11 compat_urllib_request,
19 class InfoExtractor(object):
20 """Information Extractor class.
22 Information extractors are the classes that, given a URL, extract
23 information about the video (or videos) the URL refers to. This
24 information includes the real video URL, the video title, author and
25 others. The information is stored in a dictionary which is then
26 passed to the FileDownloader. The FileDownloader processes this
27 information possibly downloading the video to the file system, among
28 other possible outcomes.
30 The dictionaries must include the following fields:
34 title: Video title, unescaped.
35 ext: Video filename extension.
37 The following fields are optional:
39 format: The video format, defaults to ext (used for --get-format)
40 thumbnails: A list of dictionaries (with the entries "resolution" and
41 "url") for the varying thumbnails
42 thumbnail: Full URL to a video thumbnail image.
43 description: One-line video description.
44 uploader: Full name of the video uploader.
45 upload_date: Video upload date (YYYYMMDD).
46 uploader_id: Nickname or id of the video uploader.
47 location: Physical location of the video.
48 player_url: SWF Player URL (used for rtmpdump).
49 subtitles: The subtitle file contents.
50 view_count: How many users have watched the video on the platform.
51 urlhandle: [internal] The urlHandle to be used to download the file,
52 like returned by urllib.request.urlopen
54 The fields should all be Unicode strings.
56 Subclasses of this one should re-define the _real_initialize() and
57 _real_extract() methods and define a _VALID_URL regexp.
58 Probably, they should also be added to the list of extractors.
60 _real_extract() must return a *list* of information dictionaries as
63 Finally, the _WORKING attribute should be set to False for broken IEs
64 in order to warn the users and skip the tests.
71 def __init__(self, downloader=None):
72 """Constructor. Receives an optional downloader."""
74 self.set_downloader(downloader)
77 def suitable(cls, url):
78 """Receives a URL and returns True if suitable for this IE."""
79 return re.match(cls._VALID_URL, url) is not None
83 """Getter method for _WORKING."""
87 """Initializes an instance (authentication, etc)."""
89 self._real_initialize()
92 def extract(self, url):
93 """Extracts URL information and returns it in list of dicts."""
95 return self._real_extract(url)
97 def set_downloader(self, downloader):
98 """Sets the downloader for this IE."""
99 self._downloader = downloader
101 def _real_initialize(self):
102 """Real initialization process. Redefine in subclasses."""
105 def _real_extract(self, url):
106 """Real extraction process. Redefine in subclasses."""
111 return type(self).__name__[:-2]
113 def _request_webpage(self, url_or_request, video_id, note=None, errnote=None):
114 """ Returns the response handle """
116 self.report_download_webpage(video_id)
117 elif note is not False:
118 self.to_screen(u'%s: %s' % (video_id, note))
120 return compat_urllib_request.urlopen(url_or_request)
121 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
123 errnote = u'Unable to download webpage'
124 raise ExtractorError(u'%s: %s' % (errnote, compat_str(err)), sys.exc_info()[2])
126 def _download_webpage_handle(self, url_or_request, video_id, note=None, errnote=None):
127 """ Returns a tuple (page content as string, URL handle) """
129 # Strip hashes from the URL (#1038)
130 if isinstance(url_or_request, (compat_str, str)):
131 url_or_request = url_or_request.partition('#')[0]
133 urlh = self._request_webpage(url_or_request, video_id, note, errnote)
134 content_type = urlh.headers.get('Content-Type', '')
135 m = re.match(r'[a-zA-Z0-9_.-]+/[a-zA-Z0-9_.-]+\s*;\s*charset=(.+)', content_type)
137 encoding = m.group(1)
140 webpage_bytes = urlh.read()
141 if self._downloader.params.get('dump_intermediate_pages', False):
143 url = url_or_request.get_full_url()
144 except AttributeError:
146 self.to_screen(u'Dumping request to ' + url)
147 dump = base64.b64encode(webpage_bytes).decode('ascii')
148 self._downloader.to_screen(dump)
149 content = webpage_bytes.decode(encoding, 'replace')
150 return (content, urlh)
152 def _download_webpage(self, url_or_request, video_id, note=None, errnote=None):
153 """ Returns the data of the page as a string """
154 return self._download_webpage_handle(url_or_request, video_id, note, errnote)[0]
156 def to_screen(self, msg):
157 """Print msg to screen, prefixing it with '[ie_name]'"""
158 self._downloader.to_screen(u'[%s] %s' % (self.IE_NAME, msg))
160 def report_extraction(self, id_or_name):
161 """Report information extraction."""
162 self.to_screen(u'%s: Extracting information' % id_or_name)
164 def report_download_webpage(self, video_id):
165 """Report webpage download."""
166 self.to_screen(u'%s: Downloading webpage' % video_id)
168 def report_age_confirmation(self):
169 """Report attempt to confirm age."""
170 self.to_screen(u'Confirming age')
172 def report_login(self):
173 """Report attempt to log in."""
