1 from __future__ import unicode_literals
13 import xml.etree.ElementTree
15 from ..compat import (
19 compat_urllib_parse_urlparse,
33 _NO_DEFAULT = object()
36 class InfoExtractor(object):
37 """Information Extractor class.
39 Information extractors are the classes that, given a URL, extract
40 information about the video (or videos) the URL refers to. This
41 information includes the real video URL, the video title, author and
42 others. The information is stored in a dictionary which is then
43 passed to the YoutubeDL. The YoutubeDL processes this
44 information possibly downloading the video to the file system, among
45 other possible outcomes.
47 The type field determines the the type of the result.
48 By far the most common value (and the default if _type is missing) is
49 "video", which indicates a single video.
51 For a video, the dictionaries must include the following fields:
54 title: Video title, unescaped.
56 Additionally, it must contain either a formats entry or a url one:
58 formats: A list of dictionaries for each format available, ordered
59 from worst to best quality.
62 * url Mandatory. The URL of the video file
63 * ext Will be calculated from url if missing
64 * format A human-readable description of the format
65 ("mp4 container with h264/opus").
66 Calculated from the format_id, width, height.
67 and format_note fields if missing.
68 * format_id A short description of the format
69 ("mp4_h264_opus" or "19").
70 Technically optional, but strongly recommended.
71 * format_note Additional info about the format
72 ("3D" or "DASH video")
73 * width Width of the video, if known
74 * height Height of the video, if known
75 * resolution Textual description of width and height
76 * tbr Average bitrate of audio and video in KBit/s
77 * abr Average audio bitrate in KBit/s
78 * acodec Name of the audio codec in use
79 * asr Audio sampling rate in Hertz
80 * vbr Average video bitrate in KBit/s
82 * vcodec Name of the video codec in use
83 * container Name of the container format
84 * filesize The number of bytes, if known in advance
85 * filesize_approx An estimate for the number of bytes
86 * player_url SWF Player URL (used for rtmpdump).
87 * protocol The protocol that will be used for the actual
89 "http", "https", "rtsp", "rtmp", "m3u8" or so.
90 * preference Order number of this format. If this field is
91 present and not None, the formats get sorted
92 by this field, regardless of all other values.
93 -1 for default (order by other properties),
94 -2 or smaller for less than default.
95 < -1000 to hide the format (if there is
96 another one which is strictly better)
97 * language_preference Is this in the correct requested
99 10 if it's what the URL is about,
100 -1 for default (don't know),
101 -10 otherwise, other values reserved for now.
102 * quality Order number of the video quality of this
103 format, irrespective of the file format.
104 -1 for default (order by other properties),
105 -2 or smaller for less than default.
106 * source_preference Order number for this video source
107 (quality takes higher priority)
108 -1 for default (order by other properties),
109 -2 or smaller for less than default.
110 * http_referer HTTP Referer header value to set.
111 * http_method HTTP method to use for the download.
112 * http_headers A dictionary of additional HTTP headers
113 to add to the request.
114 * http_post_data Additional data to send with a POST
116 url: Final video URL.
117 ext: Video filename extension.
118 format: The video format, defaults to ext (used for --get-format)
119 player_url: SWF Player URL (used for rtmpdump).
121 The following fields are optional:
123 alt_title: A secondary title of the video.
124 display_id An alternative identifier for the video, not necessarily
125 unique, but available before title. Typically, id is
126 something like "4234987", title "Dancing naked mole rats",
127 and display_id "dancing-naked-mole-rats"
128 thumbnails: A list of dictionaries, with the following entries:
130 * "width" (optional, int)
131 * "height" (optional, int)
132 * "resolution" (optional, string "{width}x{height"},
134 thumbnail: Full URL to a video thumbnail image.
135 description: Full video description.
136 uploader: Full name of the video uploader.
137 timestamp: UNIX timestamp of the moment the video became available.
138 upload_date: Video upload date (YYYYMMDD).
139 If not explicitly set, calculated from timestamp.
140 uploader_id: Nickname or id of the video uploader.
