]>
Commit | Line | Data |
---|---|---|
6a3828fd | 1 | from __future__ import unicode_literals |
f1a9d64e | 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 |
1bac3455 | 13 | import math |
d6983cb4 | 14 | |
8c25f81b | 15 | from ..compat import ( |
42939b61 | 16 | compat_cookiejar, |
799207e8 | 17 | compat_cookies, |
e9c0cdd3 | 18 | compat_etree_fromstring, |
e64b7569 | 19 | compat_getpass, |
d6983cb4 | 20 | compat_http_client, |
e9c0cdd3 YCH |
21 | compat_os_name, |
22 | compat_str, | |
d6983cb4 | 23 | compat_urllib_error, |
a107193e | 24 | compat_urllib_parse, |
f0b5d6af | 25 | compat_urlparse, |
8c25f81b PH |
26 | ) |
27 | from ..utils import ( | |
c342041f | 28 | NO_DEFAULT, |
05900629 | 29 | age_restricted, |
08f2a92c | 30 | bug_reports_message, |
d6983cb4 PH |
31 | clean_html, |
32 | compiled_regex_type, | |
70f0f5a8 | 33 | determine_ext, |
9b9c5355 | 34 | error_to_compat_str, |
d6983cb4 | 35 | ExtractorError, |
97f4aecf | 36 | fix_xml_ampersands, |
b14f3a4c | 37 | float_or_none, |
31bb8d3f | 38 | int_or_none, |
4ca2a3cf | 39 | parse_iso8601, |
55b3e45b | 40 | RegexNotFoundError, |
d41e6efc | 41 | sanitize_filename, |
5c2266df | 42 | sanitized_Request, |
f38de77f | 43 | unescapeHTML, |
647eab45 | 44 | unified_strdate, |
a107193e | 45 | url_basename, |
8d6765cf S |
46 | xpath_text, |
47 | xpath_with_ns, | |
d497a201 | 48 | determine_protocol, |
1bac3455 | 49 | parse_duration, |
cafcf657 | 50 | mimetype2ext, |
cdfee168 | 51 | update_url_query, |
d6983cb4 | 52 | ) |
c342041f | 53 | |
d6983cb4 PH |
54 | |
55 | class InfoExtractor(object): | |
56 | """Information Extractor class. | |
57 | ||
58 | Information extractors are the classes that, given a URL, extract | |
59 | information about the video (or videos) the URL refers to. This | |
60 | information includes the real video URL, the video title, author and | |
61 | others. The information is stored in a dictionary which is then | |
5d380852 | 62 | passed to the YoutubeDL. The YoutubeDL processes this |
d6983cb4 PH |
63 | information possibly downloading the video to the file system, among |
64 | other possible outcomes. | |
65 | ||
cf0649f8 | 66 | The type field determines the type of the result. |
fed5d032 PH |
67 | By far the most common value (and the default if _type is missing) is |
68 | "video", which indicates a single video. | |
69 | ||
70 | For a video, the dictionaries must include the following fields: | |
d6983cb4 PH |
71 | |
72 | id: Video identifier. | |
d6983cb4 | 73 | title: Video title, unescaped. |
d67b0b15 | 74 | |
f49d89ee | 75 | Additionally, it must contain either a formats entry or a url one: |
d67b0b15 | 76 | |
f49d89ee PH |
77 | formats: A list of dictionaries for each format available, ordered |
78 | from worst to best quality. | |
79 | ||
80 | Potential fields: | |
d67b0b15 | 81 | * url Mandatory. The URL of the video file |
10952eb2 | 82 | * ext Will be calculated from URL if missing |
d67b0b15 PH |
83 | * format A human-readable description of the format |
84 | ("mp4 container with h264/opus"). | |
85 | Calculated from the format_id, width, height. | |
86 | and format_note fields if missing. | |
87 | * format_id A short description of the format | |
5d4f3985 PH |
88 | ("mp4_h264_opus" or "19"). |
89 | Technically optional, but strongly recommended. | |
d67b0b15 PH |
90 | * format_note Additional info about the format |
91 | ("3D" or "DASH video") | |
92 | * width Width of the video, if known | |
93 | * height Height of the video, if known | |
f49d89ee | 94 | * resolution Textual description of width and height |
7217e148 | 95 | * tbr Average bitrate of audio and video in KBit/s |
d67b0b15 PH |
96 | * abr Average audio bitrate in KBit/s |
97 | * acodec Name of the audio codec in use | |
dd27fd17 | 98 | * asr Audio sampling rate in Hertz |
d67b0b15 | 99 | * vbr Average video bitrate in KBit/s |
fbb21cf5 | 100 | * fps Frame rate |
d67b0b15 | 101 | * vcodec Name of the video codec in use |
1394ce65 | 102 | * container Name of the container format |
d67b0b15 | 103 | * filesize The number of bytes, if known in advance |
9732d77e | 104 | * filesize_approx An estimate for the number of bytes |
d67b0b15 | 105 | * player_url SWF Player URL (used for rtmpdump). |
c7deaa4c PH |
106 | * protocol The protocol that will be used for the actual |
107 | download, lower-case. | |
b04b8852 | 108 | "http", "https", "rtsp", "rtmp", "rtmpe", |
af7d5a63 | 109 | "m3u8", "m3u8_native" or "http_dash_segments". |
f49d89ee | 110 | * preference Order number of this format. If this field is |
08d13955 | 111 | present and not None, the formats get sorted |
38d63d84 | 112 | by this field, regardless of all other values. |
f49d89ee PH |
113 | -1 for default (order by other properties), |
114 | -2 or smaller for less than default. | |
e65566a9 PH |
115 | < -1000 to hide the format (if there is |
116 | another one which is strictly better) | |
32f90364 PH |
117 | * language Language code, e.g. "de" or "en-US". |
118 | * language_preference Is this in the language mentioned in | |
119 | the URL? | |
aff2f4f4 PH |
120 | 10 if it's what the URL is about, |
121 | -1 for default (don't know), | |
122 | -10 otherwise, other values reserved for now. | |
5d73273f PH |
123 | * quality Order number of the video quality of this |
124 | format, irrespective of the file format. | |
125 | -1 for default (order by other properties), | |
126 | -2 or smaller for less than default. | |
c64ed2a3 PH |
127 | * source_preference Order number for this video source |
128 | (quality takes higher priority) | |
129 | -1 for default (order by other properties), | |
130 | -2 or smaller for less than default. | |
d769be6c PH |
131 | * http_headers A dictionary of additional HTTP headers |
132 | to add to the request. | |
6271f1ca | 133 | * stretched_ratio If given and not 1, indicates that the |
3dee7826 PH |
134 | video's pixels are not square. |
135 | width : height ratio as float. | |
136 | * no_resume The server does not support resuming the | |
137 | (HTTP or RTMP) download. Boolean. | |
138 | ||
c0ba0f48 | 139 | url: Final video URL. |
d6983cb4 | 140 | ext: Video filename extension. |
d67b0b15 PH |
141 | format: The video format, defaults to ext (used for --get-format) |
142 | player_url: SWF Player URL (used for rtmpdump). | |
2f5865cc | 143 | |
d6983cb4 PH |
144 | The following fields are optional: |
145 | ||
f5e43bc6 | 146 | alt_title: A secondary title of the video. |
0afef30b PH |
147 | display_id An alternative identifier for the video, not necessarily |
148 | unique, but available before title. Typically, id is | |
149 | something like "4234987", title "Dancing naked mole rats", | |
150 | and display_id "dancing-naked-mole-rats" | |
d5519808 | 151 | thumbnails: A list of dictionaries, with the following entries: |
cfb56d1a | 152 | * "id" (optional, string) - Thumbnail format ID |
d5519808 | 153 | * "url" |
cfb56d1a | 154 | * "preference" (optional, int) - quality of the image |
d5519808 PH |
155 | * "width" (optional, int) |
156 | * "height" (optional, int) | |
157 | * "resolution" (optional, string "{width}x{height"}, | |
158 | deprecated) | |
d6983cb4 | 159 | thumbnail: Full URL to a video thumbnail image. |
f5e43bc6 | 160 | description: Full video description. |
d6983cb4 | 161 | uploader: Full name of the video uploader. |
2bc0c46f | 162 | license: License name the video is licensed under. |
9bb8e0a3 | 163 | creator: The main artist who created the video. |
8aab976b | 164 | release_date: The date (YYYYMMDD) when the video was released. |
955c4514 | 165 | timestamp: UNIX timestamp of the moment the video became available. |
d6983cb4 | 166 | upload_date: Video upload date (YYYYMMDD). |
955c4514 | 167 | If not explicitly set, calculated from timestamp. |
d6983cb4 | 168 | uploader_id: Nickname or id of the video uploader. |
7bcd2830 | 169 | uploader_url: Full URL to a personal webpage of the video uploader. |
da9ec3b9 | 170 | location: Physical location where the video was filmed. |
a504ced0 JMF |
171 | subtitles: The available subtitles as a dictionary in the format |
172 | {language: subformats}. "subformats" is a list sorted from | |
173 | lower to higher preference, each element is a dictionary | |
174 | with the "ext" entry and one of: | |
175 | * "data": The subtitles file contents | |
10952eb2 | 176 | * "url": A URL pointing to the subtitles file |
4bba3716 | 177 | "ext" will be calculated from URL if missing |
360e1ca5 JMF |
178 | automatic_captions: Like 'subtitles', used by the YoutubeIE for |
179 | automatically generated captions | |
62d231c0 | 180 | duration: Length of the video in seconds, as an integer or float. |
f3d29461 | 181 | view_count: How many users have watched the video on the platform. |
19e3dfc9 PH |
182 | like_count: Number of positive ratings of the video |
183 | dislike_count: Number of negative ratings of the video | |
02835c6b | 184 | repost_count: Number of reposts of the video |
2d30521a | 185 | average_rating: Average rating give by users, the scale used depends on the webpage |
19e3dfc9 | 186 | comment_count: Number of comments on the video |
dd622d7c PH |
187 | comments: A list of comments, each with one or more of the following |
188 | properties (all but one of text or html optional): | |
189 | * "author" - human-readable name of the comment author | |
190 | * "author_id" - user ID of the comment author | |
191 | * "id" - Comment ID | |
192 | * "html" - Comment as HTML | |
193 | * "text" - Plain text of the comment | |
194 | * "timestamp" - UNIX timestamp of comment | |
195 | * "parent" - ID of the comment this one is replying to. | |
196 | Set to "root" to indicate that this is a | |
197 | comment to the original video. | |
8dbe9899 | 198 | age_limit: Age restriction for the video, as an integer (years) |
10952eb2 | 199 | webpage_url: The URL to the video webpage, if given to youtube-dl it |
9103bbc5 JMF |
200 | should allow to get the same result again. (It will be set |
201 | by YoutubeDL if it's missing) | |
ad3bc6ac PH |
202 | categories: A list of categories that the video falls in, for example |
203 | ["Sports", "Berlin"] | |
