]>
Commit | Line | Data |
---|---|---|
fd475508 | 1 | # coding: utf-8 |
6a3828fd | 2 | from __future__ import unicode_literals |
f1a9d64e | 3 | |
d6983cb4 | 4 | import base64 |
f4b1c7ad | 5 | import datetime |
3ec05685 | 6 | import hashlib |
3d3538e4 | 7 | import json |
4094b6e3 | 8 | import netrc |
d6983cb4 | 9 | import os |
773f291d | 10 | import random |
d6983cb4 | 11 | import re |
d6983cb4 | 12 | import sys |
4094b6e3 | 13 | import time |
1bac3455 | 14 | import math |
d6983cb4 | 15 | |
8c25f81b | 16 | from ..compat import ( |
6c22cee6 | 17 | compat_cookiejar_Cookie, |
f7ad7160 | 18 | compat_cookies_SimpleCookie, |
ee0ba927 | 19 | compat_etree_Element, |
e9c0cdd3 | 20 | compat_etree_fromstring, |
e64b7569 | 21 | compat_getpass, |
d6983cb4 | 22 | compat_http_client, |
e9c0cdd3 YCH |
23 | compat_os_name, |
24 | compat_str, | |
d6983cb4 | 25 | compat_urllib_error, |
98763ee3 | 26 | compat_urllib_parse_unquote, |
15707c7e | 27 | compat_urllib_parse_urlencode, |
41d06b04 | 28 | compat_urllib_request, |
f0b5d6af | 29 | compat_urlparse, |
e01c3d2e | 30 | compat_xml_parse_error, |
8c25f81b | 31 | ) |
eb8a4433 | 32 | from ..downloader import FileDownloader |
48107c19 S |
33 | from ..downloader.f4m import ( |
34 | get_base_url, | |
35 | remove_encrypted_media, | |
36 | ) | |
8c25f81b | 37 | from ..utils import ( |
c342041f | 38 | NO_DEFAULT, |
05900629 | 39 | age_restricted, |
02dc0a36 | 40 | base_url, |
08f2a92c | 41 | bug_reports_message, |
d6983cb4 PH |
42 | clean_html, |
43 | compiled_regex_type, | |
70f0f5a8 | 44 | determine_ext, |
46b18f23 | 45 | determine_protocol, |
d493f15c | 46 | dict_get, |
9b9c5355 | 47 | error_to_compat_str, |
d6983cb4 | 48 | ExtractorError, |
46b18f23 | 49 | extract_attributes, |
97f4aecf | 50 | fix_xml_ampersands, |
b14f3a4c | 51 | float_or_none, |
773f291d S |
52 | GeoRestrictedError, |
53 | GeoUtils, | |
31bb8d3f | 54 | int_or_none, |
a4a554a7 | 55 | js_to_json, |
0685d972 | 56 | JSON_LD_RE, |
46b18f23 | 57 | mimetype2ext, |
3158150c | 58 | network_exceptions, |
46b18f23 | 59 | orderedSet, |
d493f15c | 60 | parse_bitrate, |
46b18f23 JH |
61 | parse_codecs, |
62 | parse_duration, | |
4ca2a3cf | 63 | parse_iso8601, |
46b18f23 | 64 | parse_m3u8_attributes, |
d493f15c | 65 | parse_resolution, |
55b3e45b | 66 | RegexNotFoundError, |
5c2266df | 67 | sanitized_Request, |
46b18f23 | 68 | sanitize_filename, |
d493f15c | 69 | str_or_none, |
ce5b9040 | 70 | str_to_int, |
f856816b | 71 | strip_or_none, |
5d3a0e79 | 72 | traverse_obj, |
f38de77f | 73 | unescapeHTML, |
647eab45 | 74 | unified_strdate, |
6b3a3098 | 75 | unified_timestamp, |
46b18f23 JH |
76 | update_Request, |
77 | update_url_query, | |
78 | urljoin, | |
a107193e | 79 | url_basename, |
bebef109 | 80 | url_or_none, |
6606817a | 81 | variadic, |
a6571f10 | 82 | xpath_element, |
8d6765cf S |
83 | xpath_text, |
84 | xpath_with_ns, | |
d6983cb4 | 85 | ) |
c342041f | 86 | |
d6983cb4 PH |
87 | |
88 | class InfoExtractor(object): | |
89 | """Information Extractor class. | |
90 | ||
91 | Information extractors are the classes that, given a URL, extract | |
92 | information about the video (or videos) the URL refers to. This | |
93 | information includes the real video URL, the video title, author and | |
94 | others. The information is stored in a dictionary which is then | |
5d380852 | 95 | passed to the YoutubeDL. The YoutubeDL processes this |
d6983cb4 PH |
96 | information possibly downloading the video to the file system, among |
97 | other possible outcomes. | |
98 | ||
cf0649f8 | 99 | The type field determines the type of the result. |
fed5d032 PH |
100 | By far the most common value (and the default if _type is missing) is |
101 | "video", which indicates a single video. | |
102 | ||
103 | For a video, the dictionaries must include the following fields: | |
d6983cb4 PH |
104 | |
105 | id: Video identifier. | |
d6983cb4 | 106 | title: Video title, unescaped. |
d67b0b15 | 107 | |
f49d89ee | 108 | Additionally, it must contain either a formats entry or a url one: |
d67b0b15 | 109 | |
f49d89ee PH |
110 | formats: A list of dictionaries for each format available, ordered |
111 | from worst to best quality. | |
112 | ||
113 | Potential fields: | |
c790e93a S |
114 | * url The mandatory URL representing the media: |
115 | for plain file media - HTTP URL of this file, | |
116 | for RTMP - RTMP URL, | |
117 | for HLS - URL of the M3U8 media playlist, | |
118 | for HDS - URL of the F4M manifest, | |
79d2077e S |
119 | for DASH |
120 | - HTTP URL to plain file media (in case of | |
121 | unfragmented media) | |
122 | - URL of the MPD manifest or base URL | |
123 | representing the media if MPD manifest | |
8ed7a233 | 124 | is parsed from a string (in case of |
79d2077e | 125 | fragmented media) |
c790e93a | 126 | for MSS - URL of the ISM manifest. |
86f4d14f S |
127 | * manifest_url |
128 | The URL of the manifest file in case of | |
c790e93a S |
129 | fragmented media: |
130 | for HLS - URL of the M3U8 master playlist, | |
131 | for HDS - URL of the F4M manifest, | |
132 | for DASH - URL of the MPD manifest, | |
133 | for MSS - URL of the ISM manifest. | |
10952eb2 | 134 | * ext Will be calculated from URL if missing |
d67b0b15 PH |
135 | * format A human-readable description of the format |
136 | ("mp4 container with h264/opus"). | |
137 | Calculated from the format_id, width, height. | |
138 | and format_note fields if missing. | |
139 | * format_id A short description of the format | |
5d4f3985 PH |
140 | ("mp4_h264_opus" or "19"). |
141 | Technically optional, but strongly recommended. | |
d67b0b15 PH |
142 | * format_note Additional info about the format |
143 | ("3D" or "DASH video") | |
144 | * width Width of the video, if known | |
145 | * height Height of the video, if known | |
f49d89ee | 146 | * resolution Textual description of width and height |
7217e148 | 147 | * tbr Average bitrate of audio and video in KBit/s |
d67b0b15 PH |
148 | * abr Average audio bitrate in KBit/s |
149 | * acodec Name of the audio codec in use | |
dd27fd17 | 150 | * asr Audio sampling rate in Hertz |
d67b0b15 | 151 | * vbr Average video bitrate in KBit/s |
fbb21cf5 | 152 | * fps Frame rate |
d67b0b15 | 153 | * vcodec Name of the video codec in use |
1394ce65 | 154 | * container Name of the container format |
d67b0b15 | 155 | * filesize The number of bytes, if known in advance |
9732d77e | 156 | * filesize_approx An estimate for the number of bytes |
d67b0b15 | 157 | * player_url SWF Player URL (used for rtmpdump). |
c7deaa4c PH |
158 | * protocol The protocol that will be used for the actual |
159 | download, lower-case. | |
0fa9a1e2 | 160 | "http", "https", "rtsp", "rtmp", "rtmp_ffmpeg", "rtmpe", |
af7d5a63 | 161 | "m3u8", "m3u8_native" or "http_dash_segments". |
c58c2d63 S |
162 | * fragment_base_url |
163 | Base URL for fragments. Each fragment's path | |
164 | value (if present) will be relative to | |
165 | this URL. | |
166 | * fragments A list of fragments of a fragmented media. | |
167 | Each fragment entry must contain either an url | |
168 | or a path. If an url is present it should be | |
169 | considered by a client. Otherwise both path and | |
170 | fragment_base_url must be present. Here is | |
171 | the list of all potential fields: | |
172 | * "url" - fragment's URL | |
173 | * "path" - fragment's path relative to | |
174 | fragment_base_url | |
a0d5077c S |
175 | * "duration" (optional, int or float) |
176 | * "filesize" (optional, int) | |
f49d89ee | 177 | * preference Order number of this format. If this field is |
08d13955 | 178 | present and not None, the formats get sorted |
38d63d84 | 179 | by this field, regardless of all other values. |
f49d89ee PH |
180 | -1 for default (order by other properties), |
181 | -2 or smaller for less than default. | |
e65566a9 PH |
182 | < -1000 to hide the format (if there is |
183 | another one which is strictly better) | |
32f90364 PH |
184 | * language Language code, e.g. "de" or "en-US". |
185 | * language_preference Is this in the language mentioned in | |
186 | the URL? | |
aff2f4f4 PH |
187 | 10 if it's what the URL is about, |
188 | -1 for default (don't know), | |
189 | -10 otherwise, other values reserved for now. | |
5d73273f PH |
190 | * quality Order number of the video quality of this |
191 | format, irrespective of the file format. | |
192 | -1 for default (order by other properties), | |
193 | -2 or smaller for less than default. | |
c64ed2a3 PH |
194 | * source_preference Order number for this video source |
195 | (quality takes higher priority) | |
196 | -1 for default (order by other properties), | |
197 | -2 or smaller for less than default. | |
d769be6c PH |
198 | * http_headers A dictionary of additional HTTP headers |
199 | to add to the request. | |
6271f1ca | 200 | * stretched_ratio If given and not 1, indicates that the |
3dee7826 PH |
201 | video's pixels are not square. |
202 | width : height ratio as float. | |
203 | * no_resume The server does not support resuming the | |
204 | (HTTP or RTMP) download. Boolean. | |
00c97e3e S |
205 | * downloader_options A dictionary of downloader options as |
206 | described in FileDownloader | |
3b1fe47d | 207 | RTMP formats can also have the additional fields: page_url, |
208 | app, play_path, tc_url, flash_version, rtmp_live, rtmp_conn, | |
209 | rtmp_protocol, rtmp_real_time | |
3dee7826 | 210 | |
c0ba0f48 | 211 | url: Final video URL. |
d6983cb4 | 212 | ext: Video filename extension. |
d67b0b15 PH |
213 | format: The video format, defaults to ext (used for --get-format) |
214 | player_url: SWF Player URL (used for rtmpdump). | |
2f5865cc | 215 | |
d6983cb4 PH |
216 | The following fields are optional: |
217 | ||
f5e43bc6 | 218 | alt_title: A secondary title of the video. |
0afef30b PH |
219 | display_id An alternative identifier for the video, not necessarily |
220 | unique, but available before title. Typically, id is | |
221 | something like "4234987", title "Dancing naked mole rats", | |
222 | and display_id "dancing-naked-mole-rats" | |
d5519808 | 223 | thumbnails: A list of dictionaries, with the following entries: |
cfb56d1a | 224 | * "id" (optional, string) - Thumbnail format ID |
d5519808 | 225 | * "url" |
cfb56d1a | 226 | * "preference" (optional, int) - quality of the image |
d5519808 PH |
227 | * "width" (optional, int) |
228 | * "height" (optional, int) | |
5e1c39ac | 229 | * "resolution" (optional, string "{width}x{height}", |
d5519808 | 230 | deprecated) |
2de624fd | 231 | * "filesize" (optional, int) |
0ba692ac | 232 | * "_test_url" (optional, bool) - If true, test the URL |
d6983cb4 | 233 | thumbnail: Full URL to a video thumbnail image. |
f5e43bc6 | 234 | description: Full video description. |
d6983cb4 | 235 | uploader: Full name of the video uploader. |
2bc0c46f | 236 | license: License name the video is licensed under. |
8a92e51c | 237 | creator: The creator of the video. |
10db0d2f | 238 | release_timestamp: UNIX timestamp of the moment the video was released. |
8aab976b | 239 | release_date: The date (YYYYMMDD) when the video was released. |
10db0d2f | 240 | timestamp: UNIX timestamp of the moment the video was uploaded |
d6983cb4 | 241 | upload_date: Video upload date (YYYYMMDD). |
955c4514 | 242 | If not explicitly set, calculated from timestamp. |
d6983cb4 | 243 | uploader_id: Nickname or id of the video uploader. |
7bcd2830 | 244 | uploader_url: Full URL to a personal webpage of the video uploader. |
6f1f59f3 | 245 | channel: Full name of the channel the video is uploaded on. |
0e7b8d3e | 246 | Note that channel fields may or may not repeat uploader |
6f1f59f3 S |
247 | fields. This depends on a particular extractor. |
248 | channel_id: Id of the channel. | |
249 | channel_url: Full URL to a channel webpage. | |
da9ec3b9 | 250 | location: Physical location where the video was filmed. |
a504ced0 | 251 | subtitles: The available subtitles as a dictionary in the format |
4606c34e YCH |
252 | {tag: subformats}. "tag" is usually a language code, and |
253 | "subformats" is a list sorted from lower to higher | |
254 | preference, each element is a dictionary with the "ext" | |
255 | entry and one of: | |
a504ced0 | 256 | * "data": The subtitles file contents |
10952eb2 | 257 | * "url": A URL pointing to the subtitles file |
2412044c | 258 | It can optionally also have: |
259 | * "name": Name or description of the subtitles | |
4bba3716 | 260 | "ext" will be calculated from URL if missing |
e167860c | 261 | automatic_captions: Like 'subtitles'; contains automatically generated |
262 | captions instead of normal subtitles | |
62d231c0 | 263 | duration: Length of the video in seconds, as an integer or float. |
f3d29461 | 264 | view_count: How many users have watched the video on the platform. |
19e3dfc9 PH |
265 | like_count: Number of positive ratings of the video |
266 | dislike_count: Number of negative ratings of the video | |
02835c6b | 267 | repost_count: Number of reposts of the video |
2d30521a | 268 | average_rating: Average rating give by users, the scale used depends on the webpage |
19e3dfc9 | 269 | comment_count: Number of comments on the video |
dd622d7c PH |
270 | comments: A list of comments, each with one or more of the following |
271 | properties (all but one of text or html optional): | |
