]>
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 |
d6983cb4 | 13 | |
8c25f81b | 14 | from ..compat import ( |
42939b61 | 15 | compat_cookiejar, |
799207e8 | 16 | compat_cookies, |
e64b7569 | 17 | compat_getpass, |
d6983cb4 PH |
18 | compat_http_client, |
19 | compat_urllib_error, | |
a107193e | 20 | compat_urllib_parse, |
f0b5d6af | 21 | compat_urlparse, |
d6983cb4 | 22 | compat_str, |
36e6f62c | 23 | compat_etree_fromstring, |
8c25f81b PH |
24 | ) |
25 | from ..utils import ( | |
c342041f | 26 | NO_DEFAULT, |
05900629 | 27 | age_restricted, |
08f2a92c | 28 | bug_reports_message, |
d6983cb4 PH |
29 | clean_html, |
30 | compiled_regex_type, | |
70f0f5a8 | 31 | determine_ext, |
9b9c5355 | 32 | error_to_compat_str, |
d6983cb4 | 33 | ExtractorError, |
97f4aecf | 34 | fix_xml_ampersands, |
b14f3a4c | 35 | float_or_none, |
31bb8d3f | 36 | int_or_none, |
55b3e45b | 37 | RegexNotFoundError, |
d41e6efc | 38 | sanitize_filename, |
5c2266df | 39 | sanitized_Request, |
f38de77f | 40 | unescapeHTML, |
647eab45 | 41 | unified_strdate, |
a107193e | 42 | url_basename, |
8d6765cf S |
43 | xpath_text, |
44 | xpath_with_ns, | |
d497a201 | 45 | determine_protocol, |
d6983cb4 | 46 | ) |
c342041f | 47 | |
d6983cb4 PH |
48 | |
49 | class InfoExtractor(object): | |
50 | """Information Extractor class. | |
51 | ||
52 | Information extractors are the classes that, given a URL, extract | |
53 | information about the video (or videos) the URL refers to. This | |
54 | information includes the real video URL, the video title, author and | |
55 | others. The information is stored in a dictionary which is then | |
5d380852 | 56 | passed to the YoutubeDL. The YoutubeDL processes this |
d6983cb4 PH |
57 | information possibly downloading the video to the file system, among |
58 | other possible outcomes. | |
59 | ||
cf0649f8 | 60 | The type field determines the type of the result. |
fed5d032 PH |
61 | By far the most common value (and the default if _type is missing) is |
62 | "video", which indicates a single video. | |
63 | ||
64 | For a video, the dictionaries must include the following fields: | |
d6983cb4 PH |
65 | |
66 | id: Video identifier. | |
d6983cb4 | 67 | title: Video title, unescaped. |
d67b0b15 | 68 | |
f49d89ee | 69 | Additionally, it must contain either a formats entry or a url one: |
d67b0b15 | 70 | |
f49d89ee PH |
71 | formats: A list of dictionaries for each format available, ordered |
72 | from worst to best quality. | |
73 | ||
74 | Potential fields: | |
d67b0b15 | 75 | * url Mandatory. The URL of the video file |
10952eb2 | 76 | * ext Will be calculated from URL if missing |
d67b0b15 PH |
77 | * format A human-readable description of the format |
78 | ("mp4 container with h264/opus"). | |
79 | Calculated from the format_id, width, height. | |
80 | and format_note fields if missing. | |
81 | * format_id A short description of the format | |
5d4f3985 PH |
82 | ("mp4_h264_opus" or "19"). |
83 | Technically optional, but strongly recommended. | |
d67b0b15 PH |
84 | * format_note Additional info about the format |
85 | ("3D" or "DASH video") | |
86 | * width Width of the video, if known | |
87 | * height Height of the video, if known | |
f49d89ee | 88 | * resolution Textual description of width and height |
7217e148 | 89 | * tbr Average bitrate of audio and video in KBit/s |
d67b0b15 PH |
90 | * abr Average audio bitrate in KBit/s |
91 | * acodec Name of the audio codec in use | |
dd27fd17 | 92 | * asr Audio sampling rate in Hertz |
d67b0b15 | 93 | * vbr Average video bitrate in KBit/s |
fbb21cf5 | 94 | * fps Frame rate |
d67b0b15 | 95 | * vcodec Name of the video codec in use |
1394ce65 | 96 | * container Name of the container format |
d67b0b15 | 97 | * filesize The number of bytes, if known in advance |
9732d77e | 98 | * filesize_approx An estimate for the number of bytes |
d67b0b15 | 99 | * player_url SWF Player URL (used for rtmpdump). |
c7deaa4c PH |
100 | * protocol The protocol that will be used for the actual |
101 | download, lower-case. | |
b04b8852 PH |
102 | "http", "https", "rtsp", "rtmp", "rtmpe", |
103 | "m3u8", or "m3u8_native". | |
f49d89ee | 104 | * preference Order number of this format. If this field is |
08d13955 | 105 | present and not None, the formats get sorted |
38d63d84 | 106 | by this field, regardless of all other values. |
f49d89ee PH |
107 | -1 for default (order by other properties), |
108 | -2 or smaller for less than default. | |
e65566a9 PH |
109 | < -1000 to hide the format (if there is |
110 | another one which is strictly better) | |
32f90364 PH |
111 | * language Language code, e.g. "de" or "en-US". |
112 | * language_preference Is this in the language mentioned in | |
113 | the URL? | |
aff2f4f4 PH |
114 | 10 if it's what the URL is about, |
115 | -1 for default (don't know), | |
116 | -10 otherwise, other values reserved for now. | |
5d73273f PH |
117 | * quality Order number of the video quality of this |
118 | format, irrespective of the file format. | |
119 | -1 for default (order by other properties), | |
120 | -2 or smaller for less than default. | |
c64ed2a3 PH |
121 | * source_preference Order number for this video source |
122 | (quality takes higher priority) | |
123 | -1 for default (order by other properties), | |
124 | -2 or smaller for less than default. | |
d769be6c PH |
125 | * http_headers A dictionary of additional HTTP headers |
126 | to add to the request. | |
6271f1ca | 127 | * stretched_ratio If given and not 1, indicates that the |
3dee7826 PH |
128 | video's pixels are not square. |
129 | width : height ratio as float. | |
130 | * no_resume The server does not support resuming the | |
131 | (HTTP or RTMP) download. Boolean. | |
132 | ||
c0ba0f48 | 133 | url: Final video URL. |
d6983cb4 | 134 | ext: Video filename extension. |
d67b0b15 PH |
135 | format: The video format, defaults to ext (used for --get-format) |
136 | player_url: SWF Player URL (used for rtmpdump). | |
2f5865cc | 137 | |
d6983cb4 PH |
138 | The following fields are optional: |
139 | ||
f5e43bc6 | 140 | alt_title: A secondary title of the video. |
0afef30b PH |
141 | display_id An alternative identifier for the video, not necessarily |
142 | unique, but available before title. Typically, id is | |
143 | something like "4234987", title "Dancing naked mole rats", | |
144 | and display_id "dancing-naked-mole-rats" | |
d5519808 | 145 | thumbnails: A list of dictionaries, with the following entries: |
cfb56d1a | 146 | * "id" (optional, string) - Thumbnail format ID |
d5519808 | 147 | * "url" |
cfb56d1a | 148 | * "preference" (optional, int) - quality of the image |
d5519808 PH |
149 | * "width" (optional, int) |
150 | * "height" (optional, int) | |
151 | * "resolution" (optional, string "{width}x{height"}, | |
152 | deprecated) | |
d6983cb4 | 153 | thumbnail: Full URL to a video thumbnail image. |
f5e43bc6 | 154 | description: Full video description. |
d6983cb4 | 155 | uploader: Full name of the video uploader. |
9bb8e0a3 | 156 | creator: The main artist who created the video. |
8aab976b | 157 | release_date: The date (YYYYMMDD) when the video was released. |
955c4514 | 158 | timestamp: UNIX timestamp of the moment the video became available. |
d6983cb4 | 159 | upload_date: Video upload date (YYYYMMDD). |
955c4514 | 160 | If not explicitly set, calculated from timestamp. |
d6983cb4 | 161 | uploader_id: Nickname or id of the video uploader. |
da9ec3b9 | 162 | location: Physical location where the video was filmed. |
a504ced0 JMF |
163 | subtitles: The available subtitles as a dictionary in the format |
164 | {language: subformats}. "subformats" is a list sorted from | |
165 | lower to higher preference, each element is a dictionary | |
166 | with the "ext" entry and one of: | |
167 | * "data": The subtitles file contents | |
10952eb2 | 168 | * "url": A URL pointing to the subtitles file |
4bba3716 | 169 | "ext" will be calculated from URL if missing |
360e1ca5 JMF |
170 | automatic_captions: Like 'subtitles', used by the YoutubeIE for |
171 | automatically generated captions | |
62d231c0 | 172 | duration: Length of the video in seconds, as an integer or float. |
