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.

210 lines
8.6 KiB

  1. from __future__ import unicode_literals
  2. import base64
  3. import re
  4. from .common import InfoExtractor
  5. from ..compat import (
  6. compat_b64decode,
  7. compat_str,
  8. )
  9. from ..utils import (
  10. determine_ext,
  11. ExtractorError,
  12. float_or_none,
  13. int_or_none,
  14. try_get,
  15. unsmuggle_url,
  16. )
  17. class OoyalaBaseIE(InfoExtractor):
  18. _PLAYER_BASE = 'http://player.ooyala.com/'
  19. _CONTENT_TREE_BASE = _PLAYER_BASE + 'player_api/v1/content_tree/'
  20. _AUTHORIZATION_URL_TEMPLATE = _PLAYER_BASE + 'sas/player_api/v2/authorization/embed_code/%s/%s'
  21. def _extract(self, content_tree_url, video_id, domain=None, supportedformats=None, embed_token=None):
  22. content_tree = self._download_json(content_tree_url, video_id)['content_tree']
  23. metadata = content_tree[list(content_tree)[0]]
  24. embed_code = metadata['embed_code']
  25. pcode = metadata.get('asset_pcode') or embed_code
  26. title = metadata['title']
  27. auth_data = self._download_json(
  28. self._AUTHORIZATION_URL_TEMPLATE % (pcode, embed_code),
  29. video_id, headers=self.geo_verification_headers(), query={
  30. 'domain': domain or 'player.ooyala.com',
  31. 'supportedFormats': supportedformats or 'mp4,rtmp,m3u8,hds,dash,smooth',
  32. 'embedToken': embed_token,
  33. })['authorization_data'][embed_code]
  34. urls = []
  35. formats = []
  36. streams = auth_data.get('streams') or [{
  37. 'delivery_type': 'hls',
  38. 'url': {
  39. 'data': base64.b64encode(('http://player.ooyala.com/hls/player/all/%s.m3u8' % embed_code).encode()).decode(),
  40. }
  41. }]
  42. for stream in streams:
  43. url_data = try_get(stream, lambda x: x['url']['data'], compat_str)
  44. if not url_data:
  45. continue
  46. s_url = compat_b64decode(url_data).decode('utf-8')
  47. if not s_url or s_url in urls:
  48. continue
  49. urls.append(s_url)
  50. ext = determine_ext(s_url, None)
  51. delivery_type = stream.get('delivery_type')
  52. if delivery_type == 'hls' or ext == 'm3u8':
  53. formats.extend(self._extract_m3u8_formats(
  54. re.sub(r'/ip(?:ad|hone)/', '/all/', s_url), embed_code, 'mp4', 'm3u8_native',
  55. m3u8_id='hls', fatal=False))
  56. elif delivery_type == 'hds' or ext == 'f4m':
  57. formats.extend(self._extract_f4m_formats(
  58. s_url + '?hdcore=3.7.0', embed_code, f4m_id='hds', fatal=False))
  59. elif delivery_type == 'dash' or ext == 'mpd':
  60. formats.extend(self._extract_mpd_formats(
  61. s_url, embed_code, mpd_id='dash', fatal=False))
  62. elif delivery_type == 'smooth':
  63. self._extract_ism_formats(
  64. s_url, embed_code, ism_id='mss', fatal=False)
  65. elif ext == 'smil':
  66. formats.extend(self._extract_smil_formats(
  67. s_url, embed_code, fatal=False))
  68. else:
  69. formats.append({
  70. 'url': s_url,
  71. 'ext': ext or delivery_type,
  72. 'vcodec': stream.get('video_codec'),
  73. 'format_id': delivery_type,
  74. 'width': int_or_none(stream.get('width')),
  75. 'height': int_or_none(stream.get('height')),
  76. 'abr': int_or_none(stream.get('audio_bitrate')),
  77. 'vbr': int_or_none(stream.get('video_bitrate')),
  78. 'fps': float_or_none(stream.get('framerate')),
  79. })
  80. if not formats and not auth_data.get('authorized'):
  81. raise ExtractorError('%s said: %s' % (
  82. self.IE_NAME, auth_data['message']), expected=True)
  83. self._sort_formats(formats)
  84. subtitles = {}
  85. for lang, sub in metadata.get('closed_captions_vtt', {}).get('captions', {}).items():
  86. sub_url = sub.get('url')
  87. if not sub_url:
  88. continue
  89. subtitles[lang] = [{
  90. 'url': sub_url,
  91. }]
  92. return {
  93. 'id': embed_code,
  94. 'title': title,
  95. 'description': metadata.get('description'),
