]>
Commit | Line | Data |
---|---|---|
d6983cb4 | 1 | import base64 |
3ec05685 | 2 | import hashlib |
3d3538e4 | 3 | import json |
d6983cb4 PH |
4 | import os |
5 | import re | |
6 | import socket | |
7 | import sys | |
fc79158d | 8 | import netrc |
267ed0c5 | 9 | import xml.etree.ElementTree |
d6983cb4 PH |
10 | |
11 | from ..utils import ( | |
12 | compat_http_client, | |
13 | compat_urllib_error, | |
c7deaa4c | 14 | compat_urllib_parse_urlparse, |
d6983cb4 PH |
15 | compat_str, |
16 | ||
17 | clean_html, | |
18 | compiled_regex_type, | |
19 | ExtractorError, | |
55b3e45b | 20 | RegexNotFoundError, |
d41e6efc | 21 | sanitize_filename, |
f38de77f | 22 | unescapeHTML, |
d6983cb4 | 23 | ) |
46374a56 | 24 | _NO_DEFAULT = object() |
d6983cb4 | 25 | |
dca08720 | 26 | |
d6983cb4 PH |
27 | class InfoExtractor(object): |
28 | """Information Extractor class. | |
29 | ||
30 | Information extractors are the classes that, given a URL, extract | |
31 | information about the video (or videos) the URL refers to. This | |
32 | information includes the real video URL, the video title, author and | |
33 | others. The information is stored in a dictionary which is then | |
34 | passed to the FileDownloader. The FileDownloader processes this | |
35 | information possibly downloading the video to the file system, among | |
36 | other possible outcomes. | |
37 | ||
38 | The dictionaries must include the following fields: | |
39 | ||
40 | id: Video identifier. | |
d6983cb4 | 41 | title: Video title, unescaped. |
d67b0b15 | 42 | |
f49d89ee | 43 | Additionally, it must contain either a formats entry or a url one: |
d67b0b15 | 44 | |
f49d89ee PH |
45 | formats: A list of dictionaries for each format available, ordered |
46 | from worst to best quality. | |
47 | ||
48 | Potential fields: | |
d67b0b15 PH |
49 | * url Mandatory. The URL of the video file |
50 | * ext Will be calculated from url if missing | |
51 | * format A human-readable description of the format | |
52 | ("mp4 container with h264/opus"). | |
53 | Calculated from the format_id, width, height. | |
54 | and format_note fields if missing. | |
55 | * format_id A short description of the format | |
5d4f3985 PH |
56 | ("mp4_h264_opus" or "19"). |
57 | Technically optional, but strongly recommended. | |
d67b0b15 PH |
58 | * format_note Additional info about the format |
59 | ("3D" or "DASH video") | |
60 | * width Width of the video, if known | |
61 | * height Height of the video, if known | |
f49d89ee | 62 | * resolution Textual description of width and height |
7217e148 | 63 | * tbr Average bitrate of audio and video in KBit/s |
d67b0b15 PH |
64 | * abr Average audio bitrate in KBit/s |
65 | * acodec Name of the audio codec in use | |
dd27fd17 | 66 | * asr Audio sampling rate in Hertz |
d67b0b15 PH |
67 | * vbr Average video bitrate in KBit/s |
68 | * vcodec Name of the video codec in use | |
1394ce65 | 69 | * container Name of the container format |
d67b0b15 PH |
70 | * filesize The number of bytes, if known in advance |
71 | * player_url SWF Player URL (used for rtmpdump). | |
c7deaa4c PH |
72 | * protocol The protocol that will be used for the actual |
73 | download, lower-case. | |
db1f3888 | 74 | "http", "https", "rtsp", "rtmp", "m3u8" or so. |
f49d89ee | 75 | * preference Order number of this format. If this field is |
