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.

167 lines
7.2 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. urls = []
  26. formats = []
  27. for supported_format in ('mp4', 'm3u8', 'hds', 'rtmp'):
  28. auth_data = self._download_json(
  29. '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}),
  30. video_id, 'Downloading %s JSON' % supported_format)
  31. cur_auth_data = auth_data['authorization_data'][embed_code]
  32. if cur_auth_data['authorized']:
  33. for stream in cur_auth_data['streams']:
  34. url = base64.b64decode(stream['url']['data'].encode('ascii')).decode('utf-8')
  35. if url in urls:
  36. continue
  37. urls.append(url)
  38. delivery_type = stream['delivery_type']
  39. if delivery_type == 'hls' or '.m3u8' in url:
  40. m3u8_formats = self._extract_m3u8_formats(url, embed_code, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False)
  41. if m3u8_formats:
  42. formats.extend(m3u8_formats)
  43. elif delivery_type == 'hds' or '.f4m' in url:
  44. f4m_formats = self._extract_f4m_formats(url, embed_code, f4m_id='hds', fatal=False)
  45. if f4m_formats:
  46. formats.extend(f4m_formats)
  47. elif '.smil' in url:
  48. smil_formats = self._extract_smil_formats(url, embed_code, fatal=False)
  49. if smil_formats:
  50. formats.extend(smil_formats)
  51. else:
  52. formats.append({
  53. 'url': url,
  54. 'ext': stream.get('delivery_type'),
  55. 'vcodec': stream.get('video_codec'),
  56. 'format_id': delivery_type,
  57. 'width': int_or_none(stream.get('width')),
  58. 'height': int_or_none(stream.get('height')),
  59. 'abr': int_or_none(stream.get('audio_bitrate')),
  60. 'vbr': int_or_none(stream.get('video_bitrate')),
  61. 'fps': float_or_none(stream.get('framerate')),
  62. })
  63. else:
  64. raise ExtractorError('%s said: %s' % (self.IE_NAME, cur_auth_data['message']), expected=True)
  65. self._sort_formats(formats)
  66. video_info['formats'] = formats
  67. return video_info
  68. class OoyalaIE(OoyalaBaseIE):
  69. _VALID_URL = r'(?:ooyala:|https?://.+?\.ooyala\.com/.*?(?:embedCode|ec)=)(?P<id>.+?)(&|$)'
  70. _TESTS = [
  71. {
  72. # From http://it.slashdot.org/story/13/04/25/178216/recovering-data-from-broken-hard-drives-and-ssds-video
  73. 'url': 'http://player.ooyala.com/player.js?embedCode=pxczE2YjpfHfn1f3M-ykG_AmJRRn0PD8',
  74. 'info_dict': {
  75. 'id': 'pxczE2YjpfHfn1f3M-ykG_AmJRRn0PD8',
  76. 'ext': 'mp4',
  77. 'title': 'Explaining Data Recovery from Hard Drives and SSDs',
  78. 'description': 'How badly damaged does a drive have to be to defeat Russell and his crew? Apparently, smashed to bits.',
  79. 'duration': 853.386,
  80. },
  81. }, {
  82. # Only available for ipad
  83. 'url': 'http://player.ooyala.com/player.js?embedCode=x1b3lqZDq9y_7kMyC2Op5qo-p077tXD0',
  84. 'info_dict': {
  85. 'id': 'x1b3lqZDq9y_7kMyC2Op5qo-p077tXD0',
  86. 'ext': 'mp4',
  87. 'title': 'Simulation Overview - Levels of Simulation',
  88. 'duration': 194.948,
  89. },
  90. },
  91. {
  92. # Information available only through SAS api
  93. # From http://community.plm.automation.siemens.com/t5/News-NX-Manufacturing/Tool-Path-Divide/ba-p/4187
  94. 'url': 'http://player.ooyala.com/player.js?embedCode=FiOG81ZTrvckcchQxmalf4aQj590qTEx',
  95. 'md5': 'a84001441b35ea492bc03736e59e7935',
  96. 'info_dict': {
  97. 'id': 'FiOG81ZTrvckcchQxmalf4aQj590qTEx',
  98. 'ext': 'mp4',
  99. 'title': 'Divide Tool Path.mp4',
  100. 'duration': 204.405,
  101. }
  102. }
  103. ]
  104. @staticmethod
  105. def _url_for_embed_code(embed_code):
  106. return 'http://player.ooyala.com/player.js?embedCode=%s' % embed_code
  107. @classmethod
  108. def _build_url_result(cls, embed_code):
  109. return cls.url_result(cls._url_for_embed_code(embed_code),
  110. ie=cls.ie_key())
  111. def _real_extract(self, url):
  112. url, smuggled_data = unsmuggle_url(url, {})
  113. embed_code = self._match_id(url)
  114. domain = smuggled_data.get('domain')
  115. content_tree_url = 'http://player.ooyala.com/player_api/v1/content_tree/embed_code/%s/%s' % (embed_code, embed_code)
  116. return self._extract(content_tree_url, embed_code, domain)
  117. class OoyalaExternalIE(OoyalaBaseIE):
  118. _VALID_URL = r'''(?x)
  119. (?:
  120. ooyalaexternal:|
  121. https?://.+?\.ooyala\.com/.*?\bexternalId=
  122. )
  123. (?P<partner_id>[^:]+)
  124. :
  125. (?P<id>.+)
  126. (?:
  127. :|
  128. .*?&pcode=
  129. )
  130. (?P<pcode>.+?)
  131. (?:&|$)
  132. '''
  133. _TEST = {
  134. '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',
  135. 'info_dict': {
  136. 'id': 'FkYWtmazr6Ed8xmvILvKLWjd4QvYZpzG',
  137. 'ext': 'mp4',
  138. 'title': 'dm_140128_30for30Shorts___JudgingJewellv2',
  139. 'duration': 1302000,
  140. },
  141. 'params': {
  142. # m3u8 download
  143. 'skip_download': True,
  144. },
  145. }
  146. def _real_extract(self, url):
  147. partner_id, video_id, pcode = re.match(self._VALID_URL, url).groups()
  148. content_tree_url = 'http://player.ooyala.com/player_api/v1/content_tree/external_id/%s/%s:%s' % (pcode, partner_id, video_id)
  149. return self._extract(content_tree_url, video_id)