272 | * "author" - human-readable name of the comment author | |
273 | * "author_id" - user ID of the comment author | |
a1c5d2ca | 274 | * "author_thumbnail" - The thumbnail of the comment author |
dd622d7c PH |
275 | * "id" - Comment ID |
276 | * "html" - Comment as HTML | |
277 | * "text" - Plain text of the comment | |
278 | * "timestamp" - UNIX timestamp of comment | |
279 | * "parent" - ID of the comment this one is replying to. | |
280 | Set to "root" to indicate that this is a | |
281 | comment to the original video. | |
a1c5d2ca M |
282 | * "like_count" - Number of positive ratings of the comment |
283 | * "dislike_count" - Number of negative ratings of the comment | |
284 | * "is_favorited" - Whether the comment is marked as | |
285 | favorite by the video uploader | |
286 | * "author_is_uploader" - Whether the comment is made by | |
287 | the video uploader | |
8dbe9899 | 288 | age_limit: Age restriction for the video, as an integer (years) |
7a5c1cfe | 289 | webpage_url: The URL to the video webpage, if given to yt-dlp it |
9103bbc5 JMF |
290 | should allow to get the same result again. (It will be set |
291 | by YoutubeDL if it's missing) | |
ad3bc6ac PH |
292 | categories: A list of categories that the video falls in, for example |
293 | ["Sports", "Berlin"] | |
864f24bd | 294 | tags: A list of tags assigned to the video, e.g. ["sweden", "pop music"] |
d0fb4bd1 | 295 | cast: A list of the video cast |
7267bd53 PH |
296 | is_live: True, False, or None (=unknown). Whether this video is a |
297 | live stream that goes on instead of a fixed-length video. | |
f76ede8e | 298 | was_live: True, False, or None (=unknown). Whether this video was |
299 | originally a live stream. | |
ae30b840 | 300 | live_status: 'is_live', 'upcoming', 'was_live', 'not_live' or None (=unknown) |
301 | If absent, automatically set from is_live, was_live | |
7c80519c | 302 | start_time: Time in seconds where the reproduction should start, as |
10952eb2 | 303 | specified in the URL. |
297a564b | 304 | end_time: Time in seconds where the reproduction should end, as |
10952eb2 | 305 | specified in the URL. |
55949fed | 306 | chapters: A list of dictionaries, with the following entries: |
307 | * "start_time" - The start time of the chapter in seconds | |
308 | * "end_time" - The end time of the chapter in seconds | |
309 | * "title" (optional, string) | |
6cfda058 | 310 | playable_in_embed: Whether this video is allowed to play in embedded |
311 | players on other sites. Can be True (=always allowed), | |
312 | False (=never allowed), None (=unknown), or a string | |
c224251a M |
313 | specifying the criteria for embedability (Eg: 'whitelist') |
314 | availability: Under what condition the video is available. One of | |
315 | 'private', 'premium_only', 'subscriber_only', 'needs_auth', | |
316 | 'unlisted' or 'public'. Use 'InfoExtractor._availability' | |
317 | to set it | |
277d6ff5 | 318 | __post_extractor: A function to be called just before the metadata is |
319 | written to either disk, logger or console. The function | |
320 | must return a dict which will be added to the info_dict. | |
321 | This is usefull for additional information that is | |
322 | time-consuming to extract. Note that the fields thus | |
323 | extracted will not be available to output template and | |
324 | match_filter. So, only "comments" and "comment_count" are | |
325 | currently allowed to be extracted via this method. | |
d6983cb4 | 326 | |
7109903e S |
327 | The following fields should only be used when the video belongs to some logical |
328 | chapter or section: | |
329 | ||
330 | chapter: Name or title of the chapter the video belongs to. | |
27bfd4e5 S |
331 | chapter_number: Number of the chapter the video belongs to, as an integer. |
332 | chapter_id: Id of the chapter the video belongs to, as a unicode string. | |
7109903e S |
333 | |
334 | The following fields should only be used when the video is an episode of some | |
8d76bdf1 | 335 | series, programme or podcast: |
7109903e S |
336 | |
337 | series: Title of the series or programme the video episode belongs to. | |
338 | season: Title of the season the video episode belongs to. | |
27bfd4e5 S |
339 | season_number: Number of the season the video episode belongs to, as an integer. |
340 | season_id: Id of the season the video episode belongs to, as a unicode string. | |
7109903e S |
341 | episode: Title of the video episode. Unlike mandatory video title field, |
342 | this field should denote the exact title of the video episode | |
343 | without any kind of decoration. | |
27bfd4e5 S |
344 | episode_number: Number of the video episode within a season, as an integer. |
345 | episode_id: Id of the video episode, as a unicode string. | |
7109903e | 346 | |
7a93ab5f S |
347 | The following fields should only be used when the media is a track or a part of |
348 | a music album: | |
349 | ||
350 | track: Title of the track. | |
351 | track_number: Number of the track within an album or a disc, as an integer. | |
352 | track_id: Id of the track (useful in case of custom indexing, e.g. 6.iii), | |
353 | as a unicode string. | |
354 | artist: Artist(s) of the track. | |
355 | genre: Genre(s) of the track. | |
356 | album: Title of the album the track belongs to. | |
357 | album_type: Type of the album (e.g. "Demo", "Full-length", "Split", "Compilation", etc). | |
358 | album_artist: List of all artists appeared on the album (e.g. | |
359 | "Ash Borer / Fell Voices" or "Various Artists", useful for splits | |
360 | and compilations). | |
361 | disc_number: Number of the disc or other physical medium the track belongs to, | |
362 | as an integer. | |
363 | release_year: Year (YYYY) when the album was released. | |
364 | ||
deefc05b | 365 | Unless mentioned otherwise, the fields should be Unicode strings. |
d6983cb4 | 366 | |
d838b1bd PH |
367 | Unless mentioned otherwise, None is equivalent to absence of information. |
368 | ||
fed5d032 PH |
369 | |
370 | _type "playlist" indicates multiple videos. | |
b82f815f PH |
371 | There must be a key "entries", which is a list, an iterable, or a PagedList |
372 | object, each element of which is a valid dictionary by this specification. | |
fed5d032 | 373 | |
b60419c5 | 374 | Additionally, playlists can have "id", "title", and any other relevent |
375 | attributes with the same semantics as videos (see above). | |
fed5d032 PH |
376 | |
377 | ||
378 | _type "multi_video" indicates that there are multiple videos that | |
379 | form a single show, for examples multiple acts of an opera or TV episode. | |
380 | It must have an entries key like a playlist and contain all the keys | |
381 | required for a video at the same time. | |
382 | ||
383 | ||
384 | _type "url" indicates that the video must be extracted from another | |
385 | location, possibly by a different extractor. Its only required key is: | |
386 | "url" - the next URL to extract. | |
f58766ce PH |
387 | The key "ie_key" can be set to the class name (minus the trailing "IE", |
388 | e.g. "Youtube") if the extractor class is known in advance. | |
389 | Additionally, the dictionary may have any properties of the resolved entity | |
390 | known in advance, for example "title" if the title of the referred video is | |
fed5d032 PH |
391 | known ahead of time. |
392 | ||
393 | ||
394 | _type "url_transparent" entities have the same specification as "url", but | |
395 | indicate that the given additional information is more precise than the one | |
396 | associated with the resolved URL. | |
397 | This is useful when a site employs a video service that hosts the video and | |
398 | its technical metadata, but that video service does not embed a useful | |
399 | title, description etc. | |
400 | ||
401 | ||
d6983cb4 PH |
402 | Subclasses of this one should re-define the _real_initialize() and |
403 | _real_extract() methods and define a _VALID_URL regexp. | |
404 | Probably, they should also be added to the list of extractors. | |
405 | ||
4248dad9 | 406 | _GEO_BYPASS attribute may be set to False in order to disable |
773f291d S |
407 | geo restriction bypass mechanisms for a particular extractor. |
408 | Though it won't disable explicit geo restriction bypass based on | |
504f20dd | 409 | country code provided with geo_bypass_country. |
4248dad9 S |
410 | |
411 | _GEO_COUNTRIES attribute may contain a list of presumably geo unrestricted | |
412 | countries for this extractor. One of these countries will be used by | |
413 | geo restriction bypass mechanism right away in order to bypass | |
504f20dd | 414 | geo restriction, of course, if the mechanism is not disabled. |
773f291d | 415 | |
5f95927a S |
416 | _GEO_IP_BLOCKS attribute may contain a list of presumably geo unrestricted |
417 | IP blocks in CIDR notation for this extractor. One of these IP blocks | |
418 | will be used by geo restriction bypass mechanism similarly | |
504f20dd | 419 | to _GEO_COUNTRIES. |
3ccdde8c | 420 | |
d6983cb4 PH |
421 | Finally, the _WORKING attribute should be set to False for broken IEs |
422 | in order to warn the users and skip the tests. | |
423 | """ | |
424 | ||
425 | _ready = False | |
426 | _downloader = None | |
773f291d | 427 | _x_forwarded_for_ip = None |
4248dad9 S |
428 | _GEO_BYPASS = True |
429 | _GEO_COUNTRIES = None | |
5f95927a | 430 | _GEO_IP_BLOCKS = None |
d6983cb4 PH |
431 | _WORKING = True |
432 | ||
9d5d4d64 | 433 | _LOGIN_HINTS = { |
434 | 'any': 'Use --cookies, --username and --password or --netrc to provide account credentials', | |
435 | 'cookies': ( | |
436 | 'Use --cookies for the authentication. ' | |
437 | 'See https://github.com/ytdl-org/youtube-dl#how-do-i-pass-cookies-to-youtube-dl for how to pass cookies'), | |
438 | 'password': 'Use --username and --password or --netrc to provide account credentials', | |
439 | } | |
440 | ||
d6983cb4 PH |
441 | def __init__(self, downloader=None): |
442 | """Constructor. Receives an optional downloader.""" | |
443 | self._ready = False | |
773f291d | 444 | self._x_forwarded_for_ip = None |
d6983cb4 PH |
445 | self.set_downloader(downloader) |
446 | ||
447 | @classmethod | |
448 | def suitable(cls, url): | |
449 | """Receives a URL and returns True if suitable for this IE.""" | |
79cb2577 PH |
450 | |
451 | # This does not use has/getattr intentionally - we want to know whether | |
452 | # we have cached the regexp for *this* class, whereas getattr would also | |
453 | # match the superclass | |
454 | if '_VALID_URL_RE' not in cls.__dict__: | |
455 | cls._VALID_URL_RE = re.compile(cls._VALID_URL) | |
456 | return cls._VALID_URL_RE.match(url) is not None | |
d6983cb4 | 457 | |
ed9266db PH |
458 | @classmethod |
459 | def _match_id(cls, url): | |
460 | if '_VALID_URL_RE' not in cls.__dict__: | |
461 | cls._VALID_URL_RE = re.compile(cls._VALID_URL) | |
462 | m = cls._VALID_URL_RE.match(url) | |
463 | assert m | |
1afd0b0d | 464 | return compat_str(m.group('id')) |
ed9266db | 465 | |
d6983cb4 PH |
466 | @classmethod |
467 | def working(cls): | |
468 | """Getter method for _WORKING.""" | |
469 | return cls._WORKING | |
470 | ||
471 | def initialize(self): | |
472 | """Initializes an instance (authentication, etc).""" | |
5f95927a S |
473 | self._initialize_geo_bypass({ |
474 | 'countries': self._GEO_COUNTRIES, | |
475 | 'ip_blocks': self._GEO_IP_BLOCKS, | |
476 | }) | |
4248dad9 S |
477 | if not self._ready: |
478 | self._real_initialize() | |
479 | self._ready = True | |
480 | ||
5f95927a | 481 | def _initialize_geo_bypass(self, geo_bypass_context): |
e39b5d4a S |
482 | """ |
483 | Initialize geo restriction bypass mechanism. | |
484 | ||
485 | This method is used to initialize geo bypass mechanism based on faking | |
486 | X-Forwarded-For HTTP header. A random country from provided country list | |
dc0a869e | 487 | is selected and a random IP belonging to this country is generated. This |
e39b5d4a S |
488 | IP will be passed as X-Forwarded-For HTTP header in all subsequent |
489 | HTTP requests. | |
e39b5d4a S |
490 | |
491 | This method will be used for initial geo bypass mechanism initialization | |
5f95927a S |
492 | during the instance initialization with _GEO_COUNTRIES and |
493 | _GEO_IP_BLOCKS. | |
e39b5d4a | 494 | |
5f95927a | 495 | You may also manually call it from extractor's code if geo bypass |
e39b5d4a | 496 | information is not available beforehand (e.g. obtained during |
5f95927a S |
497 | extraction) or due to some other reason. In this case you should pass |
498 | this information in geo bypass context passed as first argument. It may | |
499 | contain following fields: | |
500 | ||
501 | countries: List of geo unrestricted countries (similar | |
502 | to _GEO_COUNTRIES) | |
503 | ip_blocks: List of geo unrestricted IP blocks in CIDR notation | |
504 | (similar to _GEO_IP_BLOCKS) | |
505 | ||
e39b5d4a | 506 | """ |
773f291d | 507 | if not self._x_forwarded_for_ip: |
5f95927a S |
508 | |
509 | # Geo bypass mechanism is explicitly disabled by user | |
a06916d9 | 510 | if not self.get_param('geo_bypass', True): |
5f95927a S |
511 | return |
512 | ||
513 | if not geo_bypass_context: | |
514 | geo_bypass_context = {} | |
515 | ||
516 | # Backward compatibility: previously _initialize_geo_bypass | |
517 | # expected a list of countries, some 3rd party code may still use | |
518 | # it this way | |
