',
  126. 'info_dict': {
  127. 'id': 'FkYWtmazr6Ed8xmvILvKLWjd4QvYZpzG',
  128. 'ext': 'mp4',
  129. 'title': 'dm_140128_30for30Shorts___JudgingJewellv2',
  130. 'duration': 1302000,
  131. },
  132. 'params': {
  133. # m3u8 download
  134. 'skip_download': True,
  135. },
  136. }
  137. def _real_extract(self, url):
  138. partner_id, video_id, pcode = re.match(self._VALID_URL, url).groups()
  139. content_tree_url = 'http://player.ooyala.com/player_api/v1/content_tree/external_id/%s/%s:%s' % (pcode, partner_id, video_id)
  140. return self._extract(content_tree_url, video_id)