f3d29461 | 173 | view_count: How many users have watched the video on the platform. |
19e3dfc9 PH |
174 | like_count: Number of positive ratings of the video |
175 | dislike_count: Number of negative ratings of the video | |
02835c6b | 176 | repost_count: Number of reposts of the video |
2d30521a | 177 | average_rating: Average rating give by users, the scale used depends on the webpage |
19e3dfc9 | 178 | comment_count: Number of comments on the video |
dd622d7c PH |
179 | comments: A list of comments, each with one or more of the following |
180 | properties (all but one of text or html optional): | |
181 | * "author" - human-readable name of the comment author | |
182 | * "author_id" - user ID of the comment author | |
183 | * "id" - Comment ID | |
184 | * "html" - Comment as HTML | |
185 | * "text" - Plain text of the comment | |
186 | * "timestamp" - UNIX timestamp of comment | |
187 | * "parent" - ID of the comment this one is replying to. | |
188 | Set to "root" to indicate that this is a | |
189 | comment to the original video. | |
8dbe9899 | 190 | age_limit: Age restriction for the video, as an integer (years) |
10952eb2 | 191 | webpage_url: The URL to the video webpage, if given to youtube-dl it |
9103bbc5 JMF |
192 | should allow to get the same result again. (It will be set |
193 | by YoutubeDL if it's missing) | |
ad3bc6ac PH |
194 | categories: A list of categories that the video falls in, for example |
195 | ["Sports", "Berlin"] | |
864f24bd | 196 | tags: A list of tags assigned to the video, e.g. ["sweden", "pop music"] |
7267bd53 PH |
197 | is_live: True, False, or None (=unknown). Whether this video is a |
198 | live stream that goes on instead of a fixed-length video. | |
7c80519c | 199 | start_time: Time in seconds where the reproduction should start, as |
10952eb2 | 200 | specified in the URL. |
297a564b | 201 | end_time: Time in seconds where the reproduction should end, as |
10952eb2 | 202 | specified in the URL. |
d6983cb4 | 203 | |
7109903e S |
204 | The following fields should only be used when the video belongs to some logical |
205 | chapter or section: | |
206 | ||
207 | chapter: Name or title of the chapter the video belongs to. | |
27bfd4e5 S |
208 | chapter_number: Number of the chapter the video belongs to, as an integer. |
209 | chapter_id: Id of the chapter the video belongs to, as a unicode string. | |
7109903e S |
210 | |
211 | The following fields should only be used when the video is an episode of some | |
212 | series or programme: | |
213 | ||
214 | series: Title of the series or programme the video episode belongs to. | |
215 | season: Title of the season the video episode belongs to. | |
27bfd4e5 S |
216 | season_number: Number of the season the video episode belongs to, as an integer. |
217 | season_id: Id of the season the video episode belongs to, as a unicode string. | |
7109903e S |
218 | episode: Title of the video episode. Unlike mandatory video title field, |
219 | this field should denote the exact title of the video episode | |
220 | without any kind of decoration. | |
27bfd4e5 S |
221 | episode_number: Number of the video episode within a season, as an integer. |
222 | episode_id: Id of the video episode, as a unicode string. | |
7109903e | 223 | |
deefc05b | 224 | Unless mentioned otherwise, the fields should be Unicode strings. |
d6983cb4 | 225 | |
d838b1bd PH |
226 | Unless mentioned otherwise, None is equivalent to absence of information. |
227 | ||
fed5d032 PH |
228 | |
229 | _type "playlist" indicates multiple videos. | |
b82f815f PH |
230 | There must be a key "entries", which is a list, an iterable, or a PagedList |
231 | object, each element of which is a valid dictionary by this specification. | |
fed5d032 | 232 | |
e0b9d78f S |
233 | Additionally, playlists can have "title", "description" and "id" attributes |
234 | with the same semantics as videos (see above). | |
fed5d032 PH |
235 | |
236 | ||
237 | _type "multi_video" indicates that there are multiple videos that | |
238 | form a single show, for examples multiple acts of an opera or TV episode. | |
239 | It must have an entries key like a playlist and contain all the keys | |
240 | required for a video at the same time. | |
241 | ||
242 | ||
243 | _type "url" indicates that the video must be extracted from another | |
244 | location, possibly by a different extractor. Its only required key is: | |
245 | "url" - the next URL to extract. | |
f58766ce PH |
246 | The key "ie_key" can be set to the class name (minus the trailing "IE", |
247 | e.g. "Youtube") if the extractor class is known in advance. | |
248 | Additionally, the dictionary may have any properties of the resolved entity | |
249 | known in advance, for example "title" if the title of the referred video is | |
fed5d032 PH |
250 | known ahead of time. |
251 | ||
252 | ||
253 | _type "url_transparent" entities have the same specification as "url", but | |
254 | indicate that the given additional information is more precise than the one | |
255 | associated with the resolved URL. | |
256 | This is useful when a site employs a video service that hosts the video and | |
257 | its technical metadata, but that video service does not embed a useful | |
258 | title, description etc. | |
259 | ||
260 | ||
d6983cb4 PH |
261 | Subclasses of this one should re-define the _real_initialize() and |
262 | _real_extract() methods and define a _VALID_URL regexp. | |
263 | Probably, they should also be added to the list of extractors. | |
264 | ||
d6983cb4 PH |
265 | Finally, the _WORKING attribute should be set to False for broken IEs |
266 | in order to warn the users and skip the tests. | |
267 | """ | |
268 | ||
269 | _ready = False | |
270 | _downloader = None | |
271 | _WORKING = True | |
272 | ||
273 | def __init__(self, downloader=None): | |
274 | """Constructor. Receives an optional downloader.""" | |
275 | self._ready = False | |
276 | self.set_downloader(downloader) | |
277 | ||
278 | @classmethod | |
279 | def suitable(cls, url): | |
280 | """Receives a URL and returns True if suitable for this IE.""" | |
79cb2577 PH |
281 | |
282 | # This does not use has/getattr intentionally - we want to know whether | |
283 | # we have cached the regexp for *this* class, whereas getattr would also | |
284 | # match the superclass | |
285 | if '_VALID_URL_RE' not in cls.__dict__: | |
286 | cls._VALID_URL_RE = re.compile(cls._VALID_URL) | |
287 | return cls._VALID_URL_RE.match(url) is not None | |
d6983cb4 | 288 | |
ed9266db PH |
289 | @classmethod |
290 | def _match_id(cls, url): | |
291 | if '_VALID_URL_RE' not in cls.__dict__: | |
292 | cls._VALID_URL_RE = re.compile(cls._VALID_URL) | |
293 | m = cls._VALID_URL_RE.match(url) | |
294 | assert m | |
295 | return m.group('id') | |
296 | ||
d6983cb4 PH |
297 | @classmethod |
298 | def working(cls): | |
299 | """Getter method for _WORKING.""" | |
300 | return cls._WORKING | |
301 | ||
302 | def initialize(self): | |
303 | """Initializes an instance (authentication, etc).""" | |
304 | if not self._ready: | |
305 | self._real_initialize() | |
306 | self._ready = True | |
307 | ||
308 | def extract(self, url): | |
309 | """Extracts URL information and returns it in list of dicts.""" | |
3a5bcd03 PH |
310 | try: |
311 | self.initialize() | |
312 | return self._real_extract(url) | |
313 | except ExtractorError: | |
314 | raise | |
315 | except compat_http_client.IncompleteRead as e: | |
316 | raise ExtractorError('A network error has occured.', cause=e, expected=True) | |
9650885b | 317 | except (KeyError, StopIteration) as e: |
3a5bcd03 | 318 | raise ExtractorError('An extractor error has occured.', cause=e) |
d6983cb4 PH |
319 | |
320 | def set_downloader(self, downloader): | |
321 | """Sets the downloader for this IE.""" | |
322 | self._downloader = downloader | |
323 | ||
324 | def _real_initialize(self): | |
325 | """Real initialization process. Redefine in subclasses.""" | |
326 | pass | |
327 | ||
328 | def _real_extract(self, url): | |
329 | """Real extraction process. Redefine in subclasses.""" | |
330 | pass | |
331 | ||
56c73665 JMF |
