1 from __future__ import unicode_literals
13 import xml.etree.ElementTree
15 from ..compat import (
19 compat_urllib_parse_urlparse,
34 _NO_DEFAULT = object()
37 class InfoExtractor(object):
38 """Information Extractor class.
40 Information extractors are the classes that, given a URL, extract
41 information about the video (or videos) the URL refers to. This
42 information includes the real video URL, the video title, author and
43 others. The information is stored in a dictionary which is then
44 passed to the YoutubeDL. The YoutubeDL processes this
45 information possibly downloading the video to the file system, among
46 other possible outcomes.
48 The type field determines the the type of the result.
49 By far the most common value (and the default if _type is missing) is
50 "video", which indicates a single video.
52 For a video, the dictionaries must include the following fields:
55 title: Video title, unescaped.
57 Additionally, it must contain either a formats entry or a url one:
59 formats: A list of dictionaries for each format available, ordered
60 from worst to best quality.
63 * url Mandatory. The URL of the video file
64 * ext Will be calculated from url if missing
65 * format A human-readable description of the format
66 ("mp4 container with h264/opus").
67 Calculated from the format_id, width, height.
68 and format_note fields if missing.
69 * format_id A short description of the format
70 ("mp4_h264_opus" or "19").
71 Technically optional, but strongly recommended.
72 * format_note Additional info about the format
73 ("3D" or "DASH video")
74 * width Width of the video, if known
75 * height Height of the video, if known
76 * resolution Textual description of width and height
77 * tbr Average bitrate of audio and video in KBit/s
78 * abr Average audio bitrate in KBit/s
79 * acodec Name of the audio codec in use
80 * asr Audio sampling rate in Hertz
81 * vbr Average video bitrate in KBit/s
83 * vcodec Name of the video codec in use
84 * container Name of the container format
85 * filesize The number of bytes, if known in advance
86 * filesize_approx An estimate for the number of bytes
87 * player_url SWF Player URL (used for rtmpdump).
88 * protocol The protocol that will be used for the actual
90 "http", "https", "rtsp", "rtmp", "m3u8" or so.
91 * preference Order number of this format. If this field is
92 present and not None, the formats get sorted
93 by this field, regardless of all other values.
94 -1 for default (order by other properties),
95 -2 or smaller for less than default.
96 < -1000 to hide the format (if there is
97 another one which is strictly better)
98 * language_preference Is this in the correct requested
100 10 if it's what the URL is about,
101 -1 for default (don't know),
102 -10 otherwise, other values reserved for now.
103 * quality Order number of the video quality of this
104 format, irrespective of the file format.
105 -1 for default (order by other properties),
106 -2 or smaller for less than default.
107 * source_preference Order number for this video source
108 (quality takes higher priority)
109 -1 for default (order by other properties),
110 -2 or smaller for less than default.
111 * http_referer HTTP Referer header value to set.
112 * http_method HTTP method to use for the download.
113 * http_headers A dictionary of additional HTTP headers
114 to add to the request.
115 * http_post_data Additional data to send with a POST
117 url: Final video URL.
118 ext: Video filename extension.
119 format: The video format, defaults to ext (used for --get-format)
120 player_url: SWF Player URL (used for rtmpdump).
122 The following fields are optional:
124 alt_title: A secondary title of the video.
125 display_id An alternative identifier for the video, not necessarily
126 unique, but available before title. Typically, id is
127 something like "4234987", title "Dancing naked mole rats",
128 and display_id "dancing-naked-mole-rats"
129 thumbnails: A list of dictionaries, with the following entries:
131 * "width" (optional, int)
132 * "height" (optional, int)
133 * "resolution" (optional, string "{width}x{height"},
135 thumbnail: Full URL to a video thumbnail image.
136 description: Full video description.
137 uploader: Full name of the video uploader.
138 timestamp: UNIX timestamp of the moment the video became available.
139 upload_date: Video upload date (YYYYMMDD).
140 If not explicitly set, calculated from timestamp.
141 uploader_id: Nickname or id of the video uploader.
142 location: Physical location where the video was filmed.
143 subtitles: The subtitle file contents as a dictionary in the format
144 {language: subtitles}.