141 location: Physical location where the video was filmed.
142 subtitles: The subtitle file contents as a dictionary in the format
143 {language: subtitles}.
144 duration: Length of the video in seconds, as an integer.
145 view_count: How many users have watched the video on the platform.
146 like_count: Number of positive ratings of the video
147 dislike_count: Number of negative ratings of the video
148 comment_count: Number of comments on the video
149 age_limit: Age restriction for the video, as an integer (years)
150 webpage_url: The url to the video webpage, if given to youtube-dl it
151 should allow to get the same result again. (It will be set
152 by YoutubeDL if it's missing)
153 categories: A list of categories that the video falls in, for example
155 is_live: True, False, or None (=unknown). Whether this video is a
156 live stream that goes on instead of a fixed-length video.
158 Unless mentioned otherwise, the fields should be Unicode strings.
160 Unless mentioned otherwise, None is equivalent to absence of information.
163 _type "playlist" indicates multiple videos.
164 There must be a key "entries", which is a list, an iterable, or a PagedList
165 object, each element of which is a valid dictionary by this specification.
167 Additionally, playlists can have "title" and "id" attributes with the same
168 semantics as videos (see above).
171 _type "multi_video" indicates that there are multiple videos that
172 form a single show, for examples multiple acts of an opera or TV episode.
173 It must have an entries key like a playlist and contain all the keys
174 required for a video at the same time.
177 _type "url" indicates that the video must be extracted from another
178 location, possibly by a different extractor. Its only required key is:
179 "url" - the next URL to extract.
180 The key "ie_key" can be set to the class name (minus the trailing "IE",
181 e.g. "Youtube") if the extractor class is known in advance.
182 Additionally, the dictionary may have any properties of the resolved entity
183 known in advance, for example "title" if the title of the referred video is
187 _type "url_transparent" entities have the same specification as "url", but
188 indicate that the given additional information is more precise than the one
189 associated with the resolved URL.
190 This is useful when a site employs a video service that hosts the video and
191 its technical metadata, but that video service does not embed a useful
192 title, description etc.
195 Subclasses of this one should re-define the _real_initialize() and
196 _real_extract() methods and define a _VALID_URL regexp.
197 Probably, they should also be added to the list of extractors.
199 Finally, the _WORKING attribute should be set to False for broken IEs
200 in order to warn the users and skip the tests.
207 def __init__(self, downloader=None):
208 """Constructor. Receives an optional downloader."""
210 self.set_downloader(downloader)
213 def suitable(cls, url):
214 """Receives a URL and returns True if suitable for this IE."""
216 # This does not use has/getattr intentionally - we want to know whether
217 # we have cached the regexp for *this* class, whereas getattr would also
218 # match the superclass
219 if '_VALID_URL_RE' not in cls.__dict__:
220 cls._VALID_URL_RE = re.compile(cls._VALID_URL)
221 return cls._VALID_URL_RE.match(url) is not None
224 def _match_id(cls, url):
225 if '_VALID_URL_RE' not in cls.__dict__:
226 cls._VALID_URL_RE = re.compile(cls._VALID_URL)
227 m = cls._VALID_URL_RE.match(url)
233 """Getter method for _WORKING."""
236 def initialize(self):
237 """Initializes an instance (authentication, etc)."""
239 self._real_initialize()
242 def extract(self, url):
243 """Extracts URL information and returns it in list of dicts."""
245 return self._real_extract(url)
247 def set_downloader(self, downloader):
248 """Sets the downloader for this IE."""
249 self._downloader = downloader
251 def _real_initialize(self):
252 """Real initialization process. Redefine in subclasses."""
255 def _real_extract(self, url):
256 """Real extraction process. Redefine in subclasses."""