864f24bd | 204 | tags: A list of tags assigned to the video, e.g. ["sweden", "pop music"] |
7267bd53 PH |
205 | is_live: True, False, or None (=unknown). Whether this video is a |
206 | live stream that goes on instead of a fixed-length video. | |
7c80519c | 207 | start_time: Time in seconds where the reproduction should start, as |
10952eb2 | 208 | specified in the URL. |
297a564b | 209 | end_time: Time in seconds where the reproduction should end, as |
10952eb2 | 210 | specified in the URL. |
d6983cb4 | 211 | |
7109903e S |
212 | The following fields should only be used when the video belongs to some logical |
213 | chapter or section: | |
214 | ||
215 | chapter: Name or title of the chapter the video belongs to. | |
27bfd4e5 S |
216 | chapter_number: Number of the chapter the video belongs to, as an integer. |
217 | chapter_id: Id of the chapter the video belongs to, as a unicode string. | |
7109903e S |
218 | |
219 | The following fields should only be used when the video is an episode of some | |
220 | series or programme: | |
221 | ||
222 | series: Title of the series or programme the video episode belongs to. | |
223 | season: Title of the season the video episode belongs to. | |
27bfd4e5 S |
224 | season_number: Number of the season the video episode belongs to, as an integer. |
225 | season_id: Id of the season the video episode belongs to, as a unicode string. | |
7109903e S |
226 | episode: Title of the video episode. Unlike mandatory video title field, |
227 | this field should denote the exact title of the video episode | |
228 | without any kind of decoration. | |
27bfd4e5 S |
229 | episode_number: Number of the video episode within a season, as an integer. |
230 | episode_id: Id of the video episode, as a unicode string. | |
7109903e | 231 | |
deefc05b | 232 | Unless mentioned otherwise, the fields should be Unicode strings. |
d6983cb4 | 233 | |
d838b1bd PH |
234 | Unless mentioned otherwise, None is equivalent to absence of information. |
235 | ||
fed5d032 PH |
236 | |
237 | _type "playlist" indicates multiple videos. | |
b82f815f PH |
238 | There must be a key "entries", which is a list, an iterable, or a PagedList |
239 | object, each element of which is a valid dictionary by this specification. | |
fed5d032 | 240 | |
e0b9d78f S |
241 | Additionally, playlists can have "title", "description" and "id" attributes |
242 | with the same semantics as videos (see above). | |
fed5d032 PH |
243 | |
244 | ||
245 | _type "multi_video" indicates that there are multiple videos that | |
246 | form a single show, for examples multiple acts of an opera or TV episode. | |
247 | It must have an entries key like a playlist and contain all the keys | |
248 | required for a video at the same time. | |
249 | ||
250 | ||
251 | _type "url" indicates that the video must be extracted from another | |
252 | location, possibly by a different extractor. Its only required key is: | |
253 | "url" - the next URL to extract. | |
f58766ce PH |
254 | The key "ie_key" can be set to the class name (minus the trailing "IE", |
255 | e.g. "Youtube") if the extractor class is known in advance. | |
256 | Additionally, the dictionary may have any properties of the resolved entity | |
257 | known in advance, for example "title" if the title of the referred video is | |
fed5d032 PH |
258 | known ahead of time. |
259 | ||
260 | ||
261 | _type "url_transparent" entities have the same specification as "url", but | |
262 | indicate that the given additional information is more precise than the one | |
263 | associated with the resolved URL. | |
264 | This is useful when a site employs a video service that hosts the video and | |
265 | its technical metadata, but that video service does not embed a useful | |
266 | title, description etc. | |
267 | ||
268 | ||
d6983cb4 PH |
269 | Subclasses of this one should re-define the _real_initialize() and |
270 | _real_extract() methods and define a _VALID_URL regexp. | |
271 | Probably, they should also be added to the list of extractors. | |
272 | ||
d6983cb4 PH |
273 | Finally, the _WORKING attribute should be set to False for broken IEs |
274 | in order to warn the users and skip the tests. | |
275 | """ | |
276 | ||
277 | _ready = False | |
278 | _downloader = None | |
279 | _WORKING = True | |
280 | ||
281 | def __init__(self, downloader=None): | |
282 | """Constructor. Receives an optional downloader.""" | |
283 | self._ready = False | |
284 | self.set_downloader(downloader) | |
285 | ||
286 | @classmethod | |
287 | def suitable(cls, url): | |
288 | """Receives a URL and returns True if suitable for this IE.""" | |
79cb2577 PH |
289 | |
290 | # This does not use has/getattr intentionally - we want to know whether | |
291 | # we have cached the regexp for *this* class, whereas getattr would also | |
292 | # match the superclass | |
293 | if '_VALID_URL_RE' not in cls.__dict__: | |
294 | cls._VALID_URL_RE = re.compile(cls._VALID_URL) | |
295 | return cls._VALID_URL_RE.match(url) is not None | |
d6983cb4 | 296 | |
ed9266db PH |
297 | @classmethod |
298 | def _match_id(cls, url): | |
299 | if '_VALID_URL_RE' not in cls.__dict__: | |
300 | cls._VALID_URL_RE = re.compile(cls._VALID_URL) | |
301 | m = cls._VALID_URL_RE.match(url) | |
302 | assert m | |
303 | return m.group('id') | |
304 | ||
d6983cb4 PH |
305 | @classmethod |
306 | def working(cls): | |
307 | """Getter method for _WORKING.""" | |
308 | return cls._WORKING | |
309 | ||
310 | def initialize(self): | |
311 | """Initializes an instance (authentication, etc).""" | |
312 | if not self._ready: | |
313 | self._real_initialize() | |
314 | self._ready = True | |
315 | ||
316 | def extract(self, url): | |
317 | """Extracts URL information and returns it in list of dicts.""" | |
3a5bcd03 PH |
318 | try: |
319 | self.initialize() | |
320 | return self._real_extract(url) | |
321 | except ExtractorError: | |
322 | raise | |
323 | except compat_http_client.IncompleteRead as e: | |
dfb1b146 | 324 | raise ExtractorError('A network error has occurred.', cause=e, expected=True) |
9650885b | 325 | except (KeyError, StopIteration) as e: |
dfb1b146 | 326 | raise ExtractorError('An extractor error has occurred.', cause=e) |
d6983cb4 PH |
327 | |
328 | def set_downloader(self, downloader): | |
329 | """Sets the downloader for this IE.""" | |
330 | self._downloader = downloader | |
331 | ||
332 | def _real_initialize(self): | |
333 | """Real initialization process. Redefine in subclasses.""" | |
334 | pass | |
335 | ||
336 | def _real_extract(self, url): | |
337 | """Real extraction process. Redefine in subclasses.""" | |
338 | pass | |
339 | ||
56c73665 JMF |
340 | @classmethod |
341 | def ie_key(cls): | |
342 | """A string for getting the InfoExtractor with get_info_extractor""" | |
dc519b54 | 343 | return compat_str(cls.__name__[:-2]) |
56c73665 | 344 | |
d6983cb4 PH |
345 | @property |
346 | def IE_NAME(self): | |
dc519b54 | 347 | return compat_str(type(self).__name__[:-2]) |
d6983cb4 | 348 | |
cdfee168 | 349 | def _request_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True, data=None, headers=None, query=None): |
d6983cb4 PH |
350 | """ Returns the response handle """ |
351 | if note is None: | |
352 | self.report_download_webpage(video_id) | |
353 | elif note is not False: | |
7cc3570e | 354 | if video_id is None: |
f1a9d64e | 355 | self.to_screen('%s' % (note,)) |
7cc3570e | 356 | else: |
f1a9d64e | 357 | self.to_screen('%s: %s' % (video_id, note)) |
cdfee168 | 358 | # data, headers and query params will be ignored for `Request` objects |
359 | if isinstance(url_or_request, compat_str): | |
360 | if query: | |
361 | url_or_request = update_url_query(url_or_request, query) | |
362 | if data or headers: | |
363 | url_or_request = sanitized_Request(url_or_request, data, headers or {}) | |
d6983cb4 | 364 | try: |
dca08720 | 365 | return self._downloader.urlopen(url_or_request) |
d6983cb4 | 366 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: |
aa94a6d3 PH |
367 | if errnote is False: |
368 | return False | |
d6983cb4 | 369 | if errnote is None: |
f1a9d64e | 370 | errnote = 'Unable to download webpage' |
7f8b2714 | 371 | |
9b9c5355 | 372 | errmsg = '%s: %s' % (errnote, error_to_compat_str(err)) |
7cc3570e PH |
373 | if fatal: |
374 | raise ExtractorError(errmsg, sys.exc_info()[2], cause=err) | |
375 | else: | |
376 | self._downloader.report_warning(errmsg) | |
377 | return False | |
d6983cb4 | 378 | |
cdfee168 | 379 | def _download_webpage_handle(self, url_or_request, video_id, note=None, errnote=None, fatal=True, encoding=None, data=None, headers=None, query=None): |
d6983cb4 | 380 | """ Returns a tuple (page content as string, URL handle) """ |
b9d3e163 PH |
381 | # Strip hashes from the URL (#1038) |
382 | if isinstance(url_or_request, (compat_str, str)): | |
383 | url_or_request = url_or_request.partition('#')[0] | |
384 | ||
cdfee168 | 385 | urlh = self._request_webpage(url_or_request, video_id, note, errnote, fatal, data=data, headers=headers, query=query) |
7cc3570e PH |
386 | if urlh is False: |
387 | assert not fatal | |
388 | return False | |
c9a77969 | 389 | content = self._webpage_read_content(urlh, url_or_request, video_id, note, errnote, fatal, encoding=encoding) |
23be51d8 PH |
390 | return (content, urlh) |
391 | ||
c9a77969 YCH |
392 | @staticmethod |
393 | def _guess_encoding_from_content(content_type, webpage_bytes): | |
d6983cb4 PH |
394 | m = re.match(r'[a-zA-Z0-9_.-]+/[a-zA-Z0-9_.-]+\s*;\s*charset=(.+)', content_type) |
395 | if m: | |
396 | encoding = m.group(1) | |
397 | else: | |
0d75ae2c | 398 | m = re.search(br'<meta[^>]+charset=[\'"]?([^\'")]+)[ /\'">]', |
f143d86a PH |
399 | webpage_bytes[:1024]) |
400 | if m: | |
401 | encoding = m.group(1).decode('ascii') | |
b60016e8 PH |
402 | elif webpage_bytes.startswith(b'\xff\xfe'): |
403 | encoding = 'utf-16' | |
f143d86a PH |
404 | else: |
405 | encoding = 'utf-8' | |
c9a77969 YCH |
406 | |
407 | return encoding | |
408 | ||
409 | def _webpage_read_content(self, urlh, url_or_request, video_id, note=None, errnote=None, fatal=True, prefix=None, encoding=None): | |
410 | content_type = urlh.headers.get('Content-Type', '') | |