145 duration: Length of the video in seconds, as an integer.
146 view_count: How many users have watched the video on the platform.
147 like_count: Number of positive ratings of the video
148 dislike_count: Number of negative ratings of the video
149 comment_count: Number of comments on the video
150 age_limit: Age restriction for the video, as an integer (years)
151 webpage_url: The url to the video webpage, if given to youtube-dl it
152 should allow to get the same result again. (It will be set
153 by YoutubeDL if it's missing)
154 categories: A list of categories that the video falls in, for example
156 is_live: True, False, or None (=unknown). Whether this video is a
157 live stream that goes on instead of a fixed-length video.
159 Unless mentioned otherwise, the fields should be Unicode strings.
161 Unless mentioned otherwise, None is equivalent to absence of information.
164 _type "playlist" indicates multiple videos.
165 There must be a key "entries", which is a list, an iterable, or a PagedList
166 object, each element of which is a valid dictionary by this specification.
168 Additionally, playlists can have "title" and "id" attributes with the same
169 semantics as videos (see above).
172 _type "multi_video" indicates that there are multiple videos that
173 form a single show, for examples multiple acts of an opera or TV episode.
174 It must have an entries key like a playlist and contain all the keys
175 required for a video at the same time.
178 _type "url" indicates that the video must be extracted from another
179 location, possibly by a different extractor. Its only required key is:
180 "url" - the next URL to extract.
181 The key "ie_key" can be set to the class name (minus the trailing "IE",
182 e.g. "Youtube") if the extractor class is known in advance.
183 Additionally, the dictionary may have any properties of the resolved entity
184 known in advance, for example "title" if the title of the referred video is
188 _type "url_transparent" entities have the same specification as "url", but
189 indicate that the given additional information is more precise than the one
190 associated with the resolved URL.
191 This is useful when a site employs a video service that hosts the video and
192 its technical metadata, but that video service does not embed a useful
193 title, description etc.
196 Subclasses of this one should re-define the _real_initialize() and
197 _real_extract() methods and define a _VALID_URL regexp.
198 Probably, they should also be added to the list of extractors.
200 Finally, the _WORKING attribute should be set to False for broken IEs
201 in order to warn the users and skip the tests.
208 def __init__(self, downloader=None):
209 """Constructor. Receives an optional downloader."""
211 self.set_downloader(downloader)
214 def suitable(cls, url):
215 """Receives a URL and returns True if suitable for this IE."""
217 # This does not use has/getattr intentionally - we want to know whether
218 # we have cached the regexp for *this* class, whereas getattr would also
219 # match the superclass
220 if '_VALID_URL_RE' not in cls.__dict__:
221 cls._VALID_URL_RE = re.compile(cls._VALID_URL)
222 return cls._VALID_URL_RE.match(url) is not None
225 def _match_id(cls, url):
226 if '_VALID_URL_RE' not in cls.__dict__:
227 cls._VALID_URL_RE = re.compile(cls._VALID_URL)
228 m = cls._VALID_URL_RE.match(url)
234 """Getter method for _WORKING."""
237 def initialize(self):
238 """Initializes an instance (authentication, etc)."""
240 self._real_initialize()
243 def extract(self, url):
244 """Extracts URL information and returns it in list of dicts."""
246 return self._real_extract(url)
248 def set_downloader(self, downloader):
249 """Sets the downloader for this IE."""
250 self._downloader = downloader
252 def _real_initialize(self):
253 """Real initialization process. Redefine in subclasses."""
256 def _real_extract(self, url):
257 """Real extraction process. Redefine in subclasses."""
