]>
Commit | Line | Data |
---|---|---|
f1a9d64e PH |
1 | from __future__ import unicode_literals |
2 | ||
d6983cb4 | 3 | import base64 |
f4b1c7ad | 4 | import datetime |
3ec05685 | 5 | import hashlib |
3d3538e4 | 6 | import json |
4094b6e3 | 7 | import netrc |
d6983cb4 PH |
8 | import os |
9 | import re | |
10 | import socket | |
11 | import sys | |
4094b6e3 | 12 | import time |
267ed0c5 | 13 | import xml.etree.ElementTree |
d6983cb4 | 14 | |
8c25f81b | 15 | from ..compat import ( |
42939b61 | 16 | compat_cookiejar, |
96a53167 | 17 | compat_HTTPError, |
d6983cb4 PH |
18 | compat_http_client, |
19 | compat_urllib_error, | |
c7deaa4c | 20 | compat_urllib_parse_urlparse, |
f0b5d6af | 21 | compat_urlparse, |
d6983cb4 | 22 | compat_str, |
8c25f81b PH |
23 | ) |
24 | from ..utils import ( | |
05900629 | 25 | age_restricted, |
d6983cb4 PH |
26 | clean_html, |
27 | compiled_regex_type, | |
28 | ExtractorError, | |
b14f3a4c | 29 | float_or_none, |
96a53167 | 30 | HEADRequest, |
31bb8d3f | 31 | int_or_none, |
55b3e45b | 32 | RegexNotFoundError, |
d41e6efc | 33 | sanitize_filename, |
f38de77f | 34 | unescapeHTML, |
d6983cb4 | 35 | ) |
46374a56 | 36 | _NO_DEFAULT = object() |
d6983cb4 | 37 | |
dca08720 | 38 | |
d6983cb4 PH |
39 | class InfoExtractor(object): |
40 | """Information Extractor class. | |
41 | ||
42 | Information extractors are the classes that, given a URL, extract | |
43 | information about the video (or videos) the URL refers to. This | |
44 | information includes the real video URL, the video title, author and | |
45 | others. The information is stored in a dictionary which is then | |
5d380852 | 46 | passed to the YoutubeDL. The YoutubeDL processes this |
d6983cb4 PH |
47 | information possibly downloading the video to the file system, among |
48 | other possible outcomes. | |
49 | ||
fed5d032 PH |
50 | The type field determines the the type of the result. |
51 | By far the most common value (and the default if _type is missing) is | |
52 | "video", which indicates a single video. | |
53 | ||
54 | For a video, the dictionaries must include the following fields: | |
d6983cb4 PH |
55 | |
56 | id: Video identifier. | |
d6983cb4 | 57 | title: Video title, unescaped. |
d67b0b15 | 58 | |
f49d89ee | 59 | Additionally, it must contain either a formats entry or a url one: |
d67b0b15 | 60 | |
f49d89ee PH |
61 | formats: A list of dictionaries for each format available, ordered |
62 | from worst to best quality. | |
63 | ||
64 | Potential fields: | |
d67b0b15 PH |
65 | * url Mandatory. The URL of the video file |
66 | * ext Will be calculated from url if missing | |
67 | * format A human-readable description of the format | |
68 | ("mp4 container with h264/opus"). | |
69 | Calculated from the format_id, width, height. | |
70 | and format_note fields if missing. | |
71 | * format_id A short description of the format | |
5d4f3985 PH |
72 | ("mp4_h264_opus" or "19"). |
73 | Technically optional, but strongly recommended. | |
d67b0b15 PH |
74 | * format_note Additional info about the format |
75 | ("3D" or "DASH video") | |
76 | * width Width of the video, if known | |
77 | * height Height of the video, if known | |
f49d89ee | 78 | * resolution Textual description of width and height |
7217e148 | 79 | * tbr Average bitrate of audio and video in KBit/s |
d67b0b15 PH |
80 | * abr Average audio bitrate in KBit/s |
81 | * acodec Name of the audio codec in use | |
dd27fd17 | 82 | * asr Audio sampling rate in Hertz |
d67b0b15 | 83 | * vbr Average video bitrate in KBit/s |
fbb21cf5 | 84 | * fps Frame rate |
d67b0b15 | 85 | * vcodec Name of the video codec in use |
1394ce65 | 86 | * container Name of the container format |
d67b0b15 | 87 | * filesize The number of bytes, if known in advance |
9732d77e | 88 | * filesize_approx An estimate for the number of bytes |
d67b0b15 | 89 | * player_url SWF Player URL (used for rtmpdump). |
c7deaa4c PH |
90 | * protocol The protocol that will be used for the actual |
91 | download, lower-case. | |
b04b8852 PH |
92 | "http", "https", "rtsp", "rtmp", "rtmpe", |
93 | "m3u8", or "m3u8_native". | |
f49d89ee | 94 | * preference Order number of this format. If this field is |
08d13955 | 95 | present and not None, the formats get sorted |
38d63d84 | 96 | by this field, regardless of all other values. |
f49d89ee PH |
97 | -1 for default (order by other properties), |
98 | -2 or smaller for less than default. | |
e65566a9 PH |
99 | < -1000 to hide the format (if there is |
100 | another one which is strictly better) | |
aff2f4f4 PH |
101 | * language_preference Is this in the correct requested |
102 | language? | |
103 | 10 if it's what the URL is about, | |
104 | -1 for default (don't know), | |
105 | -10 otherwise, other values reserved for now. | |
5d73273f PH |
106 | * quality Order number of the video quality of this |
107 | format, irrespective of the file format. | |
108 | -1 for default (order by other properties), | |
109 | -2 or smaller for less than default. | |
c64ed2a3 PH |
110 | * source_preference Order number for this video source |
111 | (quality takes higher priority) | |
112 | -1 for default (order by other properties), | |
113 | -2 or smaller for less than default. | |
d769be6c PH |
114 | * http_method HTTP method to use for the download. |
115 | * http_headers A dictionary of additional HTTP headers | |
116 | to add to the request. | |
117 | * http_post_data Additional data to send with a POST | |
118 | request. | |
6271f1ca | 119 | * stretched_ratio If given and not 1, indicates that the |
3dee7826 PH |
120 | video's pixels are not square. |
121 | width : height ratio as float. | |
122 | * no_resume The server does not support resuming the | |
123 | (HTTP or RTMP) download. Boolean. | |
124 | ||
c0ba0f48 | 125 | url: Final video URL. |
d6983cb4 | 126 | ext: Video filename extension. |
d67b0b15 PH |
127 | format: The video format, defaults to ext (used for --get-format) |
128 | player_url: SWF Player URL (used for rtmpdump). | |
2f5865cc | 129 | |
d6983cb4 PH |