411 | webpage_bytes = urlh.read() | |
412 | if prefix is not None: | |
413 | webpage_bytes = prefix + webpage_bytes | |
414 | if not encoding: | |
415 | encoding = self._guess_encoding_from_content(content_type, webpage_bytes) | |
d6983cb4 PH |
416 | if self._downloader.params.get('dump_intermediate_pages', False): |
417 | try: | |
418 | url = url_or_request.get_full_url() | |
419 | except AttributeError: | |
420 | url = url_or_request | |
f1a9d64e | 421 | self.to_screen('Dumping request to ' + url) |
d6983cb4 PH |
422 | dump = base64.b64encode(webpage_bytes).decode('ascii') |
423 | self._downloader.to_screen(dump) | |
d41e6efc PH |
424 | if self._downloader.params.get('write_pages', False): |
425 | try: | |
426 | url = url_or_request.get_full_url() | |
427 | except AttributeError: | |
428 | url = url_or_request | |
5afa7f8b | 429 | basen = '%s_%s' % (video_id, url) |
c1bce22f | 430 | if len(basen) > 240: |
f1a9d64e | 431 | h = '___' + hashlib.md5(basen.encode('utf-8')).hexdigest() |
c1bce22f PH |
432 | basen = basen[:240 - len(h)] + h |
433 | raw_filename = basen + '.dump' | |
d41e6efc | 434 | filename = sanitize_filename(raw_filename, restricted=True) |
f1a9d64e | 435 | self.to_screen('Saving request to ' + filename) |
5f58165d S |
436 | # Working around MAX_PATH limitation on Windows (see |
437 | # http://msdn.microsoft.com/en-us/library/windows/desktop/aa365247(v=vs.85).aspx) | |
e9c0cdd3 | 438 | if compat_os_name == 'nt': |
5f58165d S |
439 | absfilepath = os.path.abspath(filename) |
440 | if len(absfilepath) > 259: | |
441 | filename = '\\\\?\\' + absfilepath | |
d41e6efc PH |
442 | with open(filename, 'wb') as outf: |
443 | outf.write(webpage_bytes) | |
444 | ||
ec0fafbb AA |
445 | try: |
446 | content = webpage_bytes.decode(encoding, 'replace') | |
447 | except LookupError: | |
448 | content = webpage_bytes.decode('utf-8', 'replace') | |
2410c43d | 449 | |
f1a9d64e PH |
450 | if ('<title>Access to this site is blocked</title>' in content and |
451 | 'Websense' in content[:512]): | |
452 | msg = 'Access to this webpage has been blocked by Websense filtering software in your network.' | |
2410c43d PH |
453 | blocked_iframe = self._html_search_regex( |
454 | r'<iframe src="([^"]+)"', content, | |
f1a9d64e | 455 | 'Websense information URL', default=None) |
2410c43d | 456 | if blocked_iframe: |
f1a9d64e | 457 | msg += ' Visit %s for more details' % blocked_iframe |
2410c43d | 458 | raise ExtractorError(msg, expected=True) |
77b2986b PH |
459 | if '<title>The URL you requested has been blocked</title>' in content[:512]: |
460 | msg = ( | |
461 | 'Access to this webpage has been blocked by Indian censorship. ' | |
462 | 'Use a VPN or proxy server (with --proxy) to route around it.') | |
463 | block_msg = self._html_search_regex( | |
464 | r'</h1><p>(.*?)</p>', | |
465 | content, 'block message', default=None) | |
466 | if block_msg: | |
467 | msg += ' (Message: "%s")' % block_msg.replace('\n', ' ') | |
468 | raise ExtractorError(msg, expected=True) | |
2410c43d | 469 | |
23be51d8 | 470 | return content |
d6983cb4 | 471 | |
cdfee168 | 472 | def _download_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True, tries=1, timeout=5, encoding=None, data=None, headers=None, query=None): |
d6983cb4 | 473 | """ Returns the data of the page as a string """ |
995ad69c TF |
474 | success = False |
475 | try_count = 0 | |
476 | while success is False: | |
477 | try: | |
cdfee168 | 478 | res = self._download_webpage_handle(url_or_request, video_id, note, errnote, fatal, encoding=encoding, data=data, headers=headers, query=query) |
995ad69c TF |
479 | success = True |
480 | except compat_http_client.IncompleteRead as e: | |
481 | try_count += 1 | |
482 | if try_count >= tries: | |
483 | raise e | |
484 | self._sleep(timeout, video_id) | |
7cc3570e PH |
485 | if res is False: |
486 | return res | |
487 | else: | |
488 | content, _ = res | |
489 | return content | |
d6983cb4 | 490 | |
2a275ab0 | 491 | def _download_xml(self, url_or_request, video_id, |
f1a9d64e | 492 | note='Downloading XML', errnote='Unable to download XML', |
cdfee168 | 493 | transform_source=None, fatal=True, encoding=None, data=None, headers=None, query=None): |
267ed0c5 | 494 | """Return the xml as an xml.etree.ElementTree.Element""" |
28746fbd | 495 | xml_string = self._download_webpage( |
cdfee168 | 496 | url_or_request, video_id, note, errnote, fatal=fatal, encoding=encoding, data=data, headers=headers, query=query) |
28746fbd PH |
497 | if xml_string is False: |
498 | return xml_string | |
e2b38da9 PH |
499 | if transform_source: |
500 | xml_string = transform_source(xml_string) | |
36e6f62c | 501 | return compat_etree_fromstring(xml_string.encode('utf-8')) |
267ed0c5 | 502 | |
3d3538e4 | 503 | def _download_json(self, url_or_request, video_id, |
f1a9d64e PH |
504 | note='Downloading JSON metadata', |
505 | errnote='Unable to download JSON metadata', | |
b090af59 | 506 | transform_source=None, |
cdfee168 | 507 | fatal=True, encoding=None, data=None, headers=None, query=None): |
b090af59 | 508 | json_string = self._download_webpage( |
c9a77969 | 509 | url_or_request, video_id, note, errnote, fatal=fatal, |
cdfee168 | 510 | encoding=encoding, data=data, headers=headers, query=query) |
b090af59 PH |
511 | if (not fatal) and json_string is False: |
512 | return None | |
ebb64199 TF |
513 | return self._parse_json( |
514 | json_string, video_id, transform_source=transform_source, fatal=fatal) | |
515 | ||
516 | def _parse_json(self, json_string, video_id, transform_source=None, fatal=True): | |
81c2f20b PH |
517 | if transform_source: |
518 | json_string = transform_source(json_string) | |
3d3538e4 PH |
519 | try: |
520 | return json.loads(json_string) | |
521 | except ValueError as ve: | |
e7b6d122 PH |
522 | errmsg = '%s: Failed to parse JSON ' % video_id |
523 | if fatal: | |
524 | raise ExtractorError(errmsg, cause=ve) | |
525 | else: | |
526 | self.report_warning(errmsg + str(ve)) | |
3d3538e4 | 527 | |
f45f96f8 | 528 | def report_warning(self, msg, video_id=None): |
f1a9d64e | 529 | idstr = '' if video_id is None else '%s: ' % video_id |
f45f96f8 | 530 | self._downloader.report_warning( |
f1a9d64e | 531 | '[%s] %s%s' % (self.IE_NAME, idstr, msg)) |
f45f96f8 | 532 | |
d6983cb4 PH |
533 | def to_screen(self, msg): |
534 | """Print msg to screen, prefixing it with '[ie_name]'""" | |
f1a9d64e | 535 | self._downloader.to_screen('[%s] %s' % (self.IE_NAME, msg)) |
d6983cb4 PH |
536 | |
537 | def report_extraction(self, id_or_name): | |
538 | """Report information extraction.""" | |
f1a9d64e | 539 | self.to_screen('%s: Extracting information' % id_or_name) |
d6983cb4 PH |
540 | |
541 | def report_download_webpage(self, video_id): | |
542 | """Report webpage download.""" | |
f1a9d64e | 543 | self.to_screen('%s: Downloading webpage' % video_id) |
d6983cb4 PH |
544 | |
545 | def report_age_confirmation(self): | |
546 | """Report attempt to confirm age.""" | |
f1a9d64e | 547 | self.to_screen('Confirming age') |
d6983cb4 | 548 | |
fc79158d JMF |
549 | def report_login(self): |
550 | """Report attempt to log in.""" | |
f1a9d64e | 551 | self.to_screen('Logging in') |
fc79158d | 552 | |
43e7d3c9 S |
553 | @staticmethod |
554 | def raise_login_required(msg='This video is only available for registered users'): | |
555 | raise ExtractorError( | |
556 | '%s. Use --username and --password or --netrc to provide account credentials.' % msg, | |
557 | expected=True) | |
558 | ||
c430802e S |
559 | @staticmethod |
560 | def raise_geo_restricted(msg='This video is not available from your location due to geo restriction'): | |
561 | raise ExtractorError( | |
562 | '%s. You might want to use --proxy to workaround.' % msg, | |
563 | expected=True) | |
564 | ||
5f6a1245 | 565 | # Methods for following #608 |
c0d0b01f | 566 | @staticmethod |
830d53bf | 567 | def url_result(url, ie=None, video_id=None, video_title=None): |
10952eb2 | 568 | """Returns a URL that points to a page that should be processed""" |
5f6a1245 | 569 | # TODO: ie should be the class used for getting the info |
d6983cb4 PH |
570 | video_info = {'_type': 'url', |
571 | 'url': url, | |
572 | 'ie_key': ie} | |
7012b23c PH |
573 | if video_id is not None: |
574 | video_info['id'] = video_id | |
830d53bf S |
575 | if video_title is not None: |
576 | video_info['title'] = video_title | |
d6983cb4 | 577 | return video_info |
5f6a1245 | 578 | |
c0d0b01f | 579 | @staticmethod |
acf5cbfe | 580 | def playlist_result(entries, playlist_id=None, playlist_title=None, playlist_description=None): |
d6983cb4 PH |
581 | """Returns a playlist""" |
582 | video_info = {'_type': 'playlist', | |
583 | 'entries': entries} | |
584 | if playlist_id: | |
585 | video_info['id'] = playlist_id | |
586 | if playlist_title: | |
587 | video_info['title'] = playlist_title | |
acf5cbfe S |
588 | if playlist_description: |
589 | video_info['description'] = playlist_description | |
d6983cb4 PH |
590 | return video_info |
591 | ||
c342041f | 592 | def _search_regex(self, pattern, string, name, default=NO_DEFAULT, fatal=True, flags=0, group=None): |
d6983cb4 PH |
593 | """ |
594 | Perform a regex search on the given string, using a single or a list of | |
595 | patterns returning the first matching group. | |
596 | In case of failure return a default value or raise a WARNING or a | |
55b3e45b | 597 | RegexNotFoundError, depending on fatal, specifying the field name. |
d6983cb4 PH |
598 | """ |
599 | if isinstance(pattern, (str, compat_str, compiled_regex_type)): | |
600 | mobj = re.search(pattern, string, flags) | |
601 | else: | |
602 | for p in pattern: | |
603 | mobj = re.search(p, string, flags) | |
c3415d1b PH |
604 | if mobj: |
605 | break | |
d6983cb4 | 606 | |
e9c0cdd3 | 607 | if not self._downloader.params.get('no_color') and compat_os_name != 'nt' and sys.stderr.isatty(): |
f1a9d64e | 608 | _name = '\033[0;34m%s\033[0m' % name |
d6983cb4 PH |
609 | else: |
610 | _name = name | |
611 | ||
612 | if mobj: | |
711ede6e PH |
613 | if group is None: |
614 | # return the first matching group | |
615 | return next(g for g in mobj.groups() if g is not None) | |
616 | else: | |