262 """A string for getting the InfoExtractor with get_info_extractor"""
263 return cls.__name__[:-2]
267 return type(self).__name__[:-2]
269 def _request_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True):
270 """ Returns the response handle """
272 self.report_download_webpage(video_id)
273 elif note is not False:
275 self.to_screen('%s' % (note,))
277 self.to_screen('%s: %s' % (video_id, note))
279 return self._downloader.urlopen(url_or_request)
280 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
284 errnote = 'Unable to download webpage'
285 errmsg = '%s: %s' % (errnote, compat_str(err))
287 raise ExtractorError(errmsg, sys.exc_info()[2], cause=err)
289 self._downloader.report_warning(errmsg)
292 def _download_webpage_handle(self, url_or_request, video_id, note=None, errnote=None, fatal=True):
293 """ Returns a tuple (page content as string, URL handle) """
294 # Strip hashes from the URL (#1038)
295 if isinstance(url_or_request, (compat_str, str)):
296 url_or_request = url_or_request.partition('#')[0]
298 urlh = self._request_webpage(url_or_request, video_id, note, errnote, fatal)
302 content = self._webpage_read_content(urlh, url_or_request, video_id, note, errnote, fatal)
303 return (content, urlh)
305 def _webpage_read_content(self, urlh, url_or_request, video_id, note=None, errnote=None, fatal=True, prefix=None):
306 content_type = urlh.headers.get('Content-Type', '')
307 webpage_bytes = urlh.read()
308 if prefix is not None:
309 webpage_bytes = prefix + webpage_bytes
310 m = re.match(r'[a-zA-Z0-9_.-]+/[a-zA-Z0-9_.-]+\s*;\s*charset=(.+)', content_type)
312 encoding = m.group(1)
314 m = re.search(br'<meta[^>]+charset=[\'"]?([^\'")]+)[ /\'">]',
315 webpage_bytes[:1024])
317 encoding = m.group(1).decode('ascii')
318 elif webpage_bytes.startswith(b'\xff\xfe'):
322 if self._downloader.params.get('dump_intermediate_pages', False):
324 url = url_or_request.get_full_url()
325 except AttributeError:
327 self.to_screen('Dumping request to ' + url)
328 dump = base64.b64encode(webpage_bytes).decode('ascii')
329 self._downloader.to_screen(dump)
330 if self._downloader.params.get('write_pages', False):
332 url = url_or_request.get_full_url()
333 except AttributeError:
335 basen = '%s_%s' % (video_id, url)
337 h = '___' + hashlib.md5(basen.encode('utf-8')).hexdigest()
338 basen = basen[:240 - len(h)] + h
339 raw_filename = basen + '.dump'
340 filename = sanitize_filename(raw_filename, restricted=True)
341 self.to_screen('Saving request to ' + filename)
342 # Working around MAX_PATH limitation on Windows (see
343 # http://msdn.microsoft.com/en-us/library/windows/desktop/aa365247(v=vs.85).aspx)
345 absfilepath = os.path.abspath(filename)
346 if len(absfilepath) > 259:
347 filename = '\\\\?\\' + absfilepath
348 with open(filename, 'wb') as outf:
349 outf.write(webpage_bytes)
352 content = webpage_bytes.decode(encoding, 'replace')
354 content = webpage_bytes.decode('utf-8', 'replace')
356 if ('<title>Access to this site is blocked</title>' in content and
357 'Websense' in content[:512]):
358 msg = 'Access to this webpage has been blocked by Websense filtering software in your network.'
359 blocked_iframe = self._html_search_regex(
360 r'<iframe src="([^"]+)"', content,
361 'Websense information URL', default=None)
363 msg += ' Visit %s for more details' % blocked_iframe
364 raise ExtractorError(msg, expected=True)
368 def _download_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True):
369 """ Returns the data of the page as a string """
370 res = self._download_webpage_handle(url_or_request, video_id, note, errnote, fatal)
377 def _download_xml(self, url_or_request, video_id,
378 note='Downloading XML', errnote='Unable to download XML',
379 transform_source=None, fatal=True):
380 """Return the xml as an xml.etree.ElementTree.Element"""
381 xml_string = self._download_webpage(
382 url_or_request, video_id, note, errnote, fatal=fatal)
383 if xml_string is False:
386 xml_string = transform_source(xml_string)
387 return xml.etree.ElementTree.fromstring(xml_string.encode('utf-8'))
389 def _download_json(self, url_or_request, video_id,
390 note='Downloading JSON metadata',
391 errnote='Unable to download JSON metadata',
392 transform_source=None,
394 json_string = self._download_webpage(
395 url_or_request, video_id, note, errnote, fatal=fatal)
396 if (not fatal) and json_string is False:
398 return self._parse_json(
399 json_string, video_id, transform_source=transform_source, fatal=fatal)
401 def _parse_json(self, json_string, video_id, transform_source=None, fatal=True):
403 json_string = transform_source(json_string)
405 return json.loads(json_string)
406 except ValueError as ve:
407 errmsg = '%s: Failed to parse JSON ' % video_id
409 raise ExtractorError(errmsg, cause=ve)
411 self.report_warning(errmsg + str(ve))
413 def report_warning(self, msg, video_id=None):
414 idstr = '' if video_id is None else '%s: ' % video_id
415 self._downloader.report_warning(
416 '[%s] %s%s' % (self.IE_NAME, idstr, msg))
418 def to_screen(self, msg):
419 """Print msg to screen, prefixing it with '[ie_name]'"""
420 self._downloader.to_screen('[%s] %s' % (self.IE_NAME, msg))
422 def report_extraction(self, id_or_name):
423 """Report information extraction."""
