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.

338 lines
13 KiB

  1. import base64
  2. import os
  3. import re
  4. import socket
  5. import sys
  6. import netrc
  7. from ..utils import (
  8. compat_http_client,
  9. compat_urllib_error,
  10. compat_urllib_request,
  11. compat_str,
  12. clean_html,
  13. compiled_regex_type,
  14. ExtractorError,
  15. unescapeHTML,
  16. )
  17. class InfoExtractor(object):
  18. """Information Extractor class.
  19. Information extractors are the classes that, given a URL, extract
  20. information about the video (or videos) the URL refers to. This
  21. information includes the real video URL, the video title, author and
  22. others. The information is stored in a dictionary which is then
  23. passed to the FileDownloader. The FileDownloader processes this
  24. information possibly downloading the video to the file system, among
  25. other possible outcomes.
  26. The dictionaries must include the following fields:
  27. id: Video identifier.
  28. url: Final video URL.
  29. title: Video title, unescaped.
  30. ext: Video filename extension.
  31. The following fields are optional:
  32. format: The video format, defaults to ext (used for --get-format)
  33. thumbnails: A list of dictionaries (with the entries "resolution" and
  34. "url") for the varying thumbnails
  35. thumbnail: Full URL to a video thumbnail image.
  36. description: One-line video description.
  37. uploader: Full name of the video uploader.
  38. upload_date: Video upload date (YYYYMMDD).
  39. uploader_id: Nickname or id of the video uploader.
  40. location: Physical location of the video.
  41. player_url: SWF Player URL (used for rtmpdump).
  42. subtitles: The subtitle file contents.
  43. view_count: How many users have watched the video on the platform.
  44. urlhandle: [internal] The urlHandle to be used to download the file,
  45. like returned by urllib.request.urlopen
  46. The fields should all be Unicode strings.
  47. Subclasses of this one should re-define the _real_initialize() and
  48. _real_extract() methods and define a _VALID_URL regexp.
  49. Probably, they should also be added to the list of extractors.
  50. _real_extract() must return a *list* of information dictionaries as
  51. described above.
  52. Finally, the _WORKING attribute should be set to False for broken IEs
  53. in order to warn the users and skip the tests.
  54. """
  55. _ready = False
  56. _downloader = None
  57. _WORKING = True
  58. def __init__(self, downloader=None):
  59. """Constructor. Receives an optional downloader."""
  60. self._ready = False
  61. self.set_downloader(downloader)
  62. @classmethod
  63. def suitable(cls, url):
  64. """Receives a URL and returns True if suitable for this IE."""
  65. # This does not use has/getattr intentionally - we want to know whether
  66. # we have cached the regexp for *this* class, whereas getattr would also
  67. # match the superclass
  68. if '_VALID_URL_RE' not in cls.__dict__:
  69. cls._VALID_URL_RE = re.compile(cls._VALID_URL)
  70. return cls._VALID_URL_RE.match(url) is not None
  71. @classmethod
  72. def working(cls):
  73. """Getter method for _WORKING."""
  74. return cls._WORKING
  75. def initialize(self):
  76. """Initializes an instance (authentication, etc)."""
  77. if not self._ready:
  78. self._real_initialize()
  79. self._ready = True
  80. def extract(self, url):
  81. """Extracts URL information and returns it in list of dicts."""
  82. self.initialize()
  83. return self._real_extract(url)
  84. def set_downloader(self, downloader):
  85. """Sets the downloader for this IE."""
  86. self._downloader = downloader
  87. def _real_initialize(self):
  88. """Real initialization process. Redefine in subclasses."""
  89. pass
  90. def _real_extract(self, url):
  91. """Real extraction process. Redefine in subclasses."""