617 | return mobj.group(group) | |
c342041f | 618 | elif default is not NO_DEFAULT: |
d6983cb4 PH |
619 | return default |
620 | elif fatal: | |
f1a9d64e | 621 | raise RegexNotFoundError('Unable to extract %s' % _name) |
d6983cb4 | 622 | else: |
08f2a92c | 623 | self._downloader.report_warning('unable to extract %s' % _name + bug_reports_message()) |
d6983cb4 PH |
624 | return None |
625 | ||
c342041f | 626 | def _html_search_regex(self, pattern, string, name, default=NO_DEFAULT, fatal=True, flags=0, group=None): |
d6983cb4 PH |
627 | """ |
628 | Like _search_regex, but strips HTML tags and unescapes entities. | |
629 | """ | |
711ede6e | 630 | res = self._search_regex(pattern, string, name, default, fatal, flags, group) |
d6983cb4 PH |
631 | if res: |
632 | return clean_html(res).strip() | |
633 | else: | |
634 | return res | |
635 | ||
fc79158d JMF |
636 | def _get_login_info(self): |
637 | """ | |
cf0649f8 | 638 | Get the login info as (username, password) |
fc79158d JMF |
639 | It will look in the netrc file using the _NETRC_MACHINE value |
640 | If there's no info available, return (None, None) | |
641 | """ | |
642 | if self._downloader is None: | |
643 | return (None, None) | |
644 | ||
645 | username = None | |
646 | password = None | |
647 | downloader_params = self._downloader.params | |
648 | ||
649 | # Attempt to use provided username and password or .netrc data | |
d800609c | 650 | if downloader_params.get('username') is not None: |
fc79158d JMF |
651 | username = downloader_params['username'] |
652 | password = downloader_params['password'] | |
653 | elif downloader_params.get('usenetrc', False): | |
654 | try: | |
655 | info = netrc.netrc().authenticators(self._NETRC_MACHINE) | |
656 | if info is not None: | |
657 | username = info[0] | |
658 | password = info[2] | |
659 | else: | |
660 | raise netrc.NetrcParseError('No authenticators for %s' % self._NETRC_MACHINE) | |
661 | except (IOError, netrc.NetrcParseError) as err: | |
9b9c5355 | 662 | self._downloader.report_warning('parsing .netrc: %s' % error_to_compat_str(err)) |
5f6a1245 | 663 | |
fc79158d JMF |
664 | return (username, password) |
665 | ||
e64b7569 | 666 | def _get_tfa_info(self, note='two-factor verification code'): |
83317f69 | 667 | """ |
668 | Get the two-factor authentication info | |
669 | TODO - asking the user will be required for sms/phone verify | |
670 | currently just uses the command line option | |
671 | If there's no info available, return None | |
672 | """ | |
673 | if self._downloader is None: | |
83317f69 | 674 | return None |
675 | downloader_params = self._downloader.params | |
676 | ||
d800609c | 677 | if downloader_params.get('twofactor') is not None: |
83317f69 | 678 | return downloader_params['twofactor'] |
679 | ||
e64b7569 | 680 | return compat_getpass('Type %s and press [Return]: ' % note) |
83317f69 | 681 | |
46720279 JMF |
682 | # Helper functions for extracting OpenGraph info |
683 | @staticmethod | |
ab2d5247 | 684 | def _og_regexes(prop): |
448ef1f3 | 685 | content_re = r'content=(?:"([^"]+?)"|\'([^\']+?)\'|\s*([^\s"\'=<>`]+?))' |
7a6d76a6 S |
686 | property_re = (r'(?:name|property)=(?:\'og:%(prop)s\'|"og:%(prop)s"|\s*og:%(prop)s\b)' |
687 | % {'prop': re.escape(prop)}) | |
78fb87b2 | 688 | template = r'<meta[^>]+?%s[^>]+?%s' |
ab2d5247 | 689 | return [ |
78fb87b2 JMF |
690 | template % (property_re, content_re), |
691 | template % (content_re, property_re), | |
ab2d5247 | 692 | ] |
46720279 | 693 | |
864f24bd S |
694 | @staticmethod |
695 | def _meta_regex(prop): | |
696 | return r'''(?isx)<meta | |
8b9848ac | 697 | (?=[^>]+(?:itemprop|name|property|id|http-equiv)=(["\']?)%s\1) |
864f24bd S |
698 | [^>]+?content=(["\'])(?P<content>.*?)\2''' % re.escape(prop) |
699 | ||
3c4e6d83 | 700 | def _og_search_property(self, prop, html, name=None, **kargs): |
46720279 | 701 | if name is None: |
3c4e6d83 | 702 | name = 'OpenGraph %s' % prop |
ab2d5247 | 703 | escaped = self._search_regex(self._og_regexes(prop), html, name, flags=re.DOTALL, **kargs) |
eb0a8398 PH |
704 | if escaped is None: |
705 | return None | |
706 | return unescapeHTML(escaped) | |
46720279 JMF |
707 | |
708 | def _og_search_thumbnail(self, html, **kargs): | |
10952eb2 | 709 | return self._og_search_property('image', html, 'thumbnail URL', fatal=False, **kargs) |
46720279 JMF |
710 | |
711 | def _og_search_description(self, html, **kargs): | |
712 | return self._og_search_property('description', html, fatal=False, **kargs) | |
713 | ||
714 | def _og_search_title(self, html, **kargs): | |
715 | return self._og_search_property('title', html, **kargs) | |
716 | ||
8ffa13e0 | 717 | def _og_search_video_url(self, html, name='video url', secure=True, **kargs): |
a3681973 PH |
718 | regexes = self._og_regexes('video') + self._og_regexes('video:url') |
719 | if secure: | |
720 | regexes = self._og_regexes('video:secure_url') + regexes | |
8ffa13e0 | 721 | return self._html_search_regex(regexes, html, name, **kargs) |
46720279 | 722 | |
78338f71 JMF |
723 | def _og_search_url(self, html, **kargs): |
724 | return self._og_search_property('url', html, **kargs) | |
725 | ||
40c696e5 | 726 | def _html_search_meta(self, name, html, display_name=None, fatal=False, **kwargs): |
59040888 PH |
727 | if display_name is None: |
728 | display_name = name | |
729 | return self._html_search_regex( | |
864f24bd | 730 | self._meta_regex(name), |
711ede6e | 731 | html, display_name, fatal=fatal, group='content', **kwargs) |
59040888 PH |
732 | |
733 | def _dc_search_uploader(self, html): | |
734 | return self._html_search_meta('dc.creator', html, 'uploader') | |
735 | ||
8dbe9899 PH |
736 | def _rta_search(self, html): |
737 | # See http://www.rtalabel.org/index.php?content=howtofaq#single | |
738 | if re.search(r'(?ix)<meta\s+name="rating"\s+' | |
739 | r' content="RTA-5042-1996-1400-1577-RTA"', | |
740 | html): | |
741 | return 18 | |
742 | return 0 | |
743 | ||
59040888 PH |
744 | def _media_rating_search(self, html): |
745 | # See http://www.tjg-designs.com/WP/metadata-code-examples-adding-metadata-to-your-web-pages/ | |
746 | rating = self._html_search_meta('rating', html) | |
747 | ||
748 | if not rating: | |
749 | return None | |
750 | ||
751 | RATING_TABLE = { | |
752 | 'safe for kids': 0, | |
753 | 'general': 8, | |
754 | '14 years': 14, | |
755 | 'mature': 17, | |
756 | 'restricted': 19, | |
757 | } | |
d800609c | 758 | return RATING_TABLE.get(rating.lower()) |
59040888 | 759 | |
69319969 | 760 | def _family_friendly_search(self, html): |
6ca7732d | 761 | # See http://schema.org/VideoObject |
69319969 NJ |
762 | family_friendly = self._html_search_meta('isFamilyFriendly', html) |
763 | ||
764 | if not family_friendly: | |
765 | return None | |
766 | ||
767 | RATING_TABLE = { | |
768 | '1': 0, | |
769 | 'true': 0, | |
770 | '0': 18, | |
771 | 'false': 18, | |
772 | } | |
d800609c | 773 | return RATING_TABLE.get(family_friendly.lower()) |
69319969 | 774 | |
0c708f11 JMF |
775 | def _twitter_search_player(self, html): |
776 | return self._html_search_meta('twitter:player', html, | |
9e1a5b84 | 777 | 'twitter card player') |
0c708f11 | 778 | |
0b26ba3f | 779 | def _search_json_ld(self, html, video_id, **kwargs): |
4ca2a3cf S |
780 | json_ld = self._search_regex( |
781 | r'(?s)<script[^>]+type=(["\'])application/ld\+json\1[^>]*>(?P<json_ld>.+?)</script>', | |
0b26ba3f | 782 | html, 'JSON-LD', group='json_ld', **kwargs) |
4ca2a3cf S |
783 | if not json_ld: |
784 | return {} | |
0b26ba3f | 785 | return self._json_ld(json_ld, video_id, fatal=kwargs.get('fatal', True)) |
4ca2a3cf S |
786 | |
787 | def _json_ld(self, json_ld, video_id, fatal=True): | |
788 | if isinstance(json_ld, compat_str): | |
789 | json_ld = self._parse_json(json_ld, video_id, fatal=fatal) | |
790 | if not json_ld: | |
791 | return {} | |
792 | info = {} | |
793 | if json_ld.get('@context') == 'http://schema.org': | |
794 | item_type = json_ld.get('@type') | |
795 | if item_type == 'TVEpisode': | |
796 | info.update({ | |
797 | 'episode': unescapeHTML(json_ld.get('name')), | |
798 | 'episode_number': int_or_none(json_ld.get('episodeNumber')), | |
799 | 'description': unescapeHTML(json_ld.get('description')), | |
800 | }) | |
801 | part_of_season = json_ld.get('partOfSeason') | |
802 | if isinstance(part_of_season, dict) and part_of_season.get('@type') == 'TVSeason': | |
803 | info['season_number'] = int_or_none(part_of_season.get('seasonNumber')) | |
804 | part_of_series = json_ld.get('partOfSeries') | |
805 | if isinstance(part_of_series, dict) and part_of_series.get('@type') == 'TVSeries': | |
806 | info['series'] = unescapeHTML(part_of_series.get('name')) | |
807 | elif item_type == 'Article': | |
808 | info.update({ | |
809 | 'timestamp': parse_iso8601(json_ld.get('datePublished')), | |
810 | 'title': unescapeHTML(json_ld.get('headline')), | |
811 | 'description': unescapeHTML(json_ld.get('articleBody')), | |
812 | }) | |
813 | return dict((k, v) for k, v in info.items() if v is not None) | |
814 | ||
27713812 | 815 | @staticmethod |
f8da79f8 | 816 | def _hidden_inputs(html): |
586f1cc5 | 817 | html = re.sub(r'<!--(?:(?!<!--).)*-->', '', html) |
201ea3ee | 818 | hidden_inputs = {} |
73eb13df | 819 | for input in re.findall(r'(?i)<input([^>]+)>', html): |
be0e5dbd | 820 | if not re.search(r'type=(["\'])(?:hidden|submit)\1', input): |
201ea3ee S |
821 | continue |
822 | name = re.search(r'name=(["\'])(?P<value>.+?)\1', input) | |
823 | if not name: | |
824 | continue | |
825 | value = re.search(r'value=(["\'])(?P<value>.*?)\1', input) | |
826 | if not value: | |
827 | continue | |
828 | hidden_inputs[name.group('value')] = value.group('value') | |
829 | return hidden_inputs | |
27713812 | 830 | |
cf61d96d S |
831 | def _form_hidden_inputs(self, form_id, html): |
832 | form = self._search_regex( | |
73eb13df | 833 | r'(?is)<form[^>]+?id=(["\'])%s\1[^>]*>(?P<form>.+?)</form>' % form_id, |
cf61d96d S |
834 | html, '%s form' % form_id, group='form') |
835 | return self._hidden_inputs(form) | |
836 | ||
3ded7bac | 837 | def _sort_formats(self, formats, field_preference=None): |
7e8caf30 | 838 | if not formats: |
f1a9d64e | 839 | raise ExtractorError('No video formats found') |
