]>
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, |
e64b7569 | 18 | compat_getpass, |
d6983cb4 PH |
19 | compat_http_client, |
20 | compat_urllib_error, | |
a107193e | 21 | compat_urllib_parse, |
f0b5d6af | 22 | compat_urlparse, |
d6983cb4 | 23 | compat_str, |
36e6f62c | 24 | compat_etree_fromstring, |
8c25f81b PH |
25 | ) |
26 | from ..utils import ( | |
c342041f | 27 | NO_DEFAULT, |
05900629 | 28 | age_restricted, |
08f2a92c | 29 | bug_reports_message, |
d6983cb4 PH |
30 | clean_html, |
31 | compiled_regex_type, | |
70f0f5a8 | 32 | determine_ext, |
9b9c5355 | 33 | error_to_compat_str, |
d6983cb4 | 34 | ExtractorError, |
97f4aecf | 35 | fix_xml_ampersands, |
b14f3a4c | 36 | float_or_none, |
31bb8d3f | 37 | int_or_none, |
4ca2a3cf | 38 | parse_iso8601, |
55b3e45b | 39 | RegexNotFoundError, |
d41e6efc | 40 | sanitize_filename, |
5c2266df | 41 | sanitized_Request, |
f38de77f | 42 | unescapeHTML, |
647eab45 | 43 | unified_strdate, |
a107193e | 44 | url_basename, |
8d6765cf S |
45 | xpath_text, |
46 | xpath_with_ns, | |
d497a201 | 47 | determine_protocol, |
1bac3455 | 48 | parse_duration, |
cafcf657 | 49 | mimetype2ext, |
d6983cb4 | 50 | ) |
c342041f | 51 | |
d6983cb4 PH |
52 | |
53 | class InfoExtractor(object): | |
54 | """Information Extractor class. | |
55 | ||
56 | Information extractors are the classes that, given a URL, extract | |
57 | information about the video (or videos) the URL refers to. This | |
58 | information includes the real video URL, the video title, author and | |
59 | others. The information is stored in a dictionary which is then | |
5d380852 | 60 | passed to the YoutubeDL. The YoutubeDL processes this |
d6983cb4 PH |
61 | information possibly downloading the video to the file system, among |
62 | other possible outcomes. | |
63 | ||
cf0649f8 | 64 | The type field determines the type of the result. |
fed5d032 PH |
65 | By far the most common value (and the default if _type is missing) is |
66 | "video", which indicates a single video. | |
67 | ||
68 | For a video, the dictionaries must include the following fields: | |
d6983cb4 PH |
69 | |
70 | id: Video identifier. | |
d6983cb4 | 71 | title: Video title, unescaped. |
d67b0b15 | 72 | |
f49d89ee | 73 | Additionally, it must contain either a formats entry or a url one: |
d67b0b15 | 74 | |
f49d89ee PH |
75 | formats: A list of dictionaries for each format available, ordered |
76 | from worst to best quality. | |
77 | ||
78 | Potential fields: | |
d67b0b15 | 79 | * url Mandatory. The URL of the video file |
10952eb2 | 80 | * ext Will be calculated from URL if missing |
d67b0b15 PH |
81 | * format A human-readable description of the format |
82 | ("mp4 container with h264/opus"). | |
83 | Calculated from the format_id, width, height. | |
84 | and format_note fields if missing. | |
85 | * format_id A short description of the format | |
5d4f3985 PH |
86 | ("mp4_h264_opus" or "19"). |
87 | Technically optional, but strongly recommended. | |
d67b0b15 PH |
88 | * format_note Additional info about the format |
89 | ("3D" or "DASH video") | |
90 | * width Width of the video, if known | |
91 | * height Height of the video, if known | |
f49d89ee | 92 | * resolution Textual description of width and height |
7217e148 | 93 | * tbr Average bitrate of audio and video in KBit/s |
d67b0b15 PH |
94 | * abr Average audio bitrate in KBit/s |
95 | * acodec Name of the audio codec in use | |
dd27fd17 | 96 | * asr Audio sampling rate in Hertz |
d67b0b15 | 97 | * vbr Average video bitrate in KBit/s |
fbb21cf5 | 98 | * fps Frame rate |
d67b0b15 | 99 | * vcodec Name of the video codec in use |
1394ce65 | 100 | * container Name of the container format |
d67b0b15 | 101 | * filesize The number of bytes, if known in advance |
9732d77e | 102 | * filesize_approx An estimate for the number of bytes |
d67b0b15 | 103 | * player_url SWF Player URL (used for rtmpdump). |
c7deaa4c PH |
104 | * protocol The protocol that will be used for the actual |
105 | download, lower-case. | |
b04b8852 PH |
106 | "http", "https", "rtsp", "rtmp", "rtmpe", |
107 | "m3u8", or "m3u8_native". | |
f49d89ee | 108 | * preference Order number of this format. If this field is |
08d13955 | 109 | present and not None, the formats get sorted |
38d63d84 | 110 | by this field, regardless of all other values. |
f49d89ee PH |
111 | -1 for default (order by other properties), |
112 | -2 or smaller for less than default. | |
e65566a9 PH |
113 | < -1000 to hide the format (if there is |
114 | another one which is strictly better) | |
32f90364 PH |
115 | * language Language code, e.g. "de" or "en-US". |
116 | * language_preference Is this in the language mentioned in | |
117 | the URL? | |
aff2f4f4 PH |
118 | 10 if it's what the URL is about, |
119 | -1 for default (don't know), | |
120 | -10 otherwise, other values reserved for now. | |
5d73273f PH |
121 | * quality Order number of the video quality of this |
122 | format, irrespective of the file format. | |
123 | -1 for default (order by other properties), | |
124 | -2 or smaller for less than default. | |
c64ed2a3 PH |
125 | * source_preference Order number for this video source |
126 | (quality takes higher priority) | |
127 | -1 for default (order by other properties), | |
128 | -2 or smaller for less than default. | |
d769be6c PH |
129 | * http_headers A dictionary of additional HTTP headers |
130 | to add to the request. | |
6271f1ca | 131 | * stretched_ratio If given and not 1, indicates that the |
3dee7826 PH |
132 | video's pixels are not square. |
133 | width : height ratio as float. | |
134 | * no_resume The server does not support resuming the | |
135 | (HTTP or RTMP) download. Boolean. | |
136 | ||
c0ba0f48 | 137 | url: Final video URL. |
d6983cb4 | 138 | ext: Video filename extension. |
d67b0b15 PH |
139 | format: The video format, defaults to ext (used for --get-format) |
140 | player_url: SWF Player URL (used for rtmpdump). | |
2f5865cc | 141 | |
d6983cb4 PH |
142 | The following fields are optional: |
143 | ||
f5e43bc6 | 144 | alt_title: A secondary title of the video. |
0afef30b PH |
145 | display_id An alternative identifier for the video, not necessarily |
146 | unique, but available before title. Typically, id is | |
147 | something like "4234987", title "Dancing naked mole rats", | |
148 | and display_id "dancing-naked-mole-rats" | |
d5519808 | 149 | thumbnails: A list of dictionaries, with the following entries: |
cfb56d1a | 150 | * "id" (optional, string) - Thumbnail format ID |
d5519808 | 151 | * "url" |
cfb56d1a | 152 | * "preference" (optional, int) - quality of the image |
d5519808 PH |
153 | * "width" (optional, int) |
154 | * "height" (optional, int) | |
155 | * "resolution" (optional, string "{width}x{height"}, | |
156 | deprecated) | |
d6983cb4 | 157 | thumbnail: Full URL to a video thumbnail image. |
f5e43bc6 | 158 | description: Full video description. |
d6983cb4 | 159 | uploader: Full name of the video uploader. |
9bb8e0a3 | 160 | creator: The main artist who created the video. |
8aab976b | 161 | release_date: The date (YYYYMMDD) when the video was released. |
955c4514 | 162 | timestamp: UNIX timestamp of the moment the video became available. |
d6983cb4 | 163 | upload_date: Video upload date (YYYYMMDD). |
955c4514 | 164 | If not explicitly set, calculated from timestamp. |
d6983cb4 | 165 | uploader_id: Nickname or id of the video uploader. |
da9ec3b9 | 166 | location: Physical location where the video was filmed. |
a504ced0 JMF |
167 | subtitles: The available subtitles as a dictionary in the format |
168 | {language: subformats}. "subformats" is a list sorted from | |
169 | lower to higher preference, each element is a dictionary | |
170 | with the "ext" entry and one of: | |
171 | * "data": The subtitles file contents | |
10952eb2 | 172 | * "url": A URL pointing to the subtitles file |
4bba3716 | 173 | "ext" will be calculated from URL if missing |
360e1ca5 JMF |
174 | automatic_captions: Like 'subtitles', used by the YoutubeIE for |
175 | automatically generated captions | |
62d231c0 | 176 | duration: Length of the video in seconds, as an integer or float. |
f3d29461 | 177 | view_count: How many users have watched the video on the platform. |
19e3dfc9 PH |
178 | like_count: Number of positive ratings of the video |
179 | dislike_count: Number of negative ratings of the video | |
02835c6b | 180 | repost_count: Number of reposts of the video |
2d30521a | 181 | average_rating: Average rating give by users, the scale used depends on the webpage |
19e3dfc9 | 182 | comment_count: Number of comments on the video |
dd622d7c PH |
183 | comments: A list of comments, each with one or more of the following |
184 | properties (all but one of text or html optional): | |
185 | * "author" - human-readable name of the comment author | |
186 | * "author_id" - user ID of the comment author | |
187 | * "id" - Comment ID | |
188 | * "html" - Comment as HTML | |
189 | * "text" - Plain text of the comment | |
190 | * "timestamp" - UNIX timestamp of comment | |
191 | * "parent" - ID of the comment this one is replying to. | |
192 | Set to "root" to indicate that this is a | |
193 | comment to the original video. | |
8dbe9899 | 194 | age_limit: Age restriction for the video, as an integer (years) |
10952eb2 | 195 | webpage_url: The URL to the video webpage, if given to youtube-dl it |
9103bbc5 JMF |
196 | should allow to get the same result again. (It will be set |
197 | by YoutubeDL if it's missing) | |
ad3bc6ac PH |
198 | categories: A list of categories that the video falls in, for example |
199 | ["Sports", "Berlin"] | |
864f24bd | 200 | tags: A list of tags assigned to the video, e.g. ["sweden", "pop music"] |
7267bd53 PH |
201 | is_live: True, False, or None (=unknown). Whether this video is a |
202 | live stream that goes on instead of a fixed-length video. | |
7c80519c | 203 | start_time: Time in seconds where the reproduction should start, as |