424 self.to_screen('%s: Extracting information' % id_or_name)
426 def report_download_webpage(self, video_id):
427 """Report webpage download."""
428 self.to_screen('%s: Downloading webpage' % video_id)
430 def report_age_confirmation(self):
431 """Report attempt to confirm age."""
432 self.to_screen('Confirming age')
434 def report_login(self):
435 """Report attempt to log in."""
436 self.to_screen('Logging in')
438 # Methods for following #608
440 def url_result(url, ie=None, video_id=None):
441 """Returns a url that points to a page that should be processed"""
442 # TODO: ie should be the class used for getting the info
443 video_info = {'_type': 'url',
446 if video_id is not None:
447 video_info['id'] = video_id
451 def playlist_result(entries, playlist_id=None, playlist_title=None, playlist_description=None):
452 """Returns a playlist"""
453 video_info = {'_type': 'playlist',
456 video_info['id'] = playlist_id
458 video_info['title'] = playlist_title
459 if playlist_description:
460 video_info['description'] = playlist_description
463 def _search_regex(self, pattern, string, name, default=_NO_DEFAULT, fatal=True, flags=0, group=None):
465 Perform a regex search on the given string, using a single or a list of
466 patterns returning the first matching group.
467 In case of failure return a default value or raise a WARNING or a
468 RegexNotFoundError, depending on fatal, specifying the field name.
470 if isinstance(pattern, (str, compat_str, compiled_regex_type)):
471 mobj = re.search(pattern, string, flags)
474 mobj = re.search(p, string, flags)
478 if os.name != 'nt' and sys.stderr.isatty():
479 _name = '\033[0;34m%s\033[0m' % name
485 # return the first matching group
486 return next(g for g in mobj.groups() if g is not None)
488 return mobj.group(group)
489 elif default is not _NO_DEFAULT:
492 raise RegexNotFoundError('Unable to extract %s' % _name)
494 self._downloader.report_warning('unable to extract %s; '
495 'please report this issue on http://yt-dl.org/bug' % _name)
498 def _html_search_regex(self, pattern, string, name, default=_NO_DEFAULT, fatal=True, flags=0, group=None):