7e8caf30 | 840 | |
b0d21ded S |
841 | for f in formats: |
842 | # Automatically determine tbr when missing based on abr and vbr (improves | |
843 | # formats sorting in some cases) | |
350cf045 | 844 | if 'tbr' not in f and f.get('abr') is not None and f.get('vbr') is not None: |
b0d21ded S |
845 | f['tbr'] = f['abr'] + f['vbr'] |
846 | ||
4bcc7bd1 | 847 | def _formats_key(f): |
e6812ac9 PH |
848 | # TODO remove the following workaround |
849 | from ..utils import determine_ext | |
850 | if not f.get('ext') and 'url' in f: | |
851 | f['ext'] = determine_ext(f['url']) | |
852 | ||
3ded7bac S |
853 | if isinstance(field_preference, (list, tuple)): |
854 | return tuple(f.get(field) if f.get(field) is not None else -1 for field in field_preference) | |
855 | ||
4bcc7bd1 PH |
856 | preference = f.get('preference') |
857 | if preference is None: | |
d497a201 | 858 | preference = 0 |
4bcc7bd1 PH |
859 | if f.get('ext') in ['f4f', 'f4m']: # Not yet supported |
860 | preference -= 0.5 | |
861 | ||
d497a201 | 862 | proto_preference = 0 if determine_protocol(f) in ['http', 'https'] else -0.1 |
863 | ||
4bcc7bd1 PH |
864 | if f.get('vcodec') == 'none': # audio only |
865 | if self._downloader.params.get('prefer_free_formats'): | |
f1a9d64e | 866 | ORDER = ['aac', 'mp3', 'm4a', 'webm', 'ogg', 'opus'] |
4bcc7bd1 | 867 | else: |
f1a9d64e | 868 | ORDER = ['webm', 'opus', 'ogg', 'mp3', 'aac', 'm4a'] |
4bcc7bd1 PH |
869 | ext_preference = 0 |
870 | try: | |
871 | audio_ext_preference = ORDER.index(f['ext']) | |
872 | except ValueError: | |
873 | audio_ext_preference = -1 | |
874 | else: | |
875 | if self._downloader.params.get('prefer_free_formats'): | |
f1a9d64e | 876 | ORDER = ['flv', 'mp4', 'webm'] |
4bcc7bd1 | 877 | else: |
f1a9d64e | 878 | ORDER = ['webm', 'flv', 'mp4'] |
4bcc7bd1 PH |
879 | try: |
880 | ext_preference = ORDER.index(f['ext']) | |
881 | except ValueError: | |
882 | ext_preference = -1 | |
883 | audio_ext_preference = 0 | |
884 | ||
885 | return ( | |
886 | preference, | |
aff2f4f4 | 887 | f.get('language_preference') if f.get('language_preference') is not None else -1, |
5d73273f | 888 | f.get('quality') if f.get('quality') is not None else -1, |
9933b574 | 889 | f.get('tbr') if f.get('tbr') is not None else -1, |
03cd72b0 | 890 | f.get('filesize') if f.get('filesize') is not None else -1, |
4bcc7bd1 | 891 | f.get('vbr') if f.get('vbr') is not None else -1, |
1a6373ef PH |
892 | f.get('height') if f.get('height') is not None else -1, |
893 | f.get('width') if f.get('width') is not None else -1, | |
d497a201 | 894 | proto_preference, |
1e1896f2 | 895 | ext_preference, |
4bcc7bd1 PH |
896 | f.get('abr') if f.get('abr') is not None else -1, |
897 | audio_ext_preference, | |
2c8e03d9 | 898 | f.get('fps') if f.get('fps') is not None else -1, |
9732d77e | 899 | f.get('filesize_approx') if f.get('filesize_approx') is not None else -1, |
c64ed2a3 | 900 | f.get('source_preference') if f.get('source_preference') is not None else -1, |
74f72824 | 901 | f.get('format_id') if f.get('format_id') is not None else '', |
4bcc7bd1 PH |
902 | ) |
903 | formats.sort(key=_formats_key) | |
59040888 | 904 | |
96a53167 S |
905 | def _check_formats(self, formats, video_id): |
906 | if formats: | |
907 | formats[:] = filter( | |
908 | lambda f: self._is_valid_url( | |
909 | f['url'], video_id, | |
910 | item='%s video format' % f.get('format_id') if f.get('format_id') else 'video'), | |
911 | formats) | |
912 | ||
f5bdb444 S |
913 | @staticmethod |
914 | def _remove_duplicate_formats(formats): | |
915 | format_urls = set() | |
916 | unique_formats = [] | |
917 | for f in formats: | |
918 | if f['url'] not in format_urls: | |
919 | format_urls.add(f['url']) | |
920 | unique_formats.append(f) | |
921 | formats[:] = unique_formats | |
922 | ||
96a53167 | 923 | def _is_valid_url(self, url, video_id, item='video'): |
2f0f6578 S |
924 | url = self._proto_relative_url(url, scheme='http:') |
925 | # For now assume non HTTP(S) URLs always valid | |
926 | if not (url.startswith('http://') or url.startswith('https://')): | |
927 | return True | |
96a53167 | 928 | try: |
4069766c | 929 | self._request_webpage(url, video_id, 'Checking %s URL' % item) |
96a53167 S |
930 | return True |
931 | except ExtractorError as e: | |
943a1e24 | 932 | if isinstance(e.cause, compat_urllib_error.URLError): |
baa43cba S |
933 | self.to_screen( |
934 | '%s: %s URL is invalid, skipping' % (video_id, item)) | |
96a53167 S |
935 | return False |
936 | raise | |
937 | ||
20991253 | 938 | def http_scheme(self): |
1ede5b24 | 939 | """ Either "http:" or "https:", depending on the user's preferences """ |
20991253 PH |
940 | return ( |
941 | 'http:' | |
942 | if self._downloader.params.get('prefer_insecure', False) | |
943 | else 'https:') | |
944 | ||
57c7411f PH |
945 | def _proto_relative_url(self, url, scheme=None): |
946 | if url is None: | |
947 | return url | |
948 | if url.startswith('//'): | |
949 | if scheme is None: | |
950 | scheme = self.http_scheme() | |
951 | return scheme + url | |
952 | else: | |
953 | return url | |
954 | ||
4094b6e3 PH |
955 | def _sleep(self, timeout, video_id, msg_template=None): |
956 | if msg_template is None: | |
f1a9d64e | 957 | msg_template = '%(video_id)s: Waiting for %(timeout)s seconds' |
4094b6e3 PH |
958 | msg = msg_template % {'video_id': video_id, 'timeout': timeout} |
959 | self.to_screen(msg) | |
960 | time.sleep(timeout) | |
961 | ||
a38436e8 | 962 | def _extract_f4m_formats(self, manifest_url, video_id, preference=None, f4m_id=None, |
4de61310 S |
963 | transform_source=lambda s: fix_xml_ampersands(s).strip(), |
964 | fatal=True): | |
f036a632 JMF |
965 | manifest = self._download_xml( |
966 | manifest_url, video_id, 'Downloading f4m manifest', | |
97f4aecf S |
967 | 'Unable to download f4m manifest', |
968 | # Some manifests may be malformed, e.g. prosiebensat1 generated manifests | |
969 | # (see https://github.com/rg3/youtube-dl/issues/6215#issuecomment-121704244) | |
4de61310 S |
970 | transform_source=transform_source, |
971 | fatal=fatal) | |
972 | ||
973 | if manifest is False: | |
8d29e47f | 974 | return [] |
31bb8d3f | 975 | |
0fdbb332 S |
976 | return self._parse_f4m_formats( |
977 | manifest, manifest_url, video_id, preference=preference, f4m_id=f4m_id, | |
978 | transform_source=transform_source, fatal=fatal) | |
979 | ||
980 | def _parse_f4m_formats(self, manifest, manifest_url, video_id, preference=None, f4m_id=None, | |
981 | transform_source=lambda s: fix_xml_ampersands(s).strip(), | |
982 | fatal=True): | |
31bb8d3f | 983 | formats = [] |
7a47d07c | 984 | manifest_version = '1.0' |
b2527359 | 985 | media_nodes = manifest.findall('{http://ns.adobe.com/f4m/1.0}media') |
34e48bed | 986 | if not media_nodes: |
7a47d07c | 987 | manifest_version = '2.0' |
34e48bed | 988 | media_nodes = manifest.findall('{http://ns.adobe.com/f4m/2.0}media') |
019839fa S |
989 | base_url = xpath_text( |
990 | manifest, ['{http://ns.adobe.com/f4m/1.0}baseURL', '{http://ns.adobe.com/f4m/2.0}baseURL'], | |
991 | 'base URL', default=None) | |
992 | if base_url: | |
993 | base_url = base_url.strip() | |
b2527359 | 994 | for i, media_el in enumerate(media_nodes): |
7a47d07c | 995 | if manifest_version == '2.0': |
31c746e5 S |
996 | media_url = media_el.attrib.get('href') or media_el.attrib.get('url') |
997 | if not media_url: | |
998 | continue | |
cc357c4d S |
999 | manifest_url = ( |
1000 | media_url if media_url.startswith('http://') or media_url.startswith('https://') | |
019839fa | 1001 | else ((base_url or '/'.join(manifest_url.split('/')[:-1])) + '/' + media_url)) |
70f0f5a8 S |
1002 | # If media_url is itself a f4m manifest do the recursive extraction |
1003 | # since bitrates in parent manifest (this one) and media_url manifest | |
1004 | # may differ leading to inability to resolve the format by requested | |
1005 | # bitrate in f4m downloader | |
1006 | if determine_ext(manifest_url) == 'f4m': | |
7e5edcfd | 1007 | formats.extend(self._extract_f4m_formats( |
0fdbb332 S |
1008 | manifest_url, video_id, preference=preference, f4m_id=f4m_id, |
1009 | transform_source=transform_source, fatal=fatal)) | |
70f0f5a8 | 1010 | continue |
b2527359 | 1011 | tbr = int_or_none(media_el.attrib.get('bitrate')) |
31bb8d3f | 1012 | formats.append({ |
e21a55ab | 1013 | 'format_id': '-'.join(filter(None, [f4m_id, compat_str(i if tbr is None else tbr)])), |
31bb8d3f JMF |
1014 | 'url': manifest_url, |
1015 | 'ext': 'flv', | |
b2527359 | 1016 | 'tbr': tbr, |
31bb8d3f JMF |
1017 | 'width': int_or_none(media_el.attrib.get('width')), |
1018 | 'height': int_or_none(media_el.attrib.get('height')), | |
60ca389c | 1019 | 'preference': preference, |
31bb8d3f JMF |
1020 | }) |
1021 | self._sort_formats(formats) | |
1022 | ||
1023 | return formats | |
1024 | ||
f0b5d6af | 1025 | def _extract_m3u8_formats(self, m3u8_url, video_id, ext=None, |
60ca389c | 1026 | entry_protocol='m3u8', preference=None, |
13af92fd YCH |
1027 | m3u8_id=None, note=None, errnote=None, |
1028 | fatal=True): | |
f0b5d6af | 1029 | |
704df56d | 1030 | formats = [{ |
f207019c | 1031 | 'format_id': '-'.join(filter(None, [m3u8_id, 'meta'])), |
704df56d PH |
1032 | 'url': m3u8_url, |
1033 | 'ext': ext, | |
1034 | 'protocol': 'm3u8', | |
9fe6ef7a | 1035 | 'preference': preference - 1 if preference else -1, |
704df56d PH |
1036 | 'resolution': 'multiple', |
1037 | 'format_note': 'Quality selection URL', | |
1038 | }] | |
1039 | ||
f0b5d6af PH |
1040 | format_url = lambda u: ( |
1041 | u | |
1042 | if re.match(r'^https?://', u) | |
1043 | else compat_urlparse.urljoin(m3u8_url, u)) | |
1044 | ||
dbd82a1d | 1045 | res = self._download_webpage_handle( |
81515ad9 | 1046 | m3u8_url, video_id, |
621ed9f5 | 1047 | note=note or 'Downloading m3u8 information', |
13af92fd YCH |
1048 | errnote=errnote or 'Failed to download m3u8 information', |
1049 | fatal=fatal) | |
dbd82a1d | 1050 | if res is False: |
8d29e47f | 1051 | return [] |
dbd82a1d | 1052 | m3u8_doc, urlh = res |
37113045 | 1053 | m3u8_url = urlh.geturl() |
9cdffeeb S |
1054 | |
1055 | # We should try extracting formats only from master playlists [1], i.e. | |