10952eb2 | 204 | specified in the URL. |
297a564b | 205 | end_time: Time in seconds where the reproduction should end, as |
10952eb2 | 206 | specified in the URL. |
d6983cb4 | 207 | |
7109903e S |
208 | The following fields should only be used when the video belongs to some logical |
209 | chapter or section: | |
210 | ||
211 | chapter: Name or title of the chapter the video belongs to. | |
27bfd4e5 S |
212 | chapter_number: Number of the chapter the video belongs to, as an integer. |
213 | chapter_id: Id of the chapter the video belongs to, as a unicode string. | |
7109903e S |
214 | |
215 | The following fields should only be used when the video is an episode of some | |
216 | series or programme: | |
217 | ||
218 | series: Title of the series or programme the video episode belongs to. | |
219 | season: Title of the season the video episode belongs to. | |
27bfd4e5 S |
220 | season_number: Number of the season the video episode belongs to, as an integer. |
221 | season_id: Id of the season the video episode belongs to, as a unicode string. | |
7109903e S |
222 | episode: Title of the video episode. Unlike mandatory video title field, |
223 | this field should denote the exact title of the video episode | |
224 | without any kind of decoration. | |
27bfd4e5 S |
225 | episode_number: Number of the video episode within a season, as an integer. |
226 | episode_id: Id of the video episode, as a unicode string. | |
7109903e | 227 | |
deefc05b | 228 | Unless mentioned otherwise, the fields should be Unicode strings. |
d6983cb4 | 229 | |
d838b1bd PH |
230 | Unless mentioned otherwise, None is equivalent to absence of information. |
231 | ||
fed5d032 PH |
232 | |
233 | _type "playlist" indicates multiple videos. | |
b82f815f PH |
234 | There must be a key "entries", which is a list, an iterable, or a PagedList |
235 | object, each element of which is a valid dictionary by this specification. | |
fed5d032 | 236 | |
e0b9d78f S |
237 | Additionally, playlists can have "title", "description" and "id" attributes |
238 | with the same semantics as videos (see above). | |
fed5d032 PH |
239 | |
240 | ||
241 | _type "multi_video" indicates that there are multiple videos that | |
242 | form a single show, for examples multiple acts of an opera or TV episode. | |
243 | It must have an entries key like a playlist and contain all the keys | |
244 | required for a video at the same time. | |
245 | ||
246 | ||
247 | _type "url" indicates that the video must be extracted from another | |
248 | location, possibly by a different extractor. Its only required key is: | |
249 | "url" - the next URL to extract. | |
f58766ce PH |
250 | The key "ie_key" can be set to the class name (minus the trailing "IE", |
251 | e.g. "Youtube") if the extractor class is known in advance. | |
252 | Additionally, the dictionary may have any properties of the resolved entity | |
253 | known in advance, for example "title" if the title of the referred video is | |
fed5d032 PH |
254 | known ahead of time. |
255 | ||
256 | ||
257 | _type "url_transparent" entities have the same specification as "url", but | |
258 | indicate that the given additional information is more precise than the one | |
259 | associated with the resolved URL. | |
260 | This is useful when a site employs a video service that hosts the video and | |
261 | its technical metadata, but that video service does not embed a useful | |
262 | title, description etc. | |
263 | ||
264 | ||
d6983cb4 PH |
265 | Subclasses of this one should re-define the _real_initialize() and |
266 | _real_extract() methods and define a _VALID_URL regexp. | |
267 | Probably, they should also be added to the list of extractors. | |
268 | ||
d6983cb4 PH |
269 | Finally, the _WORKING attribute should be set to False for broken IEs |
270 | in order to warn the users and skip the tests. | |
271 | """ | |
272 | ||
273 | _ready = False | |
274 | _downloader = None | |
275 | _WORKING = True | |
276 | ||
277 | def __init__(self, downloader=None): | |
278 | """Constructor. Receives an optional downloader.""" | |
279 | self._ready = False | |
280 | self.set_downloader(downloader) | |
281 | ||
282 | @classmethod | |
283 | def suitable(cls, url): | |
284 | """Receives a URL and returns True if suitable for this IE.""" | |
79cb2577 PH |
285 | |
286 | # This does not use has/getattr intentionally - we want to know whether | |
287 | # we have cached the regexp for *this* class, whereas getattr would also | |
288 | # match the superclass | |
289 | if '_VALID_URL_RE' not in cls.__dict__: | |
290 | cls._VALID_URL_RE = re.compile(cls._VALID_URL) | |
291 | return cls._VALID_URL_RE.match(url) is not None | |
d6983cb4 | 292 | |
ed9266db PH |
293 | @classmethod |
294 | def _match_id(cls, url): | |
295 | if '_VALID_URL_RE' not in cls.__dict__: | |
296 | cls._VALID_URL_RE = re.compile(cls._VALID_URL) | |
297 | m = cls._VALID_URL_RE.match(url) | |
298 | assert m | |
299 | return m.group('id') | |
300 | ||
d6983cb4 PH |
301 | @classmethod |
302 | def working(cls): | |
303 | """Getter method for _WORKING.""" | |
304 | return cls._WORKING | |
305 | ||
306 | def initialize(self): | |
307 | """Initializes an instance (authentication, etc).""" | |
308 | if not self._ready: | |
309 | self._real_initialize() | |
310 | self._ready = True | |
311 | ||
312 | def extract(self, url): | |
313 | """Extracts URL information and returns it in list of dicts.""" | |
3a5bcd03 PH |
314 | try: |
315 | self.initialize() | |
316 | return self._real_extract(url) | |
317 | except ExtractorError: | |
318 | raise | |
319 | except compat_http_client.IncompleteRead as e: | |
dfb1b146 | 320 | raise ExtractorError('A network error has occurred.', cause=e, expected=True) |
9650885b | 321 | except (KeyError, StopIteration) as e: |
dfb1b146 | 322 | raise ExtractorError('An extractor error has occurred.', cause=e) |
d6983cb4 PH |
323 | |
324 | def set_downloader(self, downloader): | |
325 | """Sets the downloader for this IE.""" | |
326 | self._downloader = downloader | |
327 | ||
328 | def _real_initialize(self): | |
329 | """Real initialization process. Redefine in subclasses.""" | |
330 | pass | |
331 | ||
332 | def _real_extract(self, url): | |
333 | """Real extraction process. Redefine in subclasses.""" | |
334 | pass | |
335 | ||
56c73665 JMF |
336 | @classmethod |
337 | def ie_key(cls): | |
338 | """A string for getting the InfoExtractor with get_info_extractor""" | |
dc519b54 | 339 | return compat_str(cls.__name__[:-2]) |
56c73665 | 340 | |
d6983cb4 PH |
341 | @property |
342 | def IE_NAME(self): | |
dc519b54 | 343 | return compat_str(type(self).__name__[:-2]) |
d6983cb4 | 344 | |
7cc3570e | 345 | def _request_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True): |
d6983cb4 PH |
346 | """ Returns the response handle """ |
347 | if note is None: | |
348 | self.report_download_webpage(video_id) | |
349 | elif note is not False: | |
7cc3570e | 350 | if video_id is None: |
f1a9d64e | 351 | self.to_screen('%s' % (note,)) |
7cc3570e | 352 | else: |
f1a9d64e | 353 | self.to_screen('%s: %s' % (video_id, note)) |
d6983cb4 | 354 | try: |
dca08720 | 355 | return self._downloader.urlopen(url_or_request) |
d6983cb4 | 356 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: |
aa94a6d3 PH |
357 | if errnote is False: |
358 | return False | |
d6983cb4 | 359 | if errnote is None: |
f1a9d64e | 360 | errnote = 'Unable to download webpage' |
7f8b2714 | 361 | |
9b9c5355 | 362 | errmsg = '%s: %s' % (errnote, error_to_compat_str(err)) |
7cc3570e PH |
363 | if fatal: |
364 | raise ExtractorError(errmsg, sys.exc_info()[2], cause=err) | |
365 | else: | |
366 | self._downloader.report_warning(errmsg) | |
367 | return False | |
d6983cb4 | 368 | |
c9a77969 | 369 | def _download_webpage_handle(self, url_or_request, video_id, note=None, errnote=None, fatal=True, encoding=None): |
d6983cb4 | 370 | """ Returns a tuple (page content as string, URL handle) """ |
b9d3e163 PH |
371 | # Strip hashes from the URL (#1038) |
372 | if isinstance(url_or_request, (compat_str, str)): | |
373 | url_or_request = url_or_request.partition('#')[0] | |
374 | ||
7cc3570e PH |
375 | urlh = self._request_webpage(url_or_request, video_id, note, errnote, fatal) |
376 | if urlh is False: | |
377 | assert not fatal | |
378 | return False | |
c9a77969 | 379 | content = self._webpage_read_content(urlh, url_or_request, video_id, note, errnote, fatal, encoding=encoding) |
23be51d8 PH |
380 | return (content, urlh) |
381 | ||
c9a77969 YCH |
382 | @staticmethod |
383 | def _guess_encoding_from_content(content_type, webpage_bytes): | |
d6983cb4 PH |
384 | m = re.match(r'[a-zA-Z0-9_.-]+/[a-zA-Z0-9_.-]+\s*;\s*charset=(.+)', content_type) |
385 | if m: | |
386 | encoding = m.group(1) | |
387 | else: | |
0d75ae2c | 388 | m = re.search(br'<meta[^>]+charset=[\'"]?([^\'")]+)[ /\'">]', |
f143d86a PH |
389 | webpage_bytes[:1024]) |
390 | if m: | |
391 | encoding = m.group(1).decode('ascii') | |
b60016e8 PH |
392 | elif webpage_bytes.startswith(b'\xff\xfe'): |
393 | encoding = 'utf-16' | |
f143d86a PH |
394 | else: |
395 | encoding = 'utf-8' | |
c9a77969 YCH |
396 | |
397 | return encoding | |
398 | ||
399 | def _webpage_read_content(self, urlh, url_or_request, video_id, note=None, errnote=None, fatal=True, prefix=None, encoding=None): | |
400 | content_type = urlh.headers.get('Content-Type', '') | |
401 | webpage_bytes = urlh.read() | |
402 | if prefix is not None: | |
403 | webpage_bytes = prefix + webpage_bytes | |
404 | if not encoding: | |
405 | encoding = self._guess_encoding_from_content(content_type, webpage_bytes) | |
d6983cb4 PH |
406 | if self._downloader.params.get('dump_intermediate_pages', False): |
407 | try: | |
408 | url = url_or_request.get_full_url() | |
409 | except AttributeError: | |
410 | url = url_or_request | |
f1a9d64e | 411 | self.to_screen('Dumping request to ' + url) |