332 | @classmethod |
333 | def ie_key(cls): | |
334 | """A string for getting the InfoExtractor with get_info_extractor""" | |
dc519b54 | 335 | return compat_str(cls.__name__[:-2]) |
56c73665 | 336 | |
d6983cb4 PH |
337 | @property |
338 | def IE_NAME(self): | |
dc519b54 | 339 | return compat_str(type(self).__name__[:-2]) |
d6983cb4 | 340 | |
7cc3570e | 341 | def _request_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True): |
d6983cb4 PH |
342 | """ Returns the response handle """ |
343 | if note is None: | |
344 | self.report_download_webpage(video_id) | |
345 | elif note is not False: | |
7cc3570e | 346 | if video_id is None: |
f1a9d64e | 347 | self.to_screen('%s' % (note,)) |
7cc3570e | 348 | else: |
f1a9d64e | 349 | self.to_screen('%s: %s' % (video_id, note)) |
d6983cb4 | 350 | try: |
dca08720 | 351 | return self._downloader.urlopen(url_or_request) |
d6983cb4 | 352 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: |
aa94a6d3 PH |
353 | if errnote is False: |
354 | return False | |
d6983cb4 | 355 | if errnote is None: |
f1a9d64e | 356 | errnote = 'Unable to download webpage' |
7f8b2714 | 357 | |
9b9c5355 | 358 | errmsg = '%s: %s' % (errnote, error_to_compat_str(err)) |
7cc3570e PH |
359 | if fatal: |
360 | raise ExtractorError(errmsg, sys.exc_info()[2], cause=err) | |
361 | else: | |
362 | self._downloader.report_warning(errmsg) | |
363 | return False | |
d6983cb4 | 364 | |
c9a77969 | 365 | def _download_webpage_handle(self, url_or_request, video_id, note=None, errnote=None, fatal=True, encoding=None): |
d6983cb4 | 366 | """ Returns a tuple (page content as string, URL handle) """ |
b9d3e163 PH |
367 | # Strip hashes from the URL (#1038) |
368 | if isinstance(url_or_request, (compat_str, str)): | |
369 | url_or_request = url_or_request.partition('#')[0] | |
370 | ||
7cc3570e PH |
371 | urlh = self._request_webpage(url_or_request, video_id, note, errnote, fatal) |
372 | if urlh is False: | |
373 | assert not fatal | |
374 | return False | |
c9a77969 | 375 | content = self._webpage_read_content(urlh, url_or_request, video_id, note, errnote, fatal, encoding=encoding) |
23be51d8 PH |
376 | return (content, urlh) |
377 | ||
c9a77969 YCH |
378 | @staticmethod |
379 | def _guess_encoding_from_content(content_type, webpage_bytes): | |
d6983cb4 PH |
380 | m = re.match(r'[a-zA-Z0-9_.-]+/[a-zA-Z0-9_.-]+\s*;\s*charset=(.+)', content_type) |
381 | if m: | |
382 | encoding = m.group(1) | |
383 | else: | |
0d75ae2c | 384 | m = re.search(br'<meta[^>]+charset=[\'"]?([^\'")]+)[ /\'">]', |
f143d86a PH |
385 | webpage_bytes[:1024]) |
386 | if m: | |
387 | encoding = m.group(1).decode('ascii') | |
b60016e8 PH |
388 | elif webpage_bytes.startswith(b'\xff\xfe'): |
389 | encoding = 'utf-16' | |
f143d86a PH |
390 | else: |
391 | encoding = 'utf-8' | |
c9a77969 YCH |
392 | |
393 | return encoding | |
394 | ||
395 | def _webpage_read_content(self, urlh, url_or_request, video_id, note=None, errnote=None, fatal=True, prefix=None, encoding=None): | |
396 | content_type = urlh.headers.get('Content-Type', '') | |
397 | webpage_bytes = urlh.read() | |
398 | if prefix is not None: | |
399 | webpage_bytes = prefix + webpage_bytes | |
400 | if not encoding: | |
401 | encoding = self._guess_encoding_from_content(content_type, webpage_bytes) | |
d6983cb4 PH |
402 | if self._downloader.params.get('dump_intermediate_pages', False): |
403 | try: | |
404 | url = url_or_request.get_full_url() | |
405 | except AttributeError: | |
406 | url = url_or_request | |
f1a9d64e | 407 | self.to_screen('Dumping request to ' + url) |
d6983cb4 PH |
408 | dump = base64.b64encode(webpage_bytes).decode('ascii') |
409 | self._downloader.to_screen(dump) | |
d41e6efc PH |
410 | if self._downloader.params.get('write_pages', False): |
411 | try: | |
412 | url = url_or_request.get_full_url() | |
413 | except AttributeError: | |
414 | url = url_or_request | |
5afa7f8b | 415 | basen = '%s_%s' % (video_id, url) |
c1bce22f | 416 | if len(basen) > 240: |
f1a9d64e | 417 | h = '___' + hashlib.md5(basen.encode('utf-8')).hexdigest() |
c1bce22f PH |
418 | basen = basen[:240 - len(h)] + h |
419 | raw_filename = basen + '.dump' | |
d41e6efc | 420 | filename = sanitize_filename(raw_filename, restricted=True) |
f1a9d64e | 421 | self.to_screen('Saving request to ' + filename) |
5f58165d S |
422 | # Working around MAX_PATH limitation on Windows (see |
423 | # http://msdn.microsoft.com/en-us/library/windows/desktop/aa365247(v=vs.85).aspx) | |
424 | if os.name == 'nt': | |
425 | absfilepath = os.path.abspath(filename) | |
426 | if len(absfilepath) > 259: | |
427 | filename = '\\\\?\\' + absfilepath | |
d41e6efc PH |
428 | with open(filename, 'wb') as outf: |
429 | outf.write(webpage_bytes) | |
430 | ||
ec0fafbb AA |
431 | try: |
432 | content = webpage_bytes.decode(encoding, 'replace') | |
433 | except LookupError: | |
434 | content = webpage_bytes.decode('utf-8', 'replace') | |
2410c43d | 435 | |
f1a9d64e PH |
436 | if ('<title>Access to this site is blocked</title>' in content and |
437 | 'Websense' in content[:512]): | |
438 | msg = 'Access to this webpage has been blocked by Websense filtering software in your network.' | |
2410c43d PH |
439 | blocked_iframe = self._html_search_regex( |
440 | r'<iframe src="([^"]+)"', content, | |
f1a9d64e | 441 | 'Websense information URL', default=None) |
2410c43d | 442 | if blocked_iframe: |
f1a9d64e | 443 | msg += ' Visit %s for more details' % blocked_iframe |
2410c43d | 444 | raise ExtractorError(msg, expected=True) |
77b2986b PH |
445 | if '<title>The URL you requested has been blocked</title>' in content[:512]: |
446 | msg = ( | |
447 | 'Access to this webpage has been blocked by Indian censorship. ' | |
448 | 'Use a VPN or proxy server (with --proxy) to route around it.') | |
449 | block_msg = self._html_search_regex( | |
450 | r'</h1><p>(.*?)</p>', | |
451 | content, 'block message', default=None) | |
452 | if block_msg: | |
453 | msg += ' (Message: "%s")' % block_msg.replace('\n', ' ') | |
454 | raise ExtractorError(msg, expected=True) | |
2410c43d | 455 | |
23be51d8 | 456 | return content |
d6983cb4 | 457 | |
c9a77969 | 458 | def _download_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True, tries=1, timeout=5, encoding=None): |
d6983cb4 | 459 | """ Returns the data of the page as a string """ |
995ad69c TF |
460 | success = False |
461 | try_count = 0 | |
462 | while success is False: | |
463 | try: | |
c9a77969 | 464 | res = self._download_webpage_handle(url_or_request, video_id, note, errnote, fatal, encoding=encoding) |
995ad69c TF |
465 | success = True |
466 | except compat_http_client.IncompleteRead as e: | |
467 | try_count += 1 | |
468 | if try_count >= tries: | |
469 | raise e | |
470 | self._sleep(timeout, video_id) | |
7cc3570e PH |
471 | if res is False: |
472 | return res | |
473 | else: | |
474 | content, _ = res | |
475 | return content | |
d6983cb4 | 476 | |
2a275ab0 | 477 | def _download_xml(self, url_or_request, video_id, |
f1a9d64e | 478 | note='Downloading XML', errnote='Unable to download XML', |
c9a77969 | 479 | transform_source=None, fatal=True, encoding=None): |
267ed0c5 | 480 | """Return the xml as an xml.etree.ElementTree.Element""" |
28746fbd | 481 | xml_string = self._download_webpage( |
c9a77969 | 482 | url_or_request, video_id, note, errnote, fatal=fatal, encoding=encoding) |
28746fbd PH |
483 | if xml_string is False: |
484 | return xml_string | |
e2b38da9 PH |
485 | if transform_source: |
486 | xml_string = transform_source(xml_string) | |
36e6f62c | 487 | return compat_etree_fromstring(xml_string.encode('utf-8')) |
267ed0c5 | 488 | |
3d3538e4 | 489 | def _download_json(self, url_or_request, video_id, |
f1a9d64e PH |
490 | note='Downloading JSON metadata', |
491 | errnote='Unable to download JSON metadata', | |
b090af59 | 492 | transform_source=None, |
c9a77969 | 493 | fatal=True, encoding=None): |
b090af59 | 494 | json_string = self._download_webpage( |
c9a77969 YCH |
495 | url_or_request, video_id, note, errnote, fatal=fatal, |
496 | encoding=encoding) | |
b090af59 PH |
497 | if (not fatal) and json_string is False: |
498 | return None | |