519 | if isinstance(geo_bypass_context, (list, tuple)): | |
520 | geo_bypass_context = { | |
521 | 'countries': geo_bypass_context, | |
522 | } | |
523 | ||
524 | # The whole point of geo bypass mechanism is to fake IP | |
525 | # as X-Forwarded-For HTTP header based on some IP block or | |
526 | # country code. | |
527 | ||
528 | # Path 1: bypassing based on IP block in CIDR notation | |
529 | ||
530 | # Explicit IP block specified by user, use it right away | |
531 | # regardless of whether extractor is geo bypassable or not | |
a06916d9 | 532 | ip_block = self.get_param('geo_bypass_ip_block', None) |
5f95927a S |
533 | |
534 | # Otherwise use random IP block from geo bypass context but only | |
535 | # if extractor is known as geo bypassable | |
536 | if not ip_block: | |
537 | ip_blocks = geo_bypass_context.get('ip_blocks') | |
538 | if self._GEO_BYPASS and ip_blocks: | |
539 | ip_block = random.choice(ip_blocks) | |
540 | ||
541 | if ip_block: | |
542 | self._x_forwarded_for_ip = GeoUtils.random_ipv4(ip_block) | |
0760b0a7 | 543 | self._downloader.write_debug( |
544 | '[debug] Using fake IP %s as X-Forwarded-For' % self._x_forwarded_for_ip) | |
5f95927a S |
545 | return |
546 | ||
547 | # Path 2: bypassing based on country code | |
548 | ||
549 | # Explicit country code specified by user, use it right away | |
550 | # regardless of whether extractor is geo bypassable or not | |
a06916d9 | 551 | country = self.get_param('geo_bypass_country', None) |
5f95927a S |
552 | |
553 | # Otherwise use random country code from geo bypass context but | |
554 | # only if extractor is known as geo bypassable | |
555 | if not country: | |
556 | countries = geo_bypass_context.get('countries') | |
557 | if self._GEO_BYPASS and countries: | |
558 | country = random.choice(countries) | |
559 | ||
560 | if country: | |
561 | self._x_forwarded_for_ip = GeoUtils.random_ipv4(country) | |
0760b0a7 | 562 | self._downloader.write_debug( |
563 | 'Using fake IP %s (%s) as X-Forwarded-For' % (self._x_forwarded_for_ip, country.upper())) | |
d6983cb4 PH |
564 | |
565 | def extract(self, url): | |
566 | """Extracts URL information and returns it in list of dicts.""" | |
3a5bcd03 | 567 | try: |
773f291d S |
568 | for _ in range(2): |
569 | try: | |
570 | self.initialize() | |
a06916d9 | 571 | self.write_debug('Extracting URL: %s' % url) |
0016b84e | 572 | ie_result = self._real_extract(url) |
07cce701 | 573 | if ie_result is None: |
574 | return None | |
0016b84e S |
575 | if self._x_forwarded_for_ip: |
576 | ie_result['__x_forwarded_for_ip'] = self._x_forwarded_for_ip | |
53ed7066 | 577 | subtitles = ie_result.get('subtitles') |
578 | if (subtitles and 'live_chat' in subtitles | |
a06916d9 | 579 | and 'no-live-chat' in self.get_param('compat_opts', [])): |
53ed7066 | 580 | del subtitles['live_chat'] |
0016b84e | 581 | return ie_result |
773f291d | 582 | except GeoRestrictedError as e: |
4248dad9 S |
583 | if self.__maybe_fake_ip_and_retry(e.countries): |
584 | continue | |
773f291d | 585 | raise |
3a5bcd03 PH |
586 | except ExtractorError: |
587 | raise | |
588 | except compat_http_client.IncompleteRead as e: | |
dfb1b146 | 589 | raise ExtractorError('A network error has occurred.', cause=e, expected=True) |
9650885b | 590 | except (KeyError, StopIteration) as e: |
dfb1b146 | 591 | raise ExtractorError('An extractor error has occurred.', cause=e) |
d6983cb4 | 592 | |
4248dad9 | 593 | def __maybe_fake_ip_and_retry(self, countries): |
a06916d9 | 594 | if (not self.get_param('geo_bypass_country', None) |
3089bc74 | 595 | and self._GEO_BYPASS |
a06916d9 | 596 | and self.get_param('geo_bypass', True) |
3089bc74 S |
597 | and not self._x_forwarded_for_ip |
598 | and countries): | |
eea0716c S |
599 | country_code = random.choice(countries) |
600 | self._x_forwarded_for_ip = GeoUtils.random_ipv4(country_code) | |
4248dad9 S |
601 | if self._x_forwarded_for_ip: |
602 | self.report_warning( | |
eea0716c S |
603 | 'Video is geo restricted. Retrying extraction with fake IP %s (%s) as X-Forwarded-For.' |
604 | % (self._x_forwarded_for_ip, country_code.upper())) | |
4248dad9 S |
605 | return True |
606 | return False | |
607 | ||
d6983cb4 PH |
608 | def set_downloader(self, downloader): |
609 | """Sets the downloader for this IE.""" | |
610 | self._downloader = downloader | |
611 | ||
612 | def _real_initialize(self): | |
613 | """Real initialization process. Redefine in subclasses.""" | |
614 | pass | |
615 | ||
616 | def _real_extract(self, url): | |
617 | """Real extraction process. Redefine in subclasses.""" | |
618 | pass | |
619 | ||
56c73665 JMF |
620 | @classmethod |
621 | def ie_key(cls): | |
622 | """A string for getting the InfoExtractor with get_info_extractor""" | |
dc519b54 | 623 | return compat_str(cls.__name__[:-2]) |
56c73665 | 624 | |
d6983cb4 PH |
625 | @property |
626 | def IE_NAME(self): | |
dc519b54 | 627 | return compat_str(type(self).__name__[:-2]) |
d6983cb4 | 628 | |
d391b7e2 S |
629 | @staticmethod |
630 | def __can_accept_status_code(err, expected_status): | |
631 | assert isinstance(err, compat_urllib_error.HTTPError) | |
632 | if expected_status is None: | |
633 | return False | |
d391b7e2 S |
634 | elif callable(expected_status): |
635 | return expected_status(err.code) is True | |
636 | else: | |
6606817a | 637 | return err.code in variadic(expected_status) |
d391b7e2 S |
638 | |
639 | def _request_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True, data=None, headers={}, query={}, expected_status=None): | |
640 | """ | |
641 | Return the response handle. | |
642 | ||
643 | See _download_webpage docstring for arguments specification. | |
644 | """ | |
1cf376f5 | 645 | if not self._downloader._first_webpage_request: |
a06916d9 | 646 | sleep_interval = float_or_none(self.get_param('sleep_interval_requests')) or 0 |
1cf376f5 | 647 | if sleep_interval > 0: |
5ef7d9bd | 648 | self.to_screen('Sleeping %s seconds ...' % sleep_interval) |
1cf376f5 | 649 | time.sleep(sleep_interval) |
650 | else: | |
651 | self._downloader._first_webpage_request = False | |
652 | ||
d6983cb4 PH |
653 | if note is None: |
654 | self.report_download_webpage(video_id) | |
655 | elif note is not False: | |
7cc3570e | 656 | if video_id is None: |
f1a9d64e | 657 | self.to_screen('%s' % (note,)) |
7cc3570e | 658 | else: |
f1a9d64e | 659 | self.to_screen('%s: %s' % (video_id, note)) |
2132edaa S |
660 | |
661 | # Some sites check X-Forwarded-For HTTP header in order to figure out | |
662 | # the origin of the client behind proxy. This allows bypassing geo | |
663 | # restriction by faking this header's value to IP that belongs to some | |
664 | # geo unrestricted country. We will do so once we encounter any | |
665 | # geo restriction error. | |
666 | if self._x_forwarded_for_ip: | |
667 | if 'X-Forwarded-For' not in headers: | |
668 | headers['X-Forwarded-For'] = self._x_forwarded_for_ip | |
669 | ||
41d06b04 S |
670 | if isinstance(url_or_request, compat_urllib_request.Request): |
671 | url_or_request = update_Request( | |
672 | url_or_request, data=data, headers=headers, query=query) | |
673 | else: | |
cdfee168 | 674 | if query: |
675 | url_or_request = update_url_query(url_or_request, query) | |
2c0d9c62 | 676 | if data is not None or headers: |
41d06b04 | 677 | url_or_request = sanitized_Request(url_or_request, data, headers) |
d6983cb4 | 678 | try: |
dca08720 | 679 | return self._downloader.urlopen(url_or_request) |
3158150c | 680 | except network_exceptions as err: |
d391b7e2 S |
681 | if isinstance(err, compat_urllib_error.HTTPError): |
682 | if self.__can_accept_status_code(err, expected_status): | |
95e42d73 XDG |
683 | # Retain reference to error to prevent file object from |
684 | # being closed before it can be read. Works around the | |
685 | # effects of <https://bugs.python.org/issue15002> | |
686 | # introduced in Python 3.4.1. | |
687 | err.fp._error = err | |
d391b7e2 S |
688 | return err.fp |
689 | ||
aa94a6d3 PH |
690 | if errnote is False: |
691 | return False | |
d6983cb4 | 692 | if errnote is None: |
f1a9d64e | 693 | errnote = 'Unable to download webpage' |
7f8b2714 | 694 | |
9b9c5355 | 695 | errmsg = '%s: %s' % (errnote, error_to_compat_str(err)) |
7cc3570e PH |
696 | if fatal: |
697 | raise ExtractorError(errmsg, sys.exc_info()[2], cause=err) | |
698 | else: | |
6a39ee13 | 699 | self.report_warning(errmsg) |
7cc3570e | 700 | return False |
d6983cb4 | 701 | |
d391b7e2 S |
702 | def _download_webpage_handle(self, url_or_request, video_id, note=None, errnote=None, fatal=True, encoding=None, data=None, headers={}, query={}, expected_status=None): |
703 | """ | |
704 | Return a tuple (page content as string, URL handle). | |
705 | ||
706 | See _download_webpage docstring for arguments specification. | |
707 | """ | |
b9d3e163 PH |
708 | # Strip hashes from the URL (#1038) |
709 | if isinstance(url_or_request, (compat_str, str)): | |
710 | url_or_request = url_or_request.partition('#')[0] | |
711 | ||
d391b7e2 | 712 | urlh = self._request_webpage(url_or_request, video_id, note, errnote, fatal, data=data, headers=headers, query=query, expected_status=expected_status) |
7cc3570e PH |
713 | if urlh is False: |
714 | assert not fatal | |
715 | return False | |
c9a77969 | 716 | content = self._webpage_read_content(urlh, url_or_request, video_id, note, errnote, fatal, encoding=encoding) |
23be51d8 PH |
717 | return (content, urlh) |
718 | ||
c9a77969 YCH |
719 | @staticmethod |
720 | def _guess_encoding_from_content(content_type, webpage_bytes): | |
d6983cb4 PH |
721 | m = re.match(r'[a-zA-Z0-9_.-]+/[a-zA-Z0-9_.-]+\s*;\s*charset=(.+)', content_type) |
722 | if m: | |
723 | encoding = m.group(1) | |
724 | else: | |
0d75ae2c | 725 | m = re.search(br'<meta[^>]+charset=[\'"]?([^\'")]+)[ /\'">]', |
f143d86a PH |
726 | webpage_bytes[:1024]) |
727 | if m: | |
728 | encoding = m.group(1).decode('ascii') | |
b60016e8 PH |
729 | elif webpage_bytes.startswith(b'\xff\xfe'): |
730 | encoding = 'utf-16' | |
f143d86a PH |
731 | else: |
732 | encoding = 'utf-8' | |
c9a77969 YCH |
733 | |
734 | return encoding | |
735 | ||
4457823d S |
736 | def __check_blocked(self, content): |
737 | first_block = content[:512] | |
3089bc74 S |
738 | if ('<title>Access to this site is blocked</title>' in content |
739 | and 'Websense' in first_block): | |
4457823d S |
740 | msg = 'Access to this webpage has been blocked by Websense filtering software in your network.' |
741 | blocked_iframe = self._html_search_regex( | |
742 | r'<iframe src="([^"]+)"', content, | |
743 | 'Websense information URL', default=None) | |
744 | if blocked_iframe: | |
745 | msg += ' Visit %s for more details' % blocked_iframe | |
746 | raise ExtractorError(msg, expected=True) | |
747 | if '<title>The URL you requested has been blocked</title>' in first_block: | |
748 | msg = ( | |
749 | 'Access to this webpage has been blocked by Indian censorship. ' | |
750 | 'Use a VPN or proxy server (with --proxy) to route around it.') | |
751 | block_msg = self._html_search_regex( | |
752 | r'</h1><p>(.*?)</p>', | |
753 | content, 'block message', default=None) | |
754 | if block_msg: | |
755 | msg += ' (Message: "%s")' % block_msg.replace('\n', ' ') | |
756 | raise ExtractorError(msg, expected=True) | |
3089bc74 S |
757 | if ('<title>TTK :: Доступ к ресурсу ограничен</title>' in content |
758 | and 'blocklist.rkn.gov.ru' in content): | |
4457823d S |
759 | raise ExtractorError( |
760 | 'Access to this webpage has been blocked by decision of the Russian government. ' | |
761 | 'Visit http://blocklist.rkn.gov.ru/ for a block reason.', | |
762 | expected=True) | |
763 | ||
c9a77969 YCH |
764 | def _webpage_read_content(self, urlh, url_or_request, video_id, note=None, errnote=None, fatal=True, prefix=None, encoding=None): |
765 | content_type = urlh.headers.get('Content-Type', '') | |
766 | webpage_bytes = urlh.read() | |
767 | if prefix is not None: | |
768 | webpage_bytes = prefix + webpage_bytes | |
769 | if not encoding: | |
770 | encoding = self._guess_encoding_from_content(content_type, webpage_bytes) | |
a06916d9 | 771 | if self.get_param('dump_intermediate_pages', False): |
f610dbb0 | 772 | self.to_screen('Dumping request to ' + urlh.geturl()) |
d6983cb4 PH |
773 | dump = base64.b64encode(webpage_bytes).decode('ascii') |
774 | self._downloader.to_screen(dump) | |
a06916d9 | 775 | if self.get_param('write_pages', False): |
f610dbb0 | 776 | basen = '%s_%s' % (video_id, urlh.geturl()) |
c1bce22f | 777 | if len(basen) > 240: |
f1a9d64e | 778 | h = '___' + hashlib.md5(basen.encode('utf-8')).hexdigest() |
c1bce22f PH |
779 | basen = basen[:240 - len(h)] + h |
780 | raw_filename = basen + '.dump' | |
d41e6efc | 781 | filename = sanitize_filename(raw_filename, restricted=True) |
f1a9d64e | 782 | self.to_screen('Saving request to ' + filename) |
5f58165d S |
783 | # Working around MAX_PATH limitation on Windows (see |
784 | # http://msdn.microsoft.com/en-us/library/windows/desktop/aa365247(v=vs.85).aspx) | |
e9c0cdd3 | 785 | if compat_os_name == 'nt': |
5f58165d S |
786 | absfilepath = os.path.abspath(filename) |
787 | if len(absfilepath) > 259: | |
788 | filename = '\\\\?\\' + absfilepath | |
d41e6efc PH |
789 | with open(filename, 'wb') as outf: |
790 | outf.write(webpage_bytes) | |
791 | ||
ec0fafbb AA |
792 | try: |