d6983cb4 PH |
412 | dump = base64.b64encode(webpage_bytes).decode('ascii') |
413 | self._downloader.to_screen(dump) | |
d41e6efc PH |
414 | if self._downloader.params.get('write_pages', False): |
415 | try: | |
416 | url = url_or_request.get_full_url() | |
417 | except AttributeError: | |
418 | url = url_or_request | |
5afa7f8b | 419 | basen = '%s_%s' % (video_id, url) |
c1bce22f | 420 | if len(basen) > 240: |
f1a9d64e | 421 | h = '___' + hashlib.md5(basen.encode('utf-8')).hexdigest() |
c1bce22f PH |
422 | basen = basen[:240 - len(h)] + h |
423 | raw_filename = basen + '.dump' | |
d41e6efc | 424 | filename = sanitize_filename(raw_filename, restricted=True) |
f1a9d64e | 425 | self.to_screen('Saving request to ' + filename) |
5f58165d S |
426 | # Working around MAX_PATH limitation on Windows (see |
427 | # http://msdn.microsoft.com/en-us/library/windows/desktop/aa365247(v=vs.85).aspx) | |
428 | if os.name == 'nt': | |
429 | absfilepath = os.path.abspath(filename) | |
430 | if len(absfilepath) > 259: | |
431 | filename = '\\\\?\\' + absfilepath | |
d41e6efc PH |
432 | with open(filename, 'wb') as outf: |
433 | outf.write(webpage_bytes) | |
434 | ||
ec0fafbb AA |
435 | try: |
436 | content = webpage_bytes.decode(encoding, 'replace') | |
437 | except LookupError: | |
438 | content = webpage_bytes.decode('utf-8', 'replace') | |
2410c43d | 439 | |
f1a9d64e PH |
440 | if ('<title>Access to this site is blocked</title>' in content and |
441 | 'Websense' in content[:512]): | |
442 | msg = 'Access to this webpage has been blocked by Websense filtering software in your network.' | |
2410c43d PH |
443 | blocked_iframe = self._html_search_regex( |
444 | r'<iframe src="([^"]+)"', content, | |
f1a9d64e | 445 | 'Websense information URL', default=None) |
2410c43d | 446 | if blocked_iframe: |
f1a9d64e | 447 | msg += ' Visit %s for more details' % blocked_iframe |
2410c43d | 448 | raise ExtractorError(msg, expected=True) |
77b2986b PH |
449 | if '<title>The URL you requested has been blocked</title>' in content[:512]: |
450 | msg = ( | |
451 | 'Access to this webpage has been blocked by Indian censorship. ' | |
452 | 'Use a VPN or proxy server (with --proxy) to route around it.') | |
453 | block_msg = self._html_search_regex( | |
454 | r'</h1><p>(.*?)</p>', | |
455 | content, 'block message', default=None) | |
456 | if block_msg: | |
457 | msg += ' (Message: "%s")' % block_msg.replace('\n', ' ') | |
458 | raise ExtractorError(msg, expected=True) | |
2410c43d | 459 | |
23be51d8 | 460 | return content |
d6983cb4 | 461 | |
c9a77969 | 462 | def _download_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True, tries=1, timeout=5, encoding=None): |
d6983cb4 | 463 | """ Returns the data of the page as a string """ |
995ad69c TF |
464 | success = False |
465 | try_count = 0 | |
466 | while success is False: | |
467 | try: | |
c9a77969 | 468 | res = self._download_webpage_handle(url_or_request, video_id, note, errnote, fatal, encoding=encoding) |
995ad69c TF |
469 | success = True |
470 | except compat_http_client.IncompleteRead as e: | |
471 | try_count += 1 | |
472 | if try_count >= tries: | |
473 | raise e | |
474 | self._sleep(timeout, video_id) | |
7cc3570e PH |
475 | if res is False: |
476 | return res | |
477 | else: | |
478 | content, _ = res | |
479 | return content | |
d6983cb4 | 480 | |
2a275ab0 | 481 | def _download_xml(self, url_or_request, video_id, |
f1a9d64e | 482 | note='Downloading XML', errnote='Unable to download XML', |
c9a77969 | 483 | transform_source=None, fatal=True, encoding=None): |
267ed0c5 | 484 | """Return the xml as an xml.etree.ElementTree.Element""" |
28746fbd | 485 | xml_string = self._download_webpage( |
c9a77969 | 486 | url_or_request, video_id, note, errnote, fatal=fatal, encoding=encoding) |
28746fbd PH |
487 | if xml_string is False: |
488 | return xml_string | |
e2b38da9 PH |
489 | if transform_source: |
490 | xml_string = transform_source(xml_string) | |
36e6f62c | 491 | return compat_etree_fromstring(xml_string.encode('utf-8')) |
267ed0c5 | 492 | |
3d3538e4 | 493 | def _download_json(self, url_or_request, video_id, |
f1a9d64e PH |
494 | note='Downloading JSON metadata', |
495 | errnote='Unable to download JSON metadata', | |
b090af59 | 496 | transform_source=None, |
c9a77969 | 497 | fatal=True, encoding=None): |
b090af59 | 498 | json_string = self._download_webpage( |
c9a77969 YCH |
499 | url_or_request, video_id, note, errnote, fatal=fatal, |
500 | encoding=encoding) | |
b090af59 PH |
501 | if (not fatal) and json_string is False: |
502 | return None | |
ebb64199 TF |
503 | return self._parse_json( |
504 | json_string, video_id, transform_source=transform_source, fatal=fatal) | |
505 | ||
506 | def _parse_json(self, json_string, video_id, transform_source=None, fatal=True): | |
81c2f20b PH |
507 | if transform_source: |
508 | json_string = transform_source(json_string) | |
3d3538e4 PH |
509 | try: |
510 | return json.loads(json_string) | |
511 | except ValueError as ve: | |
e7b6d122 PH |
512 | errmsg = '%s: Failed to parse JSON ' % video_id |
513 | if fatal: | |
514 | raise ExtractorError(errmsg, cause=ve) | |
515 | else: | |
516 | self.report_warning(errmsg + str(ve)) | |
3d3538e4 | 517 | |
f45f96f8 | 518 | def report_warning(self, msg, video_id=None): |
f1a9d64e | 519 | idstr = '' if video_id is None else '%s: ' % video_id |
f45f96f8 | 520 | self._downloader.report_warning( |
f1a9d64e | 521 | '[%s] %s%s' % (self.IE_NAME, idstr, msg)) |
f45f96f8 | 522 | |
d6983cb4 PH |
523 | def to_screen(self, msg): |
524 | """Print msg to screen, prefixing it with '[ie_name]'""" | |
f1a9d64e | 525 | self._downloader.to_screen('[%s] %s' % (self.IE_NAME, msg)) |
d6983cb4 PH |
526 | |
527 | def report_extraction(self, id_or_name): | |
528 | """Report information extraction.""" | |
f1a9d64e | 529 | self.to_screen('%s: Extracting information' % id_or_name) |
d6983cb4 PH |
530 | |
531 | def report_download_webpage(self, video_id): | |
532 | """Report webpage download.""" | |
f1a9d64e | 533 | self.to_screen('%s: Downloading webpage' % video_id) |
d6983cb4 PH |
534 | |
535 | def report_age_confirmation(self): | |
536 | """Report attempt to confirm age.""" | |
f1a9d64e | 537 | self.to_screen('Confirming age') |
d6983cb4 | 538 | |
fc79158d JMF |
539 | def report_login(self): |
540 | """Report attempt to log in.""" | |
f1a9d64e | 541 | self.to_screen('Logging in') |
fc79158d | 542 | |
43e7d3c9 S |
543 | @staticmethod |
544 | def raise_login_required(msg='This video is only available for registered users'): | |
545 | raise ExtractorError( | |
546 | '%s. Use --username and --password or --netrc to provide account credentials.' % msg, | |
547 | expected=True) | |
548 | ||
c430802e S |
549 | @staticmethod |
550 | def raise_geo_restricted(msg='This video is not available from your location due to geo restriction'): | |
551 | raise ExtractorError( | |
552 | '%s. You might want to use --proxy to workaround.' % msg, | |
553 | expected=True) | |
554 | ||
5f6a1245 | 555 | # Methods for following #608 |
c0d0b01f | 556 | @staticmethod |
830d53bf | 557 | def url_result(url, ie=None, video_id=None, video_title=None): |
10952eb2 | 558 | """Returns a URL that points to a page that should be processed""" |
5f6a1245 | 559 | # TODO: ie should be the class used for getting the info |
d6983cb4 PH |
560 | video_info = {'_type': 'url', |
561 | 'url': url, | |
562 | 'ie_key': ie} | |
7012b23c PH |
563 | if video_id is not None: |
564 | video_info['id'] = video_id | |
830d53bf S |
565 | if video_title is not None: |
566 | video_info['title'] = video_title | |
d6983cb4 | 567 | return video_info |
5f6a1245 | 568 | |
c0d0b01f | 569 | @staticmethod |
acf5cbfe | 570 | def playlist_result(entries, playlist_id=None, playlist_title=None, playlist_description=None): |
d6983cb4 PH |
571 | """Returns a playlist""" |
572 | video_info = {'_type': 'playlist', | |
573 | 'entries': entries} | |
574 | if playlist_id: | |
575 | video_info['id'] = playlist_id | |
576 | if playlist_title: | |
577 | video_info['title'] = playlist_title | |
acf5cbfe S |
578 | if playlist_description: |
579 | video_info['description'] = playlist_description | |
d6983cb4 PH |
580 | return video_info |
581 | ||
c342041f | 582 | def _search_regex(self, pattern, string, name, default=NO_DEFAULT, fatal=True, flags=0, group=None): |
d6983cb4 PH |
583 | """ |
584 | Perform a regex search on the given string, using a single or a list of | |
585 | patterns returning the first matching group. | |
586 | In case of failure return a default value or raise a WARNING or a | |
55b3e45b | 587 | RegexNotFoundError, depending on fatal, specifying the field name. |
d6983cb4 PH |
588 | """ |
589 | if isinstance(pattern, (str, compat_str, compiled_regex_type)): | |
590 | mobj = re.search(pattern, string, flags) | |
591 | else: | |
592 | for p in pattern: | |
593 | mobj = re.search(p, string, flags) | |
c3415d1b PH |
594 | if mobj: |
595 | break | |
d6983cb4 | 596 | |
7e5db8c9 | 597 | if not self._downloader.params.get('no_color') and os.name != 'nt' and sys.stderr.isatty(): |
f1a9d64e | 598 | _name = '\033[0;34m%s\033[0m' % name |
d6983cb4 PH |
599 | else: |
600 | _name = name | |
601 | ||
602 | if mobj: | |
711ede6e PH |
603 | if group is None: |
604 | # return the first matching group | |
605 | return next(g for g in mobj.groups() if g is not None) | |
606 | else: | |
607 | return mobj.group(group) | |
c342041f | 608 | elif default is not NO_DEFAULT: |
d6983cb4 PH |
609 | return default |
610 | elif fatal: | |
f1a9d64e | 611 | raise RegexNotFoundError('Unable to extract %s' % _name) |
d6983cb4 | 612 | else: |
08f2a92c | 613 | self._downloader.report_warning('unable to extract %s' % _name + bug_reports_message()) |
d6983cb4 PH |
614 | return None |
615 | ||
c342041f | 616 | def _html_search_regex(self, pattern, string, name, default=NO_DEFAULT, fatal=True, flags=0, group=None): |