08d13955 | 76 | present and not None, the formats get sorted |
38d63d84 | 77 | by this field, regardless of all other values. |
f49d89ee PH |
78 | -1 for default (order by other properties), |
79 | -2 or smaller for less than default. | |
5d73273f PH |
80 | * quality Order number of the video quality of this |
81 | format, irrespective of the file format. | |
82 | -1 for default (order by other properties), | |
83 | -2 or smaller for less than default. | |
c0ba0f48 | 84 | url: Final video URL. |
d6983cb4 | 85 | ext: Video filename extension. |
d67b0b15 PH |
86 | format: The video format, defaults to ext (used for --get-format) |
87 | player_url: SWF Player URL (used for rtmpdump). | |
2f5865cc | 88 | |
d6983cb4 PH |
89 | The following fields are optional: |
90 | ||
0afef30b PH |
91 | display_id An alternative identifier for the video, not necessarily |
92 | unique, but available before title. Typically, id is | |
93 | something like "4234987", title "Dancing naked mole rats", | |
94 | and display_id "dancing-naked-mole-rats" | |
73e79f2a PH |
95 | thumbnails: A list of dictionaries (with the entries "resolution" and |
96 | "url") for the varying thumbnails | |
d6983cb4 PH |
97 | thumbnail: Full URL to a video thumbnail image. |
98 | description: One-line video description. | |
99 | uploader: Full name of the video uploader. | |
955c4514 | 100 | timestamp: UNIX timestamp of the moment the video became available. |
d6983cb4 | 101 | upload_date: Video upload date (YYYYMMDD). |
955c4514 | 102 | If not explicitly set, calculated from timestamp. |
d6983cb4 PH |
103 | uploader_id: Nickname or id of the video uploader. |
104 | location: Physical location of the video. | |
5d51a883 JMF |
105 | subtitles: The subtitle file contents as a dictionary in the format |
106 | {language: subtitles}. | |
c0ba0f48 | 107 | duration: Length of the video in seconds, as an integer. |
f3d29461 | 108 | view_count: How many users have watched the video on the platform. |
19e3dfc9 PH |
109 | like_count: Number of positive ratings of the video |
110 | dislike_count: Number of negative ratings of the video | |
111 | comment_count: Number of comments on the video | |
8dbe9899 | 112 | age_limit: Age restriction for the video, as an integer (years) |
9103bbc5 JMF |
113 | webpage_url: The url to the video webpage, if given to youtube-dl it |
114 | should allow to get the same result again. (It will be set | |
115 | by YoutubeDL if it's missing) | |
d6983cb4 | 116 | |
deefc05b | 117 | Unless mentioned otherwise, the fields should be Unicode strings. |
d6983cb4 PH |
118 | |
119 | Subclasses of this one should re-define the _real_initialize() and | |
120 | _real_extract() methods and define a _VALID_URL regexp. | |
121 | Probably, they should also be added to the list of extractors. | |
122 | ||
d6983cb4 PH |
123 | Finally, the _WORKING attribute should be set to False for broken IEs |
124 | in order to warn the users and skip the tests. | |
125 | """ | |
126 | ||
127 | _ready = False | |
128 | _downloader = None | |
129 | _WORKING = True | |
130 | ||
131 | def __init__(self, downloader=None): | |
132 | """Constructor. Receives an optional downloader.""" | |
133 | self._ready = False | |
134 | self.set_downloader(downloader) | |
135 | ||
136 | @classmethod | |
137 | def suitable(cls, url): | |
138 | """Receives a URL and returns True if suitable for this IE.""" | |
79cb2577 PH |
139 | |
140 | # This does not use has/getattr intentionally - we want to know whether | |
141 | # we have cached the regexp for *this* class, whereas getattr would also | |
142 | # match the superclass | |
143 | if '_VALID_URL_RE' not in cls.__dict__: | |
144 | cls._VALID_URL_RE = re.compile(cls._VALID_URL) | |
145 | return cls._VALID_URL_RE.match(url) is not None | |
d6983cb4 PH |
146 | |
147 | @classmethod | |
148 | def working(cls): | |
149 | """Getter method for _WORKING.""" | |
150 | return cls._WORKING | |
151 | ||
152 | def initialize(self): | |
153 | """Initializes an instance (authentication, etc).""" | |
154 | if not self._ready: | |
155 | self._real_initialize() | |
156 | self._ready = True | |
157 | ||
158 | def extract(self, url): | |
159 | """Extracts URL information and returns it in list of dicts.""" | |
160 | self.initialize() | |
161 | return self._real_extract(url) | |
162 | ||
163 | def set_downloader(self, downloader): | |
164 | """Sets the downloader for this IE.""" | |
165 | self._downloader = downloader | |
166 | ||
167 | def _real_initialize(self): | |
168 | """Real initialization process. Redefine in subclasses.""" | |
169 | pass | |
170 | ||
171 | def _real_extract(self, url): | |
172 | """Real extraction process. Redefine in subclasses.""" | |
173 | pass | |
174 | ||
56c73665 JMF |
175 | @classmethod |
176 | def ie_key(cls): | |
177 | """A string for getting the InfoExtractor with get_info_extractor""" | |
178 | return cls.__name__[:-2] | |
179 | ||
d6983cb4 PH |
180 | @property |
181 | def IE_NAME(self): | |
182 | return type(self).__name__[:-2] | |
183 | ||
7cc3570e | 184 | def _request_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True): |
d6983cb4 PH |
185 | """ Returns the response handle """ |
186 | if note is None: | |
187 | self.report_download_webpage(video_id) | |
188 | elif note is not False: | |
7cc3570e PH |
189 | if video_id is None: |
190 | self.to_screen(u'%s' % (note,)) | |
191 | else: | |
192 | self.to_screen(u'%s: %s' % (video_id, note)) | |
d6983cb4 | 193 | try: |
dca08720 | 194 | return self._downloader.urlopen(url_or_request) |
d6983cb4 | 195 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: |
aa94a6d3 PH |
196 | if errnote is False: |
197 | return False | |
d6983cb4 PH |
198 | if errnote is None: |
199 | errnote = u'Unable to download webpage' | |
7cc3570e PH |
200 | errmsg = u'%s: %s' % (errnote, compat_str(err)) |
201 | if fatal: | |
202 | raise ExtractorError(errmsg, sys.exc_info()[2], cause=err) | |
203 | else: | |
204 | self._downloader.report_warning(errmsg) | |
205 | return False | |
d6983cb4 | 206 | |
7cc3570e | 207 | def _download_webpage_handle(self, url_or_request, video_id, note=None, errnote=None, fatal=True): |
d6983cb4 | 208 | """ Returns a tuple (page content as string, URL handle) """ |
b9d3e163 PH |
209 | |
210 | # Strip hashes from the URL (#1038) | |
211 | if isinstance(url_or_request, (compat_str, str)): | |
212 | url_or_request = url_or_request.partition('#')[0] | |
213 | ||
7cc3570e PH |
214 | urlh = self._request_webpage(url_or_request, video_id, note, errnote, fatal) |
215 | if urlh is False: | |
216 | assert not fatal | |
217 | return False | |
d6983cb4 | 218 | content_type = urlh.headers.get('Content-Type', '') |
f143d86a | 219 | webpage_bytes = urlh.read() |