  92. pass
  93. @property
  94. def IE_NAME(self):
  95. return type(self).__name__[:-2]
  96. def _request_webpage(self, url_or_request, video_id, note=None, errnote=None):
  97. """ Returns the response handle """
  98. if note is None:
  99. self.report_download_webpage(video_id)
  100. elif note is not False:
  101. self.to_screen(u'%s: %s' % (video_id, note))
  102. try:
  103. return compat_urllib_request.urlopen(url_or_request)
  104. except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
  105. if errnote is None:
  106. errnote = u'Unable to download webpage'
  107. raise ExtractorError(u'%s: %s' % (errnote, compat_str(err)), sys.exc_info()[2])
  108. def _download_webpage_handle(self, url_or_request, video_id, note=None, errnote=None):
  109. """ Returns a tuple (page content as string, URL handle) """
  110. # Strip hashes from the URL (#1038)
  111. if isinstance(url_or_request, (compat_str, str)):
  112. url_or_request = url_or_request.partition('#')[0]
  113. urlh = self._request_webpage(url_or_request, video_id, note, errnote)
  114. content_type = urlh.headers.get('Content-Type', '')
  115. m = re.match(r'[a-zA-Z0-9_.-]+/[a-zA-Z0-9_.-]+\s*;\s*charset=(.+)', content_type)
  116. if m:
  117. encoding = m.group(1)
  118. else:
  119. encoding = 'utf-8'
  120. webpage_bytes = urlh.read()
  121. if self._downloader.params.get('dump_intermediate_pages', False):
  122. try:
  123. url = url_or_request.get_full_url()
  124. except AttributeError:
  125. url = url_or_request
  126. self.to_screen(u'Dumping request to ' + url)
  127. dump = base64.b64encode(webpage_bytes).decode('ascii')
  128. self._downloader.to_screen(dump)
  129. content = webpage_bytes.decode(encoding, 'replace')
  130. return (content, urlh)
  131. def _download_webpage(self, url_or_request, video_id, note=None, errnote=None):
  132. """ Returns the data of the page as a string """
  133. return self._download_webpage_handle(url_or_request, video_id, note, errnote)[0]
  134. def to_screen(self, msg):
  135. """Print msg to screen, prefixing it with '[ie_name]'"""
  136. self._downloader.to_screen(u'[%s] %s' % (self.IE_NAME, msg))
  137. def report_extraction(self, id_or_name):
  138. """Report information extraction."""
  139. self.to_screen(u'%s: Extracting information' % id_or_name)
  140. def report_download_webpage(self, video_id):
  141. """Report webpage download."""
  142. self.to_screen(u'%s: Downloading webpage' % video_id)
  143. def report_age_confirmation(self):
  144. """Report attempt to confirm age."""
  145. self.to_screen(u'Confirming age')
  146. def report_login(self):
  147. """Report attempt to log in."""
  148. self.to_screen(u'Logging in')
  149. #Methods for following #608
  150. def url_result(self, url, ie=None):
  151. """Returns a url that points to a page that should be processed"""
  152. #TODO: ie should be the class used for getting the info
  153. video_info = {'_type': 'url',
  154. 'url': url,
  155. 'ie_key': ie}
  156. return video_info
  157. def playlist_result(self, entries, playlist_id=None, playlist_title=None):
  158. """Returns a playlist"""
  159. video_info = {'_type': 'playlist',
  160. 'entries': entries}
  161. if playlist_id:
  162. video_info['id'] = playlist_id
  163. if playlist_title:
  164. video_info['title'] = playlist_title
  165. return video_info
  166. def _search_regex(self, pattern, string, name, default=None, fatal=True, flags=0):
  167. """
  168. Perform a regex search on the given string, using a single or a list of
  169. patterns returning the first matching group.
  170. In case of failure return a default value or raise a WARNING or a
  171. ExtractorError, depending on fatal, specifying the field name.
  172. """
  173. if isinstance(pattern, (str, compat_str, compiled_regex_type)):
  174. mobj = re.search(pattern, string, flags)
  175. else:
  176. for p in pattern:
  177. mobj = re.search(p, string, flags)
  178. if mobj: break
  179. if sys.stderr.isatty() and os.name != 'nt':
  180. _name = u'\033[0;34m%s\033[0m' % name
  181. else:
  182. _name = name
  183. if mobj:
  184. # return the first matching group
  185. return next(g for g in mobj.groups() if g is not None)
  186. elif default is not None:
  187. return default
  188. elif fatal:
  189. raise ExtractorError(u'Unable to extract %s' % _name)
  190. else:
  191. self._downloader.report_warning(u'unable to extract %s; '
  192. u'please report this issue on http://yt-dl.org/bug' % _name)
  193. return None
  194. def _html_search_regex(self, pattern, string, name, default=None, fatal=True, flags=0):
  195. """