130 | The following fields are optional: |
131 | ||
f5e43bc6 | 132 | alt_title: A secondary title of the video. |
0afef30b PH |
133 | display_id An alternative identifier for the video, not necessarily |
134 | unique, but available before title. Typically, id is | |
135 | something like "4234987", title "Dancing naked mole rats", | |
136 | and display_id "dancing-naked-mole-rats" | |
d5519808 | 137 | thumbnails: A list of dictionaries, with the following entries: |
cfb56d1a | 138 | * "id" (optional, string) - Thumbnail format ID |
d5519808 | 139 | * "url" |
cfb56d1a | 140 | * "preference" (optional, int) - quality of the image |
d5519808 PH |
141 | * "width" (optional, int) |
142 | * "height" (optional, int) | |
143 | * "resolution" (optional, string "{width}x{height"}, | |
144 | deprecated) | |
d6983cb4 | 145 | thumbnail: Full URL to a video thumbnail image. |
f5e43bc6 | 146 | description: Full video description. |
d6983cb4 | 147 | uploader: Full name of the video uploader. |
9bb8e0a3 | 148 | creator: The main artist who created the video. |
955c4514 | 149 | timestamp: UNIX timestamp of the moment the video became available. |
d6983cb4 | 150 | upload_date: Video upload date (YYYYMMDD). |
955c4514 | 151 | If not explicitly set, calculated from timestamp. |
d6983cb4 | 152 | uploader_id: Nickname or id of the video uploader. |
da9ec3b9 | 153 | location: Physical location where the video was filmed. |
a504ced0 JMF |
154 | subtitles: The available subtitles as a dictionary in the format |
155 | {language: subformats}. "subformats" is a list sorted from | |
156 | lower to higher preference, each element is a dictionary | |
157 | with the "ext" entry and one of: | |
158 | * "data": The subtitles file contents | |
159 | * "url": A url pointing to the subtitles file | |
360e1ca5 JMF |
160 | automatic_captions: Like 'subtitles', used by the YoutubeIE for |
161 | automatically generated captions | |
c0ba0f48 | 162 | duration: Length of the video in seconds, as an integer. |
f3d29461 | 163 | view_count: How many users have watched the video on the platform. |
19e3dfc9 PH |
164 | like_count: Number of positive ratings of the video |
165 | dislike_count: Number of negative ratings of the video | |
2d30521a | 166 | average_rating: Average rating give by users, the scale used depends on the webpage |
19e3dfc9 | 167 | comment_count: Number of comments on the video |
dd622d7c PH |
168 | comments: A list of comments, each with one or more of the following |
169 | properties (all but one of text or html optional): | |
170 | * "author" - human-readable name of the comment author | |
171 | * "author_id" - user ID of the comment author | |
172 | * "id" - Comment ID | |
173 | * "html" - Comment as HTML | |
174 | * "text" - Plain text of the comment | |
175 | * "timestamp" - UNIX timestamp of comment | |
176 | * "parent" - ID of the comment this one is replying to. | |
177 | Set to "root" to indicate that this is a | |
178 | comment to the original video. | |
8dbe9899 | 179 | age_limit: Age restriction for the video, as an integer (years) |
9103bbc5 JMF |
180 | webpage_url: The url to the video webpage, if given to youtube-dl it |
181 | should allow to get the same result again. (It will be set | |
182 | by YoutubeDL if it's missing) | |
ad3bc6ac PH |
183 | categories: A list of categories that the video falls in, for example |
184 | ["Sports", "Berlin"] | |
7267bd53 PH |
185 | is_live: True, False, or None (=unknown). Whether this video is a |
186 | live stream that goes on instead of a fixed-length video. | |
d6983cb4 | 187 | |
deefc05b | 188 | Unless mentioned otherwise, the fields should be Unicode strings. |
d6983cb4 | 189 | |
d838b1bd PH |
190 | Unless mentioned otherwise, None is equivalent to absence of information. |
191 | ||
fed5d032 PH |
192 | |
193 | _type "playlist" indicates multiple videos. | |
b82f815f PH |
194 | There must be a key "entries", which is a list, an iterable, or a PagedList |
195 | object, each element of which is a valid dictionary by this specification. | |
fed5d032 PH |
196 | |
197 | Additionally, playlists can have "title" and "id" attributes with the same | |
198 | semantics as videos (see above). | |
199 | ||
200 | ||
201 | _type "multi_video" indicates that there are multiple videos that | |
202 | form a single show, for examples multiple acts of an opera or TV episode. | |
203 | It must have an entries key like a playlist and contain all the keys | |
204 | required for a video at the same time. | |
205 | ||
206 | ||
207 | _type "url" indicates that the video must be extracted from another | |
208 | location, possibly by a different extractor. Its only required key is: | |
209 | "url" - the next URL to extract. | |
f58766ce PH |
210 | The key "ie_key" can be set to the class name (minus the trailing "IE", |
211 | e.g. "Youtube") if the extractor class is known in advance. | |
212 | Additionally, the dictionary may have any properties of the resolved entity | |
213 | known in advance, for example "title" if the title of the referred video is | |
fed5d032 PH |
214 | known ahead of time. |
215 | ||
216 | ||
217 | _type "url_transparent" entities have the same specification as "url", but | |
218 | indicate that the given additional information is more precise than the one | |
219 | associated with the resolved URL. | |
220 | This is useful when a site employs a video service that hosts the video and | |
221 | its technical metadata, but that video service does not embed a useful | |
222 | title, description etc. | |
223 | ||
224 | ||
d6983cb4 PH |
225 | Subclasses of this one should re-define the _real_initialize() and |
226 | _real_extract() methods and define a _VALID_URL regexp. | |
227 | Probably, they should also be added to the list of extractors. | |
228 | ||
d6983cb4 PH |
229 | Finally, the _WORKING attribute should be set to False for broken IEs |
230 | in order to warn the users and skip the tests. | |
231 | """ | |
232 | ||
233 | _ready = False | |
234 | _downloader = None | |