ebb64199 TF |
499 | return self._parse_json( |
500 | json_string, video_id, transform_source=transform_source, fatal=fatal) | |
501 | ||
502 | def _parse_json(self, json_string, video_id, transform_source=None, fatal=True): | |
81c2f20b PH |
503 | if transform_source: |
504 | json_string = transform_source(json_string) | |
3d3538e4 PH |
505 | try: |
506 | return json.loads(json_string) | |
507 | except ValueError as ve: | |
e7b6d122 PH |
508 | errmsg = '%s: Failed to parse JSON ' % video_id |
509 | if fatal: | |
510 | raise ExtractorError(errmsg, cause=ve) | |
511 | else: | |
512 | self.report_warning(errmsg + str(ve)) | |
3d3538e4 | 513 | |
f45f96f8 | 514 | def report_warning(self, msg, video_id=None): |
f1a9d64e | 515 | idstr = '' if video_id is None else '%s: ' % video_id |
f45f96f8 | 516 | self._downloader.report_warning( |
f1a9d64e | 517 | '[%s] %s%s' % (self.IE_NAME, idstr, msg)) |
f45f96f8 | 518 | |
d6983cb4 PH |
519 | def to_screen(self, msg): |
520 | """Print msg to screen, prefixing it with '[ie_name]'""" | |
f1a9d64e | 521 | self._downloader.to_screen('[%s] %s' % (self.IE_NAME, msg)) |
d6983cb4 PH |
522 | |
523 | def report_extraction(self, id_or_name): | |
524 | """Report information extraction.""" | |
f1a9d64e | 525 | self.to_screen('%s: Extracting information' % id_or_name) |
d6983cb4 PH |
526 | |
527 | def report_download_webpage(self, video_id): | |
528 | """Report webpage download.""" | |
f1a9d64e | 529 | self.to_screen('%s: Downloading webpage' % video_id) |
d6983cb4 PH |
530 | |
531 | def report_age_confirmation(self): | |
532 | """Report attempt to confirm age.""" | |
f1a9d64e | 533 | self.to_screen('Confirming age') |
d6983cb4 | 534 | |
fc79158d JMF |
535 | def report_login(self): |
536 | """Report attempt to log in.""" | |
f1a9d64e | 537 | self.to_screen('Logging in') |
fc79158d | 538 | |
43e7d3c9 S |
539 | @staticmethod |
540 | def raise_login_required(msg='This video is only available for registered users'): | |
541 | raise ExtractorError( | |
542 | '%s. Use --username and --password or --netrc to provide account credentials.' % msg, | |
543 | expected=True) | |
544 | ||
c430802e S |
545 | @staticmethod |
546 | def raise_geo_restricted(msg='This video is not available from your location due to geo restriction'): | |
547 | raise ExtractorError( | |
548 | '%s. You might want to use --proxy to workaround.' % msg, | |
549 | expected=True) | |
550 | ||
5f6a1245 | 551 | # Methods for following #608 |
c0d0b01f | 552 | @staticmethod |
830d53bf | 553 | def url_result(url, ie=None, video_id=None, video_title=None): |
10952eb2 | 554 | """Returns a URL that points to a page that should be processed""" |
5f6a1245 | 555 | # TODO: ie should be the class used for getting the info |
d6983cb4 PH |
556 | video_info = {'_type': 'url', |
557 | 'url': url, | |
558 | 'ie_key': ie} | |
7012b23c PH |
559 | if video_id is not None: |
560 | video_info['id'] = video_id | |
830d53bf S |
561 | if video_title is not None: |
562 | video_info['title'] = video_title | |
d6983cb4 | 563 | return video_info |
5f6a1245 | 564 | |
c0d0b01f | 565 | @staticmethod |
acf5cbfe | 566 | def playlist_result(entries, playlist_id=None, playlist_title=None, playlist_description=None): |
d6983cb4 PH |
567 | """Returns a playlist""" |
568 | video_info = {'_type': 'playlist', | |
569 | 'entries': entries} | |
570 | if playlist_id: | |
571 | video_info['id'] = playlist_id | |
572 | if playlist_title: | |
573 | video_info['title'] = playlist_title | |
acf5cbfe S |
574 | if playlist_description: |
575 | video_info['description'] = playlist_description | |
d6983cb4 PH |
576 | return video_info |
577 | ||
c342041f | 578 | def _search_regex(self, pattern, string, name, default=NO_DEFAULT, fatal=True, flags=0, group=None): |
d6983cb4 PH |
579 | """ |
580 | Perform a regex search on the given string, using a single or a list of | |
581 | patterns returning the first matching group. | |
582 | In case of failure return a default value or raise a WARNING or a | |
55b3e45b | 583 | RegexNotFoundError, depending on fatal, specifying the field name. |
d6983cb4 PH |
584 | """ |
585 | if isinstance(pattern, (str, compat_str, compiled_regex_type)): | |
586 | mobj = re.search(pattern, string, flags) | |
587 | else: | |
588 | for p in pattern: | |
589 | mobj = re.search(p, string, flags) | |
c3415d1b PH |
590 | if mobj: |
591 | break | |
d6983cb4 | 592 | |
7e5db8c9 | 593 | if not self._downloader.params.get('no_color') and os.name != 'nt' and sys.stderr.isatty(): |
f1a9d64e | 594 | _name = '\033[0;34m%s\033[0m' % name |
d6983cb4 PH |
595 | else: |
596 | _name = name | |
597 | ||
598 | if mobj: | |
711ede6e PH |
599 | if group is None: |
600 | # return the first matching group | |
601 | return next(g for g in mobj.groups() if g is not None) | |
602 | else: | |
603 | return mobj.group(group) | |
c342041f | 604 | elif default is not NO_DEFAULT: |
d6983cb4 PH |
605 | return default |
606 | elif fatal: | |
f1a9d64e | 607 | raise RegexNotFoundError('Unable to extract %s' % _name) |
d6983cb4 | 608 | else: |
08f2a92c | 609 | self._downloader.report_warning('unable to extract %s' % _name + bug_reports_message()) |
d6983cb4 PH |
610 | return None |
611 | ||
c342041f | 612 | def _html_search_regex(self, pattern, string, name, default=NO_DEFAULT, fatal=True, flags=0, group=None): |
d6983cb4 PH |
613 | """ |
614 | Like _search_regex, but strips HTML tags and unescapes entities. | |
615 | """ | |
711ede6e | 616 | res = self._search_regex(pattern, string, name, default, fatal, flags, group) |
d6983cb4 PH |
617 | if res: |
618 | return clean_html(res).strip() | |
619 | else: | |
620 | return res | |
621 | ||
fc79158d JMF |
622 | def _get_login_info(self): |
623 | """ | |
cf0649f8 | 624 | Get the login info as (username, password) |
fc79158d JMF |
625 | It will look in the netrc file using the _NETRC_MACHINE value |
626 | If there's no info available, return (None, None) | |
627 | """ | |
628 | if self._downloader is None: | |
629 | return (None, None) | |
630 | ||
631 | username = None | |
632 | password = None | |
633 | downloader_params = self._downloader.params | |
634 | ||
635 | # Attempt to use provided username and password or .netrc data | |
636 | if downloader_params.get('username', None) is not None: | |
637 | username = downloader_params['username'] | |
638 | password = downloader_params['password'] | |
639 | elif downloader_params.get('usenetrc', False): | |
640 | try: | |
641 | info = netrc.netrc().authenticators(self._NETRC_MACHINE) | |
642 | if info is not None: | |
643 | username = info[0] | |
644 | password = info[2] | |
645 | else: | |
646 | raise netrc.NetrcParseError('No authenticators for %s' % self._NETRC_MACHINE) | |
647 | except (IOError, netrc.NetrcParseError) as err: | |
9b9c5355 | 648 | self._downloader.report_warning('parsing .netrc: %s' % error_to_compat_str(err)) |
5f6a1245 | 649 | |
fc79158d JMF |
650 | return (username, password) |
651 | ||
e64b7569 | 652 | def _get_tfa_info(self, note='two-factor verification code'): |
83317f69 | 653 | """ |
654 | Get the two-factor authentication info | |
655 | TODO - asking the user will be required for sms/phone verify | |
656 | currently just uses the command line option | |
657 | If there's no info available, return None | |
658 | """ | |
659 | if self._downloader is None: | |
83317f69 | 660 | return None |
661 | downloader_params = self._downloader.params | |
662 | ||
663 | if downloader_params.get('twofactor', None) is not None: | |
664 | return downloader_params['twofactor'] | |
665 | ||
e64b7569 | 666 | return compat_getpass('Type %s and press [Return]: ' % note) |
83317f69 | 667 | |
46720279 JMF |
668 | # Helper functions for extracting OpenGraph info |
669 | @staticmethod | |
ab2d5247 | 670 | def _og_regexes(prop): |
448ef1f3 | 671 | content_re = r'content=(?:"([^"]+?)"|\'([^\']+?)\'|\s*([^\s"\'=<>`]+?))' |
7a6d76a6 S |
672 | property_re = (r'(?:name|property)=(?:\'og:%(prop)s\'|"og:%(prop)s"|\s*og:%(prop)s\b)' |
673 | % {'prop': re.escape(prop)}) | |
78fb87b2 | 674 | template = r'<meta[^>]+?%s[^>]+?%s' |
ab2d5247 | 675 | return [ |
78fb87b2 JMF |
676 | template % (property_re, content_re), |
677 | template % (content_re, property_re), | |
ab2d5247 | 678 | ] |
46720279 | 679 | |
864f24bd S |
680 | @staticmethod |
681 | def _meta_regex(prop): | |
682 | return r'''(?isx)<meta | |
8b9848ac | 683 | (?=[^>]+(?:itemprop|name|property|id|http-equiv)=(["\']?)%s\1) |
864f24bd S |
684 | [^>]+?content=(["\'])(?P<content>.*?)\2''' % re.escape(prop) |
685 | ||
3c4e6d83 | 686 | def _og_search_property(self, prop, html, name=None, **kargs): |
46720279 | 687 | if name is None: |
3c4e6d83 | 688 | name = 'OpenGraph %s' % prop |
ab2d5247 | 689 | escaped = self._search_regex(self._og_regexes(prop), html, name, flags=re.DOTALL, **kargs) |
eb0a8398 PH |
690 | if escaped is None: |
691 | return None | |
692 | return unescapeHTML(escaped) | |
46720279 JMF |
693 | |
694 | def _og_search_thumbnail(self, html, **kargs): | |
10952eb2 | 695 | return self._og_search_property('image', html, 'thumbnail URL', fatal=False, **kargs) |
46720279 JMF |
696 | |
697 | def _og_search_description(self, html, **kargs): | |
698 | return self._og_search_property('description', html, fatal=False, **kargs) | |
699 | ||
700 | def _og_search_title(self, html, **kargs): | |
701 | return self._og_search_property('title', html, **kargs) | |
702 | ||
8ffa13e0 | 703 | def _og_search_video_url(self, html, name='video url', secure=True, **kargs): |
a3681973 PH |
704 | regexes = self._og_regexes('video') + self._og_regexes('video:url') |
705 | if secure: | |
706 | regexes = self._og_regexes('video:secure_url') + regexes | |
8ffa13e0 | 707 | return self._html_search_regex(regexes, html, name, **kargs) |
46720279 | 708 | |
78338f71 JMF |
709 | def _og_search_url(self, html, **kargs): |
710 | return self._og_search_property('url', html, **kargs) | |
711 | ||
40c696e5 | 712 | def _html_search_meta(self, name, html, display_name=None, fatal=False, **kwargs): |
59040888 PH |
713 | if display_name is None: |
714 | display_name = name | |
715 | return self._html_search_regex( | |
864f24bd | 716 | self._meta_regex(name), |
711ede6e | 717 | html, display_name, fatal=fatal, group='content', **kwargs) |
59040888 PH |
718 | |
719 | def _dc_search_uploader(self, html): | |
720 | return self._html_search_meta('dc.creator', html, 'uploader') | |
721 | ||
8dbe9899 PH |
722 | def _rta_search(self, html): |
723 | # See http://www.rtalabel.org/index.php?content=howtofaq#single | |
724 | if re.search(r'(?ix)<meta\s+name="rating"\s+' | |
725 | r' content="RTA-5042-1996-1400-1577-RTA"', | |
726 | html): | |
727 | return 18 | |
728 | return 0 | |
729 | ||
59040888 PH |
730 | def _media_rating_search(self, html): |
731 | # See http://www.tjg-designs.com/WP/metadata-code-examples-adding-metadata-to-your-web-pages/ | |
732 | rating = self._html_search_meta('rating', html) | |
733 | ||
734 | if not rating: | |
735 | return None | |
736 | ||
737 | RATING_TABLE = { | |
738 | 'safe for kids': 0, | |
739 | 'general': 8, | |
740 | '14 years': 14, | |
741 | 'mature': 17, | |
742 | 'restricted': 19, | |
743 | } | |
744 | return RATING_TABLE.get(rating.lower(), None) | |
745 | ||
69319969 | 746 | def _family_friendly_search(self, html): |
6ca7732d | 747 | # See http://schema.org/VideoObject |
69319969 NJ |
748 | family_friendly = self._html_search_meta('isFamilyFriendly', html) |
749 | ||
750 | if not family_friendly: | |
751 | return None | |
752 | ||
753 | RATING_TABLE = { | |
754 | '1': 0, | |
755 | 'true': 0, | |
756 | '0': 18, | |
757 | 'false': 18, | |
758 | } | |
759 | return RATING_TABLE.get(family_friendly.lower(), None) | |
760 | ||
0c708f11 JMF |
761 | def _twitter_search_player(self, html): |
762 | return self._html_search_meta('twitter:player', html, | |
9e1a5b84 | 763 | 'twitter card player') |
0c708f11 | 764 | |
27713812 | 765 | @staticmethod |
f8da79f8 | 766 | def _hidden_inputs(html): |
586f1cc5 | 767 | html = re.sub(r'<!--(?:(?!<!--).)*-->', '', html) |
201ea3ee | 768 | hidden_inputs = {} |
73eb13df | 769 | for input in re.findall(r'(?i)<input([^>]+)>', html): |
be0e5dbd | 770 | if not re.search(r'type=(["\'])(?:hidden|submit)\1', input): |
201ea3ee S |
771 | continue |
772 | name = re.search(r'name=(["\'])(?P<value>.+?)\1', input) | |
773 | if not name: | |
774 | continue | |
775 | value = re.search(r'value=(["\'])(?P<value>.*?)\1', input) | |
776 | if not value: | |
777 | continue | |
778 | hidden_inputs[name.group('value')] = value.group('value') | |
779 | return hidden_inputs | |
27713812 | 780 | |
cf61d96d S |
781 | def _form_hidden_inputs(self, form_id, html): |
782 | form = self._search_regex( | |
73eb13df | 783 | r'(?is)<form[^>]+?id=(["\'])%s\1[^>]*>(?P<form>.+?)</form>' % form_id, |
cf61d96d S |
784 | html, '%s form' % form_id, group='form') |
785 | return self._hidden_inputs(form) | |
786 | ||
3ded7bac | 787 | def _sort_formats(self, formats, field_preference=None): |
7e8caf30 | 788 | if not formats: |
f1a9d64e | 789 | raise ExtractorError('No video formats found') |
7e8caf30 | 790 | |
4bcc7bd1 | 791 | def _formats_key(f): |
e6812ac9 PH |
792 | # TODO remove the following workaround |
793 | from ..utils import determine_ext | |
794 | if not f.get('ext') and 'url' in f: | |
795 | f['ext'] = determine_ext(f['url']) | |
796 | ||
3ded7bac S |
797 | if isinstance(field_preference, (list, tuple)): |
798 | return tuple(f.get(field) if f.get(field) is not None else -1 for field in field_preference) | |
799 | ||
4bcc7bd1 PH |
800 | preference = f.get('preference') |
801 | if preference is None: | |
d497a201 | 802 | preference = 0 |
4bcc7bd1 PH |
803 | if f.get('ext') in ['f4f', 'f4m']: # Not yet supported |
804 | preference -= 0.5 | |
805 | ||
d497a201 | 806 | proto_preference = 0 if determine_protocol(f) in ['http', 'https'] else -0.1 |
807 | ||
4bcc7bd1 PH |
808 | if f.get('vcodec') == 'none': # audio only |
809 | if self._downloader.params.get('prefer_free_formats'): | |
f1a9d64e | 810 | ORDER = ['aac', 'mp3', 'm4a', 'webm', 'ogg', 'opus'] |
4bcc7bd1 | 811 | else: |
f1a9d64e | 812 | ORDER = ['webm', 'opus', 'ogg', 'mp3', 'aac', 'm4a'] |
4bcc7bd1 PH |
813 | ext_preference = 0 |
814 | try: | |
815 | audio_ext_preference = ORDER.index(f['ext']) | |
816 | except ValueError: | |
817 | audio_ext_preference = -1 | |
818 | else: | |
819 | if self._downloader.params.get('prefer_free_formats'): | |
f1a9d64e | 820 | ORDER = ['flv', 'mp4', 'webm'] |
4bcc7bd1 | 821 | else: |
f1a9d64e | 822 | ORDER = ['webm', 'flv', 'mp4'] |
4bcc7bd1 PH |
823 | try: |
824 | ext_preference = ORDER.index(f['ext']) | |
825 | except ValueError: | |
826 | ext_preference = -1 | |
827 | audio_ext_preference = 0 | |
828 | ||
829 | return ( | |
830 | preference, | |
aff2f4f4 | 831 | f.get('language_preference') if f.get('language_preference') is not None else -1, |
5d73273f | 832 | f.get('quality') if f.get('quality') is not None else -1, |
9933b574 | 833 | f.get('tbr') if f.get('tbr') is not None else -1, |
03cd72b0 | 834 | f.get('filesize') if f.get('filesize') is not None else -1, |
4bcc7bd1 | 835 | f.get('vbr') if f.get('vbr') is not None else -1, |
1a6373ef PH |
836 | f.get('height') if f.get('height') is not None else -1, |
837 | f.get('width') if f.get('width') is not None else -1, | |
d497a201 | 838 | proto_preference, |
1e1896f2 | 839 | ext_preference, |
4bcc7bd1 PH |
840 | f.get('abr') if f.get('abr') is not None else -1, |
841 | audio_ext_preference, | |
2c8e03d9 | 842 | f.get('fps') if f.get('fps') is not None else -1, |
9732d77e | 843 | f.get('filesize_approx') if f.get('filesize_approx') is not None else -1, |
c64ed2a3 | 844 | f.get('source_preference') if f.get('source_preference') is not None else -1, |
74f72824 | 845 | f.get('format_id') if f.get('format_id') is not None else '', |
4bcc7bd1 PH |
846 | ) |
847 | formats.sort(key=_formats_key) | |
59040888 | 848 | |
96a53167 S |
849 | def _check_formats(self, formats, video_id): |
850 | if formats: | |
851 | formats[:] = filter( | |
852 | lambda f: self._is_valid_url( | |
853 | f['url'], video_id, | |
854 | item='%s video format' % f.get('format_id') if f.get('format_id') else 'video'), | |
855 | formats) | |
856 | ||