793 | content = webpage_bytes.decode(encoding, 'replace') | |
794 | except LookupError: | |
795 | content = webpage_bytes.decode('utf-8', 'replace') | |
2410c43d | 796 | |
4457823d | 797 | self.__check_blocked(content) |
2410c43d | 798 | |
23be51d8 | 799 | return content |
d6983cb4 | 800 | |
d391b7e2 S |
801 | def _download_webpage( |
802 | self, url_or_request, video_id, note=None, errnote=None, | |
803 | fatal=True, tries=1, timeout=5, encoding=None, data=None, | |
804 | headers={}, query={}, expected_status=None): | |
805 | """ | |
806 | Return the data of the page as a string. | |
807 | ||
808 | Arguments: | |
809 | url_or_request -- plain text URL as a string or | |
810 | a compat_urllib_request.Requestobject | |
811 | video_id -- Video/playlist/item identifier (string) | |
812 | ||
813 | Keyword arguments: | |
814 | note -- note printed before downloading (string) | |
815 | errnote -- note printed in case of an error (string) | |
816 | fatal -- flag denoting whether error should be considered fatal, | |
817 | i.e. whether it should cause ExtractionError to be raised, | |
818 | otherwise a warning will be reported and extraction continued | |
819 | tries -- number of tries | |
820 | timeout -- sleep interval between tries | |
821 | encoding -- encoding for a page content decoding, guessed automatically | |
822 | when not explicitly specified | |
823 | data -- POST data (bytes) | |
824 | headers -- HTTP headers (dict) | |
825 | query -- URL query (dict) | |
826 | expected_status -- allows to accept failed HTTP requests (non 2xx | |
827 | status code) by explicitly specifying a set of accepted status | |
828 | codes. Can be any of the following entities: | |
829 | - an integer type specifying an exact failed status code to | |
830 | accept | |
831 | - a list or a tuple of integer types specifying a list of | |
832 | failed status codes to accept | |
833 | - a callable accepting an actual failed status code and | |
834 | returning True if it should be accepted | |
835 | Note that this argument does not affect success status codes (2xx) | |
836 | which are always accepted. | |
837 | """ | |
838 | ||
995ad69c TF |
839 | success = False |
840 | try_count = 0 | |
841 | while success is False: | |
842 | try: | |
d391b7e2 S |
843 | res = self._download_webpage_handle( |
844 | url_or_request, video_id, note, errnote, fatal, | |
845 | encoding=encoding, data=data, headers=headers, query=query, | |
846 | expected_status=expected_status) | |
995ad69c TF |
847 | success = True |
848 | except compat_http_client.IncompleteRead as e: | |
849 | try_count += 1 | |
850 | if try_count >= tries: | |
851 | raise e | |
852 | self._sleep(timeout, video_id) | |
7cc3570e PH |
853 | if res is False: |
854 | return res | |
855 | else: | |
856 | content, _ = res | |
857 | return content | |
d6983cb4 | 858 | |
e0d198c1 S |
859 | def _download_xml_handle( |
860 | self, url_or_request, video_id, note='Downloading XML', | |
861 | errnote='Unable to download XML', transform_source=None, | |
d391b7e2 S |
862 | fatal=True, encoding=None, data=None, headers={}, query={}, |
863 | expected_status=None): | |
864 | """ | |
ee0ba927 | 865 | Return a tuple (xml as an compat_etree_Element, URL handle). |
d391b7e2 S |
866 | |
867 | See _download_webpage docstring for arguments specification. | |
868 | """ | |
e0d198c1 S |
869 | res = self._download_webpage_handle( |
870 | url_or_request, video_id, note, errnote, fatal=fatal, | |
d391b7e2 S |
871 | encoding=encoding, data=data, headers=headers, query=query, |
872 | expected_status=expected_status) | |
e0d198c1 S |
873 | if res is False: |
874 | return res | |
875 | xml_string, urlh = res | |
876 | return self._parse_xml( | |
877 | xml_string, video_id, transform_source=transform_source, | |
878 | fatal=fatal), urlh | |
879 | ||
d391b7e2 S |
880 | def _download_xml( |
881 | self, url_or_request, video_id, | |
882 | note='Downloading XML', errnote='Unable to download XML', | |
883 | transform_source=None, fatal=True, encoding=None, | |
884 | data=None, headers={}, query={}, expected_status=None): | |
885 | """ | |
ee0ba927 | 886 | Return the xml as an compat_etree_Element. |
d391b7e2 S |
887 | |
888 | See _download_webpage docstring for arguments specification. | |
889 | """ | |
e0d198c1 S |
890 | res = self._download_xml_handle( |
891 | url_or_request, video_id, note=note, errnote=errnote, | |
892 | transform_source=transform_source, fatal=fatal, encoding=encoding, | |
d391b7e2 S |
893 | data=data, headers=headers, query=query, |
894 | expected_status=expected_status) | |
e0d198c1 | 895 | return res if res is False else res[0] |
e01c3d2e S |
896 | |
897 | def _parse_xml(self, xml_string, video_id, transform_source=None, fatal=True): | |
e2b38da9 PH |
898 | if transform_source: |
899 | xml_string = transform_source(xml_string) | |
e01c3d2e S |
900 | try: |
901 | return compat_etree_fromstring(xml_string.encode('utf-8')) | |
902 | except compat_xml_parse_error as ve: | |
903 | errmsg = '%s: Failed to parse XML ' % video_id | |
904 | if fatal: | |
905 | raise ExtractorError(errmsg, cause=ve) | |
906 | else: | |
907 | self.report_warning(errmsg + str(ve)) | |
267ed0c5 | 908 | |
0fe7783e S |
909 | def _download_json_handle( |
910 | self, url_or_request, video_id, note='Downloading JSON metadata', | |
911 | errnote='Unable to download JSON metadata', transform_source=None, | |
d391b7e2 S |
912 | fatal=True, encoding=None, data=None, headers={}, query={}, |
913 | expected_status=None): | |
914 | """ | |
915 | Return a tuple (JSON object, URL handle). | |
916 | ||
917 | See _download_webpage docstring for arguments specification. | |
918 | """ | |
0fe7783e | 919 | res = self._download_webpage_handle( |
c9a77969 | 920 | url_or_request, video_id, note, errnote, fatal=fatal, |
d391b7e2 S |
921 | encoding=encoding, data=data, headers=headers, query=query, |
922 | expected_status=expected_status) | |
0fe7783e S |
923 | if res is False: |
924 | return res | |
925 | json_string, urlh = res | |
ebb64199 | 926 | return self._parse_json( |
0fe7783e S |
927 | json_string, video_id, transform_source=transform_source, |
928 | fatal=fatal), urlh | |
929 | ||
930 | def _download_json( | |
931 | self, url_or_request, video_id, note='Downloading JSON metadata', | |
932 | errnote='Unable to download JSON metadata', transform_source=None, | |
d391b7e2 S |
933 | fatal=True, encoding=None, data=None, headers={}, query={}, |
934 | expected_status=None): | |
935 | """ | |
936 | Return the JSON object as a dict. | |
937 | ||
938 | See _download_webpage docstring for arguments specification. | |
939 | """ | |
0fe7783e S |
940 | res = self._download_json_handle( |
941 | url_or_request, video_id, note=note, errnote=errnote, | |
942 | transform_source=transform_source, fatal=fatal, encoding=encoding, | |
d391b7e2 S |
943 | data=data, headers=headers, query=query, |
944 | expected_status=expected_status) | |
0fe7783e | 945 | return res if res is False else res[0] |
ebb64199 TF |
946 | |
947 | def _parse_json(self, json_string, video_id, transform_source=None, fatal=True): | |
81c2f20b PH |
948 | if transform_source: |
949 | json_string = transform_source(json_string) | |
3d3538e4 PH |
950 | try: |
951 | return json.loads(json_string) | |
952 | except ValueError as ve: | |
e7b6d122 PH |
953 | errmsg = '%s: Failed to parse JSON ' % video_id |
954 | if fatal: | |
955 | raise ExtractorError(errmsg, cause=ve) | |
956 | else: | |
957 | self.report_warning(errmsg + str(ve)) | |
3d3538e4 | 958 | |
adddc50c | 959 | def _parse_socket_response_as_json(self, data, video_id, transform_source=None, fatal=True): |
960 | return self._parse_json( | |
961 | data[data.find('{'):data.rfind('}') + 1], | |
962 | video_id, transform_source, fatal) | |
963 | ||
964 | def _download_socket_json_handle( | |
965 | self, url_or_request, video_id, note='Polling socket', | |
966 | errnote='Unable to poll socket', transform_source=None, | |
967 | fatal=True, encoding=None, data=None, headers={}, query={}, | |
968 | expected_status=None): | |
969 | """ | |
970 | Return a tuple (JSON object, URL handle). | |
971 | ||
972 | See _download_webpage docstring for arguments specification. | |
973 | """ | |
974 | res = self._download_webpage_handle( | |
975 | url_or_request, video_id, note, errnote, fatal=fatal, | |
976 | encoding=encoding, data=data, headers=headers, query=query, | |
977 | expected_status=expected_status) | |
978 | if res is False: | |
979 | return res | |
980 | webpage, urlh = res | |
981 | return self._parse_socket_response_as_json( | |
982 | webpage, video_id, transform_source=transform_source, | |
983 | fatal=fatal), urlh | |
984 | ||
985 | def _download_socket_json( | |
986 | self, url_or_request, video_id, note='Polling socket', | |
987 | errnote='Unable to poll socket', transform_source=None, | |
988 | fatal=True, encoding=None, data=None, headers={}, query={}, | |
989 | expected_status=None): | |
990 | """ | |
991 | Return the JSON object as a dict. | |
992 | ||
993 | See _download_webpage docstring for arguments specification. | |
994 | """ | |
995 | res = self._download_socket_json_handle( | |
996 | url_or_request, video_id, note=note, errnote=errnote, | |
997 | transform_source=transform_source, fatal=fatal, encoding=encoding, | |
998 | data=data, headers=headers, query=query, | |
999 | expected_status=expected_status) | |
1000 | return res if res is False else res[0] | |
1001 | ||
a06916d9 | 1002 | def report_warning(self, msg, video_id=None, *args, **kwargs): |
f1a9d64e | 1003 | idstr = '' if video_id is None else '%s: ' % video_id |
f45f96f8 | 1004 | self._downloader.report_warning( |
a06916d9 | 1005 | '[%s] %s%s' % (self.IE_NAME, idstr, msg), *args, **kwargs) |
f45f96f8 | 1006 | |
a06916d9 | 1007 | def to_screen(self, msg, *args, **kwargs): |
d6983cb4 | 1008 | """Print msg to screen, prefixing it with '[ie_name]'""" |
a06916d9 | 1009 | self._downloader.to_screen('[%s] %s' % (self.IE_NAME, msg), *args, **kwargs) |
1010 | ||
1011 | def write_debug(self, msg, *args, **kwargs): | |
1012 | self._downloader.write_debug('[%s] %s' % (self.IE_NAME, msg), *args, **kwargs) | |
1013 | ||
1014 | def get_param(self, name, default=None, *args, **kwargs): | |
1015 | if self._downloader: | |
1016 | return self._downloader.params.get(name, default, *args, **kwargs) | |
1017 | return default | |
d6983cb4 PH |
1018 | |
1019 | def report_extraction(self, id_or_name): | |
1020 | """Report information extraction.""" | |
f1a9d64e | 1021 | self.to_screen('%s: Extracting information' % id_or_name) |
d6983cb4 PH |
1022 | |
1023 | def report_download_webpage(self, video_id): | |
1024 | """Report webpage download.""" | |
f1a9d64e | 1025 | self.to_screen('%s: Downloading webpage' % video_id) |
d6983cb4 PH |
1026 | |
1027 | def report_age_confirmation(self): | |
1028 | """Report attempt to confirm age.""" | |
f1a9d64e | 1029 | self.to_screen('Confirming age') |
d6983cb4 | 1030 | |
fc79158d JMF |
1031 | def report_login(self): |
1032 | """Report attempt to log in.""" | |
f1a9d64e | 1033 | self.to_screen('Logging in') |
fc79158d | 1034 | |
b7da73eb | 1035 | def raise_login_required( |
9d5d4d64 | 1036 | self, msg='This video is only available for registered users', |
1037 | metadata_available=False, method='any'): | |
a06916d9 | 1038 | if metadata_available and self.get_param('ignore_no_formats_error'): |
b7da73eb | 1039 | self.report_warning(msg) |
46890374 | 1040 | if method is not None: |
1041 | msg = '%s. %s' % (msg, self._LOGIN_HINTS[method]) | |
1042 | raise ExtractorError(msg, expected=True) | |
43e7d3c9 | 1043 | |
b7da73eb | 1044 | def raise_geo_restricted( |
1045 | self, msg='This video is not available from your location due to geo restriction', | |
1046 | countries=None, metadata_available=False): | |
a06916d9 | 1047 | if metadata_available and self.get_param('ignore_no_formats_error'): |
b7da73eb | 1048 | self.report_warning(msg) |
1049 | else: | |
1050 | raise GeoRestrictedError(msg, countries=countries) | |
1051 | ||
1052 | def raise_no_formats(self, msg, expected=False, video_id=None): | |
a06916d9 | 1053 | if expected and self.get_param('ignore_no_formats_error'): |
b7da73eb | 1054 | self.report_warning(msg, video_id) |
1055 | else: | |
1056 | raise ExtractorError(msg, expected=expected, video_id=video_id) | |
c430802e | 1057 | |
5f6a1245 | 1058 | # Methods for following #608 |
c0d0b01f | 1059 | @staticmethod |
830d53bf | 1060 | def url_result(url, ie=None, video_id=None, video_title=None): |
10952eb2 | 1061 | """Returns a URL that points to a page that should be processed""" |
5f6a1245 | 1062 | # TODO: ie should be the class used for getting the info |
d6983cb4 PH |
1063 | video_info = {'_type': 'url', |
1064 | 'url': url, | |
1065 | 'ie_key': ie} | |
7012b23c PH |
1066 | if video_id is not None: |
1067 | video_info['id'] = video_id | |
830d53bf S |
1068 | if video_title is not None: |
1069 | video_info['title'] = video_title | |
d6983cb4 | 1070 | return video_info |
5f6a1245 | 1071 | |
749ca5ec S |
1072 | def playlist_from_matches(self, matches, playlist_id=None, playlist_title=None, getter=None, ie=None): |
1073 | urls = orderedSet( | |
46b18f23 JH |
1074 | self.url_result(self._proto_relative_url(getter(m) if getter else m), ie) |
1075 | for m in matches) | |
1076 | return self.playlist_result( | |