  196. Like _search_regex, but strips HTML tags and unescapes entities.
  197. """
  198. res = self._search_regex(pattern, string, name, default, fatal, flags)
  199. if res:
  200. return clean_html(res).strip()
  201. else:
  202. return res
  203. def _get_login_info(self):
  204. """
  205. Get the the login info as (username, password)
  206. It will look in the netrc file using the _NETRC_MACHINE value
  207. If there's no info available, return (None, None)
  208. """
  209. if self._downloader is None:
  210. return (None, None)
  211. username = None
  212. password = None
  213. downloader_params = self._downloader.params
  214. # Attempt to use provided username and password or .netrc data
  215. if downloader_params.get('username', None) is not None:
  216. username = downloader_params['username']
  217. password = downloader_params['password']
  218. elif downloader_params.get('usenetrc', False):
  219. try:
  220. info = netrc.netrc().authenticators(self._NETRC_MACHINE)
  221. if info is not None:
  222. username = info[0]
  223. password = info[2]
  224. else:
  225. raise netrc.NetrcParseError('No authenticators for %s' % self._NETRC_MACHINE)
  226. except (IOError, netrc.NetrcParseError) as err:
  227. self._downloader.report_warning(u'parsing .netrc: %s' % compat_str(err))
  228. return (username, password)
  229. # Helper functions for extracting OpenGraph info
  230. @staticmethod
  231. def _og_regex(prop):
  232. return r'<meta.+?property=[\'"]og:%s[\'"].+?content=(?:"(.+?)"|\'(.+?)\')' % re.escape(prop)
  233. def _og_search_property(self, prop, html, name=None, **kargs):
  234. if name is None:
  235. name = 'OpenGraph %s' % prop
  236. escaped = self._search_regex(self._og_regex(prop), html, name, flags=re.DOTALL, **kargs)
  237. return unescapeHTML(escaped)
  238. def _og_search_thumbnail(self, html, **kargs):
  239. return self._og_search_property('image', html, u'thumbnail url', fatal=False, **kargs)
  240. def _og_search_description(self, html, **kargs):
  241. return self._og_search_property('description', html, fatal=False, **kargs)
  242. def _og_search_title(self, html, **kargs):
  243. return self._og_search_property('title', html, **kargs)
  244. def _og_search_video_url(self, html, name='video url', **kargs):
  245. return self._html_search_regex([self._og_regex('video:secure_url'),
  246. self._og_regex('video')],
  247. html, name, **kargs)
  248. class SearchInfoExtractor(InfoExtractor):
  249. """
  250. Base class for paged search queries extractors.
  251. They accept urls in the format _SEARCH_KEY(|all|[0-9]):{query}
  252. Instances should define _SEARCH_KEY and _MAX_RESULTS.
  253. """
  254. @classmethod
  255. def _make_valid_url(cls):
  256. return r'%s(?P<prefix>|[1-9][0-9]*|all):(?P<query>[\s\S]+)' % cls._SEARCH_KEY
  257. @classmethod
  258. def suitable(cls, url):
  259. return re.match(cls._make_valid_url(), url) is not None
  260. def _real_extract(self, query):
  261. mobj = re.match(self._make_valid_url(), query)
  262. if mobj is None:
  263. raise ExtractorError(u'Invalid search query "%s"' % query)
  264. prefix = mobj.group('prefix')
  265. query = mobj.group('query')
  266. if prefix == '':
  267. return self._get_n_results(query, 1)
  268. elif prefix == 'all':
  269. return self._get_n_results(query, self._MAX_RESULTS)
  270. else:
  271. n = int(prefix)
  272. if n <= 0:
  273. raise ExtractorError(u'invalid download number %s for query "%s"' % (n, query))
  274. elif n > self._MAX_RESULTS:
  275. self._downloader.report_warning(u'%s returns max %i results (you requested %i)' % (self._SEARCH_KEY, self._MAX_RESULTS, n))
  276. n = self._MAX_RESULTS
  277. return self._get_n_results(query, n)
  278. def _get_n_results(self, query, n):
  279. """Get a specified number of results for a query"""
  280. raise NotImplementedError("This method must be implemented by sublclasses")
  281. @property
  282. def SEARCH_KEY(self):
  283. return self._SEARCH_KEY