11 compat_urllib_request,
20 class InfoExtractor(object):
21 """Information Extractor class.
23 Information extractors are the classes that, given a URL, extract
24 information about the video (or videos) the URL refers to. This
25 information includes the real video URL, the video title, author and
26 others. The information is stored in a dictionary which is then
27 passed to the FileDownloader. The FileDownloader processes this
28 information possibly downloading the video to the file system, among
29 other possible outcomes.
31 The dictionaries must include the following fields:
35 title: Video title, unescaped.
36 ext: Video filename extension.
38 The following fields are optional:
40 format: The video format, defaults to ext (used for --get-format)
41 thumbnails: A list of dictionaries (with the entries "resolution" and
42 "url") for the varying thumbnails
43 thumbnail: Full URL to a video thumbnail image.
44 description: One-line video description.
45 uploader: Full name of the video uploader.
46 upload_date: Video upload date (YYYYMMDD).
47 uploader_id: Nickname or id of the video uploader.
48 location: Physical location of the video.
49 player_url: SWF Player URL (used for rtmpdump).
50 subtitles: The subtitle file contents.
51 view_count: How many users have watched the video on the platform.
52 urlhandle: [internal] The urlHandle to be used to download the file,
53 like returned by urllib.request.urlopen
55 The fields should all be Unicode strings.
57 Subclasses of this one should re-define the _real_initialize() and
58 _real_extract() methods and define a _VALID_URL regexp.
59 Probably, they should also be added to the list of extractors.
61 _real_extract() must return a *list* of information dictionaries as
64 Finally, the _WORKING attribute should be set to False for broken IEs
65 in order to warn the users and skip the tests.
72 def __init__(self, downloader=None):
73 """Constructor. Receives an optional downloader."""
75 self.set_downloader(downloader)
78 def suitable(cls, url):
79 """Receives a URL and returns True if suitable for this IE."""
81 # This does not use has/getattr intentionally - we want to know whether
82 # we have cached the regexp for *this* class, whereas getattr would also
83 # match the superclass
84 if '_VALID_URL_RE' not in cls.__dict__:
85 cls._VALID_URL_RE = re.compile(cls._VALID_URL)
86 return cls._VALID_URL_RE.match(url) is not None
90 """Getter method for _WORKING."""
94 """Initializes an instance (authentication, etc)."""
96 self._real_initialize()
99 def extract(self, url):
100 """Extracts URL information and returns it in list of dicts."""
102 return self._real_extract(url)
104 def set_downloader(self, downloader):
105 """Sets the downloader for this IE."""
106 self._downloader = downloader
108 def _real_initialize(self):
109 """Real initialization process. Redefine in subclasses."""
112 def _real_extract(self, url):
113 """Real extraction process. Redefine in subclasses."""
118 return type(self).__name__[:-2]
120 def _request_webpage(self, url_or_request, video_id, note=None, errnote=None):
121 """ Returns the response handle """
123 self.report_download_webpage(video_id)
124 elif note is not False:
125 self.to_screen(u'%s: %s' % (video_id, note))
127 return compat_urllib_request.urlopen(url_or_request)
128 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
130 errnote = u'Unable to download webpage'
131 raise ExtractorError(u'%s: %s' % (errnote, compat_str(err)), sys.exc_info()[2])
133 def _download_webpage_handle(self, url_or_request, video_id, note=None, errnote=None):
134 """ Returns a tuple (page content as string, URL handle) """
136 # Strip hashes from the URL (#1038)
137 if isinstance(url_or_request, (compat_str, str)):
138 url_or_request = url_or_request.partition('#')[0]
140 urlh = self._request_webpage(url_or_request, video_id, note, errnote)
141 content_type = urlh.headers.get('Content-Type', '')
142 m = re.match(r'[a-zA-Z0-9_.-]+/[a-zA-Z0-9_.-]+\s*;\s*charset=(.+)', content_type)
144 encoding = m.group(1)
147 webpage_bytes = urlh.read()
148 if self._downloader.params.get('dump_intermediate_pages', False):
150 url = url_or_request.get_full_url()
151 except AttributeError:
153 self.to_screen(u'Dumping request to ' + url)
154 dump = base64.b64encode(webpage_bytes).decode('ascii')
155 self._downloader.to_screen(dump)
156 content = webpage_bytes.decode(encoding, 'replace')
157 return (content, urlh)
159 def _download_webpage(self, url_or_request, video_id, note=None, errnote=None):
160 """ Returns the data of the page as a string """
161 return self._download_webpage_handle(url_or_request, video_id, note, errnote)[0]
163 def to_screen(self, msg):
164 """Print msg to screen, prefixing it with '[ie_name]'"""
165 self._downloader.to_screen(u'[%s] %s' % (self.IE_NAME, msg))
167 def report_extraction(self, id_or_name):
168 """Report information extraction."""
169 self.to_screen(u'%s: Extracting information' % id_or_name)
171 def report_download_webpage(self, video_id):
172 """Report webpage download."""
173 self.to_screen(u'%s: Downloading webpage' % video_id)
175 def report_age_confirmation(self):
176 """Report attempt to confirm age."""