749ca5ec | 1077 | urls, playlist_id=playlist_id, playlist_title=playlist_title) |
46b18f23 | 1078 | |
c0d0b01f | 1079 | @staticmethod |
b60419c5 | 1080 | def playlist_result(entries, playlist_id=None, playlist_title=None, playlist_description=None, **kwargs): |
d6983cb4 PH |
1081 | """Returns a playlist""" |
1082 | video_info = {'_type': 'playlist', | |
1083 | 'entries': entries} | |
b60419c5 | 1084 | video_info.update(kwargs) |
d6983cb4 PH |
1085 | if playlist_id: |
1086 | video_info['id'] = playlist_id | |
1087 | if playlist_title: | |
1088 | video_info['title'] = playlist_title | |
ecc97af3 | 1089 | if playlist_description is not None: |
acf5cbfe | 1090 | video_info['description'] = playlist_description |
d6983cb4 PH |
1091 | return video_info |
1092 | ||
c342041f | 1093 | def _search_regex(self, pattern, string, name, default=NO_DEFAULT, fatal=True, flags=0, group=None): |
d6983cb4 PH |
1094 | """ |
1095 | Perform a regex search on the given string, using a single or a list of | |
1096 | patterns returning the first matching group. | |
1097 | In case of failure return a default value or raise a WARNING or a | |
55b3e45b | 1098 | RegexNotFoundError, depending on fatal, specifying the field name. |
d6983cb4 PH |
1099 | """ |
1100 | if isinstance(pattern, (str, compat_str, compiled_regex_type)): | |
1101 | mobj = re.search(pattern, string, flags) | |
1102 | else: | |
1103 | for p in pattern: | |
1104 | mobj = re.search(p, string, flags) | |
c3415d1b PH |
1105 | if mobj: |
1106 | break | |
d6983cb4 | 1107 | |
a06916d9 | 1108 | if not self.get_param('no_color') and compat_os_name != 'nt' and sys.stderr.isatty(): |
f1a9d64e | 1109 | _name = '\033[0;34m%s\033[0m' % name |
d6983cb4 PH |
1110 | else: |
1111 | _name = name | |
1112 | ||
1113 | if mobj: | |
711ede6e PH |
1114 | if group is None: |
1115 | # return the first matching group | |
1116 | return next(g for g in mobj.groups() if g is not None) | |
198f7ea8 | 1117 | elif isinstance(group, (list, tuple)): |
1118 | return tuple(mobj.group(g) for g in group) | |
711ede6e PH |
1119 | else: |
1120 | return mobj.group(group) | |
c342041f | 1121 | elif default is not NO_DEFAULT: |
d6983cb4 PH |
1122 | return default |
1123 | elif fatal: | |
f1a9d64e | 1124 | raise RegexNotFoundError('Unable to extract %s' % _name) |
d6983cb4 | 1125 | else: |
6a39ee13 | 1126 | self.report_warning('unable to extract %s' % _name + bug_reports_message()) |
d6983cb4 PH |
1127 | return None |
1128 | ||
c342041f | 1129 | def _html_search_regex(self, pattern, string, name, default=NO_DEFAULT, fatal=True, flags=0, group=None): |
d6983cb4 PH |
1130 | """ |
1131 | Like _search_regex, but strips HTML tags and unescapes entities. | |
1132 | """ | |
711ede6e | 1133 | res = self._search_regex(pattern, string, name, default, fatal, flags, group) |
d6983cb4 PH |
1134 | if res: |
1135 | return clean_html(res).strip() | |
1136 | else: | |
1137 | return res | |
1138 | ||
2118fdd1 RA |
1139 | def _get_netrc_login_info(self, netrc_machine=None): |
1140 | username = None | |
1141 | password = None | |
1142 | netrc_machine = netrc_machine or self._NETRC_MACHINE | |
1143 | ||
a06916d9 | 1144 | if self.get_param('usenetrc', False): |
2118fdd1 RA |
1145 | try: |
1146 | info = netrc.netrc().authenticators(netrc_machine) | |
1147 | if info is not None: | |
1148 | username = info[0] | |
1149 | password = info[2] | |
1150 | else: | |
dcce092e S |
1151 | raise netrc.NetrcParseError( |
1152 | 'No authenticators for %s' % netrc_machine) | |
2118fdd1 | 1153 | except (IOError, netrc.NetrcParseError) as err: |
6a39ee13 | 1154 | self.report_warning( |
dcce092e | 1155 | 'parsing .netrc: %s' % error_to_compat_str(err)) |
2118fdd1 | 1156 | |
dcce092e | 1157 | return username, password |
2118fdd1 | 1158 | |
1b6712ab | 1159 | def _get_login_info(self, username_option='username', password_option='password', netrc_machine=None): |
fc79158d | 1160 | """ |
cf0649f8 | 1161 | Get the login info as (username, password) |
32443dd3 S |
1162 | First look for the manually specified credentials using username_option |
1163 | and password_option as keys in params dictionary. If no such credentials | |
1164 | available look in the netrc file using the netrc_machine or _NETRC_MACHINE | |
1165 | value. | |
fc79158d JMF |
1166 | If there's no info available, return (None, None) |
1167 | """ | |
fc79158d JMF |
1168 | |
1169 | # Attempt to use provided username and password or .netrc data | |
a06916d9 | 1170 | username = self.get_param(username_option) |
1171 | if username is not None: | |
1172 | password = self.get_param(password_option) | |
2118fdd1 | 1173 | else: |
1b6712ab | 1174 | username, password = self._get_netrc_login_info(netrc_machine) |
5f6a1245 | 1175 | |
2133565c | 1176 | return username, password |
fc79158d | 1177 | |
e64b7569 | 1178 | def _get_tfa_info(self, note='two-factor verification code'): |
83317f69 | 1179 | """ |
1180 | Get the two-factor authentication info | |
1181 | TODO - asking the user will be required for sms/phone verify | |
1182 | currently just uses the command line option | |
1183 | If there's no info available, return None | |
1184 | """ | |
83317f69 | 1185 | |
a06916d9 | 1186 | tfa = self.get_param('twofactor') |
1187 | if tfa is not None: | |
1188 | return tfa | |
83317f69 | 1189 | |
e64b7569 | 1190 | return compat_getpass('Type %s and press [Return]: ' % note) |
83317f69 | 1191 | |
46720279 JMF |
1192 | # Helper functions for extracting OpenGraph info |
1193 | @staticmethod | |
ab2d5247 | 1194 | def _og_regexes(prop): |
448ef1f3 | 1195 | content_re = r'content=(?:"([^"]+?)"|\'([^\']+?)\'|\s*([^\s"\'=<>`]+?))' |
22f5f5c6 | 1196 | property_re = (r'(?:name|property)=(?:\'og[:-]%(prop)s\'|"og[:-]%(prop)s"|\s*og[:-]%(prop)s\b)' |
7a6d76a6 | 1197 | % {'prop': re.escape(prop)}) |
78fb87b2 | 1198 | template = r'<meta[^>]+?%s[^>]+?%s' |
ab2d5247 | 1199 | return [ |
78fb87b2 JMF |
1200 | template % (property_re, content_re), |
1201 | template % (content_re, property_re), | |
ab2d5247 | 1202 | ] |
46720279 | 1203 | |
864f24bd S |
1204 | @staticmethod |
1205 | def _meta_regex(prop): | |
1206 | return r'''(?isx)<meta | |
8b9848ac | 1207 | (?=[^>]+(?:itemprop|name|property|id|http-equiv)=(["\']?)%s\1) |
864f24bd S |
1208 | [^>]+?content=(["\'])(?P<content>.*?)\2''' % re.escape(prop) |
1209 | ||
3c4e6d83 | 1210 | def _og_search_property(self, prop, html, name=None, **kargs): |
6606817a | 1211 | prop = variadic(prop) |
46720279 | 1212 | if name is None: |
b070564e S |
1213 | name = 'OpenGraph %s' % prop[0] |
1214 | og_regexes = [] | |
1215 | for p in prop: | |
1216 | og_regexes.extend(self._og_regexes(p)) | |
1217 | escaped = self._search_regex(og_regexes, html, name, flags=re.DOTALL, **kargs) | |
eb0a8398 PH |
1218 | if escaped is None: |
1219 | return None | |
1220 | return unescapeHTML(escaped) | |
46720279 JMF |
1221 | |
1222 | def _og_search_thumbnail(self, html, **kargs): | |
10952eb2 | 1223 | return self._og_search_property('image', html, 'thumbnail URL', fatal=False, **kargs) |
46720279 JMF |
1224 | |
1225 | def _og_search_description(self, html, **kargs): | |
1226 | return self._og_search_property('description', html, fatal=False, **kargs) | |
1227 | ||
1228 | def _og_search_title(self, html, **kargs): | |
1229 | return self._og_search_property('title', html, **kargs) | |
1230 | ||
8ffa13e0 | 1231 | def _og_search_video_url(self, html, name='video url', secure=True, **kargs): |
a3681973 PH |
1232 | regexes = self._og_regexes('video') + self._og_regexes('video:url') |
1233 | if secure: | |
1234 | regexes = self._og_regexes('video:secure_url') + regexes | |
8ffa13e0 | 1235 | return self._html_search_regex(regexes, html, name, **kargs) |
46720279 | 1236 | |
78338f71 JMF |
1237 | def _og_search_url(self, html, **kargs): |
1238 | return self._og_search_property('url', html, **kargs) | |
1239 | ||
40c696e5 | 1240 | def _html_search_meta(self, name, html, display_name=None, fatal=False, **kwargs): |
6606817a | 1241 | name = variadic(name) |
59040888 | 1242 | if display_name is None: |
88d9f6c0 | 1243 | display_name = name[0] |
59040888 | 1244 | return self._html_search_regex( |
88d9f6c0 | 1245 | [self._meta_regex(n) for n in name], |
711ede6e | 1246 | html, display_name, fatal=fatal, group='content', **kwargs) |
59040888 PH |
1247 | |
1248 | def _dc_search_uploader(self, html): | |
1249 | return self._html_search_meta('dc.creator', html, 'uploader') | |
1250 | ||
8dbe9899 PH |
1251 | def _rta_search(self, html): |
1252 | # See http://www.rtalabel.org/index.php?content=howtofaq#single | |
1253 | if re.search(r'(?ix)<meta\s+name="rating"\s+' | |
1254 | r' content="RTA-5042-1996-1400-1577-RTA"', | |
1255 | html): | |
1256 | return 18 | |
1257 | return 0 | |
1258 | ||
59040888 PH |
1259 | def _media_rating_search(self, html): |
1260 | # See http://www.tjg-designs.com/WP/metadata-code-examples-adding-metadata-to-your-web-pages/ | |
1261 | rating = self._html_search_meta('rating', html) | |
1262 | ||
1263 | if not rating: | |
1264 | return None | |
1265 | ||
1266 | RATING_TABLE = { | |
1267 | 'safe for kids': 0, | |
1268 | 'general': 8, | |
1269 | '14 years': 14, | |
1270 | 'mature': 17, | |
1271 | 'restricted': 19, | |
1272 | } | |
d800609c | 1273 | return RATING_TABLE.get(rating.lower()) |
59040888 | 1274 | |
69319969 | 1275 | def _family_friendly_search(self, html): |
6ca7732d | 1276 | # See http://schema.org/VideoObject |
ac8491fc S |
1277 | family_friendly = self._html_search_meta( |
1278 | 'isFamilyFriendly', html, default=None) | |
69319969 NJ |
1279 | |
1280 | if not family_friendly: | |
1281 | return None | |
1282 | ||
1283 | RATING_TABLE = { | |
1284 | '1': 0, | |
1285 | 'true': 0, | |
1286 | '0': 18, | |
1287 | 'false': 18, | |
1288 | } | |
d800609c | 1289 | return RATING_TABLE.get(family_friendly.lower()) |
69319969 | 1290 | |
0c708f11 JMF |
1291 | def _twitter_search_player(self, html): |
1292 | return self._html_search_meta('twitter:player', html, | |
9e1a5b84 | 1293 | 'twitter card player') |
0c708f11 | 1294 | |
95b31e26 | 1295 | def _search_json_ld(self, html, video_id, expected_type=None, **kwargs): |
4433bb02 | 1296 | json_ld_list = list(re.finditer(JSON_LD_RE, html)) |
321b5e08 | 1297 | default = kwargs.get('default', NO_DEFAULT) |
321b5e08 S |
1298 | # JSON-LD may be malformed and thus `fatal` should be respected. |
1299 | # At the same time `default` may be passed that assumes `fatal=False` | |
1300 | # for _search_regex. Let's simulate the same behavior here as well. | |
1301 | fatal = kwargs.get('fatal', True) if default == NO_DEFAULT else False | |
4433bb02 S |
1302 | json_ld = [] |
1303 | for mobj in json_ld_list: | |
1304 | json_ld_item = self._parse_json( | |
1305 | mobj.group('json_ld'), video_id, fatal=fatal) | |
1306 | if not json_ld_item: | |
1307 | continue | |
1308 | if isinstance(json_ld_item, dict): | |
1309 | json_ld.append(json_ld_item) | |
1310 | elif isinstance(json_ld_item, (list, tuple)): | |
1311 | json_ld.extend(json_ld_item) | |
1312 | if json_ld: | |
1313 | json_ld = self._json_ld(json_ld, video_id, fatal=fatal, expected_type=expected_type) | |
1314 | if json_ld: | |
1315 | return json_ld | |
1316 | if default is not NO_DEFAULT: | |
1317 | return default | |
1318 | elif fatal: | |
1319 | raise RegexNotFoundError('Unable to extract JSON-LD') | |
1320 | else: | |
6a39ee13 | 1321 | self.report_warning('unable to extract JSON-LD %s' % bug_reports_message()) |
4433bb02 | 1322 | return {} |
4ca2a3cf | 1323 | |
95b31e26 | 1324 | def _json_ld(self, json_ld, video_id, fatal=True, expected_type=None): |
4ca2a3cf S |
1325 | if isinstance(json_ld, compat_str): |
1326 | json_ld = self._parse_json(json_ld, video_id, fatal=fatal) | |
1327 | if not json_ld: | |
1328 | return {} | |
1329 | info = {} | |
46933a15 S |
1330 | if not isinstance(json_ld, (list, tuple, dict)): |
1331 | return info | |
1332 | if isinstance(json_ld, dict): | |
1333 | json_ld = [json_ld] | |
bae14048 | 1334 | |
e7e4a6e0 S |
1335 | INTERACTION_TYPE_MAP = { |
1336 | 'CommentAction': 'comment', | |
1337 | 'AgreeAction': 'like', | |
1338 | 'DisagreeAction': 'dislike', | |
1339 | 'LikeAction': 'like', | |
1340 | 'DislikeAction': 'dislike', | |
1341 | 'ListenAction': 'view', | |
1342 | 'WatchAction': 'view', | |
1343 | 'ViewAction': 'view', | |
1344 | } | |
1345 | ||
29f7c58a | 1346 | def extract_interaction_type(e): |
1347 | interaction_type = e.get('interactionType') | |
1348 | if isinstance(interaction_type, dict): | |
1349 | interaction_type = interaction_type.get('@type') | |
1350 | return str_or_none(interaction_type) | |
1351 | ||
e7e4a6e0 S |
1352 | def extract_interaction_statistic(e): |
1353 | interaction_statistic = e.get('interactionStatistic') | |
29f7c58a | 1354 | if isinstance(interaction_statistic, dict): |
1355 | interaction_statistic = [interaction_statistic] | |
e7e4a6e0 S |
1356 | if not isinstance(interaction_statistic, list): |
1357 | return | |
1358 | for is_e in interaction_statistic: | |
1359 | if not isinstance(is_e, dict): | |
1360 | continue | |
1361 | if is_e.get('@type') != 'InteractionCounter': | |
1362 | continue | |
29f7c58a | 1363 | interaction_type = extract_interaction_type(is_e) |
1364 | if not interaction_type: | |
e7e4a6e0 | 1365 | continue |
ce5b9040 S |
1366 | # For interaction count some sites provide string instead of |
