10 compat_urllib_request,
18 class InfoExtractor(object):
19 """Information Extractor class.
21 Information extractors are the classes that, given a URL, extract
22 information about the video (or videos) the URL refers to. This
23 information includes the real video URL, the video title, author and
24 others. The information is stored in a dictionary which is then
25 passed to the FileDownloader. The FileDownloader processes this
26 information possibly downloading the video to the file system, among
27 other possible outcomes.
29 The dictionaries must include the following fields:
33 title: Video title, unescaped.
34 ext: Video filename extension.
36 The following fields are optional:
38 format: The video format, defaults to ext (used for --get-format)
39 thumbnails: A list of dictionaries (with the entries "resolution" and
40 "url") for the varying thumbnails
41 thumbnail: Full URL to a video thumbnail image.
42 description: One-line video description.
43 uploader: Full name of the video uploader.
44 upload_date: Video upload date (YYYYMMDD).
45 uploader_id: Nickname or id of the video uploader.
46 location: Physical location of the video.
47 player_url: SWF Player URL (used for rtmpdump).
48 subtitles: The subtitle file contents.
49 view_count: How many users have watched the video on the platform.
50 urlhandle: [internal] The urlHandle to be used to download the file,
51 like returned by urllib.request.urlopen
53 The fields should all be Unicode strings.
55 Subclasses of this one should re-define the _real_initialize() and
56 _real_extract() methods and define a _VALID_URL regexp.
57 Probably, they should also be added to the list of extractors.
59 _real_extract() must return a *list* of information dictionaries as
62 Finally, the _WORKING attribute should be set to False for broken IEs
63 in order to warn the users and skip the tests.
70 def __init__(self, downloader=None):
71 """Constructor. Receives an optional downloader."""
73 self.set_downloader(downloader)
76 def suitable(cls, url):
77 """Receives a URL and returns True if suitable for this IE."""
78 return re.match(cls._VALID_URL, url) is not None
82 """Getter method for _WORKING."""
86 """Initializes an instance (authentication, etc)."""
88 self._real_initialize()
91 def extract(self, url):
92 """Extracts URL information and returns it in list of dicts."""
94 return self._real_extract(url)
96 def set_downloader(self, downloader):
97 """Sets the downloader for this IE."""
98 self._downloader = downloader
100 def _real_initialize(self):
101 """Real initialization process. Redefine in subclasses."""
104 def _real_extract(self, url):
105 """Real extraction process. Redefine in subclasses."""
110 return type(self).__name__[:-2]
112 def _request_webpage(self, url_or_request, video_id, note=None, errnote=None):
113 """ Returns the response handle """
115 self.report_download_webpage(video_id)
116 elif note is not False:
117 self.to_screen(u'%s: %s' % (video_id, note))
119 return compat_urllib_request.urlopen(url_or_request)
120 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
122 errnote = u'Unable to download webpage'
123 raise ExtractorError(u'%s: %s' % (errnote, compat_str(err)), sys.exc_info()[2])
125 def _download_webpage_handle(self, url_or_request, video_id, note=None, errnote=None):
126 """ Returns a tuple (page content as string, URL handle) """
127 urlh = self._request_webpage(url_or_request, video_id, note, errnote)
128 content_type = urlh.headers.get('Content-Type', '')
129 m = re.match(r'[a-zA-Z0-9_.-]+/[a-zA-Z0-9_.-]+\s*;\s*charset=(.+)', content_type)
131 encoding = m.group(1)
134 webpage_bytes = urlh.read()
135 if self._downloader.params.get('dump_intermediate_pages', False):
137 url = url_or_request.get_full_url()
138 except AttributeError:
140 self.to_screen(u'Dumping request to ' + url)
141 dump = base64.b64encode(webpage_bytes).decode('ascii')
142 self._downloader.to_screen(dump)
143 content = webpage_bytes.decode(encoding, 'replace')
144 return (content, urlh)
146 def _download_webpage(self, url_or_request, video_id, note=None, errnote=None):
147 """ Returns the data of the page as a string """
148 return self._download_webpage_handle(url_or_request, video_id, note, errnote)[0]
150 def to_screen(self, msg):
151 """Print msg to screen, prefixing it with '[ie_name]'"""
152 self._downloader.to_screen(u'[%s] %s' % (self.IE_NAME, msg))
154 def report_extraction(self, id_or_name):
155 """Report information extraction."""
156 self.to_screen(u'%s: Extracting information' % id_or_name)
158 def report_download_webpage(self, video_id):
159 """Report webpage download."""
160 self.to_screen(u'%s: Downloading webpage' % video_id)
162 def report_age_confirmation(self):
163 """Report attempt to confirm age."""
164 self.to_screen(u'Confirming age')
166 #Methods for following #608
167 #They set the correct value of the '_type' key
168 def video_result(self, video_info):
169 """Returns a video"""
170 video_info['_type'] = 'video'
172 def url_result(self, url, ie=None):
173 """Returns a url that points to a page that should be processed"""
174 #TODO: ie should be the class used for getting the info
175 video_info = {'_type': 'url',
179 def playlist_result(self, entries, playlist_id=None, playlist_title=None):
180 """Returns a playlist"""
181 video_info = {'_type': 'playlist',
184 video_info['id'] = playlist_id
186 video_info['title'] = playlist_title
189 def _search_regex(self, pattern, string, name, default=None, fatal=True, flags=0):
191 Perform a regex search on the given string, using a single or a list of
192 patterns returning the first matching group.
193 In case of failure return a default value or raise a WARNING or a
194 ExtractorError, depending on fatal, specifying the field name.
196 if isinstance(pattern, (str, compat_str, compiled_regex_type)):
197 mobj = re.search(pattern, string, flags)
200 mobj = re.search(p, string, flags)
203 if sys.stderr.isatty() and os.name != 'nt':
204 _name = u'\033[0;34m%s\033[0m' % name
209 # return the first matching group
210 return next(g for g in mobj.groups() if g is not None)
211 elif default is not None:
214 raise ExtractorError(u'Unable to extract %s' % _name)
216 self._downloader.report_warning(u'unable to extract %s; '
217 u'please report this issue on http://yt-dl.org/bug' % _name)
220 def _html_search_regex(self, pattern, string, name, default=None, fatal=True, flags=0):
222 Like _search_regex, but strips HTML tags and unescapes entities.
224 res = self._search_regex(pattern, string, name, default, fatal, flags)
226 return clean_html(res).strip()
230 class SearchInfoExtractor(InfoExtractor):
232 Base class for paged search queries extractors.
233 They accept urls in the format _SEARCH_KEY(|all|[0-9]):{query}
234 Instances should define _SEARCH_KEY and _MAX_RESULTS.
238 def _make_valid_url(cls):
239 return r'%s(?P<prefix>|[1-9][0-9]*|all):(?P<query>[\s\S]+)' % cls._SEARCH_KEY
242 def suitable(cls, url):
243 return re.match(cls._make_valid_url(), url) is not None
245 def _real_extract(self, query):
246 mobj = re.match(self._make_valid_url(), query)
248 raise ExtractorError(u'Invalid search query "%s"' % query)
250 prefix = mobj.group('prefix')
251 query = mobj.group('query')
253 return self._get_n_results(query, 1)
254 elif prefix == 'all':
255 return self._get_n_results(query, self._MAX_RESULTS)
259 raise ExtractorError(u'invalid download number %s for query "%s"' % (n, query))
260 elif n > self._MAX_RESULTS:
261 self._downloader.report_warning(u'%s returns max %i results (you requested %i)' % (self._SEARCH_KEY, self._MAX_RESULTS, n))
262 n = self._MAX_RESULTS
263 return self._get_n_results(query, n)
265 def _get_n_results(self, query, n):
266 """Get a specified number of results for a query"""
267 raise NotImplementedError("This method must be implemented by sublclasses")
270 def SEARCH_KEY(self):
271 return self._SEARCH_KEY