d6983cb4 PH |
617 | """ |
618 | Like _search_regex, but strips HTML tags and unescapes entities. | |
619 | """ | |
711ede6e | 620 | res = self._search_regex(pattern, string, name, default, fatal, flags, group) |
d6983cb4 PH |
621 | if res: |
622 | return clean_html(res).strip() | |
623 | else: | |
624 | return res | |
625 | ||
fc79158d JMF |
626 | def _get_login_info(self): |
627 | """ | |
cf0649f8 | 628 | Get the login info as (username, password) |
fc79158d JMF |
629 | It will look in the netrc file using the _NETRC_MACHINE value |
630 | If there's no info available, return (None, None) | |
631 | """ | |
632 | if self._downloader is None: | |
633 | return (None, None) | |
634 | ||
635 | username = None | |
636 | password = None | |
637 | downloader_params = self._downloader.params | |
638 | ||
639 | # Attempt to use provided username and password or .netrc data | |
d800609c | 640 | if downloader_params.get('username') is not None: |
fc79158d JMF |
641 | username = downloader_params['username'] |
642 | password = downloader_params['password'] | |
643 | elif downloader_params.get('usenetrc', False): | |
644 | try: | |
645 | info = netrc.netrc().authenticators(self._NETRC_MACHINE) | |
646 | if info is not None: | |
647 | username = info[0] | |
648 | password = info[2] | |
649 | else: | |
650 | raise netrc.NetrcParseError('No authenticators for %s' % self._NETRC_MACHINE) | |
651 | except (IOError, netrc.NetrcParseError) as err: | |
9b9c5355 | 652 | self._downloader.report_warning('parsing .netrc: %s' % error_to_compat_str(err)) |
5f6a1245 | 653 | |
fc79158d JMF |
654 | return (username, password) |
655 | ||
e64b7569 | 656 | def _get_tfa_info(self, note='two-factor verification code'): |
83317f69 | 657 | """ |
658 | Get the two-factor authentication info | |
659 | TODO - asking the user will be required for sms/phone verify | |
660 | currently just uses the command line option | |
661 | If there's no info available, return None | |
662 | """ | |
663 | if self._downloader is None: | |
83317f69 | 664 | return None |
665 | downloader_params = self._downloader.params | |
666 | ||
d800609c | 667 | if downloader_params.get('twofactor') is not None: |
83317f69 | 668 | return downloader_params['twofactor'] |
669 | ||
e64b7569 | 670 | return compat_getpass('Type %s and press [Return]: ' % note) |
83317f69 | 671 | |
46720279 JMF |
672 | # Helper functions for extracting OpenGraph info |
673 | @staticmethod | |
ab2d5247 | 674 | def _og_regexes(prop): |
448ef1f3 | 675 | content_re = r'content=(?:"([^"]+?)"|\'([^\']+?)\'|\s*([^\s"\'=<>`]+?))' |
7a6d76a6 S |
676 | property_re = (r'(?:name|property)=(?:\'og:%(prop)s\'|"og:%(prop)s"|\s*og:%(prop)s\b)' |
677 | % {'prop': re.escape(prop)}) | |
78fb87b2 | 678 | template = r'<meta[^>]+?%s[^>]+?%s' |
ab2d5247 | 679 | return [ |
78fb87b2 JMF |
680 | template % (property_re, content_re), |
681 | template % (content_re, property_re), | |
ab2d5247 | 682 | ] |
46720279 | 683 | |
864f24bd S |
684 | @staticmethod |
685 | def _meta_regex(prop): | |
686 | return r'''(?isx)<meta | |
8b9848ac | 687 | (?=[^>]+(?:itemprop|name|property|id|http-equiv)=(["\']?)%s\1) |
864f24bd S |
688 | [^>]+?content=(["\'])(?P<content>.*?)\2''' % re.escape(prop) |
689 | ||
3c4e6d83 | 690 | def _og_search_property(self, prop, html, name=None, **kargs): |
46720279 | 691 | if name is None: |
3c4e6d83 | 692 | name = 'OpenGraph %s' % prop |
ab2d5247 | 693 | escaped = self._search_regex(self._og_regexes(prop), html, name, flags=re.DOTALL, **kargs) |
eb0a8398 PH |
694 | if escaped is None: |
695 | return None | |
696 | return unescapeHTML(escaped) | |
46720279 JMF |
697 | |
698 | def _og_search_thumbnail(self, html, **kargs): | |
10952eb2 | 699 | return self._og_search_property('image', html, 'thumbnail URL', fatal=False, **kargs) |
46720279 JMF |
700 | |
701 | def _og_search_description(self, html, **kargs): | |
702 | return self._og_search_property('description', html, fatal=False, **kargs) | |
703 | ||
704 | def _og_search_title(self, html, **kargs): | |
705 | return self._og_search_property('title', html, **kargs) | |
706 | ||
8ffa13e0 | 707 | def _og_search_video_url(self, html, name='video url', secure=True, **kargs): |
a3681973 PH |
708 | regexes = self._og_regexes('video') + self._og_regexes('video:url') |
709 | if secure: | |
710 | regexes = self._og_regexes('video:secure_url') + regexes | |
8ffa13e0 | 711 | return self._html_search_regex(regexes, html, name, **kargs) |
46720279 | 712 | |
78338f71 JMF |
713 | def _og_search_url(self, html, **kargs): |
714 | return self._og_search_property('url', html, **kargs) | |
715 | ||
40c696e5 | 716 | def _html_search_meta(self, name, html, display_name=None, fatal=False, **kwargs): |
59040888 PH |
717 | if display_name is None: |
718 | display_name = name | |
719 | return self._html_search_regex( | |
864f24bd | 720 | self._meta_regex(name), |
711ede6e | 721 | html, display_name, fatal=fatal, group='content', **kwargs) |
59040888 PH |
722 | |
723 | def _dc_search_uploader(self, html): | |
724 | return self._html_search_meta('dc.creator', html, 'uploader') | |
725 | ||
8dbe9899 PH |
726 | def _rta_search(self, html): |
727 | # See http://www.rtalabel.org/index.php?content=howtofaq#single | |
728 | if re.search(r'(?ix)<meta\s+name="rating"\s+' | |
729 | r' content="RTA-5042-1996-1400-1577-RTA"', | |
730 | html): | |
731 | return 18 | |
732 | return 0 | |
733 | ||
59040888 PH |
734 | def _media_rating_search(self, html): |
735 | # See http://www.tjg-designs.com/WP/metadata-code-examples-adding-metadata-to-your-web-pages/ | |
736 | rating = self._html_search_meta('rating', html) | |
737 | ||
738 | if not rating: | |
739 | return None | |
740 | ||
741 | RATING_TABLE = { | |
742 | 'safe for kids': 0, | |
743 | 'general': 8, | |
744 | '14 years': 14, | |
745 | 'mature': 17, | |
746 | 'restricted': 19, | |
747 | } | |
d800609c | 748 | return RATING_TABLE.get(rating.lower()) |
59040888 | 749 | |
69319969 | 750 | def _family_friendly_search(self, html): |
6ca7732d | 751 | # See http://schema.org/VideoObject |
69319969 NJ |
752 | family_friendly = self._html_search_meta('isFamilyFriendly', html) |
753 | ||
754 | if not family_friendly: | |
755 | return None | |
756 | ||
757 | RATING_TABLE = { | |
758 | '1': 0, | |
759 | 'true': 0, | |
760 | '0': 18, | |
761 | 'false': 18, | |
762 | } | |
d800609c | 763 | return RATING_TABLE.get(family_friendly.lower()) |
69319969 | 764 | |
0c708f11 JMF |
765 | def _twitter_search_player(self, html): |
766 | return self._html_search_meta('twitter:player', html, | |
9e1a5b84 | 767 | 'twitter card player') |
0c708f11 | 768 | |
0b26ba3f | 769 | def _search_json_ld(self, html, video_id, **kwargs): |
4ca2a3cf S |
770 | json_ld = self._search_regex( |
771 | r'(?s)<script[^>]+type=(["\'])application/ld\+json\1[^>]*>(?P<json_ld>.+?)</script>', | |
0b26ba3f | 772 | html, 'JSON-LD', group='json_ld', **kwargs) |
4ca2a3cf S |
773 | if not json_ld: |
774 | return {} | |
0b26ba3f | 775 | return self._json_ld(json_ld, video_id, fatal=kwargs.get('fatal', True)) |
4ca2a3cf S |
776 | |
777 | def _json_ld(self, json_ld, video_id, fatal=True): | |
778 | if isinstance(json_ld, compat_str): | |
779 | json_ld = self._parse_json(json_ld, video_id, fatal=fatal) | |
780 | if not json_ld: | |
781 | return {} | |
782 | info = {} | |
783 | if json_ld.get('@context') == 'http://schema.org': | |
784 | item_type = json_ld.get('@type') | |
785 | if item_type == 'TVEpisode': | |
786 | info.update({ | |
787 | 'episode': unescapeHTML(json_ld.get('name')), | |
788 | 'episode_number': int_or_none(json_ld.get('episodeNumber')), | |
789 | 'description': unescapeHTML(json_ld.get('description')), | |
790 | }) | |
791 | part_of_season = json_ld.get('partOfSeason') | |
792 | if isinstance(part_of_season, dict) and part_of_season.get('@type') == 'TVSeason': | |
793 | info['season_number'] = int_or_none(part_of_season.get('seasonNumber')) | |
794 | part_of_series = json_ld.get('partOfSeries') | |
795 | if isinstance(part_of_series, dict) and part_of_series.get('@type') == 'TVSeries': | |
796 | info['series'] = unescapeHTML(part_of_series.get('name')) | |
797 | elif item_type == 'Article': | |
798 | info.update({ | |
799 | 'timestamp': parse_iso8601(json_ld.get('datePublished')), | |
800 | 'title': unescapeHTML(json_ld.get('headline')), | |
801 | 'description': unescapeHTML(json_ld.get('articleBody')), | |
802 | }) | |
803 | return dict((k, v) for k, v in info.items() if v is not None) | |
804 | ||
27713812 | 805 | @staticmethod |
f8da79f8 | 806 | def _hidden_inputs(html): |
586f1cc5 | 807 | html = re.sub(r'<!--(?:(?!<!--).)*-->', '', html) |
201ea3ee | 808 | hidden_inputs = {} |
73eb13df | 809 | for input in re.findall(r'(?i)<input([^>]+)>', html): |
be0e5dbd | 810 | if not re.search(r'type=(["\'])(?:hidden|submit)\1', input): |
201ea3ee S |
811 | continue |
812 | name = re.search(r'name=(["\'])(?P<value>.+?)\1', input) | |
813 | if not name: | |
814 | continue | |
815 | value = re.search(r'value=(["\'])(?P<value>.*?)\1', input) | |
816 | if not value: | |
817 | continue | |
818 | hidden_inputs[name.group('value')] = value.group('value') | |
819 | return hidden_inputs | |
27713812 | 820 | |
cf61d96d S |
821 | def _form_hidden_inputs(self, form_id, html): |
822 | form = self._search_regex( | |
73eb13df | 823 | r'(?is)<form[^>]+?id=(["\'])%s\1[^>]*>(?P<form>.+?)</form>' % form_id, |
cf61d96d S |
824 | html, '%s form' % form_id, group='form') |
825 | return self._hidden_inputs(form) | |
826 | ||
3ded7bac | 827 | def _sort_formats(self, formats, field_preference=None): |
7e8caf30 | 828 | if not formats: |
f1a9d64e | 829 | raise ExtractorError('No video formats found') |
7e8caf30 | 830 | |
b0d21ded S |
831 | for f in formats: |
832 | # Automatically determine tbr when missing based on abr and vbr (improves | |
833 | # formats sorting in some cases) | |
