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.

174 lines
7.3 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_urlencode
  12. class OoyalaBaseIE(InfoExtractor):
  13. _PLAYER_BASE = 'http://player.ooyala.com/'
  14. _CONTENT_TREE_BASE = _PLAYER_BASE + 'player_api/v1/content_tree/'
  15. _AUTHORIZATION_URL_TEMPLATE = _PLAYER_BASE + 'sas/player_api/v1/authorization/embed_code/%s/%s?'
  16. def _extract(self, content_tree_url, video_id, domain='example.org'):
  17. content_tree = self._download_json(content_tree_url, video_id)['content_tree']
  18. metadata = content_tree[list(content_tree)[0]]
  19. embed_code = metadata['embed_code']
  20. pcode = metadata.get('asset_pcode') or embed_code
  21. video_info = {
  22. 'id': embed_code,
  23. 'title': metadata['title'],
  24. 'description': metadata.get('description'),
  25. 'thumbnail': metadata.get('thumbnail_image') or metadata.get('promo_image'),
  26. 'duration': float_or_none(metadata.get('duration'), 1000),
  27. }
  28. urls = []
  29. formats = []
  30. for supported_format in ('mp4', 'm3u8', 'hds', 'rtmp'):
  31. auth_data = self._download_json(
  32. self._AUTHORIZATION_URL_TEMPLATE % (pcode, embed_code) +
  33. compat_urllib_parse_urlencode({
  34. 'domain': domain,
  35. 'supportedFormats': supported_format
  36. }),
  37. video_id, 'Downloading %s JSON' % supported_format)
  38. cur_auth_data = auth_data['authorization_data'][embed_code]
  39. if cur_auth_data['authorized']:
  40. for stream in cur_auth_data['streams']:
  41. url = base64.b64decode(
  42. stream['url']['data'].encode('ascii')).decode('utf-8')
  43. if url in urls:
  44. continue
  45. urls.append(url)
  46. delivery_type = stream['delivery_type']
  47. if delivery_type == 'hls' or '.m3u8' in url:
  48. formats.extend(self._extract_m3u8_formats(
  49. url, embed_code, 'mp4', 'm3u8_native',
  50. m3u8_id='hls', fatal=False))
  51. elif delivery_type == 'hds' or '.f4m' in url:
  52. formats.extend(self._extract_f4m_formats(
  53. url + '?hdcore=3.7.0', embed_code, f4m_id='hds', fatal=False))
  54. elif '.smil' in url:
  55. formats.extend(self._extract_smil_formats(
  56. url, embed_code, fatal=False))
  57. else:
  58. formats.append({
  59. 'url': url,
  60. 'ext': stream.get('delivery_type'),
  61. 'vcodec': stream.get('video_codec'),
  62. 'format_id': delivery_type,
  63. 'width': int_or_none(stream.get('width')),
  64. 'height': int_or_none(stream.get('height')),
  65. 'abr': int_or_none(stream.get('audio_bitrate')),
  66. 'vbr': int_or_none(stream.get('video_bitrate')),
  67. 'fps': float_or_none(stream.get('framerate')),
  68. })
  69. else:
  70. raise ExtractorError('%s said: %s' % (
  71. self.IE_NAME, cur_auth_data['message']), expected=True)
  72. self._sort_formats(formats)
  73. video_info['formats'] = formats
  74. return video_info
  75. class OoyalaIE(OoyalaBaseIE):
  76. _VALID_URL = r'(?:ooyala:|https?://.+?\.ooyala\.com/.*?(?:embedCode|ec)=)(?P<id>.+?)(&|$)'
  77. _TESTS = [
  78. {
  79. # From http://it.slashdot.org/story/13/04/25/178216/recovering-data-from-broken-hard-drives-and-ssds-video
  80. 'url': 'http://player.ooyala.com/player.js?embedCode=pxczE2YjpfHfn1f3M-ykG_AmJRRn0PD8',
  81. 'info_dict': {
  82. 'id': 'pxczE2YjpfHfn1f3M-ykG_AmJRRn0PD8',
  83. 'ext': 'mp4',
  84. 'title': 'Explaining Data Recovery from Hard Drives and SSDs',
  85. 'description': 'How badly damaged does a drive have to be to defeat Russell and his crew? Apparently, smashed to bits.',
  86. 'duration': 853.386,
  87. },
  88. }, {
  89. # Only available for ipad
  90. 'url': 'http://player.ooyala.com/player.js?embedCode=x1b3lqZDq9y_7kMyC2Op5qo-p077tXD0',
  91. 'info_dict': {
  92. 'id': 'x1b3lqZDq9y_7kMyC2Op5qo-p077tXD0',
  93. 'ext': 'mp4',
  94. 'title': 'Simulation Overview - Levels of Simulation',
  95. 'duration': 194.948,
  96. },
  97. },
  98. {
  99. # Information available only through SAS api
  100. # From http://community.plm.automation.siemens.com/t5/News-NX-Manufacturing/Tool-Path-Divide/ba-p/4187
  101. 'url': 'http://player.ooyala.com/player.js?embedCode=FiOG81ZTrvckcchQxmalf4aQj590qTEx',
  102. 'md5': 'a84001441b35ea492bc03736e59e7935',
  103. 'info_dict': {
  104. 'id': 'FiOG81ZTrvckcchQxmalf4aQj590qTEx',
  105. 'ext': 'mp4',
  106. 'title': 'Divide Tool Path.mp4',
  107. 'duration': 204.405,
  108. }
  109. }
  110. ]
  111. @staticmethod
  112. def _url_for_embed_code(embed_code):
  113. return 'http://player.ooyala.com/player.js?embedCode=%s' % embed_code
  114. @classmethod
  115. def _build_url_result(cls, embed_code):
  116. return cls.url_result(cls._url_for_embed_code(embed_code),
  117. ie=cls.ie_key())
  118. def _real_extract(self, url):
  119. url, smuggled_data = unsmuggle_url(url, {})
  120. embed_code = self._match_id(url)
  121. domain = smuggled_data.get('domain')
  122. content_tree_url = self._CONTENT_TREE_BASE + 'embed_code/%s/%s' % (embed_code, embed_code)
  123. return self._extract(content_tree_url, embed_code, domain)
  124. class OoyalaExternalIE(OoyalaBaseIE):
  125. _VALID_URL = r'''(?x)
  126. (?:
  127. ooyalaexternal:|
  128. https?://.+?\.ooyala\.com/.*?\bexternalId=
  129. )
  130. (?P<partner_id>[^:]+)
  131. :
  132. (?P<id>.+)
  133. (?:
  134. :|
  135. .*?&pcode=
  136. )
  137. (?P<pcode>.+?)
  138. (?:&|$)
  139. '''
  140. _TEST = {
  141. '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',
  142. 'info_dict': {
  143. 'id': 'FkYWtmazr6Ed8xmvILvKLWjd4QvYZpzG',
  144. 'ext': 'mp4',
  145. 'title': 'dm_140128_30for30Shorts___JudgingJewellv2',
  146. 'duration': 1302.0,
  147. },
  148. 'params': {
  149. # m3u8 download
  150. 'skip_download': True,
  151. },
  152. }
  153. def _real_extract(self, url):
  154. partner_id, video_id, pcode = re.match(self._VALID_URL, url).groups()
  155. content_tree_url = self._CONTENT_TREE_BASE + 'external_id/%s/%s:%s' % (pcode, partner_id, video_id)
  156. return self._extract(content_tree_url, video_id)