500 Like _search_regex, but strips HTML tags and unescapes entities.
502 res = self._search_regex(pattern, string, name, default, fatal, flags, group)
504 return clean_html(res).strip()
508 def _get_login_info(self):
510 Get the the login info as (username, password)
511 It will look in the netrc file using the _NETRC_MACHINE value
512 If there's no info available, return (None, None)
514 if self._downloader is None:
519 downloader_params = self._downloader.params
521 # Attempt to use provided username and password or .netrc data
522 if downloader_params.get('username', None) is not None:
523 username = downloader_params['username']
524 password = downloader_params['password']
525 elif downloader_params.get('usenetrc', False):
527 info = netrc.netrc().authenticators(self._NETRC_MACHINE)
532 raise netrc.NetrcParseError('No authenticators for %s' % self._NETRC_MACHINE)
533 except (IOError, netrc.NetrcParseError) as err:
534 self._downloader.report_warning('parsing .netrc: %s' % compat_str(err))
536 return (username, password)
538 def _get_tfa_info(self):
540 Get the two-factor authentication info
541 TODO - asking the user will be required for sms/phone verify
542 currently just uses the command line option
543 If there's no info available, return None
545 if self._downloader is None:
547 downloader_params = self._downloader.params
549 if downloader_params.get('twofactor', None) is not None:
550 return downloader_params['twofactor']
554 # Helper functions for extracting OpenGraph info
556 def _og_regexes(prop):
557 content_re = r'content=(?:"([^>]+?)"|\'([^>]+?)\')'
558 property_re = r'(?:name|property)=[\'"]og:%s[\'"]' % re.escape(prop)
559 template = r'<meta[^>]+?%s[^>]+?%s'
561 template % (property_re, content_re),
562 template % (content_re, property_re),
565 def _og_search_property(self, prop, html, name=None, **kargs):
567 name = 'OpenGraph %s' % prop
568 escaped = self._search_regex(self._og_regexes(prop), html, name, flags=re.DOTALL, **kargs)
571 return unescapeHTML(escaped)
573 def _og_search_thumbnail(self, html, **kargs):
574 return self._og_search_property('image', html, 'thumbnail url', fatal=False, **kargs)
576 def _og_search_description(self, html, **kargs):
577 return self._og_search_property('description', html, fatal=False, **kargs)
579 def _og_search_title(self, html, **kargs):
580 return self._og_search_property('title', html, **kargs)
582 def _og_search_video_url(self, html, name='video url', secure=True, **kargs):
583 regexes = self._og_regexes('video') + self._og_regexes('video:url')
585 regexes = self._og_regexes('video:secure_url') + regexes
586 return self._html_search_regex(regexes, html, name, **kargs)
588 def _og_search_url(self, html, **kargs):
589 return self._og_search_property('url', html, **kargs)
591 def _html_search_meta(self, name, html, display_name=None, fatal=False, **kwargs):
592 if display_name is None:
594 return self._html_search_regex(
596 (?=[^>]+(?:itemprop|name|property)=(["\']?)%s\1)
597 [^>]+content=(["\'])(?P<content>.*?)\1''' % re.escape(name),
598 html, display_name, fatal=fatal, group='content', **kwargs)
600 def _dc_search_uploader(self, html):
601 return self._html_search_meta('dc.creator', html, 'uploader')
603 def _rta_search(self, html):
604 # See http://www.rtalabel.org/index.php?content=howtofaq#single
605 if re.search(r'(?ix)<meta\s+name="rating"\s+'
606 r' content="RTA-5042-1996-1400-1577-RTA"',
611 def _media_rating_search(self, html):
612 # See http://www.tjg-designs.com/WP/metadata-code-examples-adding-metadata-to-your-web-pages/
613 rating = self._html_search_meta('rating', html)
625 return RATING_TABLE.get(rating.lower(), None)
627 def _twitter_search_player(self, html):
628 return self._html_search_meta('twitter:player', html,
629 'twitter card player')
631 def _sort_formats(self, formats):
633 raise ExtractorError('No video formats found')
636 # TODO remove the following workaround
637 from ..utils import determine_ext
638 if not f.get('ext') and 'url' in f:
639 f['ext'] = determine_ext(f['url'])
641 preference = f.get('preference')