1367 | # an integer (as per spec) with non digit characters (e.g. ",") | |
1368 | # so extracting count with more relaxed str_to_int | |
1369 | interaction_count = str_to_int(is_e.get('userInteractionCount')) | |
e7e4a6e0 S |
1370 | if interaction_count is None: |
1371 | continue | |
1372 | count_kind = INTERACTION_TYPE_MAP.get(interaction_type.split('/')[-1]) | |
1373 | if not count_kind: | |
1374 | continue | |
1375 | count_key = '%s_count' % count_kind | |
1376 | if info.get(count_key) is not None: | |
1377 | continue | |
1378 | info[count_key] = interaction_count | |
1379 | ||
bae14048 S |
1380 | def extract_video_object(e): |
1381 | assert e['@type'] == 'VideoObject' | |
f7ad7160 | 1382 | author = e.get('author') |
bae14048 | 1383 | info.update({ |
bebef109 | 1384 | 'url': url_or_none(e.get('contentUrl')), |
bae14048 S |
1385 | 'title': unescapeHTML(e.get('name')), |
1386 | 'description': unescapeHTML(e.get('description')), | |
bebef109 | 1387 | 'thumbnail': url_or_none(e.get('thumbnailUrl') or e.get('thumbnailURL')), |
bae14048 S |
1388 | 'duration': parse_duration(e.get('duration')), |
1389 | 'timestamp': unified_timestamp(e.get('uploadDate')), | |
f7ad7160 | 1390 | # author can be an instance of 'Organization' or 'Person' types. |
1391 | # both types can have 'name' property(inherited from 'Thing' type). [1] | |
1392 | # however some websites are using 'Text' type instead. | |
1393 | # 1. https://schema.org/VideoObject | |
1394 | 'uploader': author.get('name') if isinstance(author, dict) else author if isinstance(author, compat_str) else None, | |
bae14048 S |
1395 | 'filesize': float_or_none(e.get('contentSize')), |
1396 | 'tbr': int_or_none(e.get('bitrate')), | |
1397 | 'width': int_or_none(e.get('width')), | |
1398 | 'height': int_or_none(e.get('height')), | |
33a81c2c | 1399 | 'view_count': int_or_none(e.get('interactionCount')), |
bae14048 | 1400 | }) |
e7e4a6e0 | 1401 | extract_interaction_statistic(e) |
bae14048 | 1402 | |
46933a15 | 1403 | for e in json_ld: |
4433bb02 | 1404 | if '@context' in e: |
46933a15 S |
1405 | item_type = e.get('@type') |
1406 | if expected_type is not None and expected_type != item_type: | |
4433bb02 | 1407 | continue |
c69701c6 | 1408 | if item_type in ('TVEpisode', 'Episode'): |
440863ad | 1409 | episode_name = unescapeHTML(e.get('name')) |
46933a15 | 1410 | info.update({ |
440863ad | 1411 | 'episode': episode_name, |
46933a15 S |
1412 | 'episode_number': int_or_none(e.get('episodeNumber')), |
1413 | 'description': unescapeHTML(e.get('description')), | |
1414 | }) | |
440863ad S |
1415 | if not info.get('title') and episode_name: |
1416 | info['title'] = episode_name | |
46933a15 | 1417 | part_of_season = e.get('partOfSeason') |
c69701c6 | 1418 | if isinstance(part_of_season, dict) and part_of_season.get('@type') in ('TVSeason', 'Season', 'CreativeWorkSeason'): |
458fd30f S |
1419 | info.update({ |
1420 | 'season': unescapeHTML(part_of_season.get('name')), | |
1421 | 'season_number': int_or_none(part_of_season.get('seasonNumber')), | |
1422 | }) | |
d16b3c66 | 1423 | part_of_series = e.get('partOfSeries') or e.get('partOfTVSeries') |
c69701c6 | 1424 | if isinstance(part_of_series, dict) and part_of_series.get('@type') in ('TVSeries', 'Series', 'CreativeWorkSeries'): |
46933a15 | 1425 | info['series'] = unescapeHTML(part_of_series.get('name')) |
391256dc S |
1426 | elif item_type == 'Movie': |
1427 | info.update({ | |
1428 | 'title': unescapeHTML(e.get('name')), | |
1429 | 'description': unescapeHTML(e.get('description')), | |
1430 | 'duration': parse_duration(e.get('duration')), | |
1431 | 'timestamp': unified_timestamp(e.get('dateCreated')), | |
1432 | }) | |
3931b845 | 1433 | elif item_type in ('Article', 'NewsArticle'): |
46933a15 S |
1434 | info.update({ |
1435 | 'timestamp': parse_iso8601(e.get('datePublished')), | |
1436 | 'title': unescapeHTML(e.get('headline')), | |
1437 | 'description': unescapeHTML(e.get('articleBody')), | |
1438 | }) | |
1439 | elif item_type == 'VideoObject': | |
bae14048 | 1440 | extract_video_object(e) |
4433bb02 S |
1441 | if expected_type is None: |
1442 | continue | |
1443 | else: | |
1444 | break | |
c69701c6 S |
1445 | video = e.get('video') |
1446 | if isinstance(video, dict) and video.get('@type') == 'VideoObject': | |
1447 | extract_video_object(video) | |
4433bb02 S |
1448 | if expected_type is None: |
1449 | continue | |
1450 | else: | |
1451 | break | |
4ca2a3cf S |
1452 | return dict((k, v) for k, v in info.items() if v is not None) |
1453 | ||
27713812 | 1454 | @staticmethod |
f8da79f8 | 1455 | def _hidden_inputs(html): |
586f1cc5 | 1456 | html = re.sub(r'<!--(?:(?!<!--).)*-->', '', html) |
201ea3ee | 1457 | hidden_inputs = {} |
c8498368 S |
1458 | for input in re.findall(r'(?i)(<input[^>]+>)', html): |
1459 | attrs = extract_attributes(input) | |
1460 | if not input: | |
201ea3ee | 1461 | continue |
c8498368 | 1462 | if attrs.get('type') not in ('hidden', 'submit'): |
201ea3ee | 1463 | continue |
c8498368 S |
1464 | name = attrs.get('name') or attrs.get('id') |
1465 | value = attrs.get('value') | |
1466 | if name and value is not None: | |
1467 | hidden_inputs[name] = value | |
201ea3ee | 1468 | return hidden_inputs |
27713812 | 1469 | |
cf61d96d S |
1470 | def _form_hidden_inputs(self, form_id, html): |
1471 | form = self._search_regex( | |
73eb13df | 1472 | r'(?is)<form[^>]+?id=(["\'])%s\1[^>]*>(?P<form>.+?)</form>' % form_id, |
cf61d96d S |
1473 | html, '%s form' % form_id, group='form') |
1474 | return self._hidden_inputs(form) | |
1475 | ||
eb8a4433 | 1476 | class FormatSort: |
b050d210 | 1477 | regex = r' *((?P<reverse>\+)?(?P<field>[a-zA-Z0-9_]+)((?P<separator>[~:])(?P<limit>.*?))?)? *$' |
eb8a4433 | 1478 | |
8326b00a | 1479 | default = ('hidden', 'aud_or_vid', 'hasvid', 'ie_pref', 'lang', 'quality', |
155d2b48 | 1480 | 'res', 'fps', 'codec:vp9.2', 'size', 'br', 'asr', |
53ed7066 | 1481 | 'proto', 'ext', 'hasaud', 'source', 'format_id') # These must not be aliases |
1482 | ytdl_default = ('hasaud', 'quality', 'tbr', 'filesize', 'vbr', | |
1483 | 'height', 'width', 'proto', 'vext', 'abr', 'aext', | |
1484 | 'fps', 'fs_approx', 'source', 'format_id') | |
eb8a4433 | 1485 | |
1486 | settings = { | |
1487 | 'vcodec': {'type': 'ordered', 'regex': True, | |
155d2b48 | 1488 | 'order': ['av0?1', 'vp0?9.2', 'vp0?9', '[hx]265|he?vc?', '[hx]264|avc', 'vp0?8', 'mp4v|h263', 'theora', '', None, 'none']}, |
eb8a4433 | 1489 | 'acodec': {'type': 'ordered', 'regex': True, |
1490 | 'order': ['opus', 'vorbis', 'aac', 'mp?4a?', 'mp3', 'e?a?c-?3', 'dts', '', None, 'none']}, | |
f137c99e | 1491 | 'proto': {'type': 'ordered', 'regex': True, 'field': 'protocol', |
e36d50c5 | 1492 | 'order': ['(ht|f)tps', '(ht|f)tp$', 'm3u8.+', '.*dash', 'ws|websocket', '', 'mms|rtsp', 'none', 'f4']}, |
eb8a4433 | 1493 | 'vext': {'type': 'ordered', 'field': 'video_ext', |
91ebc640 | 1494 | 'order': ('mp4', 'webm', 'flv', '', 'none'), |
eb8a4433 | 1495 | 'order_free': ('webm', 'mp4', 'flv', '', 'none')}, |
1496 | 'aext': {'type': 'ordered', 'field': 'audio_ext', | |
1497 | 'order': ('m4a', 'aac', 'mp3', 'ogg', 'opus', 'webm', '', 'none'), | |
1498 | 'order_free': ('opus', 'ogg', 'webm', 'm4a', 'mp3', 'aac', '', 'none')}, | |
1499 | 'hidden': {'visible': False, 'forced': True, 'type': 'extractor', 'max': -1000}, | |
8326b00a | 1500 | 'aud_or_vid': {'visible': False, 'forced': True, 'type': 'multiple', 'default': 1, |
1501 | 'field': ('vcodec', 'acodec'), | |
1502 | 'function': lambda it: int(any(v != 'none' for v in it))}, | |
f983b875 | 1503 | 'ie_pref': {'priority': True, 'type': 'extractor'}, |
63be1aab | 1504 | 'hasvid': {'priority': True, 'field': 'vcodec', 'type': 'boolean', 'not_in_list': ('none',)}, |
1505 | 'hasaud': {'field': 'acodec', 'type': 'boolean', 'not_in_list': ('none',)}, | |
e4beae70 | 1506 | 'lang': {'priority': True, 'convert': 'ignore', 'field': 'language_preference'}, |
6a04a74e | 1507 | 'quality': {'convert': 'float_none', 'default': -1}, |
eb8a4433 | 1508 | 'filesize': {'convert': 'bytes'}, |
f137c99e | 1509 | 'fs_approx': {'convert': 'bytes', 'field': 'filesize_approx'}, |
1510 | 'id': {'convert': 'string', 'field': 'format_id'}, | |
eb8a4433 | 1511 | 'height': {'convert': 'float_none'}, |
1512 | 'width': {'convert': 'float_none'}, | |
1513 | 'fps': {'convert': 'float_none'}, | |
1514 | 'tbr': {'convert': 'float_none'}, | |
1515 | 'vbr': {'convert': 'float_none'}, | |
1516 | 'abr': {'convert': 'float_none'}, | |
1517 | 'asr': {'convert': 'float_none'}, | |
e4beae70 | 1518 | 'source': {'convert': 'ignore', 'field': 'source_preference'}, |
63be1aab | 1519 | |
eb8a4433 | 1520 | 'codec': {'type': 'combined', 'field': ('vcodec', 'acodec')}, |
63be1aab | 1521 | 'br': {'type': 'combined', 'field': ('tbr', 'vbr', 'abr'), 'same_limit': True}, |
1522 | 'size': {'type': 'combined', 'same_limit': True, 'field': ('filesize', 'fs_approx')}, | |
1523 | 'ext': {'type': 'combined', 'field': ('vext', 'aext')}, | |
1524 | 'res': {'type': 'multiple', 'field': ('height', 'width'), 'function': min}, | |
1525 | ||
1526 | # Most of these exist only for compatibility reasons | |
1527 | 'dimension': {'type': 'alias', 'field': 'res'}, | |
1528 | 'resolution': {'type': 'alias', 'field': 'res'}, | |
1529 | 'extension': {'type': 'alias', 'field': 'ext'}, | |
1530 | 'bitrate': {'type': 'alias', 'field': 'br'}, | |
eb8a4433 | 1531 | 'total_bitrate': {'type': 'alias', 'field': 'tbr'}, |
1532 | 'video_bitrate': {'type': 'alias', 'field': 'vbr'}, | |
1533 | 'audio_bitrate': {'type': 'alias', 'field': 'abr'}, | |
1534 | 'framerate': {'type': 'alias', 'field': 'fps'}, | |
63be1aab | 1535 | 'language_preference': {'type': 'alias', 'field': 'lang'}, # not named as 'language' because such a field exists |
1536 | 'protocol': {'type': 'alias', 'field': 'proto'}, | |
1537 | 'source_preference': {'type': 'alias', 'field': 'source'}, | |
1538 | 'filesize_approx': {'type': 'alias', 'field': 'fs_approx'}, | |
1539 | 'filesize_estimate': {'type': 'alias', 'field': 'size'}, | |
eb8a4433 | 1540 | 'samplerate': {'type': 'alias', 'field': 'asr'}, |
1541 | 'video_ext': {'type': 'alias', 'field': 'vext'}, | |
1542 | 'audio_ext': {'type': 'alias', 'field': 'aext'}, | |
1543 | 'video_codec': {'type': 'alias', 'field': 'vcodec'}, | |
1544 | 'audio_codec': {'type': 'alias', 'field': 'acodec'}, | |
63be1aab | 1545 | 'video': {'type': 'alias', 'field': 'hasvid'}, |
1546 | 'has_video': {'type': 'alias', 'field': 'hasvid'}, | |
1547 | 'audio': {'type': 'alias', 'field': 'hasaud'}, | |
1548 | 'has_audio': {'type': 'alias', 'field': 'hasaud'}, | |
1549 | 'extractor': {'type': 'alias', 'field': 'ie_pref'}, | |
1550 | 'preference': {'type': 'alias', 'field': 'ie_pref'}, | |
1551 | 'extractor_preference': {'type': 'alias', 'field': 'ie_pref'}, | |
1552 | 'format_id': {'type': 'alias', 'field': 'id'}, | |
1553 | } | |
eb8a4433 | 1554 | |
1555 | _order = [] | |
1556 | ||
1557 | def _get_field_setting(self, field, key): | |
1558 | if field not in self.settings: | |
1559 | self.settings[field] = {} | |
1560 | propObj = self.settings[field] | |
1561 | if key not in propObj: | |
1562 | type = propObj.get('type') | |
1563 | if key == 'field': | |
1564 | default = 'preference' if type == 'extractor' else (field,) if type in ('combined', 'multiple') else field | |
1565 | elif key == 'convert': | |
1566 | default = 'order' if type == 'ordered' else 'float_string' if field else 'ignore' | |
4bcc7bd1 | 1567 | else: |
eb8a4433 | 1568 | default = {'type': 'field', 'visible': True, 'order': [], 'not_in_list': (None,), 'function': max}.get(key, None) |
1569 | propObj[key] = default | |
1570 | return propObj[key] | |
1571 | ||
1572 | def _resolve_field_value(self, field, value, convertNone=False): | |
1573 | if value is None: | |
1574 | if not convertNone: | |
1575 | return None | |
4bcc7bd1 | 1576 | else: |
eb8a4433 | 1577 | value = value.lower() |
1578 | conversion = self._get_field_setting(field, 'convert') | |
1579 | if conversion == 'ignore': | |
1580 | return None | |
1581 | if conversion == 'string': | |
1582 | return value | |
1583 | elif conversion == 'float_none': | |
1584 | return float_or_none(value) | |
1585 | elif conversion == 'bytes': | |
1586 | return FileDownloader.parse_bytes(value) | |
1587 | elif conversion == 'order': | |
da9be05e | 1588 | order_list = (self._use_free_order and self._get_field_setting(field, 'order_free')) or self._get_field_setting(field, 'order') |
eb8a4433 | 1589 | use_regex = self._get_field_setting(field, 'regex') |
1590 | list_length = len(order_list) | |
1591 | empty_pos = order_list.index('') if '' in order_list else list_length + 1 | |
1592 | if use_regex and value is not None: | |
da9be05e | 1593 | for i, regex in enumerate(order_list): |
eb8a4433 | 1594 | if regex and re.match(regex, value): |
1595 | return list_length - i | |
1596 | return list_length - empty_pos # not in list | |
1597 | else: # not regex or value = None | |
1598 | return list_length - (order_list.index(value) if value in order_list else empty_pos) | |
1599 | else: | |
1600 | if value.isnumeric(): | |
1601 | return float(value) | |
4bcc7bd1 | 1602 | else: |
eb8a4433 | 1603 | self.settings[field]['convert'] = 'string' |
1604 | return value | |
1605 | ||
1606 | def evaluate_params(self, params, sort_extractor): | |