  96. 'thumbnail': metadata.get('thumbnail_image') or metadata.get('promo_image'),
  97. 'duration': float_or_none(metadata.get('duration'), 1000),
  98. 'subtitles': subtitles,
  99. 'formats': formats,
  100. }
  101. class OoyalaIE(OoyalaBaseIE):
  102. _VALID_URL = r'(?:ooyala:|https?://.+?\.ooyala\.com/.*?(?:embedCode|ec)=)(?P<id>.+?)(&|$)'
  103. _TESTS = [
  104. {
  105. # From http://it.slashdot.org/story/13/04/25/178216/recovering-data-from-broken-hard-drives-and-ssds-video
  106. 'url': 'http://player.ooyala.com/player.js?embedCode=pxczE2YjpfHfn1f3M-ykG_AmJRRn0PD8',
  107. 'info_dict': {
  108. 'id': 'pxczE2YjpfHfn1f3M-ykG_AmJRRn0PD8',
  109. 'ext': 'mp4',
  110. 'title': 'Explaining Data Recovery from Hard Drives and SSDs',
  111. 'description': 'How badly damaged does a drive have to be to defeat Russell and his crew? Apparently, smashed to bits.',
  112. 'duration': 853.386,
  113. },
  114. # The video in the original webpage now uses PlayWire
  115. 'skip': 'Ooyala said: movie expired',
  116. }, {
  117. # Only available for ipad
  118. 'url': 'http://player.ooyala.com/player.js?embedCode=x1b3lqZDq9y_7kMyC2Op5qo-p077tXD0',
  119. 'info_dict': {
  120. 'id': 'x1b3lqZDq9y_7kMyC2Op5qo-p077tXD0',
  121. 'ext': 'mp4',
  122. 'title': 'Simulation Overview - Levels of Simulation',
  123. 'duration': 194.948,
  124. },
  125. },
  126. {
  127. # Information available only through SAS api
  128. # From http://community.plm.automation.siemens.com/t5/News-NX-Manufacturing/Tool-Path-Divide/ba-p/4187
  129. 'url': 'http://player.ooyala.com/player.js?embedCode=FiOG81ZTrvckcchQxmalf4aQj590qTEx',
  130. 'md5': 'a84001441b35ea492bc03736e59e7935',
  131. 'info_dict': {
  132. 'id': 'FiOG81ZTrvckcchQxmalf4aQj590qTEx',
  133. 'ext': 'mp4',
  134. 'title': 'Divide Tool Path.mp4',
  135. 'duration': 204.405,
  136. }
  137. },
  138. {
  139. # empty stream['url']['data']
  140. 'url': 'http://player.ooyala.com/player.js?embedCode=w2bnZtYjE6axZ_dw1Cd0hQtXd_ige2Is',
  141. 'only_matching': True,
  142. }
  143. ]
  144. @staticmethod
  145. def _url_for_embed_code(embed_code):
  146. return 'http://player.ooyala.com/player.js?embedCode=%s' % embed_code
  147. @classmethod
  148. def _build_url_result(cls, embed_code):
  149. return cls.url_result(cls._url_for_embed_code(embed_code),
  150. ie=cls.ie_key())
  151. def _real_extract(self, url):
  152. url, smuggled_data = unsmuggle_url(url, {})
  153. embed_code = self._match_id(url)
  154. domain = smuggled_data.get('domain')
  155. supportedformats = smuggled_data.get('supportedformats')
  156. embed_token = smuggled_data.get('embed_token')
  157. content_tree_url = self._CONTENT_TREE_BASE + 'embed_code/%s/%s' % (embed_code, embed_code)
  158. return self._extract(content_tree_url, embed_code, domain, supportedformats, embed_token)
  159. class OoyalaExternalIE(OoyalaBaseIE):
  160. _VALID_URL = r'''(?x)
  161. (?:
  162. ooyalaexternal:|
  163. https?://.+?\.ooyala\.com/.*?\bexternalId=
  164. )
  165. (?P<partner_id>[^:]+)
  166. :
  167. (?P<id>.+)
  168. (?:
  169. :|
  170. .*?&pcode=
  171. )
  172. (?P<pcode>.+?)
  173. (?:&|$)
  174. '''
  175. _TEST = {
  176. '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',
  177. 'info_dict': {
  178. 'id': 'FkYWtmazr6Ed8xmvILvKLWjd4QvYZpzG',
  179. 'ext': 'mp4',
  180. 'title': 'dm_140128_30for30Shorts___JudgingJewellv2',
  181. 'duration': 1302.0,
  182. },
  183. 'params': {
  184. # m3u8 download
  185. 'skip_download': True,
  186. },
  187. }
  188. def _real_extract(self, url):
  189. partner_id, video_id, pcode = re.match(self._VALID_URL, url).groups()
  190. content_tree_url = self._CONTENT_TREE_BASE + 'external_id/%s/%s:%s' % (pcode, partner_id, video_id)
  191. return self._extract(content_tree_url, video_id)