857 | def _is_valid_url(self, url, video_id, item='video'): | |
2f0f6578 S |
858 | url = self._proto_relative_url(url, scheme='http:') |
859 | # For now assume non HTTP(S) URLs always valid | |
860 | if not (url.startswith('http://') or url.startswith('https://')): | |
861 | return True | |
96a53167 | 862 | try: |
4069766c | 863 | self._request_webpage(url, video_id, 'Checking %s URL' % item) |
96a53167 S |
864 | return True |
865 | except ExtractorError as e: | |
943a1e24 | 866 | if isinstance(e.cause, compat_urllib_error.URLError): |
baa43cba S |
867 | self.to_screen( |
868 | '%s: %s URL is invalid, skipping' % (video_id, item)) | |
96a53167 S |
869 | return False |
870 | raise | |
871 | ||
20991253 | 872 | def http_scheme(self): |
1ede5b24 | 873 | """ Either "http:" or "https:", depending on the user's preferences """ |
20991253 PH |
874 | return ( |
875 | 'http:' | |
876 | if self._downloader.params.get('prefer_insecure', False) | |
877 | else 'https:') | |
878 | ||
57c7411f PH |
879 | def _proto_relative_url(self, url, scheme=None): |
880 | if url is None: | |
881 | return url | |
882 | if url.startswith('//'): | |
883 | if scheme is None: | |
884 | scheme = self.http_scheme() | |
885 | return scheme + url | |
886 | else: | |
887 | return url | |
888 | ||
4094b6e3 PH |
889 | def _sleep(self, timeout, video_id, msg_template=None): |
890 | if msg_template is None: | |
f1a9d64e | 891 | msg_template = '%(video_id)s: Waiting for %(timeout)s seconds' |
4094b6e3 PH |
892 | msg = msg_template % {'video_id': video_id, 'timeout': timeout} |
893 | self.to_screen(msg) | |
894 | time.sleep(timeout) | |
895 | ||
a38436e8 | 896 | def _extract_f4m_formats(self, manifest_url, video_id, preference=None, f4m_id=None, |
4de61310 S |
897 | transform_source=lambda s: fix_xml_ampersands(s).strip(), |
898 | fatal=True): | |
f036a632 JMF |
899 | manifest = self._download_xml( |
900 | manifest_url, video_id, 'Downloading f4m manifest', | |
97f4aecf S |
901 | 'Unable to download f4m manifest', |
902 | # Some manifests may be malformed, e.g. prosiebensat1 generated manifests | |
903 | # (see https://github.com/rg3/youtube-dl/issues/6215#issuecomment-121704244) | |
4de61310 S |
904 | transform_source=transform_source, |
905 | fatal=fatal) | |
906 | ||
907 | if manifest is False: | |
8d29e47f | 908 | return [] |
31bb8d3f JMF |
909 | |
910 | formats = [] | |
7a47d07c | 911 | manifest_version = '1.0' |
b2527359 | 912 | media_nodes = manifest.findall('{http://ns.adobe.com/f4m/1.0}media') |
34e48bed | 913 | if not media_nodes: |
7a47d07c | 914 | manifest_version = '2.0' |
34e48bed | 915 | media_nodes = manifest.findall('{http://ns.adobe.com/f4m/2.0}media') |
019839fa S |
916 | base_url = xpath_text( |
917 | manifest, ['{http://ns.adobe.com/f4m/1.0}baseURL', '{http://ns.adobe.com/f4m/2.0}baseURL'], | |
918 | 'base URL', default=None) | |
919 | if base_url: | |
920 | base_url = base_url.strip() | |
b2527359 | 921 | for i, media_el in enumerate(media_nodes): |
7a47d07c | 922 | if manifest_version == '2.0': |
31c746e5 S |
923 | media_url = media_el.attrib.get('href') or media_el.attrib.get('url') |
924 | if not media_url: | |
925 | continue | |
cc357c4d S |
926 | manifest_url = ( |
927 | media_url if media_url.startswith('http://') or media_url.startswith('https://') | |
019839fa | 928 | else ((base_url or '/'.join(manifest_url.split('/')[:-1])) + '/' + media_url)) |
70f0f5a8 S |
929 | # If media_url is itself a f4m manifest do the recursive extraction |
930 | # since bitrates in parent manifest (this one) and media_url manifest | |
931 | # may differ leading to inability to resolve the format by requested | |
932 | # bitrate in f4m downloader | |
933 | if determine_ext(manifest_url) == 'f4m': | |
7e5edcfd S |
934 | formats.extend(self._extract_f4m_formats( |
935 | manifest_url, video_id, preference, f4m_id, fatal=fatal)) | |
70f0f5a8 | 936 | continue |
b2527359 | 937 | tbr = int_or_none(media_el.attrib.get('bitrate')) |
31bb8d3f | 938 | formats.append({ |
e21a55ab | 939 | 'format_id': '-'.join(filter(None, [f4m_id, compat_str(i if tbr is None else tbr)])), |
31bb8d3f JMF |
940 | 'url': manifest_url, |
941 | 'ext': 'flv', | |
b2527359 | 942 | 'tbr': tbr, |
31bb8d3f JMF |
943 | 'width': int_or_none(media_el.attrib.get('width')), |
944 | 'height': int_or_none(media_el.attrib.get('height')), | |
60ca389c | 945 | 'preference': preference, |
31bb8d3f JMF |
946 | }) |
947 | self._sort_formats(formats) | |
948 | ||
949 | return formats | |
950 | ||
f0b5d6af | 951 | def _extract_m3u8_formats(self, m3u8_url, video_id, ext=None, |
60ca389c | 952 | entry_protocol='m3u8', preference=None, |
13af92fd YCH |
953 | m3u8_id=None, note=None, errnote=None, |
954 | fatal=True): | |
f0b5d6af | 955 | |
704df56d | 956 | formats = [{ |
f207019c | 957 | 'format_id': '-'.join(filter(None, [m3u8_id, 'meta'])), |
704df56d PH |
958 | 'url': m3u8_url, |
959 | 'ext': ext, | |
960 | 'protocol': 'm3u8', | |
9fe6ef7a | 961 | 'preference': preference - 1 if preference else -1, |
704df56d PH |
962 | 'resolution': 'multiple', |
963 | 'format_note': 'Quality selection URL', | |
964 | }] | |
965 | ||
f0b5d6af PH |
966 | format_url = lambda u: ( |
967 | u | |
968 | if re.match(r'^https?://', u) | |
969 | else compat_urlparse.urljoin(m3u8_url, u)) | |
970 | ||
dbd82a1d | 971 | res = self._download_webpage_handle( |
81515ad9 | 972 | m3u8_url, video_id, |
621ed9f5 | 973 | note=note or 'Downloading m3u8 information', |
13af92fd YCH |
974 | errnote=errnote or 'Failed to download m3u8 information', |
975 | fatal=fatal) | |
dbd82a1d | 976 | if res is False: |
8d29e47f | 977 | return [] |
dbd82a1d | 978 | m3u8_doc, urlh = res |
37113045 | 979 | m3u8_url = urlh.geturl() |
704df56d | 980 | last_info = None |
fa156077 | 981 | last_media = None |
704df56d PH |
982 | kv_rex = re.compile( |
983 | r'(?P<key>[a-zA-Z_-]+)=(?P<val>"[^"]+"|[^",]+)(?:,|$)') | |
984 | for line in m3u8_doc.splitlines(): | |
985 | if line.startswith('#EXT-X-STREAM-INF:'): | |
986 | last_info = {} | |
987 | for m in kv_rex.finditer(line): | |
988 | v = m.group('val') | |
989 | if v.startswith('"'): | |
990 | v = v[1:-1] | |
991 | last_info[m.group('key')] = v | |
4cd95bcb JMF |
992 | elif line.startswith('#EXT-X-MEDIA:'): |
993 | last_media = {} | |
994 | for m in kv_rex.finditer(line): | |
995 | v = m.group('val') | |
996 | if v.startswith('"'): | |
997 | v = v[1:-1] | |
998 | last_media[m.group('key')] = v | |
704df56d PH |
999 | elif line.startswith('#') or not line.strip(): |
1000 | continue | |
1001 | else: | |
daebaab6 | 1002 | if last_info is None: |
f0b5d6af | 1003 | formats.append({'url': format_url(line)}) |
3524cc25 | 1004 | continue |
704df56d | 1005 | tbr = int_or_none(last_info.get('BANDWIDTH'), scale=1000) |
8dc9d361 S |
1006 | format_id = [] |
1007 | if m3u8_id: | |
1008 | format_id.append(m3u8_id) | |
05d5392c | 1009 | last_media_name = last_media.get('NAME') if last_media and last_media.get('TYPE') != 'SUBTITLES' else None |
8dc9d361 | 1010 | format_id.append(last_media_name if last_media_name else '%d' % (tbr if tbr else len(formats))) |
704df56d | 1011 | f = { |
8dc9d361 | 1012 | 'format_id': '-'.join(format_id), |
f0b5d6af | 1013 | 'url': format_url(line.strip()), |
704df56d PH |
1014 | 'tbr': tbr, |
1015 | 'ext': ext, | |
f0b5d6af PH |
1016 | 'protocol': entry_protocol, |
1017 | 'preference': preference, | |
704df56d PH |
1018 | } |
1019 | codecs = last_info.get('CODECS') | |
1020 | if codecs: | |
9ebf22b7 S |
1021 | # TODO: looks like video codec is not always necessarily goes first |
1022 | va_codecs = codecs.split(',') | |
1023 | if va_codecs[0]: | |
1024 | f['vcodec'] = va_codecs[0].partition('.')[0] | |
1025 | if len(va_codecs) > 1 and va_codecs[1]: | |
1026 | f['acodec'] = va_codecs[1].partition('.')[0] | |
704df56d PH |
1027 | resolution = last_info.get('RESOLUTION') |
1028 | if resolution: | |
1029 | width_str, height_str = resolution.split('x') | |
1030 | f['width'] = int(width_str) | |
1031 | f['height'] = int(height_str) | |
4cd95bcb JMF |
1032 | if last_media is not None: |
1033 | f['m3u8_media'] = last_media | |
1034 | last_media = None | |
704df56d PH |