350cf045 | 834 | if 'tbr' not in f and f.get('abr') is not None and f.get('vbr') is not None: |
b0d21ded S |
835 | f['tbr'] = f['abr'] + f['vbr'] |
836 | ||
4bcc7bd1 | 837 | def _formats_key(f): |
e6812ac9 PH |
838 | # TODO remove the following workaround |
839 | from ..utils import determine_ext | |
840 | if not f.get('ext') and 'url' in f: | |
841 | f['ext'] = determine_ext(f['url']) | |
842 | ||
3ded7bac S |
843 | if isinstance(field_preference, (list, tuple)): |
844 | return tuple(f.get(field) if f.get(field) is not None else -1 for field in field_preference) | |
845 | ||
4bcc7bd1 PH |
846 | preference = f.get('preference') |
847 | if preference is None: | |
d497a201 | 848 | preference = 0 |
4bcc7bd1 PH |
849 | if f.get('ext') in ['f4f', 'f4m']: # Not yet supported |
850 | preference -= 0.5 | |
851 | ||
d497a201 | 852 | proto_preference = 0 if determine_protocol(f) in ['http', 'https'] else -0.1 |
853 | ||
4bcc7bd1 PH |
854 | if f.get('vcodec') == 'none': # audio only |
855 | if self._downloader.params.get('prefer_free_formats'): | |
f1a9d64e | 856 | ORDER = ['aac', 'mp3', 'm4a', 'webm', 'ogg', 'opus'] |
4bcc7bd1 | 857 | else: |
f1a9d64e | 858 | ORDER = ['webm', 'opus', 'ogg', 'mp3', 'aac', 'm4a'] |
4bcc7bd1 PH |
859 | ext_preference = 0 |
860 | try: | |
861 | audio_ext_preference = ORDER.index(f['ext']) | |
862 | except ValueError: | |
863 | audio_ext_preference = -1 | |
864 | else: | |
865 | if self._downloader.params.get('prefer_free_formats'): | |
f1a9d64e | 866 | ORDER = ['flv', 'mp4', 'webm'] |
4bcc7bd1 | 867 | else: |
f1a9d64e | 868 | ORDER = ['webm', 'flv', 'mp4'] |
4bcc7bd1 PH |
869 | try: |
870 | ext_preference = ORDER.index(f['ext']) | |
871 | except ValueError: | |
872 | ext_preference = -1 | |
873 | audio_ext_preference = 0 | |
874 | ||
875 | return ( | |
876 | preference, | |
aff2f4f4 | 877 | f.get('language_preference') if f.get('language_preference') is not None else -1, |
5d73273f | 878 | f.get('quality') if f.get('quality') is not None else -1, |
9933b574 | 879 | f.get('tbr') if f.get('tbr') is not None else -1, |
03cd72b0 | 880 | f.get('filesize') if f.get('filesize') is not None else -1, |
4bcc7bd1 | 881 | f.get('vbr') if f.get('vbr') is not None else -1, |
1a6373ef PH |
882 | f.get('height') if f.get('height') is not None else -1, |
883 | f.get('width') if f.get('width') is not None else -1, | |
d497a201 | 884 | proto_preference, |
1e1896f2 | 885 | ext_preference, |
4bcc7bd1 PH |
886 | f.get('abr') if f.get('abr') is not None else -1, |
887 | audio_ext_preference, | |
2c8e03d9 | 888 | f.get('fps') if f.get('fps') is not None else -1, |
9732d77e | 889 | f.get('filesize_approx') if f.get('filesize_approx') is not None else -1, |
c64ed2a3 | 890 | f.get('source_preference') if f.get('source_preference') is not None else -1, |
74f72824 | 891 | f.get('format_id') if f.get('format_id') is not None else '', |
4bcc7bd1 PH |
892 | ) |
893 | formats.sort(key=_formats_key) | |
59040888 | 894 | |
96a53167 S |
895 | def _check_formats(self, formats, video_id): |
896 | if formats: | |
897 | formats[:] = filter( | |
898 | lambda f: self._is_valid_url( | |
899 | f['url'], video_id, | |
900 | item='%s video format' % f.get('format_id') if f.get('format_id') else 'video'), | |
901 | formats) | |
902 | ||
f5bdb444 S |
903 | @staticmethod |
904 | def _remove_duplicate_formats(formats): | |
905 | format_urls = set() | |
906 | unique_formats = [] | |
907 | for f in formats: | |
908 | if f['url'] not in format_urls: | |
909 | format_urls.add(f['url']) | |
910 | unique_formats.append(f) | |
911 | formats[:] = unique_formats | |
912 | ||
96a53167 | 913 | def _is_valid_url(self, url, video_id, item='video'): |
2f0f6578 S |
914 | url = self._proto_relative_url(url, scheme='http:') |
915 | # For now assume non HTTP(S) URLs always valid | |
916 | if not (url.startswith('http://') or url.startswith('https://')): | |
917 | return True | |
96a53167 | 918 | try: |
4069766c | 919 | self._request_webpage(url, video_id, 'Checking %s URL' % item) |
96a53167 S |
920 | return True |
921 | except ExtractorError as e: | |
943a1e24 | 922 | if isinstance(e.cause, compat_urllib_error.URLError): |
baa43cba S |
923 | self.to_screen( |
924 | '%s: %s URL is invalid, skipping' % (video_id, item)) | |
96a53167 S |
925 | return False |
926 | raise | |
927 | ||
20991253 | 928 | def http_scheme(self): |
1ede5b24 | 929 | """ Either "http:" or "https:", depending on the user's preferences """ |
20991253 PH |
930 | return ( |
931 | 'http:' | |
932 | if self._downloader.params.get('prefer_insecure', False) | |
933 | else 'https:') | |
934 | ||
57c7411f PH |
935 | def _proto_relative_url(self, url, scheme=None): |
936 | if url is None: | |
937 | return url | |
938 | if url.startswith('//'): | |
939 | if scheme is None: | |
940 | scheme = self.http_scheme() | |
941 | return scheme + url | |
942 | else: | |
943 | return url | |
944 | ||
4094b6e3 PH |
945 | def _sleep(self, timeout, video_id, msg_template=None): |
946 | if msg_template is None: | |
f1a9d64e | 947 | msg_template = '%(video_id)s: Waiting for %(timeout)s seconds' |
4094b6e3 PH |
948 | msg = msg_template % {'video_id': video_id, 'timeout': timeout} |
949 | self.to_screen(msg) | |
950 | time.sleep(timeout) | |
951 | ||
a38436e8 | 952 | def _extract_f4m_formats(self, manifest_url, video_id, preference=None, f4m_id=None, |
4de61310 S |
953 | transform_source=lambda s: fix_xml_ampersands(s).strip(), |
954 | fatal=True): | |
f036a632 JMF |
955 | manifest = self._download_xml( |
956 | manifest_url, video_id, 'Downloading f4m manifest', | |
97f4aecf S |
957 | 'Unable to download f4m manifest', |
958 | # Some manifests may be malformed, e.g. prosiebensat1 generated manifests | |
959 | # (see https://github.com/rg3/youtube-dl/issues/6215#issuecomment-121704244) | |
4de61310 S |
960 | transform_source=transform_source, |
961 | fatal=fatal) | |
962 | ||
963 | if manifest is False: | |
8d29e47f | 964 | return [] |
31bb8d3f JMF |
965 | |
966 | formats = [] | |
7a47d07c | 967 | manifest_version = '1.0' |
b2527359 | 968 | media_nodes = manifest.findall('{http://ns.adobe.com/f4m/1.0}media') |
34e48bed | 969 | if not media_nodes: |
7a47d07c | 970 | manifest_version = '2.0' |
34e48bed | 971 | media_nodes = manifest.findall('{http://ns.adobe.com/f4m/2.0}media') |
019839fa S |
972 | base_url = xpath_text( |
973 | manifest, ['{http://ns.adobe.com/f4m/1.0}baseURL', '{http://ns.adobe.com/f4m/2.0}baseURL'], | |
974 | 'base URL', default=None) | |
975 | if base_url: | |
976 | base_url = base_url.strip() | |
b2527359 | 977 | for i, media_el in enumerate(media_nodes): |
7a47d07c | 978 | if manifest_version == '2.0': |
31c746e5 S |
979 | media_url = media_el.attrib.get('href') or media_el.attrib.get('url') |
980 | if not media_url: | |
981 | continue | |
cc357c4d S |
982 | manifest_url = ( |
983 | media_url if media_url.startswith('http://') or media_url.startswith('https://') | |
019839fa | 984 | else ((base_url or '/'.join(manifest_url.split('/')[:-1])) + '/' + media_url)) |
70f0f5a8 S |
985 | # If media_url is itself a f4m manifest do the recursive extraction |
986 | # since bitrates in parent manifest (this one) and media_url manifest | |
987 | # may differ leading to inability to resolve the format by requested | |
988 | # bitrate in f4m downloader | |
989 | if determine_ext(manifest_url) == 'f4m': | |
7e5edcfd S |
990 | formats.extend(self._extract_f4m_formats( |
991 | manifest_url, video_id, preference, f4m_id, fatal=fatal)) | |
70f0f5a8 | 992 | continue |
b2527359 | 993 | tbr = int_or_none(media_el.attrib.get('bitrate')) |
31bb8d3f | 994 | formats.append({ |
e21a55ab | 995 | 'format_id': '-'.join(filter(None, [f4m_id, compat_str(i if tbr is None else tbr)])), |
31bb8d3f JMF |
996 | 'url': manifest_url, |
997 | 'ext': 'flv', | |
b2527359 | 998 | 'tbr': tbr, |
31bb8d3f JMF |
999 | 'width': int_or_none(media_el.attrib.get('width')), |
1000 | 'height': int_or_none(media_el.attrib.get('height')), | |
60ca389c | 1001 | 'preference': preference, |
31bb8d3f JMF |
1002 | }) |
1003 | self._sort_formats(formats) | |
1004 | ||
1005 | return formats | |
1006 | ||
f0b5d6af | 1007 | def _extract_m3u8_formats(self, m3u8_url, video_id, ext=None, |
60ca389c | 1008 | entry_protocol='m3u8', preference=None, |
13af92fd YCH |
1009 | m3u8_id=None, note=None, errnote=None, |
1010 | fatal=True): | |
f0b5d6af | 1011 | |
704df56d | 1012 | formats = [{ |
f207019c | 1013 | 'format_id': '-'.join(filter(None, [m3u8_id, 'meta'])), |
704df56d PH |
1014 | 'url': m3u8_url, |
1015 | 'ext': ext, | |
1016 | 'protocol': 'm3u8', | |
9fe6ef7a | 1017 | 'preference': preference - 1 if preference else -1, |
704df56d PH |
1018 | 'resolution': 'multiple', |
1019 | 'format_note': 'Quality selection URL', | |
1020 | }] | |
1021 | ||
f0b5d6af PH |
1022 | format_url = lambda u: ( |
1023 | u | |
1024 | if re.match(r'^https?://', u) | |
1025 | else compat_urlparse.urljoin(m3u8_url, u)) | |
1026 | ||
dbd82a1d | 1027 | res = self._download_webpage_handle( |
81515ad9 | 1028 | m3u8_url, video_id, |
621ed9f5 | 1029 | note=note or 'Downloading m3u8 information', |
13af92fd YCH |
1030 | errnote=errnote or 'Failed to download m3u8 information', |
1031 | fatal=fatal) | |
dbd82a1d | 1032 | if res is False: |
8d29e47f | 1033 | return [] |
dbd82a1d | 1034 | m3u8_doc, urlh = res |
37113045 | 1035 | m3u8_url = urlh.geturl() |
9cdffeeb S |
1036 | |
1037 | # We should try extracting formats only from master playlists [1], i.e. | |
1038 | # playlists that describe available qualities. On the other hand media | |
1039 | # playlists [2] should be returned as is since they contain just the media | |
1040 | # without qualities renditions. | |
1041 | # Fortunately, master playlist can be easily distinguished from media | |
1042 | # playlist based on particular tags availability. As of [1, 2] master | |
1043 | # playlist tags MUST NOT appear in a media playist and vice versa. | |