261 """A string for getting the InfoExtractor with get_info_extractor"""
262 return cls.__name__[:-2]
266 return type(self).__name__[:-2]
268 def _request_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True):
269 """ Returns the response handle """
271 self.report_download_webpage(video_id)
272 elif note is not False:
274 self.to_screen('%s' % (note,))
276 self.to_screen('%s: %s' % (video_id, note))
278 return self._downloader.urlopen(url_or_request)
279 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
283 errnote = 'Unable to download webpage'
284 errmsg = '%s: %s' % (errnote, compat_str(err))
286 raise ExtractorError(errmsg, sys.exc_info()[2], cause=err)
288 self._downloader.report_warning(errmsg)
291 def _download_webpage_handle(self, url_or_request, video_id, note=None, errnote=None, fatal=True):
292 """ Returns a tuple (page content as string, URL handle) """
293 # Strip hashes from the URL (#1038)
294 if isinstance(url_or_request, (compat_str, str)):
295 url_or_request = url_or_request.partition('#')[0]
297 urlh = self._request_webpage(url_or_request, video_id, note, errnote, fatal)
301 content = self._webpage_read_content(urlh, url_or_request, video_id, note, errnote, fatal)
302 return (content, urlh)
304 def _webpage_read_content(self, urlh, url_or_request, video_id, note=None, errnote=None, fatal=True, prefix=None):
305 content_type = urlh.headers.get('Content-Type', '')
306 webpage_bytes = urlh.read()
307 if prefix is not None:
308 webpage_bytes = prefix + webpage_bytes
309 m = re.match(r'[a-zA-Z0-9_.-]+/[a-zA-Z0-9_.-]+\s*;\s*charset=(.+)', content_type)
311 encoding = m.group(1)
313 m = re.search(br'<meta[^>]+charset=[\'"]?([^\'")]+)[ /\'">]',
314 webpage_bytes[:1024])
316 encoding = m.group(1).decode('ascii')
317 elif webpage_bytes.startswith(b'\xff\xfe'):
321 if self._downloader.params.get('dump_intermediate_pages', False):
323 url = url_or_request.get_full_url()
324 except AttributeError:
326 self.to_screen('Dumping request to ' + url)
327 dump = base64.b64encode(webpage_bytes).decode('ascii')
328 self._downloader.to_screen(dump)
329 if self._downloader.params.get('write_pages', False):
331 url = url_or_request.get_full_url()
332 except AttributeError:
334 basen = '%s_%s' % (video_id, url)
336 h = '___' + hashlib.md5(basen.encode('utf-8')).hexdigest()
337 basen = basen[:240 - len(h)] + h
338 raw_filename = basen + '.dump'
339 filename = sanitize_filename(raw_filename, restricted=True)
340 self.to_screen('Saving request to ' + filename)
341 # Working around MAX_PATH limitation on Windows (see
342 # http://msdn.microsoft.com/en-us/library/windows/desktop/aa365247(v=vs.85).aspx)
344 absfilepath = os.path.abspath(filename)
345 if len(absfilepath) > 259:
346 filename = '\\\\?\\' + absfilepath
347 with open(filename, 'wb') as outf:
348 outf.write(webpage_bytes)
351 content = webpage_bytes.decode(encoding, 'replace')
353 content = webpage_bytes.decode('utf-8', 'replace')
355 if ('<title>Access to this site is blocked</title>' in content and
356 'Websense' in content[:512]):
357 msg = 'Access to this webpage has been blocked by Websense filtering software in your network.'