177 self.to_screen(u'Confirming age')
179 def report_login(self):
180 """Report attempt to log in."""
181 self.to_screen(u'Logging in')
183 #Methods for following #608
184 def url_result(self, url, ie=None):
185 """Returns a url that points to a page that should be processed"""
186 #TODO: ie should be the class used for getting the info
187 video_info = {'_type': 'url',
191 def playlist_result(self, entries, playlist_id=None, playlist_title=None):
192 """Returns a playlist"""
193 video_info = {'_type': 'playlist',
196 video_info['id'] = playlist_id
198 video_info['title'] = playlist_title
201 def _search_regex(self, pattern, string, name, default=None, fatal=True, flags=0):
203 Perform a regex search on the given string, using a single or a list of
204 patterns returning the first matching group.
205 In case of failure return a default value or raise a WARNING or a
206 ExtractorError, depending on fatal, specifying the field name.
208 if isinstance(pattern, (str, compat_str, compiled_regex_type)):
209 mobj = re.search(pattern, string, flags)
212 mobj = re.search(p, string, flags)
215 if sys.stderr.isatty() and os.name != 'nt':
216 _name = u'\033[0;34m%s\033[0m' % name
221 # return the first matching group
222 return next(g for g in mobj.groups() if g is not None)
223 elif default is not None:
226 raise ExtractorError(u'Unable to extract %s' % _name)
228 self._downloader.report_warning(u'unable to extract %s; '
229 u'please report this issue on http://yt-dl.org/bug' % _name)
232 def _html_search_regex(self, pattern, string, name, default=None, fatal=True, flags=0):
234 Like _search_regex, but strips HTML tags and unescapes entities.
236 res = self._search_regex(pattern, string, name, default, fatal, flags)
238 return clean_html(res).strip()
242 def _get_login_info(self):
244 Get the the login info as (username, password)
245 It will look in the netrc file using the _NETRC_MACHINE value
246 If there's no info available, return (None, None)
248 if self._downloader is None:
253 downloader_params = self._downloader.params
255 # Attempt to use provided username and password or .netrc data
256 if downloader_params.get('username', None) is not None:
257 username = downloader_params['username']
258 password = downloader_params['password']
259 elif downloader_params.get('usenetrc', False):
261 info = netrc.netrc().authenticators(self._NETRC_MACHINE)
266 raise netrc.NetrcParseError('No authenticators for %s' % self._NETRC_MACHINE)
267 except (IOError, netrc.NetrcParseError) as err:
268 self._downloader.report_warning(u'parsing .netrc: %s' % compat_str(err))
270 return (username, password)
272 # Helper functions for extracting OpenGraph info
275 return r'<meta.+?property=[\'"]og:%s[\'"].+?content=(?:"(.+?)"|\'(.+?)\')' % re.escape(prop)
277 def _og_search_property(self, prop, html, name=None, **kargs):
279 name = 'OpenGraph %s' % prop
280 escaped = self._search_regex(self._og_regex(prop), html, name, flags=re.DOTALL, **kargs)
281 return unescapeHTML(escaped)
283 def _og_search_thumbnail(self, html, **kargs):
284 return self._og_search_property('image', html, u'thumbnail url', fatal=False, **kargs)
286 def _og_search_description(self, html, **kargs):
287 return self._og_search_property('description', html, fatal=False, **kargs)
289 def _og_search_title(self, html, **kargs):
290 return self._og_search_property('title', html, **kargs)
292 def _og_search_video_url(self, html, name='video url', **kargs):
293 return self._html_search_regex([self._og_regex('video:secure_url'),
294 self._og_regex('video')],
297 class SearchInfoExtractor(InfoExtractor):
299 Base class for paged search queries extractors.
300 They accept urls in the format _SEARCH_KEY(|all|[0-9]):{query}
301 Instances should define _SEARCH_KEY and _MAX_RESULTS.
305 def _make_valid_url(cls):
306 return r'%s(?P<prefix>|[1-9][0-9]*|all):(?P<query>[\s\S]+)' % cls._SEARCH_KEY
309 def suitable(cls, url):
310 return re.match(cls._make_valid_url(), url) is not None
312 def _real_extract(self, query):
313 mobj = re.match(self._make_valid_url(), query)
315 raise ExtractorError(u'Invalid search query "%s"' % query)
317 prefix = mobj.group('prefix')
318 query = mobj.group('query')
320 return self._get_n_results(query, 1)
321 elif prefix == 'all':
322 return self._get_n_results(query, self._MAX_RESULTS)
326 raise ExtractorError(u'invalid download number %s for query "%s"' % (n, query))
327 elif n > self._MAX_RESULTS:
328 self._downloader.report_warning(u'%s returns max %i results (you requested %i)' % (self._SEARCH_KEY, self._MAX_RESULTS, n))
329 n = self._MAX_RESULTS
330 return self._get_n_results(query, n)
332 def _get_n_results(self, query, n):
333 """Get a specified number of results for a query"""
334 raise NotImplementedError("This method must be implemented by sublclasses")
337 def SEARCH_KEY(self):
338 return self._SEARCH_KEY