1056 | # playlists that describe available qualities. On the other hand media | |
1057 | # playlists [2] should be returned as is since they contain just the media | |
1058 | # without qualities renditions. | |
1059 | # Fortunately, master playlist can be easily distinguished from media | |
1060 | # playlist based on particular tags availability. As of [1, 2] master | |
1061 | # playlist tags MUST NOT appear in a media playist and vice versa. | |
1062 | # As of [3] #EXT-X-TARGETDURATION tag is REQUIRED for every media playlist | |
1063 | # and MUST NOT appear in master playlist thus we can clearly detect media | |
1064 | # playlist with this criterion. | |
1065 | # 1. https://tools.ietf.org/html/draft-pantos-http-live-streaming-17#section-4.3.4 | |
1066 | # 2. https://tools.ietf.org/html/draft-pantos-http-live-streaming-17#section-4.3.3 | |
1067 | # 3. https://tools.ietf.org/html/draft-pantos-http-live-streaming-17#section-4.3.3.1 | |
1068 | if '#EXT-X-TARGETDURATION' in m3u8_doc: # media playlist, return as is | |
7f32e5dc | 1069 | return [{ |
1070 | 'url': m3u8_url, | |
1071 | 'format_id': m3u8_id, | |
1072 | 'ext': ext, | |
1073 | 'protocol': entry_protocol, | |
1074 | 'preference': preference, | |
1075 | }] | |
704df56d | 1076 | last_info = None |
fa156077 | 1077 | last_media = None |
704df56d PH |
1078 | kv_rex = re.compile( |
1079 | r'(?P<key>[a-zA-Z_-]+)=(?P<val>"[^"]+"|[^",]+)(?:,|$)') | |
1080 | for line in m3u8_doc.splitlines(): | |
1081 | if line.startswith('#EXT-X-STREAM-INF:'): | |
1082 | last_info = {} | |
1083 | for m in kv_rex.finditer(line): | |
1084 | v = m.group('val') | |
1085 | if v.startswith('"'): | |
1086 | v = v[1:-1] | |
1087 | last_info[m.group('key')] = v | |
4cd95bcb JMF |
1088 | elif line.startswith('#EXT-X-MEDIA:'): |
1089 | last_media = {} | |
1090 | for m in kv_rex.finditer(line): | |
1091 | v = m.group('val') | |
1092 | if v.startswith('"'): | |
1093 | v = v[1:-1] | |
1094 | last_media[m.group('key')] = v | |
704df56d PH |
1095 | elif line.startswith('#') or not line.strip(): |
1096 | continue | |
1097 | else: | |
daebaab6 | 1098 | if last_info is None: |
f0b5d6af | 1099 | formats.append({'url': format_url(line)}) |
3524cc25 | 1100 | continue |
704df56d | 1101 | tbr = int_or_none(last_info.get('BANDWIDTH'), scale=1000) |
8dc9d361 S |
1102 | format_id = [] |
1103 | if m3u8_id: | |
1104 | format_id.append(m3u8_id) | |
05d5392c | 1105 | last_media_name = last_media.get('NAME') if last_media and last_media.get('TYPE') != 'SUBTITLES' else None |
8dc9d361 | 1106 | format_id.append(last_media_name if last_media_name else '%d' % (tbr if tbr else len(formats))) |
704df56d | 1107 | f = { |
8dc9d361 | 1108 | 'format_id': '-'.join(format_id), |
f0b5d6af | 1109 | 'url': format_url(line.strip()), |
704df56d PH |
1110 | 'tbr': tbr, |
1111 | 'ext': ext, | |
f0b5d6af PH |
1112 | 'protocol': entry_protocol, |
1113 | 'preference': preference, | |
704df56d | 1114 | } |
704df56d PH |
1115 | resolution = last_info.get('RESOLUTION') |
1116 | if resolution: | |
1117 | width_str, height_str = resolution.split('x') | |
1118 | f['width'] = int(width_str) | |
1119 | f['height'] = int(height_str) | |
fbb6edd2 S |
1120 | codecs = last_info.get('CODECS') |
1121 | if codecs: | |
1122 | vcodec, acodec = [None] * 2 | |
1123 | va_codecs = codecs.split(',') | |
1124 | if len(va_codecs) == 1: | |
1125 | # Audio only entries usually come with single codec and | |
1126 | # no resolution. For more robustness we also check it to | |
1127 | # be mp4 audio. | |
1128 | if not resolution and va_codecs[0].startswith('mp4a'): | |
1129 | vcodec, acodec = 'none', va_codecs[0] | |
1130 | else: | |
1131 | vcodec = va_codecs[0] | |
1132 | else: | |
1133 | vcodec, acodec = va_codecs[:2] | |
1134 | f.update({ | |
1135 | 'acodec': acodec, | |
1136 | 'vcodec': vcodec, | |
1137 | }) | |
4cd95bcb JMF |
1138 | if last_media is not None: |
1139 | f['m3u8_media'] = last_media | |
1140 | last_media = None | |
704df56d PH |
1141 | formats.append(f) |
1142 | last_info = {} | |
1143 | self._sort_formats(formats) | |
1144 | return formats | |
1145 | ||
a107193e S |
1146 | @staticmethod |
1147 | def _xpath_ns(path, namespace=None): | |
1148 | if not namespace: | |
1149 | return path | |
1150 | out = [] | |
1151 | for c in path.split('/'): | |
1152 | if not c or c == '.': | |
1153 | out.append(c) | |
1154 | else: | |
1155 | out.append('{%s}%s' % (namespace, c)) | |
1156 | return '/'.join(out) | |
1157 | ||
09f572fb | 1158 | def _extract_smil_formats(self, smil_url, video_id, fatal=True, f4m_params=None, transform_source=None): |
1159 | smil = self._download_smil(smil_url, video_id, fatal=fatal, transform_source=transform_source) | |
a107193e | 1160 | |
995029a1 PH |
1161 | if smil is False: |
1162 | assert not fatal | |
1163 | return [] | |
e89a2aab | 1164 | |
17712eeb | 1165 | namespace = self._parse_smil_namespace(smil) |
a107193e S |
1166 | |
1167 | return self._parse_smil_formats( | |
1168 | smil, smil_url, video_id, namespace=namespace, f4m_params=f4m_params) | |
1169 | ||
1170 | def _extract_smil_info(self, smil_url, video_id, fatal=True, f4m_params=None): | |
1171 | smil = self._download_smil(smil_url, video_id, fatal=fatal) | |
1172 | if smil is False: | |
1173 | return {} | |
1174 | return self._parse_smil(smil, smil_url, video_id, f4m_params=f4m_params) | |
1175 | ||
09f572fb | 1176 | def _download_smil(self, smil_url, video_id, fatal=True, transform_source=None): |
a107193e S |
1177 | return self._download_xml( |
1178 | smil_url, video_id, 'Downloading SMIL file', | |
09f572fb | 1179 | 'Unable to download SMIL file', fatal=fatal, transform_source=transform_source) |
a107193e S |
1180 | |
1181 | def _parse_smil(self, smil, smil_url, video_id, f4m_params=None): | |
17712eeb | 1182 | namespace = self._parse_smil_namespace(smil) |
a107193e S |
1183 | |
1184 | formats = self._parse_smil_formats( | |
1185 | smil, smil_url, video_id, namespace=namespace, f4m_params=f4m_params) | |
1186 | subtitles = self._parse_smil_subtitles(smil, namespace=namespace) | |
1187 | ||
1188 | video_id = os.path.splitext(url_basename(smil_url))[0] | |
1189 | title = None | |
1190 | description = None | |
647eab45 | 1191 | upload_date = None |
a107193e S |
1192 | for meta in smil.findall(self._xpath_ns('./head/meta', namespace)): |
1193 | name = meta.attrib.get('name') | |
1194 | content = meta.attrib.get('content') | |
1195 | if not name or not content: | |
1196 | continue | |
1197 | if not title and name == 'title': | |
1198 | title = content | |
1199 | elif not description and name in ('description', 'abstract'): | |
1200 | description = content | |
647eab45 S |
1201 | elif not upload_date and name == 'date': |
1202 | upload_date = unified_strdate(content) | |
a107193e | 1203 | |
1e5bcdec S |
1204 | thumbnails = [{ |
1205 | 'id': image.get('type'), | |
1206 | 'url': image.get('src'), | |
1207 | 'width': int_or_none(image.get('width')), | |
1208 | 'height': int_or_none(image.get('height')), | |
1209 | } for image in smil.findall(self._xpath_ns('.//image', namespace)) if image.get('src')] | |
1210 | ||
a107193e S |
1211 | return { |
1212 | 'id': video_id, | |
1213 | 'title': title or video_id, | |
1214 | 'description': description, | |
647eab45 | 1215 | 'upload_date': upload_date, |
1e5bcdec | 1216 | 'thumbnails': thumbnails, |
a107193e S |
1217 | 'formats': formats, |
1218 | 'subtitles': subtitles, | |
1219 | } | |
1220 | ||
17712eeb S |
1221 | def _parse_smil_namespace(self, smil): |
1222 | return self._search_regex( | |
1223 | r'(?i)^{([^}]+)?}smil$', smil.tag, 'namespace', default=None) | |
1224 | ||
f877c6ae | 1225 | def _parse_smil_formats(self, smil, smil_url, video_id, namespace=None, f4m_params=None, transform_rtmp_url=None): |
a107193e S |
1226 | base = smil_url |
1227 | for meta in smil.findall(self._xpath_ns('./head/meta', namespace)): | |
1228 | b = meta.get('base') or meta.get('httpBase') | |
1229 | if b: | |
1230 | base = b | |
1231 | break | |
e89a2aab S |
1232 | |
1233 | formats = [] | |
1234 | rtmp_count = 0 | |
a107193e | 1235 | http_count = 0 |
7f32e5dc | 1236 | m3u8_count = 0 |
a107193e | 1237 | |
81e1c4e2 | 1238 | srcs = [] |
a107193e S |
1239 | videos = smil.findall(self._xpath_ns('.//video', namespace)) |
1240 | for video in videos: | |
1241 | src = video.get('src') | |
81e1c4e2 | 1242 | if not src or src in srcs: |
a107193e | 1243 | continue |
81e1c4e2 | 1244 | srcs.append(src) |
a107193e | 1245 | |
e7d8e98a | 1246 | bitrate = float_or_none(video.get('system-bitrate') or video.get('systemBitrate'), 1000) |
a107193e S |
1247 | filesize = int_or_none(video.get('size') or video.get('fileSize')) |
1248 | width = int_or_none(video.get('width')) | |
1249 | height = int_or_none(video.get('height')) | |
1250 | proto = video.get('proto') | |
1251 | ext = video.get('ext') | |
1252 | src_ext = determine_ext(src) | |
1253 | streamer = video.get('streamer') or base | |
1254 | ||
1255 | if proto == 'rtmp' or streamer.startswith('rtmp'): | |
1256 | rtmp_count += 1 | |
1257 | formats.append({ | |
1258 | 'url': streamer, | |
1259 | 'play_path': src, | |
1260 | 'ext': 'flv', | |
1261 | 'format_id': 'rtmp-%d' % (rtmp_count if bitrate is None else bitrate), | |
1262 | 'tbr': bitrate, | |
1263 | 'filesize': filesize, | |
1264 | 'width': width, | |
1265 | 'height': height, | |
1266 | }) | |
f877c6ae YCH |
1267 | if transform_rtmp_url: |
1268 | streamer, src = transform_rtmp_url(streamer, src) | |
1269 | formats[-1].update({ | |
1270 | 'url': streamer, | |
1271 | 'play_path': src, | |
1272 | }) | |
a107193e S |
1273 | continue |
1274 | ||
1275 | src_url = src if src.startswith('http') else compat_urlparse.urljoin(base, src) | |
c349456e | 1276 | src_url = src_url.strip() |
a107193e S |
1277 | |
1278 | if proto == 'm3u8' or src_ext == 'm3u8': | |
7f32e5dc | 1279 | m3u8_formats = self._extract_m3u8_formats( |
1280 | src_url, video_id, ext or 'mp4', m3u8_id='hls', fatal=False) | |
1281 | if len(m3u8_formats) == 1: | |
1282 | m3u8_count += 1 | |
1283 | m3u8_formats[0].update({ | |
1284 | 'format_id': 'hls-%d' % (m3u8_count if bitrate is None else bitrate), | |
