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.

171 lines
6.9 KiB

  1. from __future__ import unicode_literals
  2. import json
  3. import re
  4. import socket
  5. from .common import InfoExtractor
  6. from ..compat import (
  7. compat_http_client,
  8. compat_str,
  9. compat_urllib_error,
  10. compat_urllib_parse_unquote,
  11. compat_urllib_request,
  12. )
  13. from ..utils import (
  14. ExtractorError,
  15. int_or_none,
  16. limit_length,
  17. urlencode_postdata,
  18. )
  19. class FacebookIE(InfoExtractor):
  20. _VALID_URL = r'''(?x)
  21. https?://(?:\w+\.)?facebook\.com/
  22. (?:[^#]*?\#!/)?
  23. (?:
  24. (?:video/video\.php|photo\.php|video\.php|video/embed)\?(?:.*?)
  25. (?:v|video_id)=|
  26. [^/]+/videos/(?:[^/]+/)?
  27. )
  28. (?P<id>[0-9]+)
  29. (?:.*)'''
  30. _LOGIN_URL = 'https://www.facebook.com/login.php?next=http%3A%2F%2Ffacebook.com%2Fhome.php&login_attempt=1'
  31. _CHECKPOINT_URL = 'https://www.facebook.com/checkpoint/?next=http%3A%2F%2Ffacebook.com%2Fhome.php&_fb_noscript=1'
  32. _NETRC_MACHINE = 'facebook'
  33. IE_NAME = 'facebook'
  34. _TESTS = [{
  35. 'url': 'https://www.facebook.com/video.php?v=637842556329505&fref=nf',
  36. 'md5': '6a40d33c0eccbb1af76cf0485a052659',
  37. 'info_dict': {
  38. 'id': '637842556329505',
  39. 'ext': 'mp4',
  40. 'title': 're:Did you know Kei Nishikori is the first Asian man to ever reach a Grand Slam',
  41. }
  42. }, {
  43. 'note': 'Video without discernible title',
  44. 'url': 'https://www.facebook.com/video.php?v=274175099429670',
  45. 'info_dict': {
  46. 'id': '274175099429670',
  47. 'ext': 'mp4',
  48. 'title': 'Facebook video #274175099429670',
  49. },
  50. 'expected_warnings': [
  51. 'title'
  52. ]
  53. }, {
  54. 'url': 'https://www.facebook.com/video.php?v=10204634152394104',
  55. 'only_matching': True,
  56. }, {
  57. 'url': 'https://www.facebook.com/amogood/videos/1618742068337349/?fref=nf',
  58. 'only_matching': True,
  59. }, {
  60. 'url': 'https://www.facebook.com/ChristyClarkForBC/videos/vb.22819070941/10153870694020942/?type=2&theater',
  61. 'only_matching': True,
  62. }]
  63. def _login(self):
  64. (useremail, password) = self._get_login_info()
  65. if useremail is None:
  66. return
  67. login_page_req = compat_urllib_request.Request(self._LOGIN_URL)
  68. login_page_req.add_header('Cookie', 'locale=en_US')
  69. login_page = self._download_webpage(login_page_req, None,
  70. note='Downloading login page',
  71. errnote='Unable to download login page')
  72. lsd = self._search_regex(
  73. r'<input type="hidden" name="lsd" value="([^"]*)"',
  74. login_page, 'lsd')
  75. lgnrnd = self._search_regex(r'name="lgnrnd" value="([^"]*?)"', login_page, 'lgnrnd')
  76. login_form = {
  77. 'email': useremail,
  78. 'pass': password,
  79. 'lsd': lsd,
  80. 'lgnrnd': lgnrnd,
  81. 'next': 'http://facebook.com/home.php',
  82. 'default_persistent': '0',
  83. 'legacy_return': '1',
  84. 'timezone': '-60',
  85. 'trynum': '1',
  86. }
  87. request = compat_urllib_request.Request(self._LOGIN_URL, urlencode_postdata(login_form))
  88. request.add_header('Content-Type', 'application/x-www-form-urlencoded')
  89. try:
  90. login_results = self._download_webpage(request, None,
  91. note='Logging in', errnote='unable to fetch login page')
  92. if re.search(r'<form(.*)name="login"(.*)</form>', login_results) is not None:
  93. self._downloader.report_warning('unable to log in: bad username/password, or exceded login rate limit (~3/min). Check credentials or wait.')
  94. return
  95. check_form = {
  96. 'fb_dtsg': self._search_regex(r'name="fb_dtsg" value="(.+?)"', login_results, 'fb_dtsg'),
  97. 'h': self._search_regex(
  98. r'name="h"\s+(?:\w+="[^"]+"\s+)*?value="([^"]+)"', login_results, 'h'),
  99. 'name_action_selected': 'dont_save',
  100. }
  101. check_req = compat_urllib_request.Request(self._CHECKPOINT_URL, urlencode_postdata(check_form))
  102. check_req.add_header('Content-Type', 'application/x-www-form-urlencoded')
  103. check_response = self._download_webpage(check_req, None,
  104. note='Confirming login')
  105. if re.search(r'id="checkpointSubmitButton"', check_response) is not None:
  106. self._downloader.report_warning('Unable to confirm login, you have to login in your brower and authorize the login.')
  107. except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
  108. self._downloader.report_warning('unable to log in: %s' % compat_str(err))
  109. return
  110. def _real_initialize(self):
  111. self._login()
  112. def _real_extract(self, url):
  113. video_id = self._match_id(url)
  114. url = 'https://www.facebook.com/video/video.php?v=%s' % video_id
  115. webpage = self._download_webpage(url, video_id)
  116. BEFORE = '{swf.addParam(param[0], param[1]);});\n'
  117. AFTER = '.forEach(function(variable) {swf.addVariable(variable[0], variable[1]);});'
  118. m = re.search(re.escape(BEFORE) + '(.*?)' + re.escape(AFTER), webpage)
  119. if not m:
  120. m_msg = re.search(r'class="[^"]*uiInterstitialContent[^"]*"><div>(.*?)</div>', webpage)
  121. if m_msg is not None:
  122. raise ExtractorError(
  123. 'The video is not available, Facebook said: "%s"' % m_msg.group(1),
  124. expected=True)
  125. else:
  126. raise ExtractorError('Cannot parse data')
  127. data = dict(json.loads(m.group(1)))
  128. params_raw = compat_urllib_parse_unquote(data['params'])
  129. params = json.loads(params_raw)
  130. video_data = params['video_data'][0]
  131. formats = []
  132. for quality in ['sd', 'hd']:
  133. src = video_data.get('%s_src' % quality)
  134. if src is not None:
  135. formats.append({
  136. 'format_id': quality,
  137. 'url': src,
  138. })
  139. if not formats:
  140. raise ExtractorError('Cannot find video formats')
  141. video_title = self._html_search_regex(
  142. r'<h2\s+[^>]*class="uiHeaderTitle"[^>]*>([^<]*)</h2>', webpage, 'title',
  143. default=None)
  144. if not video_title:
  145. video_title = self._html_search_regex(
  146. r'(?s)<span class="fbPhotosPhotoCaption".*?id="fbPhotoPageCaption"><span class="hasCaption">(.*?)</span>',
  147. webpage, 'alternative title', fatal=False)
  148. video_title = limit_length(video_title, 80)
  149. if not video_title:
  150. video_title = 'Facebook video #%s' % video_id
  151. return {
  152. 'id': video_id,
  153. 'title': video_title,
  154. 'formats': formats,
  155. 'duration': int_or_none(video_data.get('video_duration')),
  156. 'thumbnail': video_data.get('thumbnail_src'),
  157. }