642 if preference is None:
643 proto = f.get('protocol')
645 proto = compat_urllib_parse_urlparse(f.get('url', '')).scheme
647 preference = 0 if proto in ['http', 'https'] else -0.1
648 if f.get('ext') in ['f4f', 'f4m']: # Not yet supported
651 if f.get('vcodec') == 'none': # audio only
652 if self._downloader.params.get('prefer_free_formats'):
653 ORDER = ['aac', 'mp3', 'm4a', 'webm', 'ogg', 'opus']
655 ORDER = ['webm', 'opus', 'ogg', 'mp3', 'aac', 'm4a']
658 audio_ext_preference = ORDER.index(f['ext'])
660 audio_ext_preference = -1
662 if self._downloader.params.get('prefer_free_formats'):
663 ORDER = ['flv', 'mp4', 'webm']
665 ORDER = ['webm', 'flv', 'mp4']
667 ext_preference = ORDER.index(f['ext'])
670 audio_ext_preference = 0
674 f.get('language_preference') if f.get('language_preference') is not None else -1,
675 f.get('quality') if f.get('quality') is not None else -1,
676 f.get('height') if f.get('height') is not None else -1,
677 f.get('width') if f.get('width') is not None else -1,
679 f.get('tbr') if f.get('tbr') is not None else -1,
680 f.get('vbr') if f.get('vbr') is not None else -1,
681 f.get('abr') if f.get('abr') is not None else -1,
682 audio_ext_preference,
683 f.get('fps') if f.get('fps') is not None else -1,
684 f.get('filesize') if f.get('filesize') is not None else -1,
685 f.get('filesize_approx') if f.get('filesize_approx') is not None else -1,
686 f.get('source_preference') if f.get('source_preference') is not None else -1,
689 formats.sort(key=_formats_key)
691 def http_scheme(self):
692 """ Either "http:" or "https:", depending on the user's preferences """
695 if self._downloader.params.get('prefer_insecure', False)
698 def _proto_relative_url(self, url, scheme=None):
701 if url.startswith('//'):
703 scheme = self.http_scheme()
708 def _sleep(self, timeout, video_id, msg_template=None):
709 if msg_template is None:
710 msg_template = '%(video_id)s: Waiting for %(timeout)s seconds'
711 msg = msg_template % {'video_id': video_id, 'timeout': timeout}
715 def _extract_f4m_formats(self, manifest_url, video_id):
716 manifest = self._download_xml(
717 manifest_url, video_id, 'Downloading f4m manifest',
718 'Unable to download f4m manifest')
721 media_nodes = manifest.findall('{http://ns.adobe.com/f4m/1.0}media')
722 for i, media_el in enumerate(media_nodes):
723 tbr = int_or_none(media_el.attrib.get('bitrate'))
724 format_id = 'f4m-%d' % (i if tbr is None else tbr)
726 'format_id': format_id,
730 'width': int_or_none(media_el.attrib.get('width')),
731 'height': int_or_none(media_el.attrib.get('height')),
733 self._sort_formats(formats)
737 def _extract_m3u8_formats(self, m3u8_url, video_id, ext=None,
738 entry_protocol='m3u8', preference=None):
741 'format_id': 'm3u8-meta',
746 'resolution': 'multiple',
747 'format_note': 'Quality selection URL',
750 format_url = lambda u: (
752 if re.match(r'^https?://', u)
753 else compat_urlparse.urljoin(m3u8_url, u))
755 m3u8_doc = self._download_webpage(
757 note='Downloading m3u8 information',
758 errnote='Failed to download m3u8 information')
761 r'(?P<key>[a-zA-Z_-]+)=(?P<val>"[^"]+"|[^",]+)(?:,|$)')
762 for line in m3u8_doc.splitlines():
763 if line.startswith('#EXT-X-STREAM-INF:'):
765 for m in kv_rex.finditer(line):
767 if v.startswith('"'):
769 last_info[m.group('key')] = v
770 elif line.startswith('#') or not line.strip():
773 if last_info is None:
774 formats.append({'url': format_url(line)})
776 tbr = int_or_none(last_info.get('BANDWIDTH'), scale=1000)
779 'format_id': 'm3u8-%d' % (tbr if tbr else len(formats)),
780 'url': format_url(line.strip()),
783 'protocol': entry_protocol,
784 'preference': preference,
786 codecs = last_info.get('CODECS')
788 # TODO: looks like video codec is not always necessarily goes first
789 va_codecs = codecs.split(',')
791 f['vcodec'] = va_codecs[0].partition('.')[0]
792 if len(va_codecs) > 1 and va_codecs[1]:
793 f['acodec'] = va_codecs[1].partition('.')[0]
794 resolution = last_info.get('RESOLUTION')
796 width_str, height_str = resolution.split('x')
797 f['width'] = int(width_str)