235 | _WORKING = True | |
236 | ||
237 | def __init__(self, downloader=None): | |
238 | """Constructor. Receives an optional downloader.""" | |
239 | self._ready = False | |
240 | self.set_downloader(downloader) | |
241 | ||
242 | @classmethod | |
243 | def suitable(cls, url): | |
244 | """Receives a URL and returns True if suitable for this IE.""" | |
79cb2577 PH |
245 | |
246 | # This does not use has/getattr intentionally - we want to know whether | |
247 | # we have cached the regexp for *this* class, whereas getattr would also | |
248 | # match the superclass | |
249 | if '_VALID_URL_RE' not in cls.__dict__: | |
250 | cls._VALID_URL_RE = re.compile(cls._VALID_URL) | |
251 | return cls._VALID_URL_RE.match(url) is not None | |
d6983cb4 | 252 | |
ed9266db PH |
253 | @classmethod |
254 | def _match_id(cls, url): | |
255 | if '_VALID_URL_RE' not in cls.__dict__: | |
256 | cls._VALID_URL_RE = re.compile(cls._VALID_URL) | |
257 | m = cls._VALID_URL_RE.match(url) | |
258 | assert m | |
259 | return m.group('id') | |
260 | ||
d6983cb4 PH |
261 | @classmethod |
262 | def working(cls): | |
263 | """Getter method for _WORKING.""" | |
264 | return cls._WORKING | |
265 | ||
266 | def initialize(self): | |
267 | """Initializes an instance (authentication, etc).""" | |
268 | if not self._ready: | |
269 | self._real_initialize() | |
270 | self._ready = True | |
271 | ||
272 | def extract(self, url): | |
273 | """Extracts URL information and returns it in list of dicts.""" | |
3a5bcd03 PH |
274 | try: |
275 | self.initialize() | |
276 | return self._real_extract(url) | |
277 | except ExtractorError: | |
278 | raise | |
279 | except compat_http_client.IncompleteRead as e: | |
280 | raise ExtractorError('A network error has occured.', cause=e, expected=True) | |
9650885b | 281 | except (KeyError, StopIteration) as e: |
3a5bcd03 | 282 | raise ExtractorError('An extractor error has occured.', cause=e) |
d6983cb4 PH |
283 | |
284 | def set_downloader(self, downloader): | |
285 | """Sets the downloader for this IE.""" | |
286 | self._downloader = downloader | |
287 | ||
288 | def _real_initialize(self): | |
289 | """Real initialization process. Redefine in subclasses.""" | |
290 | pass | |
291 | ||
292 | def _real_extract(self, url): | |
293 | """Real extraction process. Redefine in subclasses.""" | |
294 | pass | |
295 | ||
56c73665 JMF |
296 | @classmethod |
297 | def ie_key(cls): | |
298 | """A string for getting the InfoExtractor with get_info_extractor""" | |
299 | return cls.__name__[:-2] | |
300 | ||
d6983cb4 PH |
301 | @property |
302 | def IE_NAME(self): | |
303 | return type(self).__name__[:-2] | |
304 | ||
7cc3570e | 305 | def _request_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True): |
d6983cb4 PH |
306 | """ Returns the response handle """ |
307 | if note is None: | |
308 | self.report_download_webpage(video_id) | |
309 | elif note is not False: | |
7cc3570e | 310 | if video_id is None: |
f1a9d64e | 311 | self.to_screen('%s' % (note,)) |
7cc3570e | 312 | else: |
f1a9d64e | 313 | self.to_screen('%s: %s' % (video_id, note)) |
d6983cb4 | 314 | try: |
dca08720 | 315 | return self._downloader.urlopen(url_or_request) |
d6983cb4 | 316 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: |
aa94a6d3 PH |
317 | if errnote is False: |
318 | return False | |
d6983cb4 | 319 | if errnote is None: |
f1a9d64e PH |
320 | errnote = 'Unable to download webpage' |
321 | errmsg = '%s: %s' % (errnote, compat_str(err)) | |
7cc3570e PH |
322 | if fatal: |
323 | raise ExtractorError(errmsg, sys.exc_info()[2], cause=err) | |
324 | else: | |
325 | self._downloader.report_warning(errmsg) | |
326 | return False | |
d6983cb4 | 327 | |
7cc3570e | 328 | def _download_webpage_handle(self, url_or_request, video_id, note=None, errnote=None, fatal=True): |
d6983cb4 | 329 | """ Returns a tuple (page content as string, URL handle) """ |
b9d3e163 PH |
330 | # Strip hashes from the URL (#1038) |
331 | if isinstance(url_or_request, (compat_str, str)): | |
332 | url_or_request = url_or_request.partition('#')[0] | |
333 | ||
7cc3570e PH |
334 | urlh = self._request_webpage(url_or_request, video_id, note, errnote, fatal) |
335 | if urlh is False: | |
336 | assert not fatal | |
337 | return False | |
23be51d8 PH |
338 | content = self._webpage_read_content(urlh, url_or_request, video_id, note, errnote, fatal) |
339 | return (content, urlh) | |
340 | ||
4e262a88 | 341 | def _webpage_read_content(self, urlh, url_or_request, video_id, note=None, errnote=None, fatal=True, prefix=None): |
d6983cb4 | 342 | content_type = urlh.headers.get('Content-Type', '') |
f143d86a | 343 | webpage_bytes = urlh.read() |
4e262a88 PH |
344 | if prefix is not None: |
345 | webpage_bytes = prefix + webpage_bytes | |
d6983cb4 PH |
346 | m = re.match(r'[a-zA-Z0-9_.-]+/[a-zA-Z0-9_.-]+\s*;\s*charset=(.+)', content_type) |
347 | if m: | |
348 | encoding = m.group(1) | |
349 | else: | |
0d75ae2c | 350 | m = re.search(br'<meta[^>]+charset=[\'"]?([^\'")]+)[ /\'">]', |
f143d86a PH |
351 | webpage_bytes[:1024]) |
352 | if m: | |
353 | encoding = m.group(1).decode('ascii') | |
b60016e8 PH |
354 | elif webpage_bytes.startswith(b'\xff\xfe'): |
355 | encoding = 'utf-16' | |
f143d86a PH |
356 | else: |
357 | encoding = 'utf-8' | |
d6983cb4 PH |
358 | if self._downloader.params.get('dump_intermediate_pages', False): |
359 | try: | |
360 | url = url_or_request.get_full_url() | |
361 | except AttributeError: | |
362 | url = url_or_request | |
f1a9d64e | 363 | self.to_screen('Dumping request to ' + url) |
d6983cb4 PH |
364 | dump = base64.b64encode(webpage_bytes).decode('ascii') |
365 | self._downloader.to_screen(dump) | |
d41e6efc PH |
366 | if self._downloader.params.get('write_pages', False): |
367 | try: | |
368 | url = url_or_request.get_full_url() | |
369 | except AttributeError: | |
370 | url = url_or_request | |
5afa7f8b | 371 | basen = '%s_%s' % (video_id, url) |
c1bce22f | 372 | if len(basen) > 240: |
f1a9d64e | 373 | h = '___' + hashlib.md5(basen.encode('utf-8')).hexdigest() |