1285 | 'tbr': bitrate, | |
1286 | 'width': width, | |
1287 | 'height': height, | |
1288 | }) | |
1289 | formats.extend(m3u8_formats) | |
a107193e S |
1290 | continue |
1291 | ||
1292 | if src_ext == 'f4m': | |
1293 | f4m_url = src_url | |
1294 | if not f4m_params: | |
1295 | f4m_params = { | |
1296 | 'hdcore': '3.2.0', | |
1297 | 'plugin': 'flowplayer-3.2.0.1', | |
1298 | } | |
1299 | f4m_url += '&' if '?' in f4m_url else '?' | |
41c3a5a7 | 1300 | f4m_url += compat_urllib_parse.urlencode(f4m_params) |
7e5edcfd | 1301 | formats.extend(self._extract_f4m_formats(f4m_url, video_id, f4m_id='hds', fatal=False)) |
a107193e S |
1302 | continue |
1303 | ||
c78e4817 | 1304 | if src_url.startswith('http') and self._is_valid_url(src, video_id): |
a107193e S |
1305 | http_count += 1 |
1306 | formats.append({ | |
1307 | 'url': src_url, | |
1308 | 'ext': ext or src_ext or 'flv', | |
1309 | 'format_id': 'http-%d' % (bitrate or http_count), | |
1310 | 'tbr': bitrate, | |
1311 | 'filesize': filesize, | |
1312 | 'width': width, | |
1313 | 'height': height, | |
1314 | }) | |
1315 | continue | |
63757032 | 1316 | |
e89a2aab S |
1317 | self._sort_formats(formats) |
1318 | ||
1319 | return formats | |
1320 | ||
ce00af87 | 1321 | def _parse_smil_subtitles(self, smil, namespace=None, subtitles_lang='en'): |
d413095f | 1322 | urls = [] |
a107193e S |
1323 | subtitles = {} |
1324 | for num, textstream in enumerate(smil.findall(self._xpath_ns('.//textstream', namespace))): | |
1325 | src = textstream.get('src') | |
d413095f | 1326 | if not src or src in urls: |
a107193e | 1327 | continue |
d413095f | 1328 | urls.append(src) |
cafcf657 | 1329 | ext = textstream.get('ext') or determine_ext(src) or mimetype2ext(textstream.get('type')) |
03bc7237 | 1330 | lang = textstream.get('systemLanguage') or textstream.get('systemLanguageName') or textstream.get('lang') or subtitles_lang |
a107193e S |
1331 | subtitles.setdefault(lang, []).append({ |
1332 | 'url': src, | |
1333 | 'ext': ext, | |
1334 | }) | |
1335 | return subtitles | |
63757032 | 1336 | |
942acef5 S |
1337 | def _extract_xspf_playlist(self, playlist_url, playlist_id, fatal=True): |
1338 | xspf = self._download_xml( | |
8d6765cf | 1339 | playlist_url, playlist_id, 'Downloading xpsf playlist', |
942acef5 S |
1340 | 'Unable to download xspf manifest', fatal=fatal) |
1341 | if xspf is False: | |
1342 | return [] | |
1343 | return self._parse_xspf(xspf, playlist_id) | |
8d6765cf | 1344 | |
942acef5 | 1345 | def _parse_xspf(self, playlist, playlist_id): |
8d6765cf S |
1346 | NS_MAP = { |
1347 | 'xspf': 'http://xspf.org/ns/0/', | |
1348 | 's1': 'http://static.streamone.nl/player/ns/0', | |
1349 | } | |
1350 | ||
1351 | entries = [] | |
1352 | for track in playlist.findall(xpath_with_ns('./xspf:trackList/xspf:track', NS_MAP)): | |
1353 | title = xpath_text( | |
98044462 | 1354 | track, xpath_with_ns('./xspf:title', NS_MAP), 'title', default=playlist_id) |
8d6765cf S |
1355 | description = xpath_text( |
1356 | track, xpath_with_ns('./xspf:annotation', NS_MAP), 'description') | |
1357 | thumbnail = xpath_text( | |
1358 | track, xpath_with_ns('./xspf:image', NS_MAP), 'thumbnail') | |
1359 | duration = float_or_none( | |
1360 | xpath_text(track, xpath_with_ns('./xspf:duration', NS_MAP), 'duration'), 1000) | |
1361 | ||
1362 | formats = [{ | |
1363 | 'url': location.text, | |
1364 | 'format_id': location.get(xpath_with_ns('s1:label', NS_MAP)), | |
1365 | 'width': int_or_none(location.get(xpath_with_ns('s1:width', NS_MAP))), | |
1366 | 'height': int_or_none(location.get(xpath_with_ns('s1:height', NS_MAP))), | |
1367 | } for location in track.findall(xpath_with_ns('./xspf:location', NS_MAP))] | |
1368 | self._sort_formats(formats) | |
1369 | ||
1370 | entries.append({ | |
1371 | 'id': playlist_id, | |
1372 | 'title': title, | |
1373 | 'description': description, | |
1374 | 'thumbnail': thumbnail, | |
1375 | 'duration': duration, | |
1376 | 'formats': formats, | |
1377 | }) | |
1378 | return entries | |
1379 | ||
1bac3455 | 1380 | def _extract_mpd_formats(self, mpd_url, video_id, mpd_id=None, note=None, errnote=None, fatal=True, formats_dict={}): |
1381 | res = self._download_webpage_handle( | |
1382 | mpd_url, video_id, | |
1383 | note=note or 'Downloading MPD manifest', | |
1384 | errnote=errnote or 'Failed to download MPD manifest', | |
2d2fa82d | 1385 | fatal=fatal) |
1bac3455 | 1386 | if res is False: |
2d2fa82d | 1387 | return [] |
1bac3455 | 1388 | mpd, urlh = res |
1389 | mpd_base_url = re.match(r'https?://.+/', urlh.geturl()).group() | |
1390 | ||
91cb6b50 | 1391 | return self._parse_mpd_formats( |
1bac3455 | 1392 | compat_etree_fromstring(mpd.encode('utf-8')), mpd_id, mpd_base_url, formats_dict=formats_dict) |
2d2fa82d | 1393 | |
91cb6b50 | 1394 | def _parse_mpd_formats(self, mpd_doc, mpd_id=None, mpd_base_url='', formats_dict={}): |
1bac3455 | 1395 | if mpd_doc.get('type') == 'dynamic': |
1396 | return [] | |
2d2fa82d | 1397 | |
91cb6b50 | 1398 | namespace = self._search_regex(r'(?i)^{([^}]+)?}MPD$', mpd_doc.tag, 'namespace', default=None) |
f14be228 | 1399 | |
1400 | def _add_ns(path): | |
1401 | return self._xpath_ns(path, namespace) | |
1402 | ||
675d0016 | 1403 | def is_drm_protected(element): |
1404 | return element.find(_add_ns('ContentProtection')) is not None | |
1405 | ||
1bac3455 | 1406 | def extract_multisegment_info(element, ms_parent_info): |
1407 | ms_info = ms_parent_info.copy() | |
f14be228 | 1408 | segment_list = element.find(_add_ns('SegmentList')) |
1bac3455 | 1409 | if segment_list is not None: |
f14be228 | 1410 | segment_urls_e = segment_list.findall(_add_ns('SegmentURL')) |
1bac3455 | 1411 | if segment_urls_e: |
1412 | ms_info['segment_urls'] = [segment.attrib['media'] for segment in segment_urls_e] | |
f14be228 | 1413 | initialization = segment_list.find(_add_ns('Initialization')) |
1bac3455 | 1414 | if initialization is not None: |
1415 | ms_info['initialization_url'] = initialization.attrib['sourceURL'] | |
1416 | else: | |
f14be228 | 1417 | segment_template = element.find(_add_ns('SegmentTemplate')) |
1bac3455 | 1418 | if segment_template is not None: |
1419 | start_number = segment_template.get('startNumber') | |
1420 | if start_number: | |
1421 | ms_info['start_number'] = int(start_number) | |
f14be228 | 1422 | segment_timeline = segment_template.find(_add_ns('SegmentTimeline')) |
1bac3455 | 1423 | if segment_timeline is not None: |
f14be228 | 1424 | s_e = segment_timeline.findall(_add_ns('S')) |
1bac3455 | 1425 | if s_e: |
1426 | ms_info['total_number'] = 0 | |
1427 | for s in s_e: | |
1428 | ms_info['total_number'] += 1 + int(s.get('r', '0')) | |
1429 | else: | |
1430 | timescale = segment_template.get('timescale') | |
1431 | if timescale: | |
1432 | ms_info['timescale'] = int(timescale) | |
1433 | segment_duration = segment_template.get('duration') | |
1434 | if segment_duration: | |
1435 | ms_info['segment_duration'] = int(segment_duration) | |
1436 | media_template = segment_template.get('media') | |
1437 | if media_template: | |
1438 | ms_info['media_template'] = media_template | |
1439 | initialization = segment_template.get('initialization') | |
1440 | if initialization: | |
1441 | ms_info['initialization_url'] = initialization | |
1442 | else: | |
f14be228 | 1443 | initialization = segment_template.find(_add_ns('Initialization')) |
1bac3455 | 1444 | if initialization is not None: |
1445 | ms_info['initialization_url'] = initialization.attrib['sourceURL'] | |
1446 | return ms_info | |
b323e170 | 1447 | |
1bac3455 | 1448 | mpd_duration = parse_duration(mpd_doc.get('mediaPresentationDuration')) |
17b598d3 | 1449 | formats = [] |
f14be228 | 1450 | for period in mpd_doc.findall(_add_ns('Period')): |
1bac3455 | 1451 | period_duration = parse_duration(period.get('duration')) or mpd_duration |
1452 | period_ms_info = extract_multisegment_info(period, { | |
1453 | 'start_number': 1, | |
1454 | 'timescale': 1, | |
1455 | }) | |
f14be228 | 1456 | for adaptation_set in period.findall(_add_ns('AdaptationSet')): |
675d0016 | 1457 | if is_drm_protected(adaptation_set): |
1458 | continue | |
1bac3455 | 1459 | adaption_set_ms_info = extract_multisegment_info(adaptation_set, period_ms_info) |
f14be228 | 1460 | for representation in adaptation_set.findall(_add_ns('Representation')): |
675d0016 | 1461 | if is_drm_protected(representation): |
1462 | continue | |
1bac3455 | 1463 | representation_attrib = adaptation_set.attrib.copy() |
1464 | representation_attrib.update(representation.attrib) | |
a6c8b759 YCH |
1465 | # According to page 41 of ISO/IEC 29001-1:2014, @mimeType is mandatory |
1466 | mime_type = representation_attrib['mimeType'] | |
1467 | content_type = mime_type.split('/')[0] | |
1bac3455 | 1468 | if content_type == 'text': |
1469 | # TODO implement WebVTT downloading | |
1470 | pass | |
1471 | elif content_type == 'video' or content_type == 'audio': | |
1472 | base_url = '' | |
1473 | for element in (representation, adaptation_set, period, mpd_doc): | |
f14be228 | 1474 | base_url_e = element.find(_add_ns('BaseURL')) |
1bac3455 | 1475 | if base_url_e is not None: |
1476 | base_url = base_url_e.text + base_url | |
1477 | if re.match(r'^https?://', base_url): | |
1478 | break | |
bb20526b S |
1479 | if mpd_base_url and not re.match(r'^https?://', base_url): |
1480 | if not mpd_base_url.endswith('/') and not base_url.startswith('/'): | |
1481 | mpd_base_url += '/' | |
1bac3455 | 1482 | base_url = mpd_base_url + base_url |
1483 | representation_id = representation_attrib.get('id') | |
d577c796 | 1484 | lang = representation_attrib.get('lang') |
51e9094f | 1485 | url_el = representation.find(_add_ns('BaseURL')) |
1486 | filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength') if url_el is not None else None) | |
1bac3455 | 1487 | f = { |
154c209e | 1488 | 'format_id': '%s-%s' % (mpd_id, representation_id) if mpd_id else representation_id, |
1bac3455 | 1489 | 'url': base_url, |
a6c8b759 | 1490 | 'ext': mimetype2ext(mime_type), |
1bac3455 | 1491 | 'width': int_or_none(representation_attrib.get('width')), |
1492 | 'height': int_or_none(representation_attrib.get('height')), | |
