1 from __future__ import absolute_import
12 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 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 urlhandle: [internal] The urlHandle to be used to download the file,
50 like returned by urllib.request.urlopen
52 The fields should all be Unicode strings.
54 Subclasses of this one should re-define the _real_initialize() and
55 _real_extract() methods and define a _VALID_URL regexp.
56 Probably, they should also be added to the list of extractors.
58 _real_extract() must return a *list* of information dictionaries as
61 Finally, the _WORKING attribute should be set to False for broken IEs
62 in order to warn the users and skip the tests.
69 def __init__(self, downloader=None):
70 """Constructor. Receives an optional downloader."""
72 self.set_downloader(downloader)
75 def suitable(cls, url):
76 """Receives a URL and returns True if suitable for this IE."""
77 return re.match(cls._VALID_URL, url) is not None
81 """Getter method for _WORKING."""
85 """Initializes an instance (authentication, etc)."""
87 self._real_initialize()
90 def extract(self, url):
91 """Extracts URL information and returns it in list of dicts."""
93 return self._real_extract(url)
95 def set_downloader(self, downloader):
96 """Sets the downloader for this IE."""
97 self._downloader = downloader
99 def _real_initialize(self):
100 """Real initialization process. Redefine in subclasses."""
103 def _real_extract(self, url):
104 """Real extraction process. Redefine in subclasses."""
109 return type(self).__name__[:-2]
111 def _request_webpage(self, url_or_request, video_id, note=None, errnote=None):
112 """ Returns the response handle """
114 self.report_download_webpage(video_id)
115 elif note is not False:
116 self.to_screen(u'%s: %s' % (video_id, note))
118 return compat_urllib_request.urlopen(url_or_request)
119 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
121 errnote = u'Unable to download webpage'
122 raise ExtractorError(u'%s: %s' % (errnote, compat_str(err)), sys.exc_info()[2])
124 def _download_webpage_handle(self, url_or_request, video_id, note=None, errnote=None):
125 """ Returns a tuple (page content as string, URL handle) """
126 urlh = self._request_webpage(url_or_request, video_id, note, errnote)
127 content_type = urlh.headers.get('Content-Type', '')
128 m = re.match(r'[a-zA-Z0-9_.-]+/[a-zA-Z0-9_.-]+\s*;\s*charset=(.+)', content_type)
130 encoding = m.group(1)
133 webpage_bytes = urlh.read()
134 if self._downloader.params.get('dump_intermediate_pages', False):
136 url = url_or_request.get_full_url()
137 except AttributeError:
139 self.to_screen(u'Dumping request to ' + url)
140 dump = base64.b64encode(webpage_bytes).decode('ascii')
141 self._downloader.to_screen(dump)
142 content = webpage_bytes.decode(encoding, 'replace')
143 return (content, urlh)
145 def _download_webpage(self, url_or_request, video_id, note=None, errnote=None):
146 """ Returns the data of the page as a string """
147 return self._download_webpage_handle(url_or_request, video_id, note, errnote)[0]
149 def to_screen(self, msg):
150 """Print msg to screen, prefixing it with '[ie_name]'"""
151 self._downloader.to_screen(u'[%s] %s' % (self.IE_NAME, msg))
153 def report_extraction(self, id_or_name):
154 """Report information extraction."""
155 self.to_screen(u'%s: Extracting information' % id_or_name)
157 def report_download_webpage(self, video_id):
158 """Report webpage download."""
159 self.to_screen(u'%s: Downloading webpage' % video_id)
161 def report_age_confirmation(self):
162 """Report attempt to confirm age."""
163 self.to_screen(u'Confirming age')
165 #Methods for following #608
166 #They set the correct value of the '_type' key
167 def video_result(self, video_info):
168 """Returns a video"""
169 video_info['_type'] = 'video'
171 def url_result(self, url, ie=None):
172 """Returns a url that points to a page that should be processed"""
173 #TODO: ie should be the class used for getting the info
174 video_info = {'_type': 'url',
178 def playlist_result(self, entries, playlist_id=None, playlist_title=None):
179 """Returns a playlist"""
180 video_info = {'_type': 'playlist',
183 video_info['id'] = playlist_id
185 video_info['title'] = playlist_title
188 def _search_regex(self, pattern, string, name, default=None, fatal=True, flags=0):
190 Perform a regex search on the given string, using a single or a list of
191 patterns returning the first matching group.
192 In case of failure return a default value or raise a WARNING or a
193 ExtractorError, depending on fatal, specifying the field name.
195 if isinstance(pattern, (str, compat_str, compiled_regex_type)):
196 mobj = re.search(pattern, string, flags)
199 mobj = re.search(p, string, flags)
202 if sys.stderr.isatty() and os.name != 'nt':
203 _name = u'\033[0;34m%s\033[0m' % name
208 # return the first matching group
209 return next(g for g in mobj.groups() if g is not None)
210 elif default is not None:
213 raise ExtractorError(u'Unable to extract %s' % _name)
215 self._downloader.report_warning(u'unable to extract %s; '
216 u'please report this issue on GitHub.' % _name)
219 def _html_search_regex(self, pattern, string, name, default=None, fatal=True, flags=0):
221 Like _search_regex, but strips HTML tags and unescapes entities.
223 res = self._search_regex(pattern, string, name, default, fatal, flags)
225 return clean_html(res).strip()
229 class SearchInfoExtractor(InfoExtractor):
231 Base class for paged search queries extractors.
232 They accept urls in the format _SEARCH_KEY(|all|[0-9]):{query}
233 Instances should define _SEARCH_KEY and _MAX_RESULTS.
237 def _make_valid_url(cls):
238 return r'%s(?P<prefix>|[1-9][0-9]*|all):(?P<query>[\s\S]+)' % cls._SEARCH_KEY
241 def suitable(cls, url):
242 return re.match(cls._make_valid_url(), url) is not None
244 def _real_extract(self, query):
245 mobj = re.match(self._make_valid_url(), query)
247 raise ExtractorError(u'Invalid search query "%s"' % query)
249 prefix = mobj.group('prefix')
250 query = mobj.group('query')
252 return self._get_n_results(query, 1)
253 elif prefix == 'all':
254 return self._get_n_results(query, self._MAX_RESULTS)
258 raise ExtractorError(u'invalid download number %s for query "%s"' % (n, query))
259 elif n > self._MAX_RESULTS:
260 self._downloader.report_warning(u'%s returns max %i results (you requested %i)' % (self._SEARCH_KEY, self._MAX_RESULTS, n))
261 n = self._MAX_RESULTS
262 return self._get_n_results(query, n)
264 def _get_n_results(self, query, n):
265 """Get a specified number of results for a query"""
266 raise NotImplementedError("This method must be implemented by sublclasses")