1035 | formats.append(f) |
1036 | last_info = {} | |
1037 | self._sort_formats(formats) | |
1038 | return formats | |
1039 | ||
a107193e S |
1040 | @staticmethod |
1041 | def _xpath_ns(path, namespace=None): | |
1042 | if not namespace: | |
1043 | return path | |
1044 | out = [] | |
1045 | for c in path.split('/'): | |
1046 | if not c or c == '.': | |
1047 | out.append(c) | |
1048 | else: | |
1049 | out.append('{%s}%s' % (namespace, c)) | |
1050 | return '/'.join(out) | |
1051 | ||
1052 | def _extract_smil_formats(self, smil_url, video_id, fatal=True, f4m_params=None): | |
1053 | smil = self._download_smil(smil_url, video_id, fatal=fatal) | |
1054 | ||
995029a1 PH |
1055 | if smil is False: |
1056 | assert not fatal | |
1057 | return [] | |
e89a2aab | 1058 | |
17712eeb | 1059 | namespace = self._parse_smil_namespace(smil) |
a107193e S |
1060 | |
1061 | return self._parse_smil_formats( | |
1062 | smil, smil_url, video_id, namespace=namespace, f4m_params=f4m_params) | |
1063 | ||
1064 | def _extract_smil_info(self, smil_url, video_id, fatal=True, f4m_params=None): | |
1065 | smil = self._download_smil(smil_url, video_id, fatal=fatal) | |
1066 | if smil is False: | |
1067 | return {} | |
1068 | return self._parse_smil(smil, smil_url, video_id, f4m_params=f4m_params) | |
1069 | ||
1070 | def _download_smil(self, smil_url, video_id, fatal=True): | |
1071 | return self._download_xml( | |
1072 | smil_url, video_id, 'Downloading SMIL file', | |
1073 | 'Unable to download SMIL file', fatal=fatal) | |
1074 | ||
1075 | def _parse_smil(self, smil, smil_url, video_id, f4m_params=None): | |
17712eeb | 1076 | namespace = self._parse_smil_namespace(smil) |
a107193e S |
1077 | |
1078 | formats = self._parse_smil_formats( | |
1079 | smil, smil_url, video_id, namespace=namespace, f4m_params=f4m_params) | |
1080 | subtitles = self._parse_smil_subtitles(smil, namespace=namespace) | |
1081 | ||
1082 | video_id = os.path.splitext(url_basename(smil_url))[0] | |
1083 | title = None | |
1084 | description = None | |
647eab45 | 1085 | upload_date = None |
a107193e S |
1086 | for meta in smil.findall(self._xpath_ns('./head/meta', namespace)): |
1087 | name = meta.attrib.get('name') | |
1088 | content = meta.attrib.get('content') | |
1089 | if not name or not content: | |
1090 | continue | |
1091 | if not title and name == 'title': | |
1092 | title = content | |
1093 | elif not description and name in ('description', 'abstract'): | |
1094 | description = content | |
647eab45 S |
1095 | elif not upload_date and name == 'date': |
1096 | upload_date = unified_strdate(content) | |
a107193e | 1097 | |
1e5bcdec S |
1098 | thumbnails = [{ |
1099 | 'id': image.get('type'), | |
1100 | 'url': image.get('src'), | |
1101 | 'width': int_or_none(image.get('width')), | |
1102 | 'height': int_or_none(image.get('height')), | |
1103 | } for image in smil.findall(self._xpath_ns('.//image', namespace)) if image.get('src')] | |
1104 | ||
a107193e S |
1105 | return { |
1106 | 'id': video_id, | |
1107 | 'title': title or video_id, | |
1108 | 'description': description, | |
647eab45 | 1109 | 'upload_date': upload_date, |
1e5bcdec | 1110 | 'thumbnails': thumbnails, |
a107193e S |
1111 | 'formats': formats, |
1112 | 'subtitles': subtitles, | |
1113 | } | |
1114 | ||
17712eeb S |
1115 | def _parse_smil_namespace(self, smil): |
1116 | return self._search_regex( | |
1117 | r'(?i)^{([^}]+)?}smil$', smil.tag, 'namespace', default=None) | |
1118 | ||
f877c6ae | 1119 | def _parse_smil_formats(self, smil, smil_url, video_id, namespace=None, f4m_params=None, transform_rtmp_url=None): |
a107193e S |
1120 | base = smil_url |
1121 | for meta in smil.findall(self._xpath_ns('./head/meta', namespace)): | |
1122 | b = meta.get('base') or meta.get('httpBase') | |
1123 | if b: | |
1124 | base = b | |
1125 | break | |
e89a2aab S |
1126 | |
1127 | formats = [] | |
1128 | rtmp_count = 0 | |
a107193e S |
1129 | http_count = 0 |
1130 | ||
1131 | videos = smil.findall(self._xpath_ns('.//video', namespace)) | |
1132 | for video in videos: | |
1133 | src = video.get('src') | |
1134 | if not src: | |
1135 | continue | |
1136 | ||
e7d8e98a | 1137 | bitrate = float_or_none(video.get('system-bitrate') or video.get('systemBitrate'), 1000) |
a107193e S |
1138 | filesize = int_or_none(video.get('size') or video.get('fileSize')) |
1139 | width = int_or_none(video.get('width')) | |
1140 | height = int_or_none(video.get('height')) | |
1141 | proto = video.get('proto') | |
1142 | ext = video.get('ext') | |
1143 | src_ext = determine_ext(src) | |
1144 | streamer = video.get('streamer') or base | |
1145 | ||
1146 | if proto == 'rtmp' or streamer.startswith('rtmp'): | |
1147 | rtmp_count += 1 | |
1148 | formats.append({ | |
1149 | 'url': streamer, | |
1150 | 'play_path': src, | |
1151 | 'ext': 'flv', | |
1152 | 'format_id': 'rtmp-%d' % (rtmp_count if bitrate is None else bitrate), | |
1153 | 'tbr': bitrate, | |
1154 | 'filesize': filesize, | |
1155 | 'width': width, | |
1156 | 'height': height, | |
1157 | }) | |
f877c6ae YCH |
1158 | if transform_rtmp_url: |
1159 | streamer, src = transform_rtmp_url(streamer, src) | |
1160 | formats[-1].update({ | |
1161 | 'url': streamer, | |
1162 | 'play_path': src, | |
1163 | }) | |
a107193e S |
1164 | continue |
1165 | ||
1166 | src_url = src if src.startswith('http') else compat_urlparse.urljoin(base, src) | |
1167 | ||
1168 | if proto == 'm3u8' or src_ext == 'm3u8': | |
7e5edcfd S |
1169 | formats.extend(self._extract_m3u8_formats( |
1170 | src_url, video_id, ext or 'mp4', m3u8_id='hls', fatal=False)) | |
a107193e S |
1171 | continue |
1172 | ||
1173 | if src_ext == 'f4m': | |
1174 | f4m_url = src_url | |
1175 | if not f4m_params: | |
1176 | f4m_params = { | |
1177 | 'hdcore': '3.2.0', | |
1178 | 'plugin': 'flowplayer-3.2.0.1', | |
1179 | } | |
1180 | f4m_url += '&' if '?' in f4m_url else '?' | |
41c3a5a7 | 1181 | f4m_url += compat_urllib_parse.urlencode(f4m_params) |
7e5edcfd | 1182 | formats.extend(self._extract_f4m_formats(f4m_url, video_id, f4m_id='hds', fatal=False)) |
a107193e S |
1183 | continue |
1184 | ||
c78e4817 | 1185 | if src_url.startswith('http') and self._is_valid_url(src, video_id): |
a107193e S |
1186 | http_count += 1 |
1187 | formats.append({ | |
1188 | 'url': src_url, | |
1189 | 'ext': ext or src_ext or 'flv', | |
1190 | 'format_id': 'http-%d' % (bitrate or http_count), | |
1191 | 'tbr': bitrate, | |
1192 | 'filesize': filesize, | |
1193 | 'width': width, | |
1194 | 'height': height, | |
1195 | }) | |
1196 | continue | |
63757032 | 1197 | |
e89a2aab S |
1198 | self._sort_formats(formats) |
1199 | ||
1200 | return formats | |
1201 | ||
ce00af87 | 1202 | def _parse_smil_subtitles(self, smil, namespace=None, subtitles_lang='en'): |
a107193e S |
1203 | subtitles = {} |
1204 | for num, textstream in enumerate(smil.findall(self._xpath_ns('.//textstream', namespace))): | |
1205 | src = textstream.get('src') | |
1206 | if not src: | |
1207 | continue | |
1208 | ext = textstream.get('ext') or determine_ext(src) | |
1209 | if not ext: | |
1210 | type_ = textstream.get('type') | |
5cdefc46 S |
1211 | SUBTITLES_TYPES = { |
1212 | 'text/vtt': 'vtt', | |
1213 | 'text/srt': 'srt', | |
1214 | 'application/smptett+xml': 'tt', | |
1215 | } | |
1216 | if type_ in SUBTITLES_TYPES: | |
1217 | ext = SUBTITLES_TYPES[type_] | |
03bc7237 | 1218 | lang = textstream.get('systemLanguage') or textstream.get('systemLanguageName') or textstream.get('lang') or subtitles_lang |
a107193e S |
1219 | subtitles.setdefault(lang, []).append({ |
1220 | 'url': src, | |
1221 | 'ext': ext, | |
1222 | }) | |
1223 | return subtitles | |
63757032 | 1224 | |
942acef5 S |
1225 | def _extract_xspf_playlist(self, playlist_url, playlist_id, fatal=True): |
1226 | xspf = self._download_xml( | |
8d6765cf | 1227 | playlist_url, playlist_id, 'Downloading xpsf playlist', |
942acef5 S |
1228 | 'Unable to download xspf manifest', fatal=fatal) |
1229 | if xspf is False: | |
1230 | return [] | |
1231 | return self._parse_xspf(xspf, playlist_id) | |
8d6765cf | 1232 | |
942acef5 | 1233 | def _parse_xspf(self, playlist, playlist_id): |
8d6765cf S |
1234 | NS_MAP = { |