c1bce22f PH |
374 | basen = basen[:240 - len(h)] + h |
375 | raw_filename = basen + '.dump' | |
d41e6efc | 376 | filename = sanitize_filename(raw_filename, restricted=True) |
f1a9d64e | 377 | self.to_screen('Saving request to ' + filename) |
5f58165d S |
378 | # Working around MAX_PATH limitation on Windows (see |
379 | # http://msdn.microsoft.com/en-us/library/windows/desktop/aa365247(v=vs.85).aspx) | |
380 | if os.name == 'nt': | |
381 | absfilepath = os.path.abspath(filename) | |
382 | if len(absfilepath) > 259: | |
383 | filename = '\\\\?\\' + absfilepath | |
d41e6efc PH |
384 | with open(filename, 'wb') as outf: |
385 | outf.write(webpage_bytes) | |
386 | ||
ec0fafbb AA |
387 | try: |
388 | content = webpage_bytes.decode(encoding, 'replace') | |
389 | except LookupError: | |
390 | content = webpage_bytes.decode('utf-8', 'replace') | |
2410c43d | 391 | |
f1a9d64e PH |
392 | if ('<title>Access to this site is blocked</title>' in content and |
393 | 'Websense' in content[:512]): | |
394 | msg = 'Access to this webpage has been blocked by Websense filtering software in your network.' | |
2410c43d PH |
395 | blocked_iframe = self._html_search_regex( |
396 | r'<iframe src="([^"]+)"', content, | |
f1a9d64e | 397 | 'Websense information URL', default=None) |
2410c43d | 398 | if blocked_iframe: |
f1a9d64e | 399 | msg += ' Visit %s for more details' % blocked_iframe |
2410c43d PH |
400 | raise ExtractorError(msg, expected=True) |
401 | ||
23be51d8 | 402 | return content |
d6983cb4 | 403 | |
995ad69c | 404 | def _download_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True, tries=1, timeout=5): |
d6983cb4 | 405 | """ Returns the data of the page as a string """ |
995ad69c TF |
406 | success = False |
407 | try_count = 0 | |
408 | while success is False: | |
409 | try: | |
410 | res = self._download_webpage_handle(url_or_request, video_id, note, errnote, fatal) | |
411 | success = True | |
412 | except compat_http_client.IncompleteRead as e: | |
413 | try_count += 1 | |
414 | if try_count >= tries: | |
415 | raise e | |
416 | self._sleep(timeout, video_id) | |
7cc3570e PH |
417 | if res is False: |
418 | return res | |
419 | else: | |
420 | content, _ = res | |
421 | return content | |
d6983cb4 | 422 | |
2a275ab0 | 423 | def _download_xml(self, url_or_request, video_id, |
f1a9d64e | 424 | note='Downloading XML', errnote='Unable to download XML', |
28746fbd | 425 | transform_source=None, fatal=True): |
267ed0c5 | 426 | """Return the xml as an xml.etree.ElementTree.Element""" |
28746fbd PH |
427 | xml_string = self._download_webpage( |
428 | url_or_request, video_id, note, errnote, fatal=fatal) | |
429 | if xml_string is False: | |
430 | return xml_string | |
e2b38da9 PH |
431 | if transform_source: |
432 | xml_string = transform_source(xml_string) | |
267ed0c5 JMF |
433 | return xml.etree.ElementTree.fromstring(xml_string.encode('utf-8')) |
434 | ||
3d3538e4 | 435 | def _download_json(self, url_or_request, video_id, |
f1a9d64e PH |
436 | note='Downloading JSON metadata', |
437 | errnote='Unable to download JSON metadata', | |
b090af59 PH |
438 | transform_source=None, |
439 | fatal=True): | |
440 | json_string = self._download_webpage( | |
441 | url_or_request, video_id, note, errnote, fatal=fatal) | |
442 | if (not fatal) and json_string is False: | |
443 | return None | |
ebb64199 TF |
444 | return self._parse_json( |
445 | json_string, video_id, transform_source=transform_source, fatal=fatal) | |
446 | ||
447 | def _parse_json(self, json_string, video_id, transform_source=None, fatal=True): | |
81c2f20b PH |
448 | if transform_source: |
449 | json_string = transform_source(json_string) | |
3d3538e4 PH |
450 | try: |
451 | return json.loads(json_string) | |
452 | except ValueError as ve: | |
e7b6d122 PH |
453 | errmsg = '%s: Failed to parse JSON ' % video_id |
454 | if fatal: | |
455 | raise ExtractorError(errmsg, cause=ve) | |
456 | else: | |
457 | self.report_warning(errmsg + str(ve)) | |
3d3538e4 | 458 | |
f45f96f8 | 459 | def report_warning(self, msg, video_id=None): |
f1a9d64e | 460 | idstr = '' if video_id is None else '%s: ' % video_id |
f45f96f8 | 461 | self._downloader.report_warning( |
f1a9d64e | 462 | '[%s] %s%s' % (self.IE_NAME, idstr, msg)) |
f45f96f8 | 463 | |
d6983cb4 PH |
464 | def to_screen(self, msg): |
465 | """Print msg to screen, prefixing it with '[ie_name]'""" | |
f1a9d64e | 466 | self._downloader.to_screen('[%s] %s' % (self.IE_NAME, msg)) |
d6983cb4 PH |
467 | |
468 | def report_extraction(self, id_or_name): | |
469 | """Report information extraction.""" | |
f1a9d64e | 470 | self.to_screen('%s: Extracting information' % id_or_name) |
d6983cb4 PH |
471 | |
472 | def report_download_webpage(self, video_id): | |
473 | """Report webpage download.""" | |
f1a9d64e | 474 | self.to_screen('%s: Downloading webpage' % video_id) |
d6983cb4 PH |
475 | |
476 | def report_age_confirmation(self): | |
477 | """Report attempt to confirm age.""" | |
f1a9d64e | 478 | self.to_screen('Confirming age') |
d6983cb4 | 479 | |
fc79158d JMF |
480 | def report_login(self): |
481 | """Report attempt to log in.""" | |
f1a9d64e | 482 | self.to_screen('Logging in') |
fc79158d | 483 | |
5f6a1245 | 484 | # Methods for following #608 |
c0d0b01f JMF |
485 | @staticmethod |
486 | def url_result(url, ie=None, video_id=None): | |
d6983cb4 | 487 | """Returns a url that points to a page that should be processed""" |
5f6a1245 | 488 | # TODO: ie should be the class used for getting the info |
d6983cb4 PH |
489 | video_info = {'_type': 'url', |
490 | 'url': url, | |
491 | 'ie_key': ie} | |
7012b23c PH |
492 | if video_id is not None: |
493 | video_info['id'] = video_id | |
d6983cb4 | 494 | return video_info |
5f6a1245 | 495 | |
c0d0b01f | 496 | @staticmethod |
acf5cbfe | 497 | def playlist_result(entries, playlist_id=None, playlist_title=None, playlist_description=None): |
d6983cb4 PH |
498 | """Returns a playlist""" |
499 | video_info = {'_type': 'playlist', | |
500 | 'entries': entries} | |
501 | if playlist_id: | |
502 | video_info['id'] = playlist_id | |
503 | if playlist_title: | |
504 | video_info['title'] = playlist_title | |
acf5cbfe S |
505 | if playlist_description: |
506 | video_info['description'] = playlist_description | |
d6983cb4 PH |
507 | return video_info |
508 | ||
711ede6e | 509 | def _search_regex(self, pattern, string, name, default=_NO_DEFAULT, fatal=True, flags=0, group=None): |
d6983cb4 PH |
510 | """ |
511 | Perform a regex search on the given string, using a single or a list of | |
512 | patterns returning the first matching group. | |
513 | In case of failure return a default value or raise a WARNING or a | |
55b3e45b | 514 | RegexNotFoundError, depending on fatal, specifying the field name. |
d6983cb4 PH |
515 | """ |
516 | if isinstance(pattern, (str, compat_str, compiled_regex_type)): | |
517 | mobj = re.search(pattern, string, flags) | |
518 | else: | |
519 | for p in pattern: | |
520 | mobj = re.search(p, string, flags) | |
c3415d1b PH |
521 | if mobj: |
522 | break | |
d6983cb4 | 523 | |
7e5db8c9 | 524 | if not self._downloader.params.get('no_color') and os.name != 'nt' and sys.stderr.isatty(): |
f1a9d64e | 525 | _name = '\033[0;34m%s\033[0m' % name |
d6983cb4 PH |
526 | else: |
527 | _name = name | |
528 | ||
529 | if mobj: | |
711ede6e PH |
530 | if group is None: |
531 | # return the first matching group | |
532 | return next(g for g in mobj.groups() if g is not None) | |
533 | else: | |
534 | return mobj.group(group) | |
46374a56 | 535 | elif default is not _NO_DEFAULT: |
d6983cb4 PH |
536 | return default |
537 | elif fatal: | |
f1a9d64e | 538 | raise RegexNotFoundError('Unable to extract %s' % _name) |
d6983cb4 | 539 | else: |
f1a9d64e | 540 | self._downloader.report_warning('unable to extract %s; ' |
9e1a5b84 | 541 | 'please report this issue on http://yt-dl.org/bug' % _name) |
d6983cb4 PH |
542 | return None |
543 | ||
711ede6e | 544 | def _html_search_regex(self, pattern, string, name, default=_NO_DEFAULT, fatal=True, flags=0, group=None): |
d6983cb4 PH |
545 | """ |
546 | Like _search_regex, but strips HTML tags and unescapes entities. | |
547 | """ | |
711ede6e | 548 | res = self._search_regex(pattern, string, name, default, fatal, flags, group) |
d6983cb4 PH |
549 | if res: |
550 | return clean_html(res).strip() | |
551 | else: | |
552 | return res | |
553 | ||
fc79158d JMF |
554 | def _get_login_info(self): |
555 | """ | |
556 | Get the the login info as (username, password) | |
557 | It will look in the netrc file using the _NETRC_MACHINE value | |
558 | If there's no info available, return (None, None) | |
559 | """ | |
560 | if self._downloader is None: | |
561 | return (None, None) | |
562 | ||
563 | username = None | |
564 | password = None | |
565 | downloader_params = self._downloader.params | |
566 | ||
567 | # Attempt to use provided username and password or .netrc data | |
568 | if downloader_params.get('username', None) is not None: | |
569 | username = downloader_params['username'] | |
570 | password = downloader_params['password'] | |
571 | elif downloader_params.get('usenetrc', False): | |
572 | try: | |
573 | info = netrc.netrc().authenticators(self._NETRC_MACHINE) | |
574 | if info is not None: | |
575 | username = info[0] | |
576 | password = info[2] | |
577 | else: | |
578 | raise netrc.NetrcParseError('No authenticators for %s' % self._NETRC_MACHINE) | |
579 | except (IOError, netrc.NetrcParseError) as err: | |
f1a9d64e | 580 | self._downloader.report_warning('parsing .netrc: %s' % compat_str(err)) |
5f6a1245 | 581 | |
fc79158d JMF |
582 | return (username, password) |
583 | ||
83317f69 | 584 | def _get_tfa_info(self): |
585 | """ | |
586 | Get the two-factor authentication info | |
587 | TODO - asking the user will be required for sms/phone verify | |
588 | currently just uses the command line option | |
589 | If there's no info available, return None | |
590 | """ | |
591 | if self._downloader is None: | |
83317f69 | 592 | return None |
593 | downloader_params = self._downloader.params | |
594 | ||
595 | if downloader_params.get('twofactor', None) is not None: | |
596 | return downloader_params['twofactor'] | |
597 | ||
83317f69 | 598 | return None |
599 | ||
46720279 JMF |
600 | # Helper functions for extracting OpenGraph info |
601 | @staticmethod | |
ab2d5247 | 602 | def _og_regexes(prop): |
c1206423 | 603 | content_re = r'content=(?:"([^>]+?)"|\'([^>]+?)\')' |
9887c9b2 | 604 | property_re = r'(?:name|property)=[\'"]og:%s[\'"]' % re.escape(prop) |
78fb87b2 | 605 | template = r'<meta[^>]+?%s[^>]+?%s' |
ab2d5247 | 606 | return [ |
78fb87b2 JMF |
607 | template % (property_re, content_re), |
608 | template % (content_re, property_re), | |
ab2d5247 | 609 | ] |
46720279 | 610 | |
3c4e6d83 | 611 | def _og_search_property(self, prop, html, name=None, **kargs): |
46720279 | 612 | if name is None: |
3c4e6d83 | 613 | name = 'OpenGraph %s' % prop |
ab2d5247 | 614 | escaped = self._search_regex(self._og_regexes(prop), html, name, flags=re.DOTALL, **kargs) |
eb0a8398 PH |
615 | if escaped is None: |
616 | return None | |
617 | return unescapeHTML(escaped) | |
46720279 JMF |
618 | |
619 | def _og_search_thumbnail(self, html, **kargs): | |
f1a9d64e | 620 | return self._og_search_property('image', html, 'thumbnail url', fatal=False, **kargs) |
46720279 JMF |
621 | |
622 | def _og_search_description(self, html, **kargs): | |
623 | return self._og_search_property('description', html, fatal=False, **kargs) | |
624 | ||
625 | def _og_search_title(self, html, **kargs): | |
626 | return self._og_search_property('title', html, **kargs) | |
627 | ||
8ffa13e0 | 628 | def _og_search_video_url(self, html, name='video url', secure=True, **kargs): |
a3681973 PH |
629 | regexes = self._og_regexes('video') + self._og_regexes('video:url') |
630 | if secure: | |
631 | regexes = self._og_regexes('video:secure_url') + regexes | |
8ffa13e0 | 632 | return self._html_search_regex(regexes, html, name, **kargs) |
46720279 | 633 | |
78338f71 JMF |