1607 | self._use_free_order = params.get('prefer_free_formats', False) | |
1608 | self._sort_user = params.get('format_sort', []) | |
1609 | self._sort_extractor = sort_extractor | |
1610 | ||
1611 | def add_item(field, reverse, closest, limit_text): | |
1612 | field = field.lower() | |
1613 | if field in self._order: | |
1614 | return | |
1615 | self._order.append(field) | |
1616 | limit = self._resolve_field_value(field, limit_text) | |
1617 | data = { | |
1618 | 'reverse': reverse, | |
1619 | 'closest': False if limit is None else closest, | |
1620 | 'limit_text': limit_text, | |
1621 | 'limit': limit} | |
1622 | if field in self.settings: | |
1623 | self.settings[field].update(data) | |
1624 | else: | |
1625 | self.settings[field] = data | |
1626 | ||
1627 | sort_list = ( | |
1628 | tuple(field for field in self.default if self._get_field_setting(field, 'forced')) | |
1629 | + (tuple() if params.get('format_sort_force', False) | |
1630 | else tuple(field for field in self.default if self._get_field_setting(field, 'priority'))) | |
1631 | + tuple(self._sort_user) + tuple(sort_extractor) + self.default) | |
1632 | ||
1633 | for item in sort_list: | |
1634 | match = re.match(self.regex, item) | |
1635 | if match is None: | |
1636 | raise ExtractorError('Invalid format sort string "%s" given by extractor' % item) | |
1637 | field = match.group('field') | |
1638 | if field is None: | |
1639 | continue | |
1640 | if self._get_field_setting(field, 'type') == 'alias': | |
1641 | field = self._get_field_setting(field, 'field') | |
1642 | reverse = match.group('reverse') is not None | |
b050d210 | 1643 | closest = match.group('separator') == '~' |
eb8a4433 | 1644 | limit_text = match.group('limit') |
1645 | ||
1646 | has_limit = limit_text is not None | |
1647 | has_multiple_fields = self._get_field_setting(field, 'type') == 'combined' | |
1648 | has_multiple_limits = has_limit and has_multiple_fields and not self._get_field_setting(field, 'same_limit') | |
1649 | ||
1650 | fields = self._get_field_setting(field, 'field') if has_multiple_fields else (field,) | |
1651 | limits = limit_text.split(":") if has_multiple_limits else (limit_text,) if has_limit else tuple() | |
1652 | limit_count = len(limits) | |
1653 | for (i, f) in enumerate(fields): | |
1654 | add_item(f, reverse, closest, | |
1655 | limits[i] if i < limit_count | |
1656 | else limits[0] if has_limit and not has_multiple_limits | |
1657 | else None) | |
1658 | ||
0760b0a7 | 1659 | def print_verbose_info(self, write_debug): |
b31fdeed | 1660 | if self._sort_user: |
0760b0a7 | 1661 | write_debug('Sort order given by user: %s' % ', '.join(self._sort_user)) |
eb8a4433 | 1662 | if self._sort_extractor: |
0760b0a7 | 1663 | write_debug('Sort order given by extractor: %s' % ', '.join(self._sort_extractor)) |
1664 | write_debug('Formats sorted by: %s' % ', '.join(['%s%s%s' % ( | |
eb8a4433 | 1665 | '+' if self._get_field_setting(field, 'reverse') else '', field, |
1666 | '%s%s(%s)' % ('~' if self._get_field_setting(field, 'closest') else ':', | |
1667 | self._get_field_setting(field, 'limit_text'), | |
1668 | self._get_field_setting(field, 'limit')) | |
1669 | if self._get_field_setting(field, 'limit_text') is not None else '') | |
1670 | for field in self._order if self._get_field_setting(field, 'visible')])) | |
1671 | ||
1672 | def _calculate_field_preference_from_value(self, format, field, type, value): | |
1673 | reverse = self._get_field_setting(field, 'reverse') | |
1674 | closest = self._get_field_setting(field, 'closest') | |
1675 | limit = self._get_field_setting(field, 'limit') | |
1676 | ||
1677 | if type == 'extractor': | |
1678 | maximum = self._get_field_setting(field, 'max') | |
1679 | if value is None or (maximum is not None and value >= maximum): | |
f983b875 | 1680 | value = -1 |
eb8a4433 | 1681 | elif type == 'boolean': |
1682 | in_list = self._get_field_setting(field, 'in_list') | |
1683 | not_in_list = self._get_field_setting(field, 'not_in_list') | |
1684 | value = 0 if ((in_list is None or value in in_list) and (not_in_list is None or value not in not_in_list)) else -1 | |
1685 | elif type == 'ordered': | |
1686 | value = self._resolve_field_value(field, value, True) | |
1687 | ||
1688 | # try to convert to number | |
6a04a74e | 1689 | val_num = float_or_none(value, default=self._get_field_setting(field, 'default')) |
eb8a4433 | 1690 | is_num = self._get_field_setting(field, 'convert') != 'string' and val_num is not None |
1691 | if is_num: | |
1692 | value = val_num | |
1693 | ||
1694 | return ((-10, 0) if value is None | |
1695 | else (1, value, 0) if not is_num # if a field has mixed strings and numbers, strings are sorted higher | |
1696 | else (0, -abs(value - limit), value - limit if reverse else limit - value) if closest | |
1697 | else (0, value, 0) if not reverse and (limit is None or value <= limit) | |
1698 | else (0, -value, 0) if limit is None or (reverse and value == limit) or value > limit | |
1699 | else (-1, value, 0)) | |
1700 | ||
1701 | def _calculate_field_preference(self, format, field): | |
1702 | type = self._get_field_setting(field, 'type') # extractor, boolean, ordered, field, multiple | |
1703 | get_value = lambda f: format.get(self._get_field_setting(f, 'field')) | |
1704 | if type == 'multiple': | |
1705 | type = 'field' # Only 'field' is allowed in multiple for now | |
1706 | actual_fields = self._get_field_setting(field, 'field') | |
1707 | ||
1708 | def wrapped_function(values): | |
1709 | values = tuple(filter(lambda x: x is not None, values)) | |
8326b00a | 1710 | return self._get_field_setting(field, 'function')(values) if values else None |
eb8a4433 | 1711 | |
1712 | value = wrapped_function((get_value(f) for f in actual_fields)) | |
1713 | else: | |
1714 | value = get_value(field) | |
1715 | return self._calculate_field_preference_from_value(format, field, type, value) | |
1716 | ||
1717 | def calculate_preference(self, format): | |
1718 | # Determine missing protocol | |
1719 | if not format.get('protocol'): | |
1720 | format['protocol'] = determine_protocol(format) | |
1721 | ||
1722 | # Determine missing ext | |
1723 | if not format.get('ext') and 'url' in format: | |
1724 | format['ext'] = determine_ext(format['url']) | |
1725 | if format.get('vcodec') == 'none': | |
8326b00a | 1726 | format['audio_ext'] = format['ext'] if format.get('acodec') != 'none' else 'none' |
eb8a4433 | 1727 | format['video_ext'] = 'none' |
1728 | else: | |
1729 | format['video_ext'] = format['ext'] | |
1730 | format['audio_ext'] = 'none' | |
1731 | # if format.get('preference') is None and format.get('ext') in ('f4f', 'f4m'): # Not supported? | |
1732 | # format['preference'] = -1000 | |
1733 | ||
1734 | # Determine missing bitrates | |
1735 | if format.get('tbr') is None: | |
1736 | if format.get('vbr') is not None and format.get('abr') is not None: | |
1737 | format['tbr'] = format.get('vbr', 0) + format.get('abr', 0) | |
1738 | else: | |
1739 | if format.get('vcodec') != "none" and format.get('vbr') is None: | |
1740 | format['vbr'] = format.get('tbr') - format.get('abr', 0) | |
1741 | if format.get('acodec') != "none" and format.get('abr') is None: | |
1742 | format['abr'] = format.get('tbr') - format.get('vbr', 0) | |
1743 | ||
1744 | return tuple(self._calculate_field_preference(format, field) for field in self._order) | |
1745 | ||
1746 | def _sort_formats(self, formats, field_preference=[]): | |
1747 | if not formats: | |
a06916d9 | 1748 | if self.get_param('ignore_no_formats_error'): |
b7da73eb | 1749 | return |
eb8a4433 | 1750 | raise ExtractorError('No video formats found') |
1751 | format_sort = self.FormatSort() # params and to_screen are taken from the downloader | |
1752 | format_sort.evaluate_params(self._downloader.params, field_preference) | |
a06916d9 | 1753 | if self.get_param('verbose', False): |
0760b0a7 | 1754 | format_sort.print_verbose_info(self._downloader.write_debug) |
eb8a4433 | 1755 | formats.sort(key=lambda f: format_sort.calculate_preference(f)) |
59040888 | 1756 | |
96a53167 S |
1757 | def _check_formats(self, formats, video_id): |
1758 | if formats: | |
1759 | formats[:] = filter( | |
1760 | lambda f: self._is_valid_url( | |
1761 | f['url'], video_id, | |
1762 | item='%s video format' % f.get('format_id') if f.get('format_id') else 'video'), | |
1763 | formats) | |
1764 | ||
f5bdb444 S |
1765 | @staticmethod |
1766 | def _remove_duplicate_formats(formats): | |
1767 | format_urls = set() | |
1768 | unique_formats = [] | |
1769 | for f in formats: | |
1770 | if f['url'] not in format_urls: | |
1771 | format_urls.add(f['url']) | |
1772 | unique_formats.append(f) | |
1773 | formats[:] = unique_formats | |
1774 | ||
45024183 | 1775 | def _is_valid_url(self, url, video_id, item='video', headers={}): |
2f0f6578 S |
1776 | url = self._proto_relative_url(url, scheme='http:') |
1777 | # For now assume non HTTP(S) URLs always valid | |
1778 | if not (url.startswith('http://') or url.startswith('https://')): | |
1779 | return True | |
96a53167 | 1780 | try: |
45024183 | 1781 | self._request_webpage(url, video_id, 'Checking %s URL' % item, headers=headers) |
96a53167 | 1782 | return True |
8bdd16b4 | 1783 | except ExtractorError as e: |
25e911a9 | 1784 | self.to_screen( |
8bdd16b4 | 1785 | '%s: %s URL is invalid, skipping: %s' |
1786 | % (video_id, item, error_to_compat_str(e.cause))) | |
25e911a9 | 1787 | return False |
96a53167 | 1788 | |
20991253 | 1789 | def http_scheme(self): |
1ede5b24 | 1790 | """ Either "http:" or "https:", depending on the user's preferences """ |
20991253 PH |
1791 | return ( |
1792 | 'http:' | |
a06916d9 | 1793 | if self.get_param('prefer_insecure', False) |
20991253 PH |
1794 | else 'https:') |
1795 | ||
57c7411f PH |
1796 | def _proto_relative_url(self, url, scheme=None): |
1797 | if url is None: | |
1798 | return url | |
1799 | if url.startswith('//'): | |
1800 | if scheme is None: | |
1801 | scheme = self.http_scheme() | |
1802 | return scheme + url | |
1803 | else: | |
1804 | return url | |
1805 | ||
4094b6e3 PH |
1806 | def _sleep(self, timeout, video_id, msg_template=None): |
1807 | if msg_template is None: | |
f1a9d64e | 1808 | msg_template = '%(video_id)s: Waiting for %(timeout)s seconds' |
4094b6e3 PH |
1809 | msg = msg_template % {'video_id': video_id, 'timeout': timeout} |
1810 | self.to_screen(msg) | |
1811 | time.sleep(timeout) | |
1812 | ||
f983b875 | 1813 | def _extract_f4m_formats(self, manifest_url, video_id, preference=None, quality=None, f4m_id=None, |
4de61310 | 1814 | transform_source=lambda s: fix_xml_ampersands(s).strip(), |
7360c06f | 1815 | fatal=True, m3u8_id=None, data=None, headers={}, query={}): |
f036a632 JMF |
1816 | manifest = self._download_xml( |
1817 | manifest_url, video_id, 'Downloading f4m manifest', | |
97f4aecf S |
1818 | 'Unable to download f4m manifest', |
1819 | # Some manifests may be malformed, e.g. prosiebensat1 generated manifests | |
067aa17e | 1820 | # (see https://github.com/ytdl-org/youtube-dl/issues/6215#issuecomment-121704244) |
4de61310 | 1821 | transform_source=transform_source, |
7360c06f | 1822 | fatal=fatal, data=data, headers=headers, query=query) |
4de61310 S |
1823 | |
1824 | if manifest is False: | |
8d29e47f | 1825 | return [] |
31bb8d3f | 1826 | |
0fdbb332 | 1827 | return self._parse_f4m_formats( |
f983b875 | 1828 | manifest, manifest_url, video_id, preference=preference, quality=quality, f4m_id=f4m_id, |
448bb5f3 | 1829 | transform_source=transform_source, fatal=fatal, m3u8_id=m3u8_id) |
0fdbb332 | 1830 | |
f983b875 | 1831 | def _parse_f4m_formats(self, manifest, manifest_url, video_id, preference=None, quality=None, f4m_id=None, |
0fdbb332 | 1832 | transform_source=lambda s: fix_xml_ampersands(s).strip(), |
448bb5f3 | 1833 | fatal=True, m3u8_id=None): |
ee0ba927 | 1834 | if not isinstance(manifest, compat_etree_Element) and not fatal: |
d9eb580a S |
1835 | return [] |
1836 | ||
7a5c1cfe | 1837 | # currently yt-dlp cannot decode the playerVerificationChallenge as Akamai uses Adobe Alchemy |
fb72ec58 | 1838 | akamai_pv = manifest.find('{http://ns.adobe.com/f4m/1.0}pv-2.0') |
1839 | if akamai_pv is not None and ';' in akamai_pv.text: | |
1840 | playerVerificationChallenge = akamai_pv.text.split(';')[0] | |
1841 | if playerVerificationChallenge.strip() != '': | |
1842 | return [] | |
1843 | ||
31bb8d3f | 1844 | formats = [] |
7a47d07c | 1845 | manifest_version = '1.0' |
b2527359 | 1846 | media_nodes = manifest.findall('{http://ns.adobe.com/f4m/1.0}media') |
34e48bed | 1847 | if not media_nodes: |
7a47d07c | 1848 | manifest_version = '2.0' |
34e48bed | 1849 | media_nodes = manifest.findall('{http://ns.adobe.com/f4m/2.0}media') |
b22ca762 | 1850 | # Remove unsupported DRM protected media from final formats |
067aa17e | 1851 | # rendition (see https://github.com/ytdl-org/youtube-dl/issues/8573). |
b22ca762 S |
1852 | media_nodes = remove_encrypted_media(media_nodes) |
1853 | if not media_nodes: | |
1854 | return formats | |
48107c19 S |
1855 | |
1856 | manifest_base_url = get_base_url(manifest) | |
0a5685b2 | 1857 | |
a6571f10 | 1858 | bootstrap_info = xpath_element( |
0a5685b2 YCH |
1859 | manifest, ['{http://ns.adobe.com/f4m/1.0}bootstrapInfo', '{http://ns.adobe.com/f4m/2.0}bootstrapInfo'], |
1860 | 'bootstrap info', default=None) | |