1235 | 'xspf': 'http://xspf.org/ns/0/', | |
1236 | 's1': 'http://static.streamone.nl/player/ns/0', | |
1237 | } | |
1238 | ||
1239 | entries = [] | |
1240 | for track in playlist.findall(xpath_with_ns('./xspf:trackList/xspf:track', NS_MAP)): | |
1241 | title = xpath_text( | |
98044462 | 1242 | track, xpath_with_ns('./xspf:title', NS_MAP), 'title', default=playlist_id) |
8d6765cf S |
1243 | description = xpath_text( |
1244 | track, xpath_with_ns('./xspf:annotation', NS_MAP), 'description') | |
1245 | thumbnail = xpath_text( | |
1246 | track, xpath_with_ns('./xspf:image', NS_MAP), 'thumbnail') | |
1247 | duration = float_or_none( | |
1248 | xpath_text(track, xpath_with_ns('./xspf:duration', NS_MAP), 'duration'), 1000) | |
1249 | ||
1250 | formats = [{ | |
1251 | 'url': location.text, | |
1252 | 'format_id': location.get(xpath_with_ns('s1:label', NS_MAP)), | |
1253 | 'width': int_or_none(location.get(xpath_with_ns('s1:width', NS_MAP))), | |
1254 | 'height': int_or_none(location.get(xpath_with_ns('s1:height', NS_MAP))), | |
1255 | } for location in track.findall(xpath_with_ns('./xspf:location', NS_MAP))] | |
1256 | self._sort_formats(formats) | |
1257 | ||
1258 | entries.append({ | |
1259 | 'id': playlist_id, | |
1260 | 'title': title, | |
1261 | 'description': description, | |
1262 | 'thumbnail': thumbnail, | |
1263 | 'duration': duration, | |
1264 | 'formats': formats, | |
1265 | }) | |
1266 | return entries | |
1267 | ||
f4b1c7ad PH |
1268 | def _live_title(self, name): |
1269 | """ Generate the title for a live video """ | |
1270 | now = datetime.datetime.now() | |
1271 | now_str = now.strftime("%Y-%m-%d %H:%M") | |
1272 | return name + ' ' + now_str | |
1273 | ||
b14f3a4c PH |
1274 | def _int(self, v, name, fatal=False, **kwargs): |
1275 | res = int_or_none(v, **kwargs) | |
1276 | if 'get_attr' in kwargs: | |
1277 | print(getattr(v, kwargs['get_attr'])) | |
1278 | if res is None: | |
1279 | msg = 'Failed to extract %s: Could not parse value %r' % (name, v) | |
1280 | if fatal: | |
1281 | raise ExtractorError(msg) | |
1282 | else: | |
1283 | self._downloader.report_warning(msg) | |
1284 | return res | |
1285 | ||
1286 | def _float(self, v, name, fatal=False, **kwargs): | |
1287 | res = float_or_none(v, **kwargs) | |
1288 | if res is None: | |
1289 | msg = 'Failed to extract %s: Could not parse value %r' % (name, v) | |
1290 | if fatal: | |
1291 | raise ExtractorError(msg) | |
1292 | else: | |
1293 | self._downloader.report_warning(msg) | |
1294 | return res | |
1295 | ||
42939b61 | 1296 | def _set_cookie(self, domain, name, value, expire_time=None): |
810fb84d PH |
1297 | cookie = compat_cookiejar.Cookie( |
1298 | 0, name, value, None, None, domain, None, | |
42939b61 JMF |
1299 | None, '/', True, False, expire_time, '', None, None, None) |
1300 | self._downloader.cookiejar.set_cookie(cookie) | |
1301 | ||
799207e8 | 1302 | def _get_cookies(self, url): |
1303 | """ Return a compat_cookies.SimpleCookie with the cookies for the url """ | |
5c2266df | 1304 | req = sanitized_Request(url) |
799207e8 | 1305 | self._downloader.cookiejar.add_cookie_header(req) |
1306 | return compat_cookies.SimpleCookie(req.get_header('Cookie')) | |
1307 | ||
05900629 PH |
1308 | def get_testcases(self, include_onlymatching=False): |
1309 | t = getattr(self, '_TEST', None) | |
1310 | if t: | |
1311 | assert not hasattr(self, '_TESTS'), \ | |
1312 | '%s has _TEST and _TESTS' % type(self).__name__ | |
1313 | tests = [t] | |
1314 | else: | |
1315 | tests = getattr(self, '_TESTS', []) | |
1316 | for t in tests: | |
1317 | if not include_onlymatching and t.get('only_matching', False): | |
1318 | continue | |
1319 | t['name'] = type(self).__name__[:-len('IE')] | |
1320 | yield t | |
1321 | ||
1322 | def is_suitable(self, age_limit): | |
1323 | """ Test whether the extractor is generally suitable for the given | |
1324 | age limit (i.e. pornographic sites are not, all others usually are) """ | |
1325 | ||
1326 | any_restricted = False | |
1327 | for tc in self.get_testcases(include_onlymatching=False): | |
1328 | if 'playlist' in tc: | |
1329 | tc = tc['playlist'][0] | |
1330 | is_restricted = age_restricted( | |
1331 | tc.get('info_dict', {}).get('age_limit'), age_limit) | |
1332 | if not is_restricted: | |
1333 | return True | |
1334 | any_restricted = any_restricted or is_restricted | |
1335 | return not any_restricted | |
1336 | ||
a504ced0 | 1337 | def extract_subtitles(self, *args, **kwargs): |
9868ea49 JMF |
1338 | if (self._downloader.params.get('writesubtitles', False) or |
1339 | self._downloader.params.get('listsubtitles')): | |
1340 | return self._get_subtitles(*args, **kwargs) | |
1341 | return {} | |
a504ced0 JMF |
1342 | |
1343 | def _get_subtitles(self, *args, **kwargs): | |
1344 | raise NotImplementedError("This method must be implemented by subclasses") | |
1345 | ||
912e0b7e YCH |
1346 | @staticmethod |
1347 | def _merge_subtitle_items(subtitle_list1, subtitle_list2): | |
1348 | """ Merge subtitle items for one language. Items with duplicated URLs | |
1349 | will be dropped. """ | |
1350 | list1_urls = set([item['url'] for item in subtitle_list1]) | |
1351 | ret = list(subtitle_list1) | |
1352 | ret.extend([item for item in subtitle_list2 if item['url'] not in list1_urls]) | |
1353 | return ret | |
1354 | ||
1355 | @classmethod | |
8c97f819 | 1356 | def _merge_subtitles(cls, subtitle_dict1, subtitle_dict2): |
912e0b7e | 1357 | """ Merge two subtitle dictionaries, language by language. """ |
912e0b7e YCH |
1358 | ret = dict(subtitle_dict1) |
1359 | for lang in subtitle_dict2: | |
8c97f819 | 1360 | ret[lang] = cls._merge_subtitle_items(subtitle_dict1.get(lang, []), subtitle_dict2[lang]) |
912e0b7e YCH |
1361 | return ret |
1362 | ||
360e1ca5 | 1363 | def extract_automatic_captions(self, *args, **kwargs): |
9868ea49 JMF |
1364 | if (self._downloader.params.get('writeautomaticsub', False) or |
1365 | self._downloader.params.get('listsubtitles')): | |
1366 | return self._get_automatic_captions(*args, **kwargs) | |
1367 | return {} | |
360e1ca5 JMF |
1368 | |
1369 | def _get_automatic_captions(self, *args, **kwargs): | |
1370 | raise NotImplementedError("This method must be implemented by subclasses") | |
1371 | ||
8dbe9899 | 1372 | |
d6983cb4 PH |
1373 | class SearchInfoExtractor(InfoExtractor): |
1374 | """ | |
1375 | Base class for paged search queries extractors. | |
10952eb2 | 1376 | They accept URLs in the format _SEARCH_KEY(|all|[0-9]):{query} |
d6983cb4 PH |
1377 | Instances should define _SEARCH_KEY and _MAX_RESULTS. |
1378 | """ | |
1379 | ||
1380 | @classmethod | |
1381 | def _make_valid_url(cls): | |
1382 | return r'%s(?P<prefix>|[1-9][0-9]*|all):(?P<query>[\s\S]+)' % cls._SEARCH_KEY | |
1383 | ||
1384 | @classmethod | |
1385 | def suitable(cls, url): | |
1386 | return re.match(cls._make_valid_url(), url) is not None | |
1387 | ||
1388 | def _real_extract(self, query): | |
1389 | mobj = re.match(self._make_valid_url(), query) | |
1390 | if mobj is None: | |
f1a9d64e | 1391 | raise ExtractorError('Invalid search query "%s"' % query) |
d6983cb4 PH |
1392 | |
1393 | prefix = mobj.group('prefix') | |
1394 | query = mobj.group('query') | |
1395 | if prefix == '': | |
1396 | return self._get_n_results(query, 1) | |
1397 | elif prefix == 'all': | |
1398 | return self._get_n_results(query, self._MAX_RESULTS) | |
1399 | else: | |
1400 | n = int(prefix) | |
1401 | if n <= 0: | |
f1a9d64e | 1402 | raise ExtractorError('invalid download number %s for query "%s"' % (n, query)) |
d6983cb4 | 1403 | elif n > self._MAX_RESULTS: |
f1a9d64e | 1404 | self._downloader.report_warning('%s returns max %i results (you requested %i)' % (self._SEARCH_KEY, self._MAX_RESULTS, n)) |
d6983cb4 PH |
1405 | n = self._MAX_RESULTS |
1406 | return self._get_n_results(query, n) | |
1407 | ||
1408 | def _get_n_results(self, query, n): | |
1409 | """Get a specified number of results for a query""" | |
416a5efc | 1410 | raise NotImplementedError("This method must be implemented by subclasses") |
0f818663 PH |
1411 | |
1412 | @property | |
1413 | def SEARCH_KEY(self): | |
1414 | return self._SEARCH_KEY |