d6983cb4 PH |
220 | m = re.match(r'[a-zA-Z0-9_.-]+/[a-zA-Z0-9_.-]+\s*;\s*charset=(.+)', content_type) |
221 | if m: | |
222 | encoding = m.group(1) | |
223 | else: | |
0d75ae2c | 224 | m = re.search(br'<meta[^>]+charset=[\'"]?([^\'")]+)[ /\'">]', |
f143d86a PH |
225 | webpage_bytes[:1024]) |
226 | if m: | |
227 | encoding = m.group(1).decode('ascii') | |
b60016e8 PH |
228 | elif webpage_bytes.startswith(b'\xff\xfe'): |
229 | encoding = 'utf-16' | |
f143d86a PH |
230 | else: |
231 | encoding = 'utf-8' | |
d6983cb4 PH |
232 | if self._downloader.params.get('dump_intermediate_pages', False): |
233 | try: | |
234 | url = url_or_request.get_full_url() | |
235 | except AttributeError: | |
236 | url = url_or_request | |
237 | self.to_screen(u'Dumping request to ' + url) | |
238 | dump = base64.b64encode(webpage_bytes).decode('ascii') | |
239 | self._downloader.to_screen(dump) | |
d41e6efc PH |
240 | if self._downloader.params.get('write_pages', False): |
241 | try: | |
242 | url = url_or_request.get_full_url() | |
243 | except AttributeError: | |
244 | url = url_or_request | |
3ec05685 | 245 | if len(url) > 200: |
944d65c7 | 246 | h = u'___' + hashlib.md5(url.encode('utf-8')).hexdigest() |
3ec05685 | 247 | url = url[:200 - len(h)] + h |
d41e6efc PH |
248 | raw_filename = ('%s_%s.dump' % (video_id, url)) |
249 | filename = sanitize_filename(raw_filename, restricted=True) | |
250 | self.to_screen(u'Saving request to ' + filename) | |
251 | with open(filename, 'wb') as outf: | |
252 | outf.write(webpage_bytes) | |
253 | ||
ec0fafbb AA |
254 | try: |
255 | content = webpage_bytes.decode(encoding, 'replace') | |
256 | except LookupError: | |
257 | content = webpage_bytes.decode('utf-8', 'replace') | |
2410c43d PH |
258 | |
259 | if (u'<title>Access to this site is blocked</title>' in content and | |
260 | u'Websense' in content[:512]): | |
b6cfde99 | 261 | msg = u'Access to this webpage has been blocked by Websense filtering software in your network.' |
2410c43d PH |
262 | blocked_iframe = self._html_search_regex( |
263 | r'<iframe src="([^"]+)"', content, | |
264 | u'Websense information URL', default=None) | |
265 | if blocked_iframe: | |
266 | msg += u' Visit %s for more details' % blocked_iframe | |
267 | raise ExtractorError(msg, expected=True) | |
268 | ||
d6983cb4 PH |
269 | return (content, urlh) |
270 | ||
7cc3570e | 271 | def _download_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True): |
d6983cb4 | 272 | """ Returns the data of the page as a string """ |
7cc3570e PH |
273 | res = self._download_webpage_handle(url_or_request, video_id, note, errnote, fatal) |
274 | if res is False: | |
275 | return res | |
276 | else: | |
277 | content, _ = res | |
278 | return content | |
d6983cb4 | 279 | |
2a275ab0 | 280 | def _download_xml(self, url_or_request, video_id, |
e2b38da9 | 281 | note=u'Downloading XML', errnote=u'Unable to download XML', |
28746fbd | 282 | transform_source=None, fatal=True): |
267ed0c5 | 283 | """Return the xml as an xml.etree.ElementTree.Element""" |
28746fbd PH |
284 | xml_string = self._download_webpage( |
285 | url_or_request, video_id, note, errnote, fatal=fatal) | |
286 | if xml_string is False: | |
287 | return xml_string | |
e2b38da9 PH |
288 | if transform_source: |
289 | xml_string = transform_source(xml_string) | |
267ed0c5 JMF |
290 | return xml.etree.ElementTree.fromstring(xml_string.encode('utf-8')) |
291 | ||
3d3538e4 PH |
292 | def _download_json(self, url_or_request, video_id, |
293 | note=u'Downloading JSON metadata', | |
81c2f20b PH |
294 | errnote=u'Unable to download JSON metadata', |
295 | transform_source=None): | |
3d3538e4 | 296 | json_string = self._download_webpage(url_or_request, video_id, note, errnote) |
81c2f20b PH |
297 | if transform_source: |
298 | json_string = transform_source(json_string) | |
3d3538e4 PH |
299 | try: |
300 | return json.loads(json_string) | |
301 | except ValueError as ve: | |
302 | raise ExtractorError('Failed to download JSON', cause=ve) | |
303 | ||
f45f96f8 PH |
304 | def report_warning(self, msg, video_id=None): |
305 | idstr = u'' if video_id is None else u'%s: ' % video_id | |
306 | self._downloader.report_warning( | |
307 | u'[%s] %s%s' % (self.IE_NAME, idstr, msg)) | |
308 | ||
d6983cb4 PH |
309 | def to_screen(self, msg): |
310 | """Print msg to screen, prefixing it with '[ie_name]'""" | |
311 | self._downloader.to_screen(u'[%s] %s' % (self.IE_NAME, msg)) | |
312 | ||
313 | def report_extraction(self, id_or_name): | |
314 | """Report information extraction.""" | |
315 | self.to_screen(u'%s: Extracting information' % id_or_name) | |
316 | ||
317 | def report_download_webpage(self, video_id): | |
318 | """Report webpage download.""" | |
319 | self.to_screen(u'%s: Downloading webpage' % video_id) | |
320 | ||
321 | def report_age_confirmation(self): | |
322 | """Report attempt to confirm age.""" | |
323 | self.to_screen(u'Confirming age') | |
324 | ||
fc79158d JMF |
325 | def report_login(self): |
326 | """Report attempt to log in.""" | |
327 | self.to_screen(u'Logging in') | |
328 | ||
d6983cb4 | 329 | #Methods for following #608 |
c0d0b01f JMF |
330 | @staticmethod |
331 | def url_result(url, ie=None, video_id=None): | |
d6983cb4 PH |
332 | """Returns a url that points to a page that should be processed""" |
333 | #TODO: ie should be the class used for getting the info | |
334 | video_info = {'_type': 'url', | |
335 | 'url': url, | |
336 | 'ie_key': ie} | |
7012b23c PH |
337 | if video_id is not None: |
338 | video_info['id'] = video_id | |
d6983cb4 | 339 | return video_info |
c0d0b01f JMF |
340 | @staticmethod |
341 | def playlist_result(entries, playlist_id=None, playlist_title=None): | |
d6983cb4 PH |
342 | """Returns a playlist""" |
343 | video_info = {'_type': 'playlist', | |
344 | 'entries': entries} | |
345 | if playlist_id: | |
346 | video_info['id'] = playlist_id | |
347 | if playlist_title: | |
348 | video_info['title'] = playlist_title | |
349 | return video_info | |
350 | ||
46374a56 | 351 | def _search_regex(self, pattern, string, name, default=_NO_DEFAULT, fatal=True, flags=0): |
d6983cb4 PH |
352 | """ |
353 | Perform a regex search on the given string, using a single or a list of | |
354 | patterns returning the first matching group. | |
355 | In case of failure return a default value or raise a WARNING or a | |
55b3e45b | 356 | RegexNotFoundError, depending on fatal, specifying the field name. |
d6983cb4 PH |
357 | """ |
358 | if isinstance(pattern, (str, compat_str, compiled_regex_type)): | |
359 | mobj = re.search(pattern, string, flags) | |
360 | else: | |
361 | for p in pattern: | |
362 | mobj = re.search(p, string, flags) | |
363 | if mobj: break | |
364 | ||
87a28127 | 365 | if os.name != 'nt' and sys.stderr.isatty(): |
d6983cb4 PH |
366 | _name = u'\033[0;34m%s\033[0m' % name |
367 | else: | |
368 | _name = name | |
369 | ||
370 | if mobj: | |
371 | # return the first matching group | |
372 | return next(g for g in mobj.groups() if g is not None) | |
46374a56 | 373 | elif default is not _NO_DEFAULT: |
d6983cb4 PH |
374 | return default |
375 | elif fatal: | |
55b3e45b | 376 | raise RegexNotFoundError(u'Unable to extract %s' % _name) |
d6983cb4 PH |
377 | else: |
378 | self._downloader.report_warning(u'unable to extract %s; ' | |
98bcd283 | 379 | u'please report this issue on http://yt-dl.org/bug' % _name) |
d6983cb4 PH |
380 | return None |
381 | ||
46374a56 | 382 | def _html_search_regex(self, pattern, string, name, default=_NO_DEFAULT, fatal=True, flags=0): |
d6983cb4 PH |
383 | """ |
384 | Like _search_regex, but strips HTML tags and unescapes entities. | |
385 | """ | |
386 | res = self._search_regex(pattern, string, name, default, fatal, flags) | |
387 | if res: | |
388 | return clean_html(res).strip() | |
389 | else: | |
390 | return res | |
391 | ||
fc79158d JMF |
392 | def _get_login_info(self): |
393 | """ | |
394 | Get the the login info as (username, password) | |
395 | It will look in the netrc file using the _NETRC_MACHINE value | |
396 | If there's no info available, return (None, None) | |
397 | """ | |
398 | if self._downloader is None: | |
399 | return (None, None) | |
400 | ||
401 | username = None | |
402 | password = None | |
403 | downloader_params = self._downloader.params | |
404 | ||
405 | # Attempt to use provided username and password or .netrc data | |
406 | if downloader_params.get('username', None) is not None: | |
407 | username = downloader_params['username'] | |
408 | password = downloader_params['password'] | |
409 | elif downloader_params.get('usenetrc', False): | |
410 | try: | |
411 | info = netrc.netrc().authenticators(self._NETRC_MACHINE) | |
412 | if info is not None: | |
413 | username = info[0] | |
414 | password = info[2] | |
415 | else: | |
416 | raise netrc.NetrcParseError('No authenticators for %s' % self._NETRC_MACHINE) | |
417 | except (IOError, netrc.NetrcParseError) as err: | |
418 | self._downloader.report_warning(u'parsing .netrc: %s' % compat_str(err)) | |
419 | ||
420 | return (username, password) | |
421 | ||
46720279 JMF |
422 | # Helper functions for extracting OpenGraph info |
423 | @staticmethod | |
ab2d5247 | 424 | def _og_regexes(prop): |
c1206423 | 425 | content_re = r'content=(?:"([^>]+?)"|\'([^>]+?)\')' |
9887c9b2 | 426 | property_re = r'(?:name|property)=[\'"]og:%s[\'"]' % re.escape(prop) |
78fb87b2 | 427 | template = r'<meta[^>]+?%s[^>]+?%s' |
ab2d5247 | 428 | return [ |
78fb87b2 JMF |
429 | template % (property_re, content_re), |
430 | template % (content_re, property_re), | |
ab2d5247 | 431 | ] |
46720279 | 432 | |
3c4e6d83 | 433 | def _og_search_property(self, prop, html, name=None, **kargs): |
46720279 | 434 | if name is None: |
3c4e6d83 | 435 | name = 'OpenGraph %s' % prop |
ab2d5247 | 436 | escaped = self._search_regex(self._og_regexes(prop), html, name, flags=re.DOTALL, **kargs) |
eb0a8398 PH |
437 | if escaped is None: |
438 | return None | |
439 | return unescapeHTML(escaped) | |
46720279 JMF |
440 | |
441 | def _og_search_thumbnail(self, html, **kargs): | |
3c4e6d83 | 442 | return self._og_search_property('image', html, u'thumbnail url', fatal=False, **kargs) |
46720279 JMF |
443 | |
444 | def _og_search_description(self, html, **kargs): | |
445 | return self._og_search_property('description', html, fatal=False, **kargs) | |
446 | ||
447 | def _og_search_title(self, html, **kargs): | |
448 | return self._og_search_property('title', html, **kargs) | |
449 | ||
8ffa13e0 | 450 | def _og_search_video_url(self, html, name='video url', secure=True, **kargs): |
ab2d5247 JMF |
451 | regexes = self._og_regexes('video') |
452 | if secure: regexes = self._og_regexes('video:secure_url') + regexes | |
8ffa13e0 | 453 | return self._html_search_regex(regexes, html, name, **kargs) |
46720279 | 454 | |
9f62eaf4 | 455 | def _html_search_meta(self, name, html, display_name=None, fatal=False): |
59040888 PH |
456 | if display_name is None: |
457 | display_name = name | |
458 | return self._html_search_regex( | |
aaebed13 PH |
459 | r'''(?ix)<meta |
460 | (?=[^>]+(?:itemprop|name|property)=["\']%s["\']) | |
59040888 | 461 | [^>]+content=["\']([^"\']+)["\']''' % re.escape(name), |
9f62eaf4 | 462 | html, display_name, fatal=fatal) |
59040888 PH |
463 | |
464 | def _dc_search_uploader(self, html): | |
465 | return self._html_search_meta('dc.creator', html, 'uploader') | |
466 | ||
8dbe9899 PH |
467 | def _rta_search(self, html): |
468 | # See http://www.rtalabel.org/index.php?content=howtofaq#single | |
469 | if re.search(r'(?ix)<meta\s+name="rating"\s+' | |
470 | r' content="RTA-5042-1996-1400-1577-RTA"', | |
471 | html): | |
472 | return 18 | |
473 | return 0 | |
474 | ||
59040888 PH |
475 | def _media_rating_search(self, html): |
476 | # See http://www.tjg-designs.com/WP/metadata-code-examples-adding-metadata-to-your-web-pages/ | |
477 | rating = self._html_search_meta('rating', html) | |
478 | ||
479 | if not rating: | |
480 | return None | |
481 | ||
482 | RATING_TABLE = { | |
483 | 'safe for kids': 0, | |
484 | 'general': 8, | |
485 | '14 years': 14, | |
486 | 'mature': 17, | |
487 | 'restricted': 19, | |
488 | } | |
489 | return RATING_TABLE.get(rating.lower(), None) | |
490 | ||
0c708f11 JMF |
491 | def _twitter_search_player(self, html): |
492 | return self._html_search_meta('twitter:player', html, | |
493 | 'twitter card player') | |
494 | ||
4bcc7bd1 | 495 | def _sort_formats(self, formats): |
7e8caf30 PH |
496 | if not formats: |
497 | raise ExtractorError(u'No video formats found') | |
498 | ||
4bcc7bd1 | 499 | def _formats_key(f): |
e6812ac9 PH |
500 | # TODO remove the following workaround |
501 | from ..utils import determine_ext | |
502 | if not f.get('ext') and 'url' in f: | |
503 | f['ext'] = determine_ext(f['url']) | |
504 | ||
4bcc7bd1 PH |
505 | preference = f.get('preference') |
506 | if preference is None: | |
c7deaa4c PH |
507 | proto = f.get('protocol') |
508 | if proto is None: | |
509 | proto = compat_urllib_parse_urlparse(f.get('url', '')).scheme | |
510 | ||
511 | preference = 0 if proto in ['http', 'https'] else -0.1 | |
4bcc7bd1 PH |
512 | if f.get('ext') in ['f4f', 'f4m']: # Not yet supported |
513 | preference -= 0.5 | |
514 | ||
515 | if f.get('vcodec') == 'none': # audio only | |
516 | if self._downloader.params.get('prefer_free_formats'): | |
517 | ORDER = [u'aac', u'mp3', u'm4a', u'webm', u'ogg', u'opus'] | |