798 f['height'] = int(height_str)
801 self._sort_formats(formats)
804 # TODO: improve extraction
805 def _extract_smil_formats(self, smil_url, video_id):
806 smil = self._download_xml(
807 smil_url, video_id, 'Downloading SMIL file',
808 'Unable to download SMIL file')
810 base = smil.find('./head/meta').get('base')
814 for video in smil.findall('./body/switch/video'):
815 src = video.get('src')
818 bitrate = int_or_none(video.get('system-bitrate') or video.get('systemBitrate'), 1000)
819 width = int_or_none(video.get('width'))
820 height = int_or_none(video.get('height'))
821 proto = video.get('proto')
824 if base.startswith('rtmp'):
826 elif base.startswith('http'):
828 ext = video.get('ext')
830 formats.extend(self._extract_m3u8_formats(src, video_id, ext))
831 elif proto == 'rtmp':
833 streamer = video.get('streamer') or base
838 'format_id': 'rtmp-%d' % (rtmp_count if bitrate is None else bitrate),
843 self._sort_formats(formats)
847 def _live_title(self, name):
848 """ Generate the title for a live video """
849 now = datetime.datetime.now()
850 now_str = now.strftime("%Y-%m-%d %H:%M")
851 return name + ' ' + now_str
853 def _int(self, v, name, fatal=False, **kwargs):
854 res = int_or_none(v, **kwargs)
855 if 'get_attr' in kwargs:
856 print(getattr(v, kwargs['get_attr']))
858 msg = 'Failed to extract %s: Could not parse value %r' % (name, v)
860 raise ExtractorError(msg)
862 self._downloader.report_warning(msg)
865 def _float(self, v, name, fatal=False, **kwargs):
866 res = float_or_none(v, **kwargs)
868 msg = 'Failed to extract %s: Could not parse value %r' % (name, v)
870 raise ExtractorError(msg)
872 self._downloader.report_warning(msg)
875 def _set_cookie(self, domain, name, value, expire_time=None):
876 cookie = compat_cookiejar.Cookie(
877 0, name, value, None, None, domain, None,
878 None, '/', True, False, expire_time, '', None, None, None)
879 self._downloader.cookiejar.set_cookie(cookie)
881 def get_testcases(self, include_onlymatching=False):
882 t = getattr(self, '_TEST', None)
884 assert not hasattr(self, '_TESTS'), \
885 '%s has _TEST and _TESTS' % type(self).__name__
888 tests = getattr(self, '_TESTS', [])
890 if not include_onlymatching and t.get('only_matching', False):
892 t['name'] = type(self).__name__[:-len('IE')]
895 def is_suitable(self, age_limit):
896 """ Test whether the extractor is generally suitable for the given
897 age limit (i.e. pornographic sites are not, all others usually are) """
899 any_restricted = False
900 for tc in self.get_testcases(include_onlymatching=False):
902 tc = tc['playlist'][0]
903 is_restricted = age_restricted(
904 tc.get('info_dict', {}).get('age_limit'), age_limit)
905 if not is_restricted:
907 any_restricted = any_restricted or is_restricted
908 return not any_restricted
911 class SearchInfoExtractor(InfoExtractor):
913 Base class for paged search queries extractors.
914 They accept urls in the format _SEARCH_KEY(|all|[0-9]):{query}
915 Instances should define _SEARCH_KEY and _MAX_RESULTS.
919 def _make_valid_url(cls):
920 return r'%s(?P<prefix>|[1-9][0-9]*|all):(?P<query>[\s\S]+)' % cls._SEARCH_KEY
923 def suitable(cls, url):
924 return re.match(cls._make_valid_url(), url) is not None
926 def _real_extract(self, query):
927 mobj = re.match(self._make_valid_url(), query)
929 raise ExtractorError('Invalid search query "%s"' % query)
931 prefix = mobj.group('prefix')
932 query = mobj.group('query')
934 return self._get_n_results(query, 1)
935 elif prefix == 'all':
936 return self._get_n_results(query, self._MAX_RESULTS)
940 raise ExtractorError('invalid download number %s for query "%s"' % (n, query))
941 elif n > self._MAX_RESULTS:
942 self._downloader.report_warning('%s returns max %i results (you requested %i)' % (self._SEARCH_KEY, self._MAX_RESULTS, n))
943 n = self._MAX_RESULTS
944 return self._get_n_results(query, n)
946 def _get_n_results(self, query, n):
947 """Get a specified number of results for a query"""
948 raise NotImplementedError("This method must be implemented by subclasses")
951 def SEARCH_KEY(self):
952 return self._SEARCH_KEY