1861 | ||
edd6074c RA |
1862 | vcodec = None |
1863 | mime_type = xpath_text( | |
1864 | manifest, ['{http://ns.adobe.com/f4m/1.0}mimeType', '{http://ns.adobe.com/f4m/2.0}mimeType'], | |
1865 | 'base URL', default=None) | |
1866 | if mime_type and mime_type.startswith('audio/'): | |
1867 | vcodec = 'none' | |
1868 | ||
b2527359 | 1869 | for i, media_el in enumerate(media_nodes): |
77b8b4e6 S |
1870 | tbr = int_or_none(media_el.attrib.get('bitrate')) |
1871 | width = int_or_none(media_el.attrib.get('width')) | |
1872 | height = int_or_none(media_el.attrib.get('height')) | |
1873 | format_id = '-'.join(filter(None, [f4m_id, compat_str(i if tbr is None else tbr)])) | |
448bb5f3 YCH |
1874 | # If <bootstrapInfo> is present, the specified f4m is a |
1875 | # stream-level manifest, and only set-level manifests may refer to | |
1876 | # external resources. See section 11.4 and section 4 of F4M spec | |
1877 | if bootstrap_info is None: | |
1878 | media_url = None | |
1879 | # @href is introduced in 2.0, see section 11.6 of F4M spec | |
1880 | if manifest_version == '2.0': | |
1881 | media_url = media_el.attrib.get('href') | |
1882 | if media_url is None: | |
1883 | media_url = media_el.attrib.get('url') | |
31c746e5 S |
1884 | if not media_url: |
1885 | continue | |
cc357c4d S |
1886 | manifest_url = ( |
1887 | media_url if media_url.startswith('http://') or media_url.startswith('https://') | |
48107c19 | 1888 | else ((manifest_base_url or '/'.join(manifest_url.split('/')[:-1])) + '/' + media_url)) |
70f0f5a8 S |
1889 | # If media_url is itself a f4m manifest do the recursive extraction |
1890 | # since bitrates in parent manifest (this one) and media_url manifest | |
1891 | # may differ leading to inability to resolve the format by requested | |
1892 | # bitrate in f4m downloader | |
240b6045 YCH |
1893 | ext = determine_ext(manifest_url) |
1894 | if ext == 'f4m': | |
77b8b4e6 | 1895 | f4m_formats = self._extract_f4m_formats( |
f983b875 | 1896 | manifest_url, video_id, preference=preference, quality=quality, f4m_id=f4m_id, |
77b8b4e6 S |
1897 | transform_source=transform_source, fatal=fatal) |
1898 | # Sometimes stream-level manifest contains single media entry that | |
1899 | # does not contain any quality metadata (e.g. http://matchtv.ru/#live-player). | |
1900 | # At the same time parent's media entry in set-level manifest may | |
1901 | # contain it. We will copy it from parent in such cases. | |
1902 | if len(f4m_formats) == 1: | |
1903 | f = f4m_formats[0] | |
1904 | f.update({ | |
1905 | 'tbr': f.get('tbr') or tbr, | |
1906 | 'width': f.get('width') or width, | |
1907 | 'height': f.get('height') or height, | |
1908 | 'format_id': f.get('format_id') if not tbr else format_id, | |
edd6074c | 1909 | 'vcodec': vcodec, |
77b8b4e6 S |
1910 | }) |
1911 | formats.extend(f4m_formats) | |
70f0f5a8 | 1912 | continue |
240b6045 YCH |
1913 | elif ext == 'm3u8': |
1914 | formats.extend(self._extract_m3u8_formats( | |
1915 | manifest_url, video_id, 'mp4', preference=preference, | |
f983b875 | 1916 | quality=quality, m3u8_id=m3u8_id, fatal=fatal)) |
240b6045 | 1917 | continue |
31bb8d3f | 1918 | formats.append({ |
77b8b4e6 | 1919 | 'format_id': format_id, |
31bb8d3f | 1920 | 'url': manifest_url, |
30d0b549 | 1921 | 'manifest_url': manifest_url, |
a6571f10 | 1922 | 'ext': 'flv' if bootstrap_info is not None else None, |
187ee66c | 1923 | 'protocol': 'f4m', |
b2527359 | 1924 | 'tbr': tbr, |
77b8b4e6 S |
1925 | 'width': width, |
1926 | 'height': height, | |
edd6074c | 1927 | 'vcodec': vcodec, |
60ca389c | 1928 | 'preference': preference, |
f983b875 | 1929 | 'quality': quality, |
31bb8d3f | 1930 | }) |
31bb8d3f JMF |
1931 | return formats |
1932 | ||
f983b875 | 1933 | def _m3u8_meta_format(self, m3u8_url, ext=None, preference=None, quality=None, m3u8_id=None): |
16da9bbc | 1934 | return { |
f207019c | 1935 | 'format_id': '-'.join(filter(None, [m3u8_id, 'meta'])), |
704df56d PH |
1936 | 'url': m3u8_url, |
1937 | 'ext': ext, | |
1938 | 'protocol': 'm3u8', | |
37768f92 | 1939 | 'preference': preference - 100 if preference else -100, |
f983b875 | 1940 | 'quality': quality, |
704df56d PH |
1941 | 'resolution': 'multiple', |
1942 | 'format_note': 'Quality selection URL', | |
16da9bbc YCH |
1943 | } |
1944 | ||
a0c3b2d5 F |
1945 | def _extract_m3u8_formats(self, *args, **kwargs): |
1946 | fmts, subs = self._extract_m3u8_formats_and_subtitles(*args, **kwargs) | |
1947 | if subs: | |
1948 | self.report_warning(bug_reports_message( | |
1949 | "Ignoring subtitle tracks found in the HLS manifest; " | |
1950 | "if any subtitle tracks are missing," | |
1951 | )) | |
1952 | return fmts | |
1953 | ||
1954 | def _extract_m3u8_formats_and_subtitles( | |
177877c5 | 1955 | self, m3u8_url, video_id, ext=None, entry_protocol='m3u8_native', |
a0c3b2d5 F |
1956 | preference=None, quality=None, m3u8_id=None, note=None, |
1957 | errnote=None, fatal=True, live=False, data=None, headers={}, | |
1958 | query={}): | |
1959 | ||
dbd82a1d | 1960 | res = self._download_webpage_handle( |
81515ad9 | 1961 | m3u8_url, video_id, |
37a3bb66 | 1962 | note='Downloading m3u8 information' if note is None else note, |
1963 | errnote='Failed to download m3u8 information' if errnote is None else errnote, | |
7360c06f | 1964 | fatal=fatal, data=data, headers=headers, query=query) |
cb252080 | 1965 | |
dbd82a1d | 1966 | if res is False: |
a0c3b2d5 | 1967 | return [], {} |
cb252080 | 1968 | |
dbd82a1d | 1969 | m3u8_doc, urlh = res |
37113045 | 1970 | m3u8_url = urlh.geturl() |
9cdffeeb | 1971 | |
a0c3b2d5 | 1972 | return self._parse_m3u8_formats_and_subtitles( |
cb252080 | 1973 | m3u8_doc, m3u8_url, ext=ext, entry_protocol=entry_protocol, |
310c2ed2 | 1974 | preference=preference, quality=quality, m3u8_id=m3u8_id, |
1975 | note=note, errnote=errnote, fatal=fatal, live=live, data=data, | |
1976 | headers=headers, query=query, video_id=video_id) | |
cb252080 | 1977 | |
a0c3b2d5 | 1978 | def _parse_m3u8_formats_and_subtitles( |
177877c5 | 1979 | self, m3u8_doc, m3u8_url, ext=None, entry_protocol='m3u8_native', |
a0c3b2d5 F |
1980 | preference=None, quality=None, m3u8_id=None, live=False, note=None, |
1981 | errnote=None, fatal=True, data=None, headers={}, query={}, | |
1982 | video_id=None): | |
60755938 | 1983 | formats, subtitles = [], {} |
a0c3b2d5 | 1984 | |
08a00eef | 1985 | if '#EXT-X-FAXS-CM:' in m3u8_doc: # Adobe Flash Access |
60755938 | 1986 | return formats, subtitles |
08a00eef | 1987 | |
a06916d9 | 1988 | if (not self.get_param('allow_unplayable_formats') |
73d4343e | 1989 | and re.search(r'#EXT-X-SESSION-KEY:.*?URI="skd://', m3u8_doc)): # Apple FairPlay |
60755938 | 1990 | return formats, subtitles |
a0c3b2d5 | 1991 | |
60755938 | 1992 | def format_url(url): |
1993 | return url if re.match(r'^https?://', url) else compat_urlparse.urljoin(m3u8_url, url) | |
1994 | ||
1995 | if self.get_param('hls_split_discontinuity', False): | |
1996 | def _extract_m3u8_playlist_indices(manifest_url=None, m3u8_doc=None): | |
1997 | if not m3u8_doc: | |
1998 | if not manifest_url: | |
1999 | return [] | |
2000 | m3u8_doc = self._download_webpage( | |
2001 | manifest_url, video_id, fatal=fatal, data=data, headers=headers, | |
2002 | note=False, errnote='Failed to download m3u8 playlist information') | |
2003 | if m3u8_doc is False: | |
2004 | return [] | |
2005 | return range(1 + sum(line.startswith('#EXT-X-DISCONTINUITY') for line in m3u8_doc.splitlines())) | |
0def7587 | 2006 | |
60755938 | 2007 | else: |
2008 | def _extract_m3u8_playlist_indices(*args, **kwargs): | |
2009 | return [None] | |
310c2ed2 | 2010 | |
cb252080 S |
2011 | # References: |
2012 | # 1. https://tools.ietf.org/html/draft-pantos-http-live-streaming-21 | |
067aa17e S |
2013 | # 2. https://github.com/ytdl-org/youtube-dl/issues/12211 |
2014 | # 3. https://github.com/ytdl-org/youtube-dl/issues/18923 | |
cb252080 S |
2015 | |
2016 | # We should try extracting formats only from master playlists [1, 4.3.4], | |
2017 | # i.e. playlists that describe available qualities. On the other hand | |
2018 | # media playlists [1, 4.3.3] should be returned as is since they contain | |
2019 | # just the media without qualities renditions. | |
9cdffeeb | 2020 | # Fortunately, master playlist can be easily distinguished from media |
cb252080 | 2021 | # playlist based on particular tags availability. As of [1, 4.3.3, 4.3.4] |
a0566bbf | 2022 | # master playlist tags MUST NOT appear in a media playlist and vice versa. |
cb252080 S |
2023 | # As of [1, 4.3.3.1] #EXT-X-TARGETDURATION tag is REQUIRED for every |
2024 | # media playlist and MUST NOT appear in master playlist thus we can | |
2025 | # clearly detect media playlist with this criterion. | |
2026 | ||
9cdffeeb | 2027 | if '#EXT-X-TARGETDURATION' in m3u8_doc: # media playlist, return as is |
60755938 | 2028 | formats = [{ |
2029 | 'format_id': '-'.join(map(str, filter(None, [m3u8_id, idx]))), | |
2030 | 'format_index': idx, | |
2031 | 'url': m3u8_url, | |
2032 | 'ext': ext, | |
2033 | 'protocol': entry_protocol, | |
2034 | 'preference': preference, | |
2035 | 'quality': quality, | |
2036 | } for idx in _extract_m3u8_playlist_indices(m3u8_doc=m3u8_doc)] | |
310c2ed2 | 2037 | |
a0c3b2d5 | 2038 | return formats, subtitles |
cb252080 S |
2039 | |
2040 | groups = {} | |
2041 | last_stream_inf = {} | |
2042 | ||
2043 | def extract_media(x_media_line): | |
2044 | media = parse_m3u8_attributes(x_media_line) | |
2045 | # As per [1, 4.3.4.1] TYPE, GROUP-ID and NAME are REQUIRED | |
2046 | media_type, group_id, name = media.get('TYPE'), media.get('GROUP-ID'), media.get('NAME') | |
2047 | if not (media_type and group_id and name): | |
2048 | return | |
2049 | groups.setdefault(group_id, []).append(media) | |
a0c3b2d5 F |
2050 | # <https://tools.ietf.org/html/rfc8216#section-4.3.4.1> |
2051 | if media_type == 'SUBTITLES': | |
3907333c | 2052 | # According to RFC 8216 §4.3.4.2.1, URI is REQUIRED in the |
2053 | # EXT-X-MEDIA tag if the media type is SUBTITLES. | |
2054 | # However, lack of URI has been spotted in the wild. | |
2055 | # e.g. NebulaIE; see https://github.com/yt-dlp/yt-dlp/issues/339 | |
2056 | if not media.get('URI'): | |
2057 | return | |
a0c3b2d5 F |
2058 | url = format_url(media['URI']) |
2059 | sub_info = { | |
2060 | 'url': url, | |
2061 | 'ext': determine_ext(url), | |
2062 | } | |
4a2f19ab F |
2063 | if sub_info['ext'] == 'm3u8': |
2064 | # Per RFC 8216 §3.1, the only possible subtitle format m3u8 | |
2065 | # files may contain is WebVTT: | |
2066 | # <https://tools.ietf.org/html/rfc8216#section-3.1> | |
2067 | sub_info['ext'] = 'vtt' | |
2068 | sub_info['protocol'] = 'm3u8_native' | |
37a3bb66 | 2069 | lang = media.get('LANGUAGE') or 'und' |
a0c3b2d5 | 2070 | subtitles.setdefault(lang, []).append(sub_info) |
cb252080 S |
2071 | if media_type not in ('VIDEO', 'AUDIO'): |
2072 | return | |
2073 | media_url = media.get('URI') | |
2074 | if media_url: | |
310c2ed2 | 2075 | manifest_url = format_url(media_url) |
60755938 | 2076 | formats.extend({ |
2077 | 'format_id': '-'.join(map(str, filter(None, (m3u8_id, group_id, name, idx)))), | |
2078 | 'format_note': name, | |
2079 | 'format_index': idx, | |
2080 | 'url': manifest_url, | |
2081 | 'manifest_url': m3u8_url, | |
2082 | 'language': media.get('LANGUAGE'), | |
2083 | 'ext': ext, | |
2084 | 'protocol': entry_protocol, | |
2085 | 'preference': preference, | |
2086 | 'quality': quality, | |
2087 | 'vcodec': 'none' if media_type == 'AUDIO' else None, | |
2088 | } for idx in _extract_m3u8_playlist_indices(manifest_url)) | |
cb252080 S |
2089 | |
2090 | def build_stream_name(): | |
2091 | # Despite specification does not mention NAME attribute for | |
3019cb0c S |
2092 | # EXT-X-STREAM-INF tag it still sometimes may be present (see [1] |
2093 | # or vidio test in TestInfoExtractor.test_parse_m3u8_formats) | |
ddd258f9 | 2094 | # 1. http://www.vidio.com/watch/165683-dj_ambred-booyah-live-2015 |
cb252080 S |
2095 | stream_name = last_stream_inf.get('NAME') |
2096 | if stream_name: | |
2097 | return stream_name | |
2098 | # If there is no NAME in EXT-X-STREAM-INF it will be obtained | |
2099 | # from corresponding rendition group | |
2100 | stream_group_id = last_stream_inf.get('VIDEO') | |
2101 | if not stream_group_id: | |
2102 | return | |
2103 | stream_group = groups.get(stream_group_id) | |
2104 | if not stream_group: | |
2105 | return stream_group_id | |
2106 | rendition = stream_group[0] | |
2107 | return rendition.get('NAME') or stream_group_id | |
2108 | ||
379306ef | 2109 | # parse EXT-X-MEDIA tags before EXT-X-STREAM-INF in order to have the |
2bfc1d9d RA |
2110 | # chance to detect video only formats when EXT-X-STREAM-INF tags |
2111 | # precede EXT-X-MEDIA tags in HLS manifest such as [3]. | |
2112 | for line in m3u8_doc.splitlines(): | |
2113 | if line.startswith('#EXT-X-MEDIA:'): | |
2114 | extract_media(line) | |
2115 | ||
704df56d PH |
2116 | for line in m3u8_doc.splitlines(): |
2117 | if line.startswith('#EXT-X-STREAM-INF:'): | |
cb252080 | 2118 | last_stream_inf = parse_m3u8_attributes(line) |
704df56d PH |
2119 | elif line.startswith('#') or not line.strip(): |
2120 | continue | |
2121 | else: | |
9c99bef7 | 2122 | tbr = float_or_none( |
3089bc74 S |
2123 | last_stream_inf.get('AVERAGE-BANDWIDTH') |
2124 | or last_stream_inