]>
Commit | Line | Data |
---|---|---|
f1a9d64e PH |
1 | from __future__ import unicode_literals |
2 | ||
d6983cb4 | 3 | import base64 |
3ec05685 | 4 | import hashlib |
3d3538e4 | 5 | import json |
4094b6e3 | 6 | import netrc |
d6983cb4 PH |
7 | import os |
8 | import re | |
9 | import socket | |
10 | import sys | |
4094b6e3 | 11 | import time |
267ed0c5 | 12 | import xml.etree.ElementTree |
d6983cb4 PH |
13 | |
14 | from ..utils import ( | |
15 | compat_http_client, | |
16 | compat_urllib_error, | |
c7deaa4c | 17 | compat_urllib_parse_urlparse, |
d6983cb4 PH |
18 | compat_str, |
19 | ||
20 | clean_html, | |
21 | compiled_regex_type, | |
22 | ExtractorError, | |
31bb8d3f | 23 | int_or_none, |
55b3e45b | 24 | RegexNotFoundError, |
d41e6efc | 25 | sanitize_filename, |
f38de77f | 26 | unescapeHTML, |
d6983cb4 | 27 | ) |
46374a56 | 28 | _NO_DEFAULT = object() |
d6983cb4 | 29 | |
dca08720 | 30 | |
d6983cb4 PH |
31 | class InfoExtractor(object): |
32 | """Information Extractor class. | |
33 | ||
34 | Information extractors are the classes that, given a URL, extract | |
35 | information about the video (or videos) the URL refers to. This | |
36 | information includes the real video URL, the video title, author and | |
37 | others. The information is stored in a dictionary which is then | |
38 | passed to the FileDownloader. The FileDownloader processes this | |
39 | information possibly downloading the video to the file system, among | |
40 | other possible outcomes. | |
41 | ||
42 | The dictionaries must include the following fields: | |
43 | ||
44 | id: Video identifier. | |
d6983cb4 | 45 | title: Video title, unescaped. |
d67b0b15 | 46 | |
f49d89ee | 47 | Additionally, it must contain either a formats entry or a url one: |
d67b0b15 | 48 | |
f49d89ee PH |
49 | formats: A list of dictionaries for each format available, ordered |
50 | from worst to best quality. | |
51 | ||
52 | Potential fields: | |
d67b0b15 PH |
53 | * url Mandatory. The URL of the video file |
54 | * ext Will be calculated from url if missing | |
55 | * format A human-readable description of the format | |
56 | ("mp4 container with h264/opus"). | |
57 | Calculated from the format_id, width, height. | |
58 | and format_note fields if missing. | |
59 | * format_id A short description of the format | |
5d4f3985 PH |
60 | ("mp4_h264_opus" or "19"). |
61 | Technically optional, but strongly recommended. | |
d67b0b15 PH |
62 | * format_note Additional info about the format |
63 | ("3D" or "DASH video") | |
64 | * width Width of the video, if known | |
65 | * height Height of the video, if known | |
f49d89ee | 66 | * resolution Textual description of width and height |
7217e148 | 67 | * tbr Average bitrate of audio and video in KBit/s |
d67b0b15 PH |
68 | * abr Average audio bitrate in KBit/s |
69 | * acodec Name of the audio codec in use | |
dd27fd17 | 70 | * asr Audio sampling rate in Hertz |
d67b0b15 PH |
71 | * vbr Average video bitrate in KBit/s |
72 | * vcodec Name of the video codec in use | |
1394ce65 | 73 | * container Name of the container format |
d67b0b15 | 74 | * filesize The number of bytes, if known in advance |
9732d77e | 75 | * filesize_approx An estimate for the number of bytes |
d67b0b15 | 76 | * player_url SWF Player URL (used for rtmpdump). |
c7deaa4c PH |
77 | * protocol The protocol that will be used for the actual |
78 | download, lower-case. | |
db1f3888 | 79 | "http", "https", "rtsp", "rtmp", "m3u8" or so. |
f49d89ee | 80 | * preference Order number of this format. If this field is |
08d13955 | 81 | present and not None, the formats get sorted |
38d63d84 | 82 | by this field, regardless of all other values. |
f49d89ee PH |
83 | -1 for default (order by other properties), |
84 | -2 or smaller for less than default. | |
5d73273f PH |
85 | * quality Order number of the video quality of this |
86 | format, irrespective of the file format. | |
87 | -1 for default (order by other properties), | |
88 | -2 or smaller for less than default. | |
d769be6c PH |
89 | * http_referer HTTP Referer header value to set. |
90 | * http_method HTTP method to use for the download. | |
91 | * http_headers A dictionary of additional HTTP headers | |
92 | to add to the request. | |
93 | * http_post_data Additional data to send with a POST | |
94 | request. | |
c0ba0f48 | 95 | url: Final video URL. |
d6983cb4 | 96 | ext: Video filename extension. |
d67b0b15 PH |
97 | format: The video format, defaults to ext (used for --get-format) |
98 | player_url: SWF Player URL (used for rtmpdump). | |
2f5865cc | 99 | |
d6983cb4 PH |
100 | The following fields are optional: |
101 | ||
0afef30b PH |
102 | display_id An alternative identifier for the video, not necessarily |
103 | unique, but available before title. Typically, id is | |
104 | something like "4234987", title "Dancing naked mole rats", | |
105 | and display_id "dancing-naked-mole-rats" | |
d5519808 PH |
106 | thumbnails: A list of dictionaries, with the following entries: |
107 | * "url" | |
108 | * "width" (optional, int) | |
109 | * "height" (optional, int) | |
110 | * "resolution" (optional, string "{width}x{height"}, | |
111 | deprecated) | |
d6983cb4 PH |
112 | thumbnail: Full URL to a video thumbnail image. |
113 | description: One-line video description. | |
114 | uploader: Full name of the video uploader. | |
955c4514 | 115 | timestamp: UNIX timestamp of the moment the video became available. |
d6983cb4 | 116 | upload_date: Video upload date (YYYYMMDD). |
955c4514 | 117 | If not explicitly set, calculated from timestamp. |
d6983cb4 | 118 | uploader_id: Nickname or id of the video uploader. |
da9ec3b9 | 119 | location: Physical location where the video was filmed. |
5d51a883 JMF |
120 | subtitles: The subtitle file contents as a dictionary in the format |
121 | {language: subtitles}. | |
c0ba0f48 | 122 | duration: Length of the video in seconds, as an integer. |
f3d29461 | 123 | view_count: How many users have watched the video on the platform. |
19e3dfc9 PH |
124 | like_count: Number of positive ratings of the video |
125 | dislike_count: Number of negative ratings of the video | |
126 | comment_count: Number of comments on the video | |
8dbe9899 | 127 | age_limit: Age restriction for the video, as an integer (years) |
9103bbc5 JMF |
128 | webpage_url: The url to the video webpage, if given to youtube-dl it |
129 | should allow to get the same result again. (It will be set | |
130 | by YoutubeDL if it's missing) | |
ad3bc6ac PH |
131 | categories: A list of categories that the video falls in, for example |
132 | ["Sports", "Berlin"] | |
d6983cb4 | 133 | |
deefc05b | 134 | Unless mentioned otherwise, the fields should be Unicode strings. |
d6983cb4 PH |
135 | |
136 | Subclasses of this one should re-define the _real_initialize() and | |
137 | _real_extract() methods and define a _VALID_URL regexp. | |
138 | Probably, they should also be added to the list of extractors. | |
139 | ||
d6983cb4 PH |
140 | Finally, the _WORKING attribute should be set to False for broken IEs |
141 | in order to warn the users and skip the tests. | |
142 | """ | |
143 | ||
144 | _ready = False | |
145 | _downloader = None | |
146 | _WORKING = True | |
147 | ||
148 | def __init__(self, downloader=None): | |
149 | """Constructor. Receives an optional downloader.""" | |
150 | self._ready = False | |
151 | self.set_downloader(downloader) | |
152 | ||
153 | @classmethod | |
154 | def suitable(cls, url): | |
155 | """Receives a URL and returns True if suitable for this IE.""" | |
79cb2577 PH |
156 | |
157 | # This does not use has/getattr intentionally - we want to know whether | |
158 | # we have cached the regexp for *this* class, whereas getattr would also | |
159 | # match the superclass | |
160 | if '_VALID_URL_RE' not in cls.__dict__: | |
161 | cls._VALID_URL_RE = re.compile(cls._VALID_URL) | |
162 | return cls._VALID_URL_RE.match(url) is not None | |
d6983cb4 PH |
163 | |
164 | @classmethod | |
165 | def working(cls): | |
166 | """Getter method for _WORKING.""" | |
167 | return cls._WORKING | |
168 | ||
169 | def initialize(self): | |
170 | """Initializes an instance (authentication, etc).""" | |