634 | def _og_search_url(self, html, **kargs): |
635 | return self._og_search_property('url', html, **kargs) | |
636 | ||
40c696e5 | 637 | def _html_search_meta(self, name, html, display_name=None, fatal=False, **kwargs): |
59040888 PH |
638 | if display_name is None: |
639 | display_name = name | |
640 | return self._html_search_regex( | |
6c6f1408 | 641 | r'''(?isx)<meta |
711ede6e | 642 | (?=[^>]+(?:itemprop|name|property)=(["\']?)%s\1) |
bec22481 | 643 | [^>]+?content=(["\'])(?P<content>.*?)\2''' % re.escape(name), |
711ede6e | 644 | html, display_name, fatal=fatal, group='content', **kwargs) |
59040888 PH |
645 | |
646 | def _dc_search_uploader(self, html): | |
647 | return self._html_search_meta('dc.creator', html, 'uploader') | |
648 | ||
8dbe9899 PH |
649 | def _rta_search(self, html): |
650 | # See http://www.rtalabel.org/index.php?content=howtofaq#single | |
651 | if re.search(r'(?ix)<meta\s+name="rating"\s+' | |
652 | r' content="RTA-5042-1996-1400-1577-RTA"', | |
653 | html): | |
654 | return 18 | |
655 | return 0 | |
656 | ||
59040888 PH |
657 | def _media_rating_search(self, html): |
658 | # See http://www.tjg-designs.com/WP/metadata-code-examples-adding-metadata-to-your-web-pages/ | |
659 | rating = self._html_search_meta('rating', html) | |
660 | ||
661 | if not rating: | |
662 | return None | |
663 | ||
664 | RATING_TABLE = { | |
665 | 'safe for kids': 0, | |
666 | 'general': 8, | |
667 | '14 years': 14, | |
668 | 'mature': 17, | |
669 | 'restricted': 19, | |
670 | } | |
671 | return RATING_TABLE.get(rating.lower(), None) | |
672 | ||
69319969 | 673 | def _family_friendly_search(self, html): |
6ca7732d | 674 | # See http://schema.org/VideoObject |
69319969 NJ |
675 | family_friendly = self._html_search_meta('isFamilyFriendly', html) |
676 | ||
677 | if not family_friendly: | |
678 | return None | |
679 | ||
680 | RATING_TABLE = { | |
681 | '1': 0, | |
682 | 'true': 0, | |
683 | '0': 18, | |
684 | 'false': 18, | |
685 | } | |
686 | return RATING_TABLE.get(family_friendly.lower(), None) | |
687 | ||
0c708f11 JMF |
688 | def _twitter_search_player(self, html): |
689 | return self._html_search_meta('twitter:player', html, | |
9e1a5b84 | 690 | 'twitter card player') |
0c708f11 | 691 | |
4bcc7bd1 | 692 | def _sort_formats(self, formats): |
7e8caf30 | 693 | if not formats: |
f1a9d64e | 694 | raise ExtractorError('No video formats found') |
7e8caf30 | 695 | |
4bcc7bd1 | 696 | def _formats_key(f): |
e6812ac9 PH |
697 | # TODO remove the following workaround |
698 | from ..utils import determine_ext | |
699 | if not f.get('ext') and 'url' in f: | |
700 | f['ext'] = determine_ext(f['url']) | |
701 | ||
4bcc7bd1 PH |
702 | preference = f.get('preference') |
703 | if preference is None: | |
c7deaa4c PH |
704 | proto = f.get('protocol') |
705 | if proto is None: | |
706 | proto = compat_urllib_parse_urlparse(f.get('url', '')).scheme | |
707 | ||
708 | preference = 0 if proto in ['http', 'https'] else -0.1 | |
4bcc7bd1 PH |
709 | if f.get('ext') in ['f4f', 'f4m']: # Not yet supported |
710 | preference -= 0.5 | |
711 | ||
712 | if f.get('vcodec') == 'none': # audio only | |
713 | if self._downloader.params.get('prefer_free_formats'): | |
f1a9d64e | 714 | ORDER = ['aac', 'mp3', 'm4a', 'webm', 'ogg', 'opus'] |
4bcc7bd1 | 715 | else: |
f1a9d64e | 716 | ORDER = ['webm', 'opus', 'ogg', 'mp3', 'aac', 'm4a'] |
4bcc7bd1 PH |
717 | ext_preference = 0 |
718 | try: | |
719 | audio_ext_preference = ORDER.index(f['ext']) | |
720 | except ValueError: | |
721 | audio_ext_preference = -1 | |
722 | else: | |
723 | if self._downloader.params.get('prefer_free_formats'): | |
f1a9d64e | 724 | ORDER = ['flv', 'mp4', 'webm'] |
4bcc7bd1 | 725 | else: |
f1a9d64e | 726 | ORDER = ['webm', 'flv', 'mp4'] |
4bcc7bd1 PH |
727 | try: |
728 | ext_preference = ORDER.index(f['ext']) | |
729 | except ValueError: | |
730 | ext_preference = -1 | |
731 | audio_ext_preference = 0 | |
732 | ||
733 | return ( | |
734 | preference, | |
aff2f4f4 | 735 | f.get('language_preference') if f.get('language_preference') is not None else -1, |
5d73273f | 736 | f.get('quality') if f.get('quality') is not None else -1, |
9933b574 | 737 | f.get('tbr') if f.get('tbr') is not None else -1, |
4bcc7bd1 | 738 | f.get('vbr') if f.get('vbr') is not None else -1, |
1a6373ef PH |
739 | f.get('height') if f.get('height') is not None else -1, |
740 | f.get('width') if f.get('width') is not None else -1, | |
1e1896f2 | 741 | ext_preference, |
4bcc7bd1 PH |
742 | f.get('abr') if f.get('abr') is not None else -1, |
743 | audio_ext_preference, | |
2c8e03d9 | 744 | f.get('fps') if f.get('fps') is not None else -1, |
4bcc7bd1 | 745 | f.get('filesize') if f.get('filesize') is not None else -1, |
9732d77e | 746 | f.get('filesize_approx') if f.get('filesize_approx') is not None else -1, |
c64ed2a3 | 747 | f.get('source_preference') if f.get('source_preference') is not None else -1, |
4bcc7bd1 PH |
748 | f.get('format_id'), |
749 | ) | |
750 | formats.sort(key=_formats_key) | |
59040888 | 751 | |
96a53167 S |
752 | def _check_formats(self, formats, video_id): |
753 | if formats: | |
754 | formats[:] = filter( | |
755 | lambda f: self._is_valid_url( | |
756 | f['url'], video_id, | |
757 | item='%s video format' % f.get('format_id') if f.get('format_id') else 'video'), | |
758 | formats) | |
759 | ||
760 | def _is_valid_url(self, url, video_id, item='video'): | |
761 | try: | |
762 | self._request_webpage( | |
763 | HEADRequest(url), video_id, | |
764 | 'Checking %s URL' % item) | |
765 | return True | |
766 | except ExtractorError as e: | |
767 | if isinstance(e.cause, compat_HTTPError): | |
768 | self.report_warning( | |
769 | '%s URL is invalid, skipping' % item, video_id) | |
770 | return False | |
771 | raise | |
772 | ||
20991253 | 773 | def http_scheme(self): |
1ede5b24 | 774 | """ Either "http:" or "https:", depending on the user's preferences """ |
20991253 PH |
775 | return ( |
776 | 'http:' | |
777 | if self._downloader.params.get('prefer_insecure', False) | |