358 blocked_iframe = self._html_search_regex(
359 r'<iframe src="([^"]+)"', content,
360 'Websense information URL', default=None)
362 msg += ' Visit %s for more details' % blocked_iframe
363 raise ExtractorError(msg, expected=True)
367 def _download_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True):
368 """ Returns the data of the page as a string """
369 res = self._download_webpage_handle(url_or_request, video_id, note, errnote, fatal)
376 def _download_xml(self, url_or_request, video_id,
377 note='Downloading XML', errnote='Unable to download XML',
378 transform_source=None, fatal=True):
379 """Return the xml as an xml.etree.ElementTree.Element"""
380 xml_string = self._download_webpage(
381 url_or_request, video_id, note, errnote, fatal=fatal)
382 if xml_string is False:
385 xml_string = transform_source(xml_string)
386 return xml.etree.ElementTree.fromstring(xml_string.encode('utf-8'))
388 def _download_json(self, url_or_request, video_id,
389 note='Downloading JSON metadata',
390 errnote='Unable to download JSON metadata',
391 transform_source=None,
393 json_string = self._download_webpage(
394 url_or_request, video_id, note, errnote, fatal=fatal)
395 if (not fatal) and json_string is False:
397 return self._parse_json(
398 json_string, video_id, transform_source=transform_source, fatal=fatal)
400 def _parse_json(self, json_string, video_id, transform_source=None, fatal=True):
402 json_string = transform_source(json_string)
404 return json.loads(json_string)
405 except ValueError as ve:
406 errmsg = '%s: Failed to parse JSON ' % video_id
408 raise ExtractorError(errmsg, cause=ve)
410 self.report_warning(errmsg + str(ve))
412 def report_warning(self, msg, video_id=None):
413 idstr = '' if video_id is None else '%s: ' % video_id
414 self._downloader.report_warning(
415 '[%s] %s%s' % (self.IE_NAME, idstr, msg))
417 def to_screen(self, msg):
418 """Print msg to screen, prefixing it with '[ie_name]'"""
419 self._downloader.to_screen('[%s] %s' % (self.IE_NAME, msg))
421 def report_extraction(self, id_or_name):
422 """Report information extraction."""
423 self.to_screen('%s: Extracting information' % id_or_name)
425 def report_download_webpage(self, video_id):
426 """Report webpage download."""
427 self.to_screen('%s: Downloading webpage' % video_id)
429 def report_age_confirmation(self):
430 """Report attempt to confirm age."""
431 self.to_screen('Confirming age')
433 def report_login(self):
434 """Report attempt to log in."""
435 self.to_screen('Logging in')
437 # Methods for following #608
439 def url_result(url, ie=None, video_id=None):
440 """Returns a url that points to a page that should be processed"""
441 # TODO: ie should be the class used for getting the info
442 video_info = {'_type': 'url',
445 if video_id is not None:
446 video_info['id'] = video_id
450 def playlist_result(entries, playlist_id=None, playlist_title=None, playlist_description=None):
451 """Returns a playlist"""
452 video_info = {'_type': 'playlist',
455 video_info['id'] = playlist_id
457 video_info['title'] = playlist_title
458 if playlist_description:
459 video_info['description'] = playlist_description
462 def _search_regex(self, pattern, string, name, default=_NO_DEFAULT, fatal=True, flags=0, group=None):
464 Perform a regex search on the given string, using a single or a list of
465 patterns returning the first matching group.
466 In case of failure return a default value or raise a WARNING or a
467 RegexNotFoundError, depending on fatal, specifying the field name.
469 if isinstance(pattern, (str, compat_str, compiled_regex_type)):
470 mobj = re.search(pattern, string, flags)
473 mobj = re.search(p, string, flags)
477 if os.name != 'nt' and sys.stderr.isatty():
478 _name = '\033[0;34m%s\033[0m' % name
484 # return the first matching group
485 return next(g for g in mobj.groups() if g is not None)
487 return mobj.group(group)
488 elif default is not _NO_DEFAULT:
491 raise RegexNotFoundError('Unable to extract %s' % _name)
493 self._downloader.report_warning('unable to extract %s; '
494 'please report this issue on http://yt-dl.org/bug' % _name)
497 def _html_search_regex(self, pattern, string, name, default=_NO_DEFAULT, fatal=True, flags=0, group=None):
499 Like _search_regex, but strips HTML tags and unescapes entities.
501 res = self._search_regex(pattern, string, name, default, fatal, flags, group)
503 return clean_html(res).strip()
507 def _get_login_info(self):
509 Get the the login info as (username, password)
510 It will look in the netrc file using the _NETRC_MACHINE value
511 If there's no info available, return (None, None)
513 if self._downloader is None:
518 downloader_params = self._downloader.params
520 # Attempt to use provided username and password or .netrc data
521 if downloader_params.get('username', None) is not None:
522 username = downloader_params['username']
523 password = downloader_params['password']
524 elif downloader_params.get('usenetrc', False):
526 info = netrc.netrc().authenticators(self._NETRC_MACHINE)
531 raise netrc.NetrcParseError('No authenticators for %s' % self._NETRC_MACHINE)
532 except (IOError, netrc.NetrcParseError) as err:
533 self._downloader.report_warning('parsing .netrc: %s' % compat_str(err))
535 return (username, password)
537 def _get_tfa_info(self):
539 Get the two-factor authentication info
540 TODO - asking the user will be required for sms/phone verify
541 currently just uses the command line option
542 If there's no info available, return None
544 if self._downloader is None:
546 downloader_params = self._downloader.params
548 if downloader_params.get('twofactor', None) is not None:
549 return downloader_params['twofactor']
553 # Helper functions for extracting OpenGraph info
555 def _og_regexes(prop):
556 content_re = r'content=(?:"([^>]+?)"|\'([^>]+?)\')'
557 property_re = r'(?:name|property)=[\'"]og:%s[\'"]' % re.escape(prop)
558 template = r'<meta[^>]+?%s[^>]+?%s'
560 template % (property_re, content_re),
561 template % (content_re, property_re),
564 def _og_search_property(self, prop, html, name=None, **kargs):
566 name = 'OpenGraph %s' % prop
567 escaped = self._search_regex(self._og_regexes(prop), html, name, flags=re.DOTALL, **kargs)
570 return unescapeHTML(escaped)
572 def _og_search_thumbnail(self, html, **kargs):
573 return self._og_search_property('image', html, 'thumbnail url', fatal=False, **kargs)
575 def _og_search_description(self, html, **kargs):
576 return self._og_search_property('description', html, fatal=False, **kargs)
578 def _og_search_title(self, html, **kargs):
579 return self._og_search_property('title', html, **kargs)
581 def _og_search_video_url(self, html, name='video url', secure=True, **kargs):
582 regexes = self._og_regexes('video') + self._og_regexes('video:url')
584 regexes = self._og_regexes('video:secure_url') + regexes
585 return self._html_search_regex(regexes, html, name, **kargs)
587 def _og_search_url(self, html, **kargs):
588 return self._og_search_property('url', html, **kargs)
590 def _html_search_meta(self, name, html, display_name=None, fatal=False, **kwargs):
591 if display_name is None:
593 return self._html_search_regex(
595 (?=[^>]+(?:itemprop|name|property)=(["\']?)%s\1)
596 [^>]+content=(["\'])(?P<content>.*?)\1''' % re.escape(name),
597 html, display_name, fatal=fatal, group='content', **kwargs)
599 def _dc_search_uploader(self, html):
600 return self._html_search_meta('dc.creator', html, 'uploader')
602 def _rta_search(self, html):
603 # See http://www.rtalabel.org/index.php?content=howtofaq#single
604 if re.search(r'(?ix)<meta\s+name="rating"\s+'
605 r' content="RTA-5042-1996-1400-1577-RTA"',
610 def _media_rating_search(self, html):
611 # See http://www.tjg-designs.com/WP/metadata-code-examples-adding-metadata-to-your-web-pages/
612 rating = self._html_search_meta('rating', html)
624 return RATING_TABLE.get(rating.lower(), None)
626 def _twitter_search_player(self, html):
627 return self._html_search_meta('twitter:player', html,
628 'twitter card player')