171 | if not self._ready: | |
172 | self._real_initialize() | |
173 | self._ready = True | |
174 | ||
175 | def extract(self, url): | |
176 | """Extracts URL information and returns it in list of dicts.""" | |
177 | self.initialize() | |
178 | return self._real_extract(url) | |
179 | ||
180 | def set_downloader(self, downloader): | |
181 | """Sets the downloader for this IE.""" | |
182 | self._downloader = downloader | |
183 | ||
184 | def _real_initialize(self): | |
185 | """Real initialization process. Redefine in subclasses.""" | |
186 | pass | |
187 | ||
188 | def _real_extract(self, url): | |
189 | """Real extraction process. Redefine in subclasses.""" | |
190 | pass | |
191 | ||
56c73665 JMF |
192 | @classmethod |
193 | def ie_key(cls): | |
194 | """A string for getting the InfoExtractor with get_info_extractor""" | |
195 | return cls.__name__[:-2] | |
196 | ||
d6983cb4 PH |
197 | @property |
198 | def IE_NAME(self): | |
199 | return type(self).__name__[:-2] | |
200 | ||
7cc3570e | 201 | def _request_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True): |
d6983cb4 PH |
202 | """ Returns the response handle """ |
203 | if note is None: | |
204 | self.report_download_webpage(video_id) | |
205 | elif note is not False: | |
7cc3570e | 206 | if video_id is None: |
f1a9d64e | 207 | self.to_screen('%s' % (note,)) |
7cc3570e | 208 | else: |
f1a9d64e | 209 | self.to_screen('%s: %s' % (video_id, note)) |
d6983cb4 | 210 | try: |
dca08720 | 211 | return self._downloader.urlopen(url_or_request) |
d6983cb4 | 212 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: |
aa94a6d3 PH |
213 | if errnote is False: |
214 | return False | |
d6983cb4 | 215 | if errnote is None: |
f1a9d64e PH |
216 | errnote = 'Unable to download webpage' |
217 | errmsg = '%s: %s' % (errnote, compat_str(err)) | |
7cc3570e PH |
218 | if fatal: |
219 | raise ExtractorError(errmsg, sys.exc_info()[2], cause=err) | |
220 | else: | |
221 | self._downloader.report_warning(errmsg) | |
222 | return False | |
d6983cb4 | 223 | |
7cc3570e | 224 | def _download_webpage_handle(self, url_or_request, video_id, note=None, errnote=None, fatal=True): |
d6983cb4 | 225 | """ Returns a tuple (page content as string, URL handle) """ |
b9d3e163 PH |
226 | |
227 | # Strip hashes from the URL (#1038) | |
228 | if isinstance(url_or_request, (compat_str, str)): | |
229 | url_or_request = url_or_request.partition('#')[0] | |
230 | ||
7cc3570e PH |
231 | urlh = self._request_webpage(url_or_request, video_id, note, errnote, fatal) |
232 | if urlh is False: | |
233 | assert not fatal | |
234 | return False | |
d6983cb4 | 235 | content_type = urlh.headers.get('Content-Type', '') |
f143d86a | 236 | webpage_bytes = urlh.read() |
d6983cb4 PH |
237 | m = re.match(r'[a-zA-Z0-9_.-]+/[a-zA-Z0-9_.-]+\s*;\s*charset=(.+)', content_type) |
238 | if m: | |
239 | encoding = m.group(1) | |
240 | else: | |
0d75ae2c | 241 | m = re.search(br'<meta[^>]+charset=[\'"]?([^\'")]+)[ /\'">]', |
f143d86a PH |
242 | webpage_bytes[:1024]) |
243 | if m: | |
244 | encoding = m.group(1).decode('ascii') | |
b60016e8 PH |
245 | elif webpage_bytes.startswith(b'\xff\xfe'): |
246 | encoding = 'utf-16' | |
f143d86a PH |
247 | else: |
248 | encoding = 'utf-8' | |
d6983cb4 PH |
249 | if self._downloader.params.get('dump_intermediate_pages', False): |
250 | try: | |
251 | url = url_or_request.get_full_url() | |
252 | except AttributeError: | |
253 | url = url_or_request | |
f1a9d64e | 254 | self.to_screen('Dumping request to ' + url) |
d6983cb4 PH |
255 | dump = base64.b64encode(webpage_bytes).decode('ascii') |
256 | self._downloader.to_screen(dump) | |
d41e6efc PH |
257 | if self._downloader.params.get('write_pages', False): |
258 | try: | |
259 | url = url_or_request.get_full_url() | |
260 | except AttributeError: | |
261 | url = url_or_request | |
5afa7f8b | 262 | basen = '%s_%s' % (video_id, url) |
c1bce22f | 263 | if len(basen) > 240: |
f1a9d64e | 264 | h = '___' + hashlib.md5(basen.encode('utf-8')).hexdigest() |
c1bce22f PH |
265 | basen = basen[:240 - len(h)] + h |
266 | raw_filename = basen + '.dump' | |
d41e6efc | 267 | filename = sanitize_filename(raw_filename, restricted=True) |
f1a9d64e | 268 | self.to_screen('Saving request to ' + filename) |
d41e6efc PH |
269 | with open(filename, 'wb') as outf: |
270 | outf.write(webpage_bytes) | |
271 | ||
ec0fafbb AA |
272 | try: |
273 | content = webpage_bytes.decode(encoding, 'replace') | |
274 | except LookupError: | |
275 | content = webpage_bytes.decode('utf-8', 'replace') | |
2410c43d | 276 | |
f1a9d64e PH |
277 | if ('<title>Access to this site is blocked</title>' in content and |
278 | 'Websense' in content[:512]): | |
279 | msg = 'Access to this webpage has been blocked by Websense filtering software in your network.' | |
2410c43d PH |
280 | blocked_iframe = self._html_search_regex( |
281 | r'<iframe src="([^"]+)"', content, | |
f1a9d64e | 282 | 'Websense information URL', default=None) |
2410c43d | 283 | if blocked_iframe: |
f1a9d64e | 284 | msg += ' Visit %s for more details' % blocked_iframe |
2410c43d PH |
285 | raise ExtractorError(msg, expected=True) |
286 | ||
d6983cb4 PH |
287 | return (content, urlh) |
288 | ||
7cc3570e | 289 | def _download_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True): |
d6983cb4 | 290 | """ Returns the data of the page as a string """ |
7cc3570e PH |
291 | res = self._download_webpage_handle(url_or_request, video_id, note, errnote, fatal) |
292 | if res is False: | |
293 | return res | |
294 | else: | |
295 | content, _ = res | |
296 | return content | |
d6983cb4 | 297 | |
2a275ab0 | 298 | def _download_xml(self, url_or_request, video_id, |
f1a9d64e | 299 | note='Downloading XML', errnote='Unable to download XML', |
28746fbd | 300 | transform_source=None, fatal=True): |
267ed0c5 | 301 | """Return the xml as an xml.etree.ElementTree.Element""" |
28746fbd PH |
302 | xml_string = self._download_webpage( |
303 | url_or_request, video_id, note, errnote, fatal=fatal) | |
304 | if xml_string is False: | |
305 | return xml_string | |
e2b38da9 PH |
306 | if transform_source: |
307 | xml_string = transform_source(xml_string) | |
267ed0c5 JMF |
308 | return xml.etree.ElementTree.fromstring(xml_string.encode('utf-8')) |
309 | ||
3d3538e4 | 310 | def _download_json(self, url_or_request, video_id, |
f1a9d64e PH |
311 | note='Downloading JSON metadata', |
312 | errnote='Unable to download JSON metadata', | |
b090af59 PH |
313 | transform_source=None, |
314 | fatal=True): | |
315 | json_string = self._download_webpage( | |
316 | url_or_request, video_id, note, errnote, fatal=fatal) | |
317 | if (not fatal) and json_string is False: | |
318 | return None | |
81c2f20b PH |
319 | if transform_source: |
320 | json_string = transform_source(json_string) | |
3d3538e4 PH |
321 | try: |
322 | return json.loads(json_string) | |
323 | except ValueError as ve: | |
324 | raise ExtractorError('Failed to download JSON', cause=ve) | |
325 | ||
f45f96f8 | 326 | def report_warning(self, msg, video_id=None): |
f1a9d64e | 327 | idstr = '' if video_id is None else '%s: ' % video_id |
f45f96f8 | 328 | self._downloader.report_warning( |
f1a9d64e | 329 | '[%s] %s%s' % (self.IE_NAME, idstr, msg)) |
f45f96f8 | 330 | |
d6983cb4 PH |
331 | def to_screen(self, msg): |
332 | """Print msg to screen, prefixing it with '[ie_name]'""" | |
f1a9d64e | 333 | self._downloader.to_screen('[%s] %s' % (self.IE_NAME, msg)) |
d6983cb4 PH |
334 | |
335 | def report_extraction(self, id_or_name): | |
336 | """Report information extraction.""" | |
f1a9d64e | 337 | self.to_screen('%s: Extracting information' % id_or_name) |
d6983cb4 PH |
338 | |
339 | def report_download_webpage(self, video_id): | |
340 | """Report webpage download.""" | |
f1a9d64e | 341 | self.to_screen('%s: Downloading webpage' % video_id) |
d6983cb4 PH |
342 | |
343 | def report_age_confirmation(self): | |
344 | """Report attempt to confirm age.""" | |
f1a9d64e | 345 | self.to_screen('Confirming age') |
d6983cb4 | 346 | |
fc79158d JMF |
347 | def report_login(self): |
348 | """Report attempt to log in.""" | |
f1a9d64e | 349 | self.to_screen('Logging in') |
fc79158d | 350 | |
d6983cb4 | 351 | #Methods for following #608 |
c0d0b01f JMF |
352 | @staticmethod |
353 | def url_result(url, ie=None, video_id=None): | |
d6983cb4 PH |
354 | """Returns a url that points to a page that should be processed""" |
355 | #TODO: ie should be the class used for getting the info | |
356 | video_info = {'_type': 'url', | |
357 | 'url': url, | |
358 | 'ie_key': ie} | |
7012b23c PH |
359 | if video_id is not None: |
360 | video_info['id'] = video_id | |
d6983cb4 | 361 | return video_info |
c0d0b01f JMF |
362 | @staticmethod |
363 | def playlist_result(entries, playlist_id=None, playlist_title=None): | |
d6983cb4 PH |
364 | """Returns a playlist""" |
365 | video_info = {'_type': 'playlist', | |
366 | 'entries': entries} | |
367 | if playlist_id: | |
368 | video_info['id'] = playlist_id | |
369 | if playlist_title: | |
370 | video_info['title'] = playlist_title | |
371 | return video_info | |
372 | ||
46374a56 | 373 | def _search_regex(self, pattern, string, name, default=_NO_DEFAULT, fatal=True, flags=0): |
d6983cb4 PH |
374 | """ |
375 | Perform a regex search on the given string, using a single or a list of | |
376 | patterns returning the first matching group. | |
377 | In case of failure return a default value or raise a WARNING or a | |
55b3e45b | 378 | RegexNotFoundError, depending on fatal, specifying the field name. |
d6983cb4 PH |
379 | """ |
380 | if isinstance(pattern, (str, compat_str, compiled_regex_type)): | |
381 | mobj = re.search(pattern, string, flags) | |
382 | else: | |
383 | for p in pattern: | |
384 | mobj = re.search(p, string, flags) | |
c3415d1b PH |
385 | if mobj: |
386 | break | |
d6983cb4 | 387 | |
87a28127 | 388 | if os.name != 'nt' and sys.stderr.isatty(): |
f1a9d64e | 389 | _name = '\033[0;34m%s\033[0m' % name |
d6983cb4 PH |
390 | else: |
391 | _name = name | |
392 | ||
393 | if mobj: | |
394 | # return the first matching group | |
395 | return next(g for g in mobj.groups() if g is not None) | |
46374a56 | 396 | elif default is not _NO_DEFAULT: |
d6983cb4 PH |
397 | return default |
398 | elif fatal: | |
f1a9d64e | 399 | raise RegexNotFoundError('Unable to extract %s' % _name) |
d6983cb4 | 400 | else: |
f1a9d64e PH |
401 | self._downloader.report_warning('unable to extract %s; ' |
402 | 'please report this issue on http://yt-dl.org/bug' % _name) | |
d6983cb4 PH |
403 | return None |
404 | ||
46374a56 | 405 | def _html_search_regex(self, pattern, string, name, default=_NO_DEFAULT, fatal=True, flags=0): |
d6983cb4 PH |
406 | """ |
407 | Like _search_regex, but strips HTML tags and unescapes entities. | |
408 | """ | |
409 | res = self._search_regex(pattern, string, name, default, fatal, flags) | |
410 | if res: | |
411 | return clean_html(res).strip() | |
412 | else: | |
413 | return res | |
414 | ||
fc79158d JMF |
415 | def _get_login_info(self): |
416 | """ | |
417 | Get the the login info as (username, password) | |
418 | It will look in the netrc file using the _NETRC_MACHINE value | |
419 | If there's no info available, return (None, None) | |
420 | """ | |
421 | if self._downloader is None: | |
422 | return (None, None) | |
423 | ||
424 | username = None | |
425 | password = None | |
426 | downloader_params = self._downloader.params | |
427 | ||
428 | # Attempt to use provided username and password or .netrc data | |
429 | if downloader_params.get('username', None) is not None: | |
430 | username = downloader_params['username'] | |
431 | password = downloader_params['password'] | |
432 | elif downloader_params.get('usenetrc', False): | |
433 | try: | |
434 | info = netrc.netrc().authenticators(self._NETRC_MACHINE) | |
435 | if info is not None: | |
436 | username = info[0] | |
437 | password = info[2] | |
438 | else: | |
439 | raise netrc.NetrcParseError('No authenticators for %s' % self._NETRC_MACHINE) | |
440 | except (IOError, netrc.NetrcParseError) as err: | |
f1a9d64e | 441 | self._downloader.report_warning('parsing .netrc: %s' % compat_str(err)) |
fc79158d JMF |
442 | |
443 | return (username, password) | |
444 | ||
83317f69 | 445 | def _get_tfa_info(self): |
446 | """ | |
447 | Get the two-factor authentication info | |
448 | TODO - asking the user will be required for sms/phone verify | |
449 | currently just uses the command line option | |
450 | If there's no info available, return None | |
451 | """ | |
452 | if self._downloader is None: | |
83317f69 | 453 | return None |
454 | downloader_params = self._downloader.params | |
455 | ||
456 | if downloader_params.get('twofactor', None) is not None: | |
457 | return downloader_params['twofactor'] | |
458 | ||
83317f69 | 459 | return None |
460 | ||
46720279 JMF |
461 | # Helper functions for extracting OpenGraph info |
462 | @staticmethod | |
ab2d5247 | 463 | def _og_regexes(prop): |
c1206423 | 464 | content_re = r'content=(?:"([^>]+?)"|\'([^>]+?)\')' |
9887c9b2 | 465 | property_re = r'(?:name|property)=[\'"]og:%s[\'"]' % re.escape(prop) |
78fb87b2 | 466 | template = r'<meta[^>]+?%s[^>]+?%s' |
ab2d5247 | 467 | return [ |
78fb87b2 JMF |
468 | template % (property_re, content_re), |
469 | template % (content_re, property_re), | |
ab2d5247 | 470 | ] |
46720279 | 471 | |
3c4e6d83 | 472 | def _og_search_property(self, prop, html, name=None, **kargs): |
46720279 | 473 | if name is None: |
3c4e6d83 | 474 | name = 'OpenGraph %s' % prop |
ab2d5247 | 475 | escaped = self._search_regex(self._og_regexes(prop), html, name, flags=re.DOTALL, **kargs) |
eb0a8398 PH |
476 | if escaped is None: |
477 | return None | |
478 | return unescapeHTML(escaped) | |
46720279 JMF |
479 | |
480 | def _og_search_thumbnail(self, html, **kargs): | |
f1a9d64e | 481 | return self._og_search_property('image', html, 'thumbnail url', fatal=False, **kargs) |
46720279 JMF |
482 | |
483 | def _og_search_description(self, html, **kargs): | |
484 | return self._og_search_property('description', html, fatal=False, **kargs) | |
485 | ||
486 | def _og_search_title(self, html, **kargs): | |
487 | return self._og_search_property('title', html, **kargs) | |
488 | ||
8ffa13e0 | 489 | def _og_search_video_url(self, html, name='video url', secure=True, **kargs): |
a3681973 PH |
490 | regexes = self._og_regexes('video') + self._og_regexes('video:url') |
491 | if secure: | |
492 | regexes = self._og_regexes('video:secure_url') + regexes | |
8ffa13e0 | 493 | return self._html_search_regex(regexes, html, name, **kargs) |
46720279 | 494 | |
78338f71 JMF |
495 | def _og_search_url(self, html, **kargs): |
496 | return self._og_search_property('url', html, **kargs) | |
497 | ||
40c696e5 | 498 | def _html_search_meta(self, name, html, display_name=None, fatal=False, **kwargs): |
59040888 PH |
499 | if display_name is None: |
500 | display_name = name | |
501 | return self._html_search_regex( | |
aaebed13 | 502 | r'''(?ix)<meta |
1a30deca | 503 | (?=[^>]+(?:itemprop|name|property)=["\']?%s["\']?) |
59040888 | 504 | [^>]+content=["\']([^"\']+)["\']''' % re.escape(name), |
40c696e5 | 505 | html, display_name, fatal=fatal, **kwargs) |
59040888 PH |
506 | |
507 | def _dc_search_uploader(self, html): | |
508 | return self._html_search_meta('dc.creator', html, 'uploader') | |
509 | ||
8dbe9899 PH |
510 | def _rta_search(self, html): |
511 | # See http://www.rtalabel.org/index.php?content=howtofaq#single | |
512 | if re.search(r'(?ix)<meta\s+name="rating"\s+' | |
513 | r' content="RTA-5042-1996-1400-1577-RTA"', | |
514 | html): | |
515 | return 18 | |
516 | return 0 | |
517 | ||
59040888 PH |
518 | def _media_rating_search(self, html): |
519 | # See http://www.tjg-designs.com/WP/metadata-code-examples-adding-metadata-to-your-web-pages/ | |
520 | rating = self._html_search_meta('rating', html) | |
521 | ||
522 | if not rating: | |
523 | return None | |
524 | ||
525 | RATING_TABLE = { | |
526 | 'safe for kids': 0, | |
527 | 'general': 8, | |
528 | '14 years': 14, | |
529 | 'mature': 17, | |
530 | 'restricted': 19, | |
531 | } | |
532 | return RATING_TABLE.get(rating.lower(), None) | |
533 | ||
0c708f11 JMF |
534 | def _twitter_search_player(self, html): |
535 | return self._html_search_meta('twitter:player', html, | |
536 | 'twitter card player') | |
537 | ||
4bcc7bd1 | 538 | def _sort_formats(self, formats): |
7e8caf30 | 539 | if not formats: |
f1a9d64e | 540 | raise ExtractorError('No video formats found') |
7e8caf30 | 541 | |
4bcc7bd1 | 542 | def _formats_key(f): |
e6812ac9 PH |
543 | # TODO remove the following workaround |
544 | from ..utils import determine_ext | |
545 | if not f.get('ext') and 'url' in f: | |
546 | f['ext'] = determine_ext(f['url']) | |
547 | ||
4bcc7bd1 PH |
548 | preference = f.get('preference') |
549 | if preference is None: | |
c7deaa4c PH |
550 | proto = f.get('protocol') |
551 | if proto is None: | |
552 | proto = compat_urllib_parse_urlparse(f.get('url', '')).scheme | |
553 | ||
554 | preference = 0 if proto in ['http', 'https'] else -0.1 | |
4bcc7bd1 PH |
555 | if f.get('ext') in ['f4f', 'f4m']: # Not yet supported |
556 | preference -= 0.5 | |
557 | ||
558 | if f.get('vcodec') == 'none': # audio only | |
559 | if self._downloader.params.get('prefer_free_formats'): | |
f1a9d64e | 560 | ORDER = ['aac', 'mp3', 'm4a', 'webm', 'ogg', 'opus'] |
4bcc7bd1 | 561 | else: |
f1a9d64e | 562 | ORDER = ['webm', 'opus', 'ogg', 'mp3', 'aac', 'm4a'] |
4bcc7bd1 PH |
563 | ext_preference = 0 |
564 | try: | |
565 | audio_ext_preference = ORDER.index(f['ext']) | |
566 | except ValueError: | |
567 | audio_ext_preference = -1 | |
568 | else: | |
569 | if self._downloader.params.get('prefer_free_formats'): | |
f1a9d64e | 570 | ORDER = ['flv', 'mp4', 'webm'] |
4bcc7bd1 | 571 | else: |
f1a9d64e | 572 | ORDER = ['webm', 'flv', 'mp4'] |
4bcc7bd1 PH |
573 | try: |
574 | ext_preference = ORDER.index(f['ext']) | |
575 | except ValueError: | |
576 | ext_preference = -1 | |
577 | audio_ext_preference = 0 | |
578 | ||
579 | return ( | |
580 | preference, | |
5d73273f | 581 | f.get('quality') if f.get('quality') is not None else -1, |
4bcc7bd1 PH |
582 | f.get('height') if f.get('height') is not None else -1, |
583 | f.get('width') if f.get('width') is not None else -1, | |
584 | ext_preference, | |
9933b574 | 585 | f.get('tbr') if f.get('tbr') is not None else -1, |
4bcc7bd1 PH |
586 | f.get('vbr') if f.get('vbr') is not None else -1, |
587 | f.get('abr') if f.get('abr') is not None else -1, | |
588 | audio_ext_preference, | |
589 | f.get('filesize') if f.get('filesize') is not None else -1, | |
9732d77e | 590 | f.get('filesize_approx') if f.get('filesize_approx') is not None else -1, |
4bcc7bd1 PH |
591 | f.get('format_id'), |
592 | ) | |
593 | formats.sort(key=_formats_key) | |
59040888 | 594 | |
20991253 PH |
595 | def http_scheme(self): |
596 | """ Either "https:" or "https:", depending on the user's preferences """ | |
597 | return ( | |
598 | 'http:' | |
599 | if self._downloader.params.get('prefer_insecure', False) | |
600 | else 'https:') | |
601 | ||
57c7411f PH |
602 | def _proto_relative_url(self, url, scheme=None): |
603 | if url is None: | |
604 | return url | |
605 | if url.startswith('//'): | |
606 | if scheme is None: | |
607 | scheme = self.http_scheme() | |
608 | return scheme + url | |
609 | else: | |
610 | return url | |
611 | ||
4094b6e3 PH |
612 | def _sleep(self, timeout, video_id, msg_template=None): |
613 | if msg_template is None: | |
f1a9d64e | 614 | msg_template = '%(video_id)s: Waiting for %(timeout)s seconds' |
4094b6e3 PH |
615 | msg = msg_template % {'video_id': video_id, 'timeout': timeout} |
616 | self.to_screen(msg) | |
617 | time.sleep(timeout) | |
618 | ||
31bb8d3f | 619 | def _extract_f4m_formats(self, manifest_url, video_id): |
f036a632 JMF |
620 | manifest = self._download_xml( |
621 | manifest_url, video_id, 'Downloading f4m manifest', | |
622 | 'Unable to download f4m manifest') | |
31bb8d3f JMF |
623 | |
624 | formats = [] | |
b2527359 PH |
625 | media_nodes = manifest.findall('{http://ns.adobe.com/f4m/1.0}media') |
626 | for i, media_el in enumerate(media_nodes): | |
627 | tbr = int_or_none(media_el.attrib.get('bitrate')) | |
628 | format_id = 'f4m-%d' % (i if tbr is None else tbr) | |
31bb8d3f | 629 | formats.append({ |