1493 | 'tbr': int_or_none(representation_attrib.get('bandwidth'), 1000), | |
1494 | 'asr': int_or_none(representation_attrib.get('audioSamplingRate')), | |
1495 | 'fps': int_or_none(representation_attrib.get('frameRate')), | |
1496 | 'vcodec': 'none' if content_type == 'audio' else representation_attrib.get('codecs'), | |
1497 | 'acodec': 'none' if content_type == 'video' else representation_attrib.get('codecs'), | |
d577c796 | 1498 | 'language': lang if lang not in ('mul', 'und', 'zxx', 'mis') else None, |
1bac3455 | 1499 | 'format_note': 'DASH %s' % content_type, |
51e9094f | 1500 | 'filesize': filesize, |
1bac3455 | 1501 | } |
1502 | representation_ms_info = extract_multisegment_info(representation, adaption_set_ms_info) | |
1503 | if 'segment_urls' not in representation_ms_info and 'media_template' in representation_ms_info: | |
1504 | if 'total_number' not in representation_ms_info and 'segment_duration': | |
6a3828fd | 1505 | segment_duration = float(representation_ms_info['segment_duration']) / float(representation_ms_info['timescale']) |
1506 | representation_ms_info['total_number'] = int(math.ceil(float(period_duration) / segment_duration)) | |
1bac3455 | 1507 | media_template = representation_ms_info['media_template'] |
1508 | media_template = media_template.replace('$RepresentationID$', representation_id) | |
53c269c6 | 1509 | media_template = re.sub(r'\$(Number|Bandwidth)(?:%(0\d+)d)?\$', r'%(\1)\2d', media_template) |
1bac3455 | 1510 | media_template.replace('$$', '$') |
53c269c6 | 1511 | representation_ms_info['segment_urls'] = [media_template % {'Number': segment_number, 'Bandwidth': representation_attrib.get('bandwidth')} for segment_number in range(representation_ms_info['start_number'], representation_ms_info['total_number'] + representation_ms_info['start_number'])] |
1bac3455 | 1512 | if 'segment_urls' in representation_ms_info: |
1513 | f.update({ | |
1514 | 'segment_urls': representation_ms_info['segment_urls'], | |
1515 | 'protocol': 'http_dash_segments', | |
df374b52 | 1516 | }) |
1bac3455 | 1517 | if 'initialization_url' in representation_ms_info: |
1518 | initialization_url = representation_ms_info['initialization_url'].replace('$RepresentationID$', representation_id) | |
1519 | f.update({ | |
1520 | 'initialization_url': initialization_url, | |
1521 | }) | |
1522 | if not f.get('url'): | |
1523 | f['url'] = initialization_url | |
1524 | try: | |
1525 | existing_format = next( | |
1526 | fo for fo in formats | |
1527 | if fo['format_id'] == representation_id) | |
1528 | except StopIteration: | |
1529 | full_info = formats_dict.get(representation_id, {}).copy() | |
1530 | full_info.update(f) | |
1531 | formats.append(full_info) | |
1532 | else: | |
1533 | existing_format.update(f) | |
17b598d3 | 1534 | else: |
1bac3455 | 1535 | self.report_warning('Unknown MIME type %s in DASH manifest' % mime_type) |
0826a0b5 | 1536 | self._sort_formats(formats) |
17b598d3 YCH |
1537 | return formats |
1538 | ||
f4b1c7ad PH |
1539 | def _live_title(self, name): |
1540 | """ Generate the title for a live video """ | |
1541 | now = datetime.datetime.now() | |
611c1dd9 | 1542 | now_str = now.strftime('%Y-%m-%d %H:%M') |
f4b1c7ad PH |
1543 | return name + ' ' + now_str |
1544 | ||
b14f3a4c PH |
1545 | def _int(self, v, name, fatal=False, **kwargs): |
1546 | res = int_or_none(v, **kwargs) | |
1547 | if 'get_attr' in kwargs: | |
1548 | print(getattr(v, kwargs['get_attr'])) | |
1549 | if res is None: | |
1550 | msg = 'Failed to extract %s: Could not parse value %r' % (name, v) | |
1551 | if fatal: | |
1552 | raise ExtractorError(msg) | |
1553 | else: | |
1554 | self._downloader.report_warning(msg) | |
1555 | return res | |
1556 | ||
1557 | def _float(self, v, name, fatal=False, **kwargs): | |
1558 | res = float_or_none(v, **kwargs) | |
1559 | if res is None: | |
1560 | msg = 'Failed to extract %s: Could not parse value %r' % (name, v) | |
1561 | if fatal: | |
1562 | raise ExtractorError(msg) | |
1563 | else: | |
1564 | self._downloader.report_warning(msg) | |
1565 | return res | |
1566 | ||
42939b61 | 1567 | def _set_cookie(self, domain, name, value, expire_time=None): |
810fb84d PH |
1568 | cookie = compat_cookiejar.Cookie( |
1569 | 0, name, value, None, None, domain, None, | |
42939b61 JMF |
1570 | None, '/', True, False, expire_time, '', None, None, None) |
1571 | self._downloader.cookiejar.set_cookie(cookie) | |
1572 | ||
799207e8 | 1573 | def _get_cookies(self, url): |
1574 | """ Return a compat_cookies.SimpleCookie with the cookies for the url """ | |
5c2266df | 1575 | req = sanitized_Request(url) |
799207e8 | 1576 | self._downloader.cookiejar.add_cookie_header(req) |
1577 | return compat_cookies.SimpleCookie(req.get_header('Cookie')) | |
1578 | ||
05900629 PH |
1579 | def get_testcases(self, include_onlymatching=False): |
1580 | t = getattr(self, '_TEST', None) | |
1581 | if t: | |
1582 | assert not hasattr(self, '_TESTS'), \ | |
1583 | '%s has _TEST and _TESTS' % type(self).__name__ | |
1584 | tests = [t] | |
1585 | else: | |
1586 | tests = getattr(self, '_TESTS', []) | |
1587 | for t in tests: | |
1588 | if not include_onlymatching and t.get('only_matching', False): | |
1589 | continue | |
1590 | t['name'] = type(self).__name__[:-len('IE')] | |
1591 | yield t | |
1592 | ||
1593 | def is_suitable(self, age_limit): | |
1594 | """ Test whether the extractor is generally suitable for the given | |
1595 | age limit (i.e. pornographic sites are not, all others usually are) """ | |
1596 | ||
1597 | any_restricted = False | |
1598 | for tc in self.get_testcases(include_onlymatching=False): | |
1599 | if 'playlist' in tc: | |
1600 | tc = tc['playlist'][0] | |
1601 | is_restricted = age_restricted( | |
1602 | tc.get('info_dict', {}).get('age_limit'), age_limit) | |
1603 | if not is_restricted: | |
1604 | return True | |
1605 | any_restricted = any_restricted or is_restricted | |
1606 | return not any_restricted | |
1607 | ||
a504ced0 | 1608 | def extract_subtitles(self, *args, **kwargs): |
9868ea49 JMF |
1609 | if (self._downloader.params.get('writesubtitles', False) or |
1610 | self._downloader.params.get('listsubtitles')): | |
1611 | return self._get_subtitles(*args, **kwargs) | |
1612 | return {} | |
a504ced0 JMF |
1613 | |
1614 | def _get_subtitles(self, *args, **kwargs): | |
611c1dd9 | 1615 | raise NotImplementedError('This method must be implemented by subclasses') |
a504ced0 | 1616 | |
912e0b7e YCH |
1617 | @staticmethod |
1618 | def _merge_subtitle_items(subtitle_list1, subtitle_list2): | |
1619 | """ Merge subtitle items for one language. Items with duplicated URLs | |
1620 | will be dropped. """ | |
1621 | list1_urls = set([item['url'] for item in subtitle_list1]) | |
1622 | ret = list(subtitle_list1) | |
1623 | ret.extend([item for item in subtitle_list2 if item['url'] not in list1_urls]) | |
1624 | return ret | |
1625 | ||
1626 | @classmethod | |
8c97f819 | 1627 | def _merge_subtitles(cls, subtitle_dict1, subtitle_dict2): |
912e0b7e | 1628 | """ Merge two subtitle dictionaries, language by language. """ |
912e0b7e YCH |
1629 | ret = dict(subtitle_dict1) |
1630 | for lang in subtitle_dict2: | |
8c97f819 | 1631 | ret[lang] = cls._merge_subtitle_items(subtitle_dict1.get(lang, []), subtitle_dict2[lang]) |
912e0b7e YCH |
1632 | return ret |
1633 | ||
360e1ca5 | 1634 | def extract_automatic_captions(self, *args, **kwargs): |
9868ea49 JMF |
1635 | if (self._downloader.params.get('writeautomaticsub', False) or |
1636 | self._downloader.params.get('listsubtitles')): | |
1637 | return self._get_automatic_captions(*args, **kwargs) | |
1638 | return {} | |
360e1ca5 JMF |
1639 | |
1640 | def _get_automatic_captions(self, *args, **kwargs): | |
611c1dd9 | 1641 | raise NotImplementedError('This method must be implemented by subclasses') |
360e1ca5 | 1642 | |
d77ab8e2 S |
1643 | def mark_watched(self, *args, **kwargs): |
1644 | if (self._downloader.params.get('mark_watched', False) and | |
1645 | (self._get_login_info()[0] is not None or | |
1646 | self._downloader.params.get('cookiefile') is not None)): | |
1647 | self._mark_watched(*args, **kwargs) | |
1648 | ||
1649 | def _mark_watched(self, *args, **kwargs): | |
1650 | raise NotImplementedError('This method must be implemented by subclasses') | |
1651 | ||
8dbe9899 | 1652 | |
d6983cb4 PH |
1653 | class SearchInfoExtractor(InfoExtractor): |
1654 | """ | |
1655 | Base class for paged search queries extractors. | |
10952eb2 | 1656 | They accept URLs in the format _SEARCH_KEY(|all|[0-9]):{query} |
d6983cb4 PH |
1657 | Instances should define _SEARCH_KEY and _MAX_RESULTS. |
1658 | """ | |
1659 | ||
1660 | @classmethod | |
1661 | def _make_valid_url(cls): | |
1662 | return r'%s(?P<prefix>|[1-9][0-9]*|all):(?P<query>[\s\S]+)' % cls._SEARCH_KEY | |
1663 | ||
1664 | @classmethod | |
1665 | def suitable(cls, url): | |
1666 | return re.match(cls._make_valid_url(), url) is not None | |
1667 | ||
1668 | def _real_extract(self, query): | |
1669 | mobj = re.match(self._make_valid_url(), query) | |
1670 | if mobj is None: | |
f1a9d64e | 1671 | raise ExtractorError('Invalid search query "%s"' % query) |
d6983cb4 PH |
1672 | |
1673 | prefix = mobj.group('prefix') | |
1674 | query = mobj.group('query') | |
1675 | if prefix == '': | |
1676 | return self._get_n_results(query, 1) | |
1677 | elif prefix == 'all': | |
1678 | return self._get_n_results(query, self._MAX_RESULTS) | |
1679 | else: | |
1680 | n = int(prefix) | |
1681 | if n <= 0: | |
f1a9d64e | 1682 | raise ExtractorError('invalid download number %s for query "%s"' % (n, query)) |
d6983cb4 | 1683 | elif n > self._MAX_RESULTS: |
f1a9d64e | 1684 | self._downloader.report_warning('%s returns max %i results (you requested %i)' % (self._SEARCH_KEY, self._MAX_RESULTS, n)) |
d6983cb4 PH |
1685 | n = self._MAX_RESULTS |
1686 | return self._get_n_results(query, n) | |
1687 | ||
1688 | def _get_n_results(self, query, n): | |
1689 | """Get a specified number of results for a query""" | |
611c1dd9 | 1690 | raise NotImplementedError('This method must be implemented by subclasses') |
0f818663 PH |
1691 | |
1692 | @property | |
1693 | def SEARCH_KEY(self): | |
1694 | return self._SEARCH_KEY |