1044 | # As of [3] #EXT-X-TARGETDURATION tag is REQUIRED for every media playlist | |
1045 | # and MUST NOT appear in master playlist thus we can clearly detect media | |
1046 | # playlist with this criterion. | |
1047 | # 1. https://tools.ietf.org/html/draft-pantos-http-live-streaming-17#section-4.3.4 | |
1048 | # 2. https://tools.ietf.org/html/draft-pantos-http-live-streaming-17#section-4.3.3 | |
1049 | # 3. https://tools.ietf.org/html/draft-pantos-http-live-streaming-17#section-4.3.3.1 | |
1050 | if '#EXT-X-TARGETDURATION' in m3u8_doc: # media playlist, return as is | |
7f32e5dc | 1051 | return [{ |
1052 | 'url': m3u8_url, | |
1053 | 'format_id': m3u8_id, | |
1054 | 'ext': ext, | |
1055 | 'protocol': entry_protocol, | |
1056 | 'preference': preference, | |
1057 | }] | |
704df56d | 1058 | last_info = None |
fa156077 | 1059 | last_media = None |
704df56d PH |
1060 | kv_rex = re.compile( |
1061 | r'(?P<key>[a-zA-Z_-]+)=(?P<val>"[^"]+"|[^",]+)(?:,|$)') | |
1062 | for line in m3u8_doc.splitlines(): | |
1063 | if line.startswith('#EXT-X-STREAM-INF:'): | |
1064 | last_info = {} | |
1065 | for m in kv_rex.finditer(line): | |
1066 | v = m.group('val') | |
1067 | if v.startswith('"'): | |
1068 | v = v[1:-1] | |
1069 | last_info[m.group('key')] = v | |
4cd95bcb JMF |
1070 | elif line.startswith('#EXT-X-MEDIA:'): |
1071 | last_media = {} | |
1072 | for m in kv_rex.finditer(line): | |
1073 | v = m.group('val') | |
1074 | if v.startswith('"'): | |
1075 | v = v[1:-1] | |
1076 | last_media[m.group('key')] = v | |
704df56d PH |
1077 | elif line.startswith('#') or not line.strip(): |
1078 | continue | |
1079 | else: | |
daebaab6 | 1080 | if last_info is None: |
f0b5d6af | 1081 | formats.append({'url': format_url(line)}) |
3524cc25 | 1082 | continue |
704df56d | 1083 | tbr = int_or_none(last_info.get('BANDWIDTH'), scale=1000) |
8dc9d361 S |
1084 | format_id = [] |
1085 | if m3u8_id: | |
1086 | format_id.append(m3u8_id) | |
05d5392c | 1087 | last_media_name = last_media.get('NAME') if last_media and last_media.get('TYPE') != 'SUBTITLES' else None |
8dc9d361 | 1088 | format_id.append(last_media_name if last_media_name else '%d' % (tbr if tbr else len(formats))) |
704df56d | 1089 | f = { |
8dc9d361 | 1090 | 'format_id': '-'.join(format_id), |
f0b5d6af | 1091 | 'url': format_url(line.strip()), |
704df56d PH |
1092 | 'tbr': tbr, |
1093 | 'ext': ext, | |
f0b5d6af PH |
1094 | 'protocol': entry_protocol, |
1095 | 'preference': preference, | |
704df56d | 1096 | } |
704df56d PH |
1097 | resolution = last_info.get('RESOLUTION') |
1098 | if resolution: | |
1099 | width_str, height_str = resolution.split('x') | |
1100 | f['width'] = int(width_str) | |
1101 | f['height'] = int(height_str) | |
fbb6edd2 S |
1102 | codecs = last_info.get('CODECS') |
1103 | if codecs: | |
1104 | vcodec, acodec = [None] * 2 | |
1105 | va_codecs = codecs.split(',') | |
1106 | if len(va_codecs) == 1: | |
1107 | # Audio only entries usually come with single codec and | |
1108 | # no resolution. For more robustness we also check it to | |
1109 | # be mp4 audio. | |
1110 | if not resolution and va_codecs[0].startswith('mp4a'): | |
1111 | vcodec, acodec = 'none', va_codecs[0] | |
1112 | else: | |
1113 | vcodec = va_codecs[0] | |
1114 | else: | |
1115 | vcodec, acodec = va_codecs[:2] | |
1116 | f.update({ | |
1117 | 'acodec': acodec, | |
1118 | 'vcodec': vcodec, | |
1119 | }) | |
4cd95bcb JMF |
1120 | if last_media is not None: |
1121 | f['m3u8_media'] = last_media | |
1122 | last_media = None | |
704df56d PH |
1123 | formats.append(f) |
1124 | last_info = {} | |
1125 | self._sort_formats(formats) | |
1126 | return formats | |
1127 | ||
a107193e S |
1128 | @staticmethod |
1129 | def _xpath_ns(path, namespace=None): | |
1130 | if not namespace: | |
1131 | return path | |
1132 | out = [] | |
1133 | for c in path.split('/'): | |
1134 | if not c or c == '.': | |
1135 | out.append(c) | |
1136 | else: | |
1137 | out.append('{%s}%s' % (namespace, c)) | |
1138 | return '/'.join(out) | |
1139 | ||
1140 | def _extract_smil_formats(self, smil_url, video_id, fatal=True, f4m_params=None): | |
1141 | smil = self._download_smil(smil_url, video_id, fatal=fatal) | |
1142 | ||
995029a1 PH |
1143 | if smil is False: |
1144 | assert not fatal | |
1145 | return [] | |
e89a2aab | 1146 | |
17712eeb | 1147 | namespace = self._parse_smil_namespace(smil) |
a107193e S |
1148 | |
1149 | return self._parse_smil_formats( | |
1150 | smil, smil_url, video_id, namespace=namespace, f4m_params=f4m_params) | |
1151 | ||
1152 | def _extract_smil_info(self, smil_url, video_id, fatal=True, f4m_params=None): | |
1153 | smil = self._download_smil(smil_url, video_id, fatal=fatal) | |
1154 | if smil is False: | |
1155 | return {} | |
1156 | return self._parse_smil(smil, smil_url, video_id, f4m_params=f4m_params) | |
1157 | ||
1158 | def _download_smil(self, smil_url, video_id, fatal=True): | |
1159 | return self._download_xml( | |
1160 | smil_url, video_id, 'Downloading SMIL file', | |
1161 | 'Unable to download SMIL file', fatal=fatal) | |
1162 | ||
1163 | def _parse_smil(self, smil, smil_url, video_id, f4m_params=None): | |
17712eeb | 1164 | namespace = self._parse_smil_namespace(smil) |
a107193e S |
1165 | |
1166 | formats = self._parse_smil_formats( | |
1167 | smil, smil_url, video_id, namespace=namespace, f4m_params=f4m_params) | |
1168 | subtitles = self._parse_smil_subtitles(smil, namespace=namespace) | |
1169 | ||
1170 | video_id = os.path.splitext(url_basename(smil_url))[0] | |
1171 | title = None | |
1172 | description = None | |
647eab45 | 1173 | upload_date = None |
a107193e S |
1174 | for meta in smil.findall(self._xpath_ns('./head/meta', namespace)): |
1175 | name = meta.attrib.get('name') | |
1176 | content = meta.attrib.get('content') | |
1177 | if not name or not content: | |
1178 | continue | |
1179 | if not title and name == 'title': | |
1180 | title = content | |
1181 | elif not description and name in ('description', 'abstract'): | |
1182 | description = content | |
647eab45 S |
1183 | elif not upload_date and name == 'date': |
1184 | upload_date = unified_strdate(content) | |
a107193e | 1185 | |
1e5bcdec S |
1186 | thumbnails = [{ |
1187 | 'id': image.get('type'), | |
1188 | 'url': image.get('src'), | |
1189 | 'width': int_or_none(image.get('width')), | |
1190 | 'height': int_or_none(image.get('height')), | |
1191 | } for image in smil.findall(self._xpath_ns('.//image', namespace)) if image.get('src')] | |
1192 | ||
a107193e S |
1193 | return { |
1194 | 'id': video_id, | |
1195 | 'title': title or video_id, | |
1196 | 'description': description, | |
647eab45 | 1197 | 'upload_date': upload_date, |
1e5bcdec | 1198 | 'thumbnails': thumbnails, |
a107193e S |
1199 | 'formats': formats, |
1200 | 'subtitles': subtitles, | |
1201 | } | |
1202 | ||
17712eeb S |
1203 | def _parse_smil_namespace(self, smil): |
1204 | return self._search_regex( | |
1205 | r'(?i)^{([^}]+)?}smil$', smil.tag, 'namespace', default=None) | |
1206 | ||
f877c6ae | 1207 | def _parse_smil_formats(self, smil, smil_url, video_id, namespace=None, f4m_params=None, transform_rtmp_url=None): |
a107193e S |
1208 | base = smil_url |
1209 | for meta in smil.findall(self._xpath_ns('./head/meta', namespace)): | |
1210 | b = meta.get('base') or meta.get('httpBase') | |
1211 | if b: | |
1212 | base = b | |
1213 | break | |
e89a2aab S |
1214 | |
1215 | formats = [] | |
1216 | rtmp_count = 0 | |
a107193e | 1217 | http_count = 0 |
7f32e5dc | 1218 | m3u8_count = 0 |
a107193e | 1219 | |
81e1c4e2 | 1220 | srcs = [] |
a107193e S |
1221 | videos = smil.findall(self._xpath_ns('.//video', namespace)) |
1222 | for video in videos: | |
1223 | src = video.get('src') | |
81e1c4e2 | 1224 | if not src or src in srcs: |
a107193e | 1225 | continue |
81e1c4e2 | 1226 | srcs.append(src) |
a107193e | 1227 | |
e7d8e98a | 1228 | bitrate = float_or_none(video.get('system-bitrate') or video.get('systemBitrate'), 1000) |
a107193e S |
1229 | filesize = int_or_none(video.get('size') or video.get('fileSize')) |
1230 | width = int_or_none(video.get('width')) | |
1231 | height = int_or_none(video.get('height')) | |
1232 | proto = video.get('proto') | |
1233 | ext = video.get('ext') | |
1234 | src_ext = determine_ext(src) | |
1235 | streamer = video.get('streamer') or base | |
1236 | ||
1237 | if proto == 'rtmp' or streamer.startswith('rtmp'): | |
1238 | rtmp_count += 1 | |
1239 | formats.append({ | |
1240 | 'url': streamer, | |
1241 | 'play_path': src, | |
1242 | 'ext': 'flv', | |
1243 | 'format_id': 'rtmp-%d' % (rtmp_count if bitrate is None else bitrate), | |
1244 | 'tbr': bitrate, | |
1245 | 'filesize': filesize, | |
1246 | 'width': width, | |
1247 | 'height': height, | |
1248 | }) | |
f877c6ae YCH |
1249 | if transform_rtmp_url: |
1250 | streamer, src = transform_rtmp_url(streamer, src) | |
1251 | formats[-1].update({ | |
1252 | 'url': streamer, | |
1253 | 'play_path': src, | |
1254 | }) | |
a107193e S |
1255 | continue |
1256 | ||
1257 | src_url = src if src.startswith('http') else compat_urlparse.urljoin(base, src) | |
c349456e | 1258 | src_url = src_url.strip() |
a107193e S |
1259 | |
1260 | if proto == 'm3u8' or src_ext == 'm3u8': | |
7f32e5dc | 1261 | m3u8_formats = self._extract_m3u8_formats( |
1262 | src_url, video_id, ext or 'mp4', m3u8_id='hls', fatal=False) | |
1263 | if len(m3u8_formats) == 1: | |
1264 | m3u8_count += 1 | |
1265 | m3u8_formats[0].update({ | |
1266 | 'format_id': 'hls-%d' % (m3u8_count if bitrate is None else bitrate), | |
1267 | 'tbr': bitrate, | |
1268 | 'width': width, | |
1269 | 'height': height, | |
1270 | }) | |
1271 | formats.extend(m3u8_formats) | |
a107193e S |
1272 | continue |
1273 | ||
1274 | if src_ext == 'f4m': | |
1275 | f4m_url = src_url | |
1276 | if not f4m_params: | |
1277 | f4m_params = { | |
1278 | 'hdcore': '3.2.0', | |
1279 | 'plugin': 'flowplayer-3.2.0.1', | |