b2527359 | 630 | 'format_id': format_id, |
31bb8d3f JMF |
631 | 'url': manifest_url, |
632 | 'ext': 'flv', | |
b2527359 | 633 | 'tbr': tbr, |
31bb8d3f JMF |
634 | 'width': int_or_none(media_el.attrib.get('width')), |
635 | 'height': int_or_none(media_el.attrib.get('height')), | |
636 | }) | |
637 | self._sort_formats(formats) | |
638 | ||
639 | return formats | |
640 | ||
704df56d PH |
641 | def _extract_m3u8_formats(self, m3u8_url, video_id, ext=None): |
642 | formats = [{ | |
643 | 'format_id': 'm3u8-meta', | |
644 | 'url': m3u8_url, | |
645 | 'ext': ext, | |
646 | 'protocol': 'm3u8', | |
647 | 'preference': -1, | |
648 | 'resolution': 'multiple', | |
649 | 'format_note': 'Quality selection URL', | |
650 | }] | |
651 | ||
652 | m3u8_doc = self._download_webpage(m3u8_url, video_id) | |
653 | last_info = None | |
654 | kv_rex = re.compile( | |
655 | r'(?P<key>[a-zA-Z_-]+)=(?P<val>"[^"]+"|[^",]+)(?:,|$)') | |
656 | for line in m3u8_doc.splitlines(): | |
657 | if line.startswith('#EXT-X-STREAM-INF:'): | |
658 | last_info = {} | |
659 | for m in kv_rex.finditer(line): | |
660 | v = m.group('val') | |
661 | if v.startswith('"'): | |
662 | v = v[1:-1] | |
663 | last_info[m.group('key')] = v | |
664 | elif line.startswith('#') or not line.strip(): | |
665 | continue | |
666 | else: | |
daebaab6 | 667 | if last_info is None: |
3524cc25 PH |
668 | formats.append({'url': line}) |
669 | continue | |
704df56d PH |
670 | tbr = int_or_none(last_info.get('BANDWIDTH'), scale=1000) |
671 | ||
672 | f = { | |
673 | 'format_id': 'm3u8-%d' % (tbr if tbr else len(formats)), | |
674 | 'url': line.strip(), | |
675 | 'tbr': tbr, | |
676 | 'ext': ext, | |
677 | } | |
678 | codecs = last_info.get('CODECS') | |
679 | if codecs: | |
680 | video, audio = codecs.split(',') | |
681 | f['vcodec'] = video.partition('.')[0] | |
682 | f['acodec'] = audio.partition('.')[0] | |
683 | resolution = last_info.get('RESOLUTION') | |
684 | if resolution: | |
685 | width_str, height_str = resolution.split('x') | |
686 | f['width'] = int(width_str) | |
687 | f['height'] = int(height_str) | |
688 | formats.append(f) | |
689 | last_info = {} | |
690 | self._sort_formats(formats) | |
691 | return formats | |
692 | ||
8dbe9899 | 693 | |
d6983cb4 PH |
694 | class SearchInfoExtractor(InfoExtractor): |
695 | """ | |
696 | Base class for paged search queries extractors. | |
697 | They accept urls in the format _SEARCH_KEY(|all|[0-9]):{query} | |
698 | Instances should define _SEARCH_KEY and _MAX_RESULTS. | |
699 | """ | |
700 | ||
701 | @classmethod | |
702 | def _make_valid_url(cls): | |
703 | return r'%s(?P<prefix>|[1-9][0-9]*|all):(?P<query>[\s\S]+)' % cls._SEARCH_KEY | |
704 | ||
705 | @classmethod | |
706 | def suitable(cls, url): | |
707 | return re.match(cls._make_valid_url(), url) is not None | |
708 | ||
709 | def _real_extract(self, query): | |
710 | mobj = re.match(self._make_valid_url(), query) | |
711 | if mobj is None: | |
f1a9d64e | 712 | raise ExtractorError('Invalid search query "%s"' % query) |
d6983cb4 PH |
713 | |
714 | prefix = mobj.group('prefix') | |
715 | query = mobj.group('query') | |
716 | if prefix == '': | |
717 | return self._get_n_results(query, 1) | |
718 | elif prefix == 'all': | |
719 | return self._get_n_results(query, self._MAX_RESULTS) | |
720 | else: | |
721 | n = int(prefix) | |
722 | if n <= 0: | |
f1a9d64e | 723 | raise ExtractorError('invalid download number %s for query "%s"' % (n, query)) |
d6983cb4 | 724 | elif n > self._MAX_RESULTS: |
f1a9d64e | 725 | self._downloader.report_warning('%s returns max %i results (you requested %i)' % (self._SEARCH_KEY, self._MAX_RESULTS, n)) |
d6983cb4 PH |
726 | n = self._MAX_RESULTS |
727 | return self._get_n_results(query, n) | |
728 | ||
729 | def _get_n_results(self, query, n): | |
730 | """Get a specified number of results for a query""" | |
416a5efc | 731 | raise NotImplementedError("This method must be implemented by subclasses") |
0f818663 PH |
732 | |
733 | @property | |
734 | def SEARCH_KEY(self): | |
735 | return self._SEARCH_KEY |