518 | else: | |
519 | ORDER = [u'webm', u'opus', u'ogg', u'mp3', u'aac', u'm4a'] | |
520 | ext_preference = 0 | |
521 | try: | |
522 | audio_ext_preference = ORDER.index(f['ext']) | |
523 | except ValueError: | |
524 | audio_ext_preference = -1 | |
525 | else: | |
526 | if self._downloader.params.get('prefer_free_formats'): | |
527 | ORDER = [u'flv', u'mp4', u'webm'] | |
528 | else: | |
529 | ORDER = [u'webm', u'flv', u'mp4'] | |
530 | try: | |
531 | ext_preference = ORDER.index(f['ext']) | |
532 | except ValueError: | |
533 | ext_preference = -1 | |
534 | audio_ext_preference = 0 | |
535 | ||
536 | return ( | |
537 | preference, | |
5d73273f | 538 | f.get('quality') if f.get('quality') is not None else -1, |
4bcc7bd1 PH |
539 | f.get('height') if f.get('height') is not None else -1, |
540 | f.get('width') if f.get('width') is not None else -1, | |
541 | ext_preference, | |
9933b574 | 542 | f.get('tbr') if f.get('tbr') is not None else -1, |
4bcc7bd1 PH |
543 | f.get('vbr') if f.get('vbr') is not None else -1, |
544 | f.get('abr') if f.get('abr') is not None else -1, | |
545 | audio_ext_preference, | |
546 | f.get('filesize') if f.get('filesize') is not None else -1, | |
547 | f.get('format_id'), | |
548 | ) | |
549 | formats.sort(key=_formats_key) | |
59040888 | 550 | |
20991253 PH |
551 | def http_scheme(self): |
552 | """ Either "https:" or "https:", depending on the user's preferences """ | |
553 | return ( | |
554 | 'http:' | |
555 | if self._downloader.params.get('prefer_insecure', False) | |
556 | else 'https:') | |
557 | ||
8dbe9899 | 558 | |
d6983cb4 PH |
559 | class SearchInfoExtractor(InfoExtractor): |
560 | """ | |
561 | Base class for paged search queries extractors. | |
562 | They accept urls in the format _SEARCH_KEY(|all|[0-9]):{query} | |
563 | Instances should define _SEARCH_KEY and _MAX_RESULTS. | |
564 | """ | |
565 | ||
566 | @classmethod | |
567 | def _make_valid_url(cls): | |
568 | return r'%s(?P<prefix>|[1-9][0-9]*|all):(?P<query>[\s\S]+)' % cls._SEARCH_KEY | |
569 | ||
570 | @classmethod | |
571 | def suitable(cls, url): | |
572 | return re.match(cls._make_valid_url(), url) is not None | |
573 | ||
574 | def _real_extract(self, query): | |
575 | mobj = re.match(self._make_valid_url(), query) | |
576 | if mobj is None: | |
577 | raise ExtractorError(u'Invalid search query "%s"' % query) | |
578 | ||
579 | prefix = mobj.group('prefix') | |
580 | query = mobj.group('query') | |
581 | if prefix == '': | |
582 | return self._get_n_results(query, 1) | |
583 | elif prefix == 'all': | |
584 | return self._get_n_results(query, self._MAX_RESULTS) | |
585 | else: | |
586 | n = int(prefix) | |
587 | if n <= 0: | |
588 | raise ExtractorError(u'invalid download number %s for query "%s"' % (n, query)) | |
589 | elif n > self._MAX_RESULTS: | |
590 | self._downloader.report_warning(u'%s returns max %i results (you requested %i)' % (self._SEARCH_KEY, self._MAX_RESULTS, n)) | |
591 | n = self._MAX_RESULTS | |
592 | return self._get_n_results(query, n) | |
593 | ||
594 | def _get_n_results(self, query, n): | |
595 | """Get a specified number of results for a query""" | |
416a5efc | 596 | raise NotImplementedError("This method must be implemented by subclasses") |
0f818663 PH |
597 | |
598 | @property | |
599 | def SEARCH_KEY(self): | |
600 | return self._SEARCH_KEY | |
20991253 | 601 |