174 self.to_screen(u'Logging in')
176 #Methods for following #608
177 def url_result(self, url, ie=None):
178 """Returns a url that points to a page that should be processed"""
179 #TODO: ie should be the class used for getting the info
180 video_info = {'_type': 'url',
184 def playlist_result(self, entries, playlist_id=None, playlist_title=None):
185 """Returns a playlist"""
186 video_info = {'_type': 'playlist',
189 video_info['id'] = playlist_id
191 video_info['title'] = playlist_title
194 def _search_regex(self, pattern, string, name, default=None, fatal=True, flags=0):
196 Perform a regex search on the given string, using a single or a list of
197 patterns returning the first matching group.
198 In case of failure return a default value or raise a WARNING or a
199 ExtractorError, depending on fatal, specifying the field name.
201 if isinstance(pattern, (str, compat_str, compiled_regex_type)):
202 mobj = re.search(pattern, string, flags)
205 mobj = re.search(p, string, flags)
208 if sys.stderr.isatty() and os.name != 'nt':
209 _name = u'\033[0;34m%s\033[0m' % name
214 # return the first matching group
215 return next(g for g in mobj.groups() if g is not None)
216 elif default is not None:
219 raise ExtractorError(u'Unable to extract %s' % _name)
221 self._downloader.report_warning(u'unable to extract %s; '
222 u'please report this issue on http://yt-dl.org/bug' % _name)
225 def _html_search_regex(self, pattern, string, name, default=None, fatal=True, flags=0):
227 Like _search_regex, but strips HTML tags and unescapes entities.
229 res = self._search_regex(pattern, string, name, default, fatal, flags)
231 return clean_html(res).strip()
235 def _get_login_info(self):
237 Get the the login info as (username, password)
238 It will look in the netrc file using the _NETRC_MACHINE value
239 If there's no info available, return (None, None)
241 if self._downloader is None:
246 downloader_params = self._downloader.params
248 # Attempt to use provided username and password or .netrc data
249 if downloader_params.get('username', None) is not None:
250 username = downloader_params['username']
251 password = downloader_params['password']
252 elif downloader_params.get('usenetrc', False):
254 info = netrc.netrc().authenticators(self._NETRC_MACHINE)
259 raise netrc.NetrcParseError('No authenticators for %s' % self._NETRC_MACHINE)
260 except (IOError, netrc.NetrcParseError) as err:
261 self._downloader.report_warning(u'parsing .netrc: %s' % compat_str(err))
263 return (username, password)
265 # Helper functions for extracting OpenGraph info
268 return r'<meta.+?property=[\'"]og:%s[\'"].+?content=(?:"(.+?)"|\'(.+?)\')' % re.escape(prop)
270 def _og_search_property(self, prop, html, name=None, **kargs):
272 name = 'OpenGraph %s' % prop
273 return self._html_search_regex(self._og_regex(prop), html, name, flags=re.DOTALL, **kargs)
275 def _og_search_thumbnail(self, html, **kargs):
276 return self._og_search_property('image', html, u'thumbnail url', fatal=False, **kargs)
278 def _og_search_description(self, html, **kargs):
279 return self._og_search_property('description', html, fatal=False, **kargs)
281 def _og_search_title(self, html, **kargs):
282 return self._og_search_property('title', html, **kargs)
284 def _og_search_video_url(self, html, name='video url', **kargs):
285 return self._html_search_regex([self._og_regex('video:secure_url'),
286 self._og_regex('video')],
289 class SearchInfoExtractor(InfoExtractor):
291 Base class for paged search queries extractors.
292 They accept urls in the format _SEARCH_KEY(|all|[0-9]):{query}
293 Instances should define _SEARCH_KEY and _MAX_RESULTS.
297 def _make_valid_url(cls):
298 return r'%s(?P<prefix>|[1-9][0-9]*|all):(?P<query>[\s\S]+)' % cls._SEARCH_KEY
301 def suitable(cls, url):
302 return re.match(cls._make_valid_url(), url) is not None
304 def _real_extract(self, query):
305 mobj = re.match(self._make_valid_url(), query)
307 raise ExtractorError(u'Invalid search query "%s"' % query)
309 prefix = mobj.group('prefix')
310 query = mobj.group('query')
312 return self._get_n_results(query, 1)
313 elif prefix == 'all':
314 return self._get_n_results(query, self._MAX_RESULTS)
318 raise ExtractorError(u'invalid download number %s for query "%s"' % (n, query))
319 elif n > self._MAX_RESULTS:
320 self._downloader.report_warning(u'%s returns max %i results (you requested %i)' % (self._SEARCH_KEY, self._MAX_RESULTS, n))
321 n = self._MAX_RESULTS
322 return self._get_n_results(query, n)
324 def _get_n_results(self, query, n):
325 """Get a specified number of results for a query"""
326 raise NotImplementedError("This method must be implemented by sublclasses")
329 def SEARCH_KEY(self):
330 return self._SEARCH_KEY