1280 | } | |
1281 | f4m_url += '&' if '?' in f4m_url else '?' | |
41c3a5a7 | 1282 | f4m_url += compat_urllib_parse.urlencode(f4m_params) |
7e5edcfd | 1283 | formats.extend(self._extract_f4m_formats(f4m_url, video_id, f4m_id='hds', fatal=False)) |
a107193e S |
1284 | continue |
1285 | ||
c78e4817 | 1286 | if src_url.startswith('http') and self._is_valid_url(src, video_id): |
a107193e S |
1287 | http_count += 1 |
1288 | formats.append({ | |
1289 | 'url': src_url, | |
1290 | 'ext': ext or src_ext or 'flv', | |
1291 | 'format_id': 'http-%d' % (bitrate or http_count), | |
1292 | 'tbr': bitrate, | |
1293 | 'filesize': filesize, | |
1294 | 'width': width, | |
1295 | 'height': height, | |
1296 | }) | |
1297 | continue | |
63757032 | 1298 | |
e89a2aab S |
1299 | self._sort_formats(formats) |
1300 | ||
1301 | return formats | |
1302 | ||
ce00af87 | 1303 | def _parse_smil_subtitles(self, smil, namespace=None, subtitles_lang='en'): |
d413095f | 1304 | urls = [] |
a107193e S |
1305 | subtitles = {} |
1306 | for num, textstream in enumerate(smil.findall(self._xpath_ns('.//textstream', namespace))): | |
1307 | src = textstream.get('src') | |
d413095f | 1308 | if not src or src in urls: |
a107193e | 1309 | continue |
d413095f | 1310 | urls.append(src) |
cafcf657 | 1311 | ext = textstream.get('ext') or determine_ext(src) or mimetype2ext(textstream.get('type')) |
03bc7237 | 1312 | lang = textstream.get('systemLanguage') or textstream.get('systemLanguageName') or textstream.get('lang') or subtitles_lang |
a107193e S |
1313 | subtitles.setdefault(lang, []).append({ |
1314 | 'url': src, | |
1315 | 'ext': ext, | |
1316 | }) | |
1317 | return subtitles | |
63757032 | 1318 | |
942acef5 S |
1319 | def _extract_xspf_playlist(self, playlist_url, playlist_id, fatal=True): |
1320 | xspf = self._download_xml( | |
8d6765cf | 1321 | playlist_url, playlist_id, 'Downloading xpsf playlist', |
942acef5 S |
1322 | 'Unable to download xspf manifest', fatal=fatal) |
1323 | if xspf is False: | |
1324 | return [] | |
1325 | return self._parse_xspf(xspf, playlist_id) | |
8d6765cf | 1326 | |
942acef5 | 1327 | def _parse_xspf(self, playlist, playlist_id): |
8d6765cf S |
1328 | NS_MAP = { |
1329 | 'xspf': 'http://xspf.org/ns/0/', | |
1330 | 's1': 'http://static.streamone.nl/player/ns/0', | |
1331 | } | |
1332 | ||
1333 | entries = [] | |
1334 | for track in playlist.findall(xpath_with_ns('./xspf:trackList/xspf:track', NS_MAP)): | |
1335 | title = xpath_text( | |
98044462 | 1336 | track, xpath_with_ns('./xspf:title', NS_MAP), 'title', default=playlist_id) |
8d6765cf S |
1337 | description = xpath_text( |
1338 | track, xpath_with_ns('./xspf:annotation', NS_MAP), 'description') | |
1339 | thumbnail = xpath_text( | |
1340 | track, xpath_with_ns('./xspf:image', NS_MAP), 'thumbnail') | |
1341 | duration = float_or_none( | |
1342 | xpath_text(track, xpath_with_ns('./xspf:duration', NS_MAP), 'duration'), 1000) | |
1343 | ||
1344 | formats = [{ | |
1345 | 'url': location.text, | |
1346 | 'format_id': location.get(xpath_with_ns('s1:label', NS_MAP)), | |
1347 | 'width': int_or_none(location.get(xpath_with_ns('s1:width', NS_MAP))), | |
1348 | 'height': int_or_none(location.get(xpath_with_ns('s1:height', NS_MAP))), | |
1349 | } for location in track.findall(xpath_with_ns('./xspf:location', NS_MAP))] | |
1350 | self._sort_formats(formats) | |
1351 | ||
1352 | entries.append({ | |
1353 | 'id': playlist_id, | |
1354 | 'title': title, | |
1355 | 'description': description, | |
1356 | 'thumbnail': thumbnail, | |
1357 | 'duration': duration, | |
1358 | 'formats': formats, | |
1359 | }) | |
1360 | return entries | |
1361 | ||
1bac3455 | 1362 | def _extract_mpd_formats(self, mpd_url, video_id, mpd_id=None, note=None, errnote=None, fatal=True, formats_dict={}): |
1363 | res = self._download_webpage_handle( | |
1364 | mpd_url, video_id, | |
1365 | note=note or 'Downloading MPD manifest', | |
1366 | errnote=errnote or 'Failed to download MPD manifest', | |
2d2fa82d | 1367 | fatal=fatal) |
1bac3455 | 1368 | if res is False: |
2d2fa82d | 1369 | return [] |
1bac3455 | 1370 | mpd, urlh = res |
1371 | mpd_base_url = re.match(r'https?://.+/', urlh.geturl()).group() | |
1372 | ||
91cb6b50 | 1373 | return self._parse_mpd_formats( |
1bac3455 | 1374 | compat_etree_fromstring(mpd.encode('utf-8')), mpd_id, mpd_base_url, formats_dict=formats_dict) |
2d2fa82d | 1375 | |
91cb6b50 | 1376 | def _parse_mpd_formats(self, mpd_doc, mpd_id=None, mpd_base_url='', formats_dict={}): |
1bac3455 | 1377 | if mpd_doc.get('type') == 'dynamic': |
1378 | return [] | |
2d2fa82d | 1379 | |
91cb6b50 | 1380 | namespace = self._search_regex(r'(?i)^{([^}]+)?}MPD$', mpd_doc.tag, 'namespace', default=None) |
f14be228 | 1381 | |
1382 | def _add_ns(path): | |
1383 | return self._xpath_ns(path, namespace) | |
1384 | ||
675d0016 | 1385 | def is_drm_protected(element): |
1386 | return element.find(_add_ns('ContentProtection')) is not None | |
1387 | ||
1bac3455 | 1388 | def extract_multisegment_info(element, ms_parent_info): |
1389 | ms_info = ms_parent_info.copy() | |
f14be228 | 1390 | segment_list = element.find(_add_ns('SegmentList')) |
1bac3455 | 1391 | if segment_list is not None: |
f14be228 | 1392 | segment_urls_e = segment_list.findall(_add_ns('SegmentURL')) |
1bac3455 | 1393 | if segment_urls_e: |
1394 | ms_info['segment_urls'] = [segment.attrib['media'] for segment in segment_urls_e] | |
f14be228 | 1395 | initialization = segment_list.find(_add_ns('Initialization')) |
1bac3455 | 1396 | if initialization is not None: |
1397 | ms_info['initialization_url'] = initialization.attrib['sourceURL'] | |
1398 | else: | |
f14be228 | 1399 | segment_template = element.find(_add_ns('SegmentTemplate')) |
1bac3455 | 1400 | if segment_template is not None: |
1401 | start_number = segment_template.get('startNumber') | |
1402 | if start_number: | |
1403 | ms_info['start_number'] = int(start_number) | |
f14be228 | 1404 | segment_timeline = segment_template.find(_add_ns('SegmentTimeline')) |
1bac3455 | 1405 | if segment_timeline is not None: |
f14be228 | 1406 | s_e = segment_timeline.findall(_add_ns('S')) |
1bac3455 | 1407 | if s_e: |
1408 | ms_info['total_number'] = 0 | |
1409 | for s in s_e: | |
1410 | ms_info['total_number'] += 1 + int(s.get('r', '0')) | |
1411 | else: | |
1412 | timescale = segment_template.get('timescale') | |
1413 | if timescale: | |
1414 | ms_info['timescale'] = int(timescale) | |
1415 | segment_duration = segment_template.get('duration') | |
1416 | if segment_duration: | |
1417 | ms_info['segment_duration'] = int(segment_duration) | |
1418 | media_template = segment_template.get('media') | |
1419 | if media_template: | |
1420 | ms_info['media_template'] = media_template | |
1421 | initialization = segment_template.get('initialization') | |
1422 | if initialization: | |
1423 | ms_info['initialization_url'] = initialization | |
1424 | else: | |
f14be228 | 1425 | initialization = segment_template.find(_add_ns('Initialization')) |
1bac3455 | 1426 | if initialization is not None: |
1427 | ms_info['initialization_url'] = initialization.attrib['sourceURL'] | |
1428 | return ms_info | |
b323e170 | 1429 | |
1bac3455 | 1430 | mpd_duration = parse_duration(mpd_doc.get('mediaPresentationDuration')) |
17b598d3 | 1431 | formats = [] |
f14be228 | 1432 | for period in mpd_doc.findall(_add_ns('Period')): |
1bac3455 | 1433 | period_duration = parse_duration(period.get('duration')) or mpd_duration |
1434 | period_ms_info = extract_multisegment_info(period, { | |
1435 | 'start_number': 1, | |
1436 | 'timescale': 1, | |
1437 | }) | |
f14be228 | 1438 | for adaptation_set in period.findall(_add_ns('AdaptationSet')): |
675d0016 | 1439 | if is_drm_protected(adaptation_set): |
1440 | continue | |
1bac3455 | 1441 | adaption_set_ms_info = extract_multisegment_info(adaptation_set, period_ms_info) |
f14be228 | 1442 | for representation in adaptation_set.findall(_add_ns('Representation')): |
675d0016 | 1443 | if is_drm_protected(representation): |
1444 | continue | |
1bac3455 | 1445 | representation_attrib = adaptation_set.attrib.copy() |
1446 | representation_attrib.update(representation.attrib) | |
1447 | mime_type = representation_attrib.get('mimeType') | |
1448 | content_type = mime_type.split('/')[0] if mime_type else representation_attrib.get('contentType') | |
1449 | if content_type == 'text': | |
1450 | # TODO implement WebVTT downloading | |
1451 | pass | |
1452 | elif content_type == 'video' or content_type == 'audio': | |
1453 | base_url = '' | |
1454 | for element in (representation, adaptation_set, period, mpd_doc): | |
f14be228 | 1455 | base_url_e = element.find(_add_ns('BaseURL')) |
1bac3455 | 1456 | if base_url_e is not None: |
1457 | base_url = base_url_e.text + base_url | |
1458 | if re.match(r'^https?://', base_url): | |
1459 | break | |
bb20526b S |
1460 | if mpd_base_url and not re.match(r'^https?://', base_url): |
1461 | if not mpd_base_url.endswith('/') and not base_url.startswith('/'): | |
1462 | mpd_base_url += '/' | |
1bac3455 | 1463 | base_url = mpd_base_url + base_url |
1464 | representation_id = representation_attrib.get('id') | |
d577c796 | 1465 | lang = representation_attrib.get('lang') |
51e9094f | 1466 | url_el = representation.find(_add_ns('BaseURL')) |
1467 | 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 | 1468 | f = { |
154c209e | 1469 | 'format_id': '%s-%s' % (mpd_id, representation_id) if mpd_id else representation_id, |
1bac3455 | 1470 | 'url': base_url, |
1471 | 'width': int_or_none(representation_attrib.get('width')), | |
1472 | 'height': int_or_none(representation_attrib.get('height')), | |
1473 | 'tbr': int_or_none(representation_attrib.get('bandwidth'), 1000), | |
1474 | 'asr': int_or_none(representation_attrib.get('audioSamplingRate')), | |
1475 | 'fps': int_or_none(representation_attrib.get('frameRate')), | |
1476 | 'vcodec': 'none' if content_type == 'audio' else representation_attrib.get('codecs'), | |