778 | else 'https:') | |
779 | ||
57c7411f PH |
780 | def _proto_relative_url(self, url, scheme=None): |
781 | if url is None: | |
782 | return url | |
783 | if url.startswith('//'): | |
784 | if scheme is None: | |
785 | scheme = self.http_scheme() | |
786 | return scheme + url | |
787 | else: | |
788 | return url | |
789 | ||
4094b6e3 PH |
790 | def _sleep(self, timeout, video_id, msg_template=None): |
791 | if msg_template is None: | |
f1a9d64e | 792 | msg_template = '%(video_id)s: Waiting for %(timeout)s seconds' |
4094b6e3 PH |
793 | msg = msg_template % {'video_id': video_id, 'timeout': timeout} |
794 | self.to_screen(msg) | |
795 | time.sleep(timeout) | |
796 | ||
60ca389c | 797 | def _extract_f4m_formats(self, manifest_url, video_id, preference=None, f4m_id=None): |
f036a632 JMF |
798 | manifest = self._download_xml( |
799 | manifest_url, video_id, 'Downloading f4m manifest', | |
800 | 'Unable to download f4m manifest') | |
31bb8d3f JMF |
801 | |
802 | formats = [] | |
7a47d07c | 803 | manifest_version = '1.0' |
b2527359 | 804 | media_nodes = manifest.findall('{http://ns.adobe.com/f4m/1.0}media') |
34e48bed | 805 | if not media_nodes: |
7a47d07c | 806 | manifest_version = '2.0' |
34e48bed | 807 | media_nodes = manifest.findall('{http://ns.adobe.com/f4m/2.0}media') |
b2527359 | 808 | for i, media_el in enumerate(media_nodes): |
7a47d07c | 809 | if manifest_version == '2.0': |
3900eec2 S |
810 | manifest_url = ('/'.join(manifest_url.split('/')[:-1]) + '/' |
811 | + (media_el.attrib.get('href') or media_el.attrib.get('url'))) | |
b2527359 | 812 | tbr = int_or_none(media_el.attrib.get('bitrate')) |
31bb8d3f | 813 | formats.append({ |
60ca389c | 814 | 'format_id': '-'.join(filter(None, [f4m_id, 'f4m-%d' % (i if tbr is None else tbr)])), |
31bb8d3f JMF |
815 | 'url': manifest_url, |
816 | 'ext': 'flv', | |
b2527359 | 817 | 'tbr': tbr, |
31bb8d3f JMF |
818 | 'width': int_or_none(media_el.attrib.get('width')), |
819 | 'height': int_or_none(media_el.attrib.get('height')), | |
60ca389c | 820 | 'preference': preference, |
31bb8d3f JMF |
821 | }) |
822 | self._sort_formats(formats) | |
823 | ||
824 | return formats | |
825 | ||
f0b5d6af | 826 | def _extract_m3u8_formats(self, m3u8_url, video_id, ext=None, |
60ca389c S |
827 | entry_protocol='m3u8', preference=None, |
828 | m3u8_id=None): | |
f0b5d6af | 829 | |
704df56d | 830 | formats = [{ |
60ca389c | 831 | 'format_id': '-'.join(filter(None, [m3u8_id, 'm3u8-meta'])), |
704df56d PH |
832 | 'url': m3u8_url, |
833 | 'ext': ext, | |
834 | 'protocol': 'm3u8', | |
835 | 'preference': -1, | |
836 | 'resolution': 'multiple', | |
837 | 'format_note': 'Quality selection URL', | |
838 | }] | |
839 | ||
f0b5d6af PH |
840 | format_url = lambda u: ( |
841 | u | |
842 | if re.match(r'^https?://', u) | |
843 | else compat_urlparse.urljoin(m3u8_url, u)) | |
844 | ||
81515ad9 PH |
845 | m3u8_doc = self._download_webpage( |
846 | m3u8_url, video_id, | |
847 | note='Downloading m3u8 information', | |
848 | errnote='Failed to download m3u8 information') | |
704df56d PH |
849 | last_info = None |
850 | kv_rex = re.compile( | |
851 | r'(?P<key>[a-zA-Z_-]+)=(?P<val>"[^"]+"|[^",]+)(?:,|$)') | |
852 | for line in m3u8_doc.splitlines(): | |
853 | if line.startswith('#EXT-X-STREAM-INF:'): | |
854 | last_info = {} | |
855 | for m in kv_rex.finditer(line): | |
856 | v = m.group('val') | |
857 | if v.startswith('"'): | |
858 | v = v[1:-1] | |
859 | last_info[m.group('key')] = v | |
860 | elif line.startswith('#') or not line.strip(): | |
861 | continue | |
862 | else: | |
daebaab6 | 863 | if last_info is None: |
f0b5d6af | 864 | formats.append({'url': format_url(line)}) |
3524cc25 | 865 | continue |
704df56d | 866 | tbr = int_or_none(last_info.get('BANDWIDTH'), scale=1000) |
704df56d | 867 | f = { |
60ca389c | 868 | 'format_id': '-'.join(filter(None, [m3u8_id, 'm3u8-%d' % (tbr if tbr else len(formats))])), |
f0b5d6af | 869 | 'url': format_url(line.strip()), |
704df56d PH |
870 | 'tbr': tbr, |
871 | 'ext': ext, | |
f0b5d6af PH |
872 | 'protocol': entry_protocol, |
873 | 'preference': preference, | |
704df56d PH |
874 | } |
875 | codecs = last_info.get('CODECS') | |
876 | if codecs: | |
9ebf22b7 S |
877 | # TODO: looks like video codec is not always necessarily goes first |
878 | va_codecs = codecs.split(',') | |
879 | if va_codecs[0]: | |
880 | f['vcodec'] = va_codecs[0].partition('.')[0] | |
881 | if len(va_codecs) > 1 and va_codecs[1]: | |
882 | f['acodec'] = va_codecs[1].partition('.')[0] | |
704df56d PH |
883 | resolution = last_info.get('RESOLUTION') |
884 | if resolution: | |
885 | width_str, height_str = resolution.split('x') | |
886 | f['width'] = int(width_str) | |
887 | f['height'] = int(height_str) | |
888 | formats.append(f) | |
889 | last_info = {} | |
890 | self._sort_formats(formats) | |
891 | return formats | |
892 | ||
e89a2aab | 893 | # TODO: improve extraction |
995029a1 | 894 | def _extract_smil_formats(self, smil_url, video_id, fatal=True): |
e89a2aab S |
895 | smil = self._download_xml( |
896 | smil_url, video_id, 'Downloading SMIL file', | |
995029a1 PH |
897 | 'Unable to download SMIL file', fatal=fatal) |
898 | if smil is False: | |
899 | assert not fatal | |
900 | return [] | |
e89a2aab S |
901 | |
902 | base = smil.find('./head/meta').get('base') | |
903 | ||
904 | formats = [] | |
905 | rtmp_count = 0 | |
906 | for video in smil.findall('./body/switch/video'): | |
907 | src = video.get('src') | |
908 | if not src: | |
909 | continue | |
910 | bitrate = int_or_none(video.get('system-bitrate') or video.get('systemBitrate'), 1000) | |
911 | width = int_or_none(video.get('width')) | |
912 | height = int_or_none(video.get('height')) | |
913 | proto = video.get('proto') | |
914 | if not proto: | |
915 | if base: | |
916 | if base.startswith('rtmp'): | |
917 | proto = 'rtmp' | |
918 | elif base.startswith('http'): | |
919 | proto = 'http' | |
920 | ext = video.get('ext') | |
921 | if proto == 'm3u8': | |