630 def _sort_formats(self, formats):
632 raise ExtractorError('No video formats found')
635 # TODO remove the following workaround
636 from ..utils import determine_ext
637 if not f.get('ext') and 'url' in f:
638 f['ext'] = determine_ext(f['url'])
640 preference = f.get('preference')
641 if preference is None:
642 proto = f.get('protocol')
644 proto = compat_urllib_parse_urlparse(f.get('url', '')).scheme
646 preference = 0 if proto in ['http', 'https'] else -0.1
647 if f.get('ext') in ['f4f', 'f4m']: # Not yet supported
650 if f.get('vcodec') == 'none': # audio only
651 if self._downloader.params.get('prefer_free_formats'):
652 ORDER = ['aac', 'mp3', 'm4a', 'webm', 'ogg', 'opus']
654 ORDER = ['webm', 'opus', 'ogg', 'mp3', 'aac', 'm4a']
657 audio_ext_preference = ORDER.index(f['ext'])
659 audio_ext_preference = -1
661 if self._downloader.params.get('prefer_free_formats'):
662 ORDER = ['flv', 'mp4', 'webm']
664 ORDER = ['webm', 'flv', 'mp4']
666 ext_preference = ORDER.index(f['ext'])
669 audio_ext_preference = 0
673 f.get('language_preference') if f.get('language_preference') is not None else -1,
674 f.get('quality') if f.get('quality') is not None else -1,
675 f.get('height') if f.get('height') is not None else -1,
676 f.get('width') if f.get('width') is not None else -1,
678 f.get('tbr') if f.get('tbr') is not None else -1,
679 f.get('vbr') if f.get('vbr') is not None else -1,
680 f.get('abr') if f.get('abr') is not None else -1,
681 audio_ext_preference,
682 f.get('fps') if f.get('fps') is not None else -1,
683 f.get('filesize') if f.get('filesize') is not None else -1,
684 f.get('filesize_approx') if f.get('filesize_approx') is not None else -1,
685 f.get('source_preference') if f.get('source_preference') is not None else -1,
688 formats.sort(key=_formats_key)
690 def http_scheme(self):
691 """ Either "http:" or "https:", depending on the user's preferences """
694 if self._downloader.params.get('prefer_insecure', False)
697 def _proto_relative_url(self, url, scheme=None):
700 if url.startswith('//'):
702 scheme = self.http_scheme()
707 def _sleep(self, timeout, video_id, msg_template=None):
708 if msg_template is None:
709 msg_template = '%(video_id)s: Waiting for %(timeout)s seconds'
710 msg = msg_template % {'video_id': video_id, 'timeout': timeout}
714 def _extract_f4m_formats(self, manifest_url, video_id):
715 manifest = self._download_xml(
716 manifest_url, video_id, 'Downloading f4m manifest',
717 'Unable to download f4m manifest')
720 media_nodes = manifest.findall('{http://ns.adobe.com/f4m/1.0}media')
721 for i, media_el in enumerate(media_nodes):
722 tbr = int_or_none(media_el.attrib.get('bitrate'))
723 format_id = 'f4m-%d' % (i if tbr is None else tbr)
725 'format_id': format_id,
729 'width': int_or_none(media_el.attrib.get('width')),
730 'height': int_or_none(media_el.attrib.get('height')),
732 self._sort_formats(formats)
736 def _extract_m3u8_formats(self, m3u8_url, video_id, ext=None,
737 entry_protocol='m3u8', preference=None):
740 'format_id': 'm3u8-meta',
745 'resolution': 'multiple',
746 'format_note': 'Quality selection URL',
749 format_url = lambda u: (
751 if re.match(r'^https?://', u)
752 else compat_urlparse.urljoin(m3u8_url, u))
754 m3u8_doc = self._download_webpage(
756 note='Downloading m3u8 information',
757 errnote='Failed to download m3u8 information')
760 r'(?P<key>[a-zA-Z_-]+)=(?P<val>"[^"]+"|[^",]+)(?:,|$)')
761 for line in m3u8_doc.splitlines():
762 if line.startswith('#EXT-X-STREAM-INF:'):
764 for m in kv_rex.finditer(line):
766 if v.startswith('"'):
768 last_info[m.group('key')] = v
769 elif line.startswith('#') or not line.strip():
772 if last_info is None:
773 formats.append({'url': format_url(line)})
775 tbr = int_or_none(last_info.get('BANDWIDTH'), scale=1000)