1477 | 'acodec': 'none' if content_type == 'video' else representation_attrib.get('codecs'), | |
d577c796 | 1478 | 'language': lang if lang not in ('mul', 'und', 'zxx', 'mis') else None, |
1bac3455 | 1479 | 'format_note': 'DASH %s' % content_type, |
51e9094f | 1480 | 'filesize': filesize, |
1bac3455 | 1481 | } |
1482 | representation_ms_info = extract_multisegment_info(representation, adaption_set_ms_info) | |
1483 | if 'segment_urls' not in representation_ms_info and 'media_template' in representation_ms_info: | |
1484 | if 'total_number' not in representation_ms_info and 'segment_duration': | |
6a3828fd | 1485 | segment_duration = float(representation_ms_info['segment_duration']) / float(representation_ms_info['timescale']) |
1486 | representation_ms_info['total_number'] = int(math.ceil(float(period_duration) / segment_duration)) | |
1bac3455 | 1487 | media_template = representation_ms_info['media_template'] |
1488 | media_template = media_template.replace('$RepresentationID$', representation_id) | |
53c269c6 | 1489 | media_template = re.sub(r'\$(Number|Bandwidth)(?:%(0\d+)d)?\$', r'%(\1)\2d', media_template) |
1bac3455 | 1490 | media_template.replace('$$', '$') |
53c269c6 | 1491 | representation_ms_info['segment_urls'] = [media_template % {'Number': segment_number, 'Bandwidth': representation_attrib.get('bandwidth')} for segment_number in range(representation_ms_info['start_number'], representation_ms_info['total_number'] + representation_ms_info['start_number'])] |
1bac3455 | 1492 | if 'segment_urls' in representation_ms_info: |
1493 | f.update({ | |
1494 | 'segment_urls': representation_ms_info['segment_urls'], | |
1495 | 'protocol': 'http_dash_segments', | |
df374b52 | 1496 | }) |
1bac3455 | 1497 | if 'initialization_url' in representation_ms_info: |
1498 | initialization_url = representation_ms_info['initialization_url'].replace('$RepresentationID$', representation_id) | |
1499 | f.update({ | |
1500 | 'initialization_url': initialization_url, | |
1501 | }) | |
1502 | if not f.get('url'): | |
1503 | f['url'] = initialization_url | |
1504 | try: | |
1505 | existing_format = next( | |
1506 | fo for fo in formats | |
1507 | if fo['format_id'] == representation_id) | |
1508 | except StopIteration: | |
1509 | full_info = formats_dict.get(representation_id, {}).copy() | |
1510 | full_info.update(f) | |
1511 | formats.append(full_info) | |
1512 | else: | |
1513 | existing_format.update(f) | |
17b598d3 | 1514 | else: |
1bac3455 | 1515 | self.report_warning('Unknown MIME type %s in DASH manifest' % mime_type) |
0826a0b5 | 1516 | self._sort_formats(formats) |
17b598d3 YCH |
1517 | return formats |
1518 | ||
f4b1c7ad PH |
1519 | def _live_title(self, name): |
1520 | """ Generate the title for a live video """ | |
1521 | now = datetime.datetime.now() | |
611c1dd9 | 1522 | now_str = now.strftime('%Y-%m-%d %H:%M') |
f4b1c7ad PH |
1523 | return name + ' ' + now_str |
1524 | ||
b14f3a4c PH |
1525 | def _int(self, v, name, fatal=False, **kwargs): |
1526 | res = int_or_none(v, **kwargs) | |
1527 | if 'get_attr' in kwargs: | |
1528 | print(getattr(v, kwargs['get_attr'])) | |
1529 | if res is None: | |
1530 | msg = 'Failed to extract %s: Could not parse value %r' % (name, v) | |
1531 | if fatal: | |
1532 | raise ExtractorError(msg) | |
1533 | else: | |
1534 | self._downloader.report_warning(msg) | |
1535 | return res | |
1536 | ||
1537 | def _float(self, v, name, fatal=False, **kwargs): | |
1538 | res = float_or_none(v, **kwargs) | |
1539 | if res is None: | |
1540 | msg = 'Failed to extract %s: Could not parse value %r' % (name, v) | |
1541 | if fatal: | |
1542 | raise ExtractorError(msg) | |
1543 | else: | |
1544 | self._downloader.report_warning(msg) | |
1545 | return res | |
1546 | ||
42939b61 | 1547 | def _set_cookie(self, domain, name, value, expire_time=None): |
810fb84d PH |
1548 | cookie = compat_cookiejar.Cookie( |
1549 | 0, name, value, None, None, domain, None, | |
42939b61 JMF |
1550 | None, '/', True, False, expire_time, '', None, None, None) |
1551 | self._downloader.cookiejar.set_cookie(cookie) | |
1552 | ||
799207e8 | 1553 | def _get_cookies(self, url): |
1554 | """ Return a compat_cookies.SimpleCookie with the cookies for the url """ | |
5c2266df | 1555 | req = sanitized_Request(url) |
799207e8 | 1556 | self._downloader.cookiejar.add_cookie_header(req) |
1557 | return compat_cookies.SimpleCookie(req.get_header('Cookie')) | |
1558 | ||
05900629 PH |
1559 | def get_testcases(self, include_onlymatching=False): |
1560 | t = getattr(self, '_TEST', None) | |
1561 | if t: | |
1562 | assert not hasattr(self, '_TESTS'), \ | |
1563 | '%s has _TEST and _TESTS' % type(self).__name__ | |
1564 | tests = [t] | |
1565 | else: | |
1566 | tests = getattr(self, '_TESTS', []) | |
1567 | for t in tests: | |
1568 | if not include_onlymatching and t.get('only_matching', False): | |
1569 | continue | |
1570 | t['name'] = type(self).__name__[:-len('IE')] | |
1571 | yield t | |
1572 | ||
1573 | def is_suitable(self, age_limit): | |
1574 | """ Test whether the extractor is generally suitable for the given | |
1575 | age limit (i.e. pornographic sites are not, all others usually are) """ | |
1576 | ||
1577 | any_restricted = False | |
1578 | for tc in self.get_testcases(include_onlymatching=False): | |
1579 | if 'playlist' in tc: | |
1580 | tc = tc['playlist'][0] | |
1581 | is_restricted = age_restricted( | |
1582 | tc.get('info_dict', {}).get('age_limit'), age_limit) | |
1583 | if not is_restricted: | |
1584 | return True | |
1585 | any_restricted = any_restricted or is_restricted | |
1586 | return not any_restricted | |
1587 | ||
a504ced0 | 1588 | def extract_subtitles(self, *args, **kwargs): |
9868ea49 JMF |
1589 | if (self._downloader.params.get('writesubtitles', False) or |
1590 | self._downloader.params.get('listsubtitles')): | |
1591 | return self._get_subtitles(*args, **kwargs) | |
1592 | return {} | |
a504ced0 JMF |
1593 | |
1594 | def _get_subtitles(self, *args, **kwargs): | |
611c1dd9 | 1595 | raise NotImplementedError('This method must be implemented by subclasses') |
a504ced0 | 1596 | |
912e0b7e YCH |
1597 | @staticmethod |
1598 | def _merge_subtitle_items(subtitle_list1, subtitle_list2): | |
1599 | """ Merge subtitle items for one language. Items with duplicated URLs | |
1600 | will be dropped. """ | |
1601 | list1_urls = set([item['url'] for item in subtitle_list1]) | |
1602 | ret = list(subtitle_list1) | |
1603 | ret.extend([item for item in subtitle_list2 if item['url'] not in list1_urls]) | |
1604 | return ret | |
1605 | ||
1606 | @classmethod | |
8c97f819 | 1607 | def _merge_subtitles(cls, subtitle_dict1, subtitle_dict2): |
912e0b7e | 1608 | """ Merge two subtitle dictionaries, language by language. """ |
912e0b7e YCH |
1609 | ret = dict(subtitle_dict1) |
1610 | for lang in subtitle_dict2: | |
8c97f819 | 1611 | ret[lang] = cls._merge_subtitle_items(subtitle_dict1.get(lang, []), subtitle_dict2[lang]) |
912e0b7e YCH |
1612 | return ret |
1613 | ||
360e1ca5 | 1614 | def extract_automatic_captions(self, *args, **kwargs): |
9868ea49 JMF |
1615 | if (self._downloader.params.get('writeautomaticsub', False) or |
1616 | self._downloader.params.get('listsubtitles')): | |
1617 | return self._get_automatic_captions(*args, **kwargs) | |
1618 | return {} | |
360e1ca5 JMF |
1619 | |
1620 | def _get_automatic_captions(self, *args, **kwargs): | |
611c1dd9 | 1621 | raise NotImplementedError('This method must be implemented by subclasses') |
360e1ca5 | 1622 | |
d77ab8e2 S |
1623 | def mark_watched(self, *args, **kwargs): |
1624 | if (self._downloader.params.get('mark_watched', False) and | |
1625 | (self._get_login_info()[0] is not None or | |
1626 | self._downloader.params.get('cookiefile') is not None)): | |
1627 | self._mark_watched(*args, **kwargs) | |
1628 | ||
1629 | def _mark_watched(self, *args, **kwargs): | |
1630 | raise NotImplementedError('This method must be implemented by subclasses') | |
1631 | ||
8dbe9899 | 1632 | |
d6983cb4 PH |
1633 | class SearchInfoExtractor(InfoExtractor): |
1634 | """ | |
1635 | Base class for paged search queries extractors. | |
10952eb2 | 1636 | They accept URLs in the format _SEARCH_KEY(|all|[0-9]):{query} |
d6983cb4 PH |
1637 | Instances should define _SEARCH_KEY and _MAX_RESULTS. |
1638 | """ | |
1639 | ||
1640 | @classmethod | |
1641 | def _make_valid_url(cls): | |
1642 | return r'%s(?P<prefix>|[1-9][0-9]*|all):(?P<query>[\s\S]+)' % cls._SEARCH_KEY | |
1643 | ||
1644 | @classmethod | |
1645 | def suitable(cls, url): | |
1646 | return re.match(cls._make_valid_url(), url) is not None | |
1647 | ||
1648 | def _real_extract(self, query): | |
1649 | mobj = re.match(self._make_valid_url(), query) | |
1650 | if mobj is None: | |
f1a9d64e | 1651 | raise ExtractorError('Invalid search query "%s"' % query) |
d6983cb4 PH |
1652 | |
1653 | prefix = mobj.group('prefix') | |
1654 | query = mobj.group('query') | |
1655 | if prefix == '': | |
1656 | return self._get_n_results(query, 1) | |
1657 | elif prefix == 'all': | |
1658 | return self._get_n_results(query, self._MAX_RESULTS) | |
1659 | else: | |
1660 | n = int(prefix) | |
1661 | if n <= 0: | |
f1a9d64e | 1662 | raise ExtractorError('invalid download number %s for query "%s"' % (n, query)) |
d6983cb4 | 1663 | elif n > self._MAX_RESULTS: |
f1a9d64e | 1664 | self._downloader.report_warning('%s returns max %i results (you requested %i)' % (self._SEARCH_KEY, self._MAX_RESULTS, n)) |
d6983cb4 PH |
1665 | n = self._MAX_RESULTS |
1666 | return self._get_n_results(query, n) | |
1667 | ||
1668 | def _get_n_results(self, query, n): | |
1669 | """Get a specified number of results for a query""" | |
611c1dd9 | 1670 | raise NotImplementedError('This method must be implemented by subclasses') |
0f818663 PH |
1671 | |
1672 | @property | |
1673 | def SEARCH_KEY(self): | |
1674 | return self._SEARCH_KEY |