922 | formats.extend(self._extract_m3u8_formats(src, video_id, ext)) | |
923 | elif proto == 'rtmp': | |
924 | rtmp_count += 1 | |
925 | streamer = video.get('streamer') or base | |
926 | formats.append({ | |
927 | 'url': streamer, | |
928 | 'play_path': src, | |
929 | 'ext': 'flv', | |
930 | 'format_id': 'rtmp-%d' % (rtmp_count if bitrate is None else bitrate), | |
931 | 'tbr': bitrate, | |
932 | 'width': width, | |
933 | 'height': height, | |
934 | }) | |
935 | self._sort_formats(formats) | |
936 | ||
937 | return formats | |
938 | ||
f4b1c7ad PH |
939 | def _live_title(self, name): |
940 | """ Generate the title for a live video """ | |
941 | now = datetime.datetime.now() | |
942 | now_str = now.strftime("%Y-%m-%d %H:%M") | |
943 | return name + ' ' + now_str | |
944 | ||
b14f3a4c PH |
945 | def _int(self, v, name, fatal=False, **kwargs): |
946 | res = int_or_none(v, **kwargs) | |
947 | if 'get_attr' in kwargs: | |
948 | print(getattr(v, kwargs['get_attr'])) | |
949 | if res is None: | |
950 | msg = 'Failed to extract %s: Could not parse value %r' % (name, v) | |
951 | if fatal: | |
952 | raise ExtractorError(msg) | |
953 | else: | |
954 | self._downloader.report_warning(msg) | |
955 | return res | |
956 | ||
957 | def _float(self, v, name, fatal=False, **kwargs): | |
958 | res = float_or_none(v, **kwargs) | |
959 | if res is None: | |
960 | msg = 'Failed to extract %s: Could not parse value %r' % (name, v) | |
961 | if fatal: | |
962 | raise ExtractorError(msg) | |
963 | else: | |
964 | self._downloader.report_warning(msg) | |
965 | return res | |
966 | ||
42939b61 | 967 | def _set_cookie(self, domain, name, value, expire_time=None): |
810fb84d PH |
968 | cookie = compat_cookiejar.Cookie( |
969 | 0, name, value, None, None, domain, None, | |
42939b61 JMF |
970 | None, '/', True, False, expire_time, '', None, None, None) |
971 | self._downloader.cookiejar.set_cookie(cookie) | |
972 | ||
05900629 PH |
973 | def get_testcases(self, include_onlymatching=False): |
974 | t = getattr(self, '_TEST', None) | |
975 | if t: | |
976 | assert not hasattr(self, '_TESTS'), \ | |
977 | '%s has _TEST and _TESTS' % type(self).__name__ | |
978 | tests = [t] | |
979 | else: | |
980 | tests = getattr(self, '_TESTS', []) | |
981 | for t in tests: | |
982 | if not include_onlymatching and t.get('only_matching', False): | |
983 | continue | |
984 | t['name'] = type(self).__name__[:-len('IE')] | |
985 | yield t | |
986 | ||
987 | def is_suitable(self, age_limit): | |
988 | """ Test whether the extractor is generally suitable for the given | |
989 | age limit (i.e. pornographic sites are not, all others usually are) """ | |
990 | ||
991 | any_restricted = False | |
992 | for tc in self.get_testcases(include_onlymatching=False): | |
993 | if 'playlist' in tc: | |
994 | tc = tc['playlist'][0] | |
995 | is_restricted = age_restricted( | |
996 | tc.get('info_dict', {}).get('age_limit'), age_limit) | |
997 | if not is_restricted: | |
998 | return True | |
999 | any_restricted = any_restricted or is_restricted | |
1000 | return not any_restricted | |
1001 | ||
a504ced0 JMF |
1002 | def extract_subtitles(self, *args, **kwargs): |
1003 | subtitles = {} | |
1004 | list_subtitles = self._downloader.params.get('listsubtitles') | |
1005 | if self._downloader.params.get('writesubtitles', False) or list_subtitles: | |
1006 | subtitles.update(self._get_subtitles(*args, **kwargs)) | |
1007 | return subtitles | |
1008 | ||
1009 | def _get_subtitles(self, *args, **kwargs): | |
1010 | raise NotImplementedError("This method must be implemented by subclasses") | |
1011 | ||
360e1ca5 JMF |
1012 | def extract_automatic_captions(self, *args, **kwargs): |
1013 | automatic_captions = {} | |
1014 | list_subtitles = self._downloader.params.get('listsubtitles') | |
1015 | if self._downloader.params.get('writeautomaticsub', False) or list_subtitles: | |
1016 | automatic_captions.update(self._get_automatic_captions(*args, **kwargs)) | |
1017 | return automatic_captions | |
1018 | ||
1019 | def _get_automatic_captions(self, *args, **kwargs): | |
1020 | raise NotImplementedError("This method must be implemented by subclasses") | |
1021 | ||
8dbe9899 | 1022 | |
d6983cb4 PH |
1023 | class SearchInfoExtractor(InfoExtractor): |
1024 | """ | |
1025 | Base class for paged search queries extractors. | |
1026 | They accept urls in the format _SEARCH_KEY(|all|[0-9]):{query} | |
1027 | Instances should define _SEARCH_KEY and _MAX_RESULTS. | |
1028 | """ | |
1029 | ||
1030 | @classmethod | |
1031 | def _make_valid_url(cls): | |
1032 | return r'%s(?P<prefix>|[1-9][0-9]*|all):(?P<query>[\s\S]+)' % cls._SEARCH_KEY | |
1033 | ||
1034 | @classmethod | |
1035 | def suitable(cls, url): | |
1036 | return re.match(cls._make_valid_url(), url) is not None | |
1037 | ||
1038 | def _real_extract(self, query): | |
1039 | mobj = re.match(self._make_valid_url(), query) | |
1040 | if mobj is None: | |
f1a9d64e | 1041 | raise ExtractorError('Invalid search query "%s"' % query) |
d6983cb4 PH |
1042 | |
1043 | prefix = mobj.group('prefix') | |
1044 | query = mobj.group('query') | |
1045 | if prefix == '': | |
1046 | return self._get_n_results(query, 1) | |
1047 | elif prefix == 'all': | |
1048 | return self._get_n_results(query, self._MAX_RESULTS) | |
1049 | else: | |
1050 | n = int(prefix) | |
1051 | if n <= 0: | |
f1a9d64e | 1052 | raise ExtractorError('invalid download number %s for query "%s"' % (n, query)) |
d6983cb4 | 1053 | elif n > self._MAX_RESULTS: |
f1a9d64e | 1054 | self._downloader.report_warning('%s returns max %i results (you requested %i)' % (self._SEARCH_KEY, self._MAX_RESULTS, n)) |
d6983cb4 PH |
1055 | n = self._MAX_RESULTS |
1056 | return self._get_n_results(query, n) | |
1057 | ||
1058 | def _get_n_results(self, query, n): | |
1059 | """Get a specified number of results for a query""" | |
416a5efc | 1060 | raise NotImplementedError("This method must be implemented by subclasses") |
0f818663 PH |
1061 | |
1062 | @property | |
1063 | def SEARCH_KEY(self): | |
1064 | return self._SEARCH_KEY |