778 'format_id': 'm3u8-%d' % (tbr if tbr else len(formats)),
779 'url': format_url(line.strip()),
782 'protocol': entry_protocol,
783 'preference': preference,
785 codecs = last_info.get('CODECS')
787 # TODO: looks like video codec is not always necessarily goes first
788 va_codecs = codecs.split(',')
790 f['vcodec'] = va_codecs[0].partition('.')[0]
791 if len(va_codecs) > 1 and va_codecs[1]:
792 f['acodec'] = va_codecs[1].partition('.')[0]
793 resolution = last_info.get('RESOLUTION')
795 width_str, height_str = resolution.split('x')
796 f['width'] = int(width_str)
797 f['height'] = int(height_str)
800 self._sort_formats(formats)
803 # TODO: improve extraction
804 def _extract_smil_formats(self, smil_url, video_id):
805 smil = self._download_xml(
806 smil_url, video_id, 'Downloading SMIL file',
807 'Unable to download SMIL file')
809 base = smil.find('./head/meta').get('base')
813 for video in smil.findall('./body/switch/video'):
814 src = video.get('src')
817 bitrate = int_or_none(video.get('system-bitrate') or video.get('systemBitrate'), 1000)
818 width = int_or_none(video.get('width'))
819 height = int_or_none(video.get('height'))
820 proto = video.get('proto')
823 if base.startswith('rtmp'):
825 elif base.startswith('http'):
827 ext = video.get('ext')
829 formats.extend(self._extract_m3u8_formats(src, video_id, ext))
830 elif proto == 'rtmp':
832 streamer = video.get('streamer') or base
837 'format_id': 'rtmp-%d' % (rtmp_count if bitrate is None else bitrate),
842 self._sort_formats(formats)
846 def _live_title(self, name):
847 """ Generate the title for a live video """
848 now = datetime.datetime.now()
849 now_str = now.strftime("%Y-%m-%d %H:%M")
850 return name + ' ' + now_str
852 def _int(self, v, name, fatal=False, **kwargs):
853 res = int_or_none(v, **kwargs)
854 if 'get_attr' in kwargs:
855 print(getattr(v, kwargs['get_attr']))
857 msg = 'Failed to extract %s: Could not parse value %r' % (name, v)
859 raise ExtractorError(msg)
861 self._downloader.report_warning(msg)
864 def _float(self, v, name, fatal=False, **kwargs):
865 res = float_or_none(v, **kwargs)
867 msg = 'Failed to extract %s: Could not parse value %r' % (name, v)
869 raise ExtractorError(msg)
871 self._downloader.report_warning(msg)
874 def _set_cookie(self, domain, name, value, expire_time=None):
875 cookie = compat_cookiejar.Cookie(
876 0, name, value, None, None, domain, None,
877 None, '/', True, False, expire_time, '', None, None, None)
878 self._downloader.cookiejar.set_cookie(cookie)
881 class SearchInfoExtractor(InfoExtractor):
883 Base class for paged search queries extractors.
884 They accept urls in the format _SEARCH_KEY(|all|[0-9]):{query}
885 Instances should define _SEARCH_KEY and _MAX_RESULTS.
889 def _make_valid_url(cls):
890 return r'%s(?P<prefix>|[1-9][0-9]*|all):(?P<query>[\s\S]+)' % cls._SEARCH_KEY
893 def suitable(cls, url):
894 return re.match(cls._make_valid_url(), url) is not None
896 def _real_extract(self, query):
897 mobj = re.match(self._make_valid_url(), query)
899 raise ExtractorError('Invalid search query "%s"' % query)
901 prefix = mobj.group('prefix')
902 query = mobj.group('query')
904 return self._get_n_results(query, 1)
905 elif prefix == 'all':
906 return self._get_n_results(query, self._MAX_RESULTS)
910 raise ExtractorError('invalid download number %s for query "%s"' % (n, query))
911 elif n > self._MAX_RESULTS:
912 self._downloader.report_warning('%s returns max %i results (you requested %i)' % (self._SEARCH_KEY, self._MAX_RESULTS, n))
913 n = self._MAX_RESULTS
914 return self._get_n_results(query, n)
916 def _get_n_results(self, query, n):
917 """Get a specified number of results for a query"""
918 raise NotImplementedError("This method must be implemented by subclasses")
921 def SEARCH_KEY(self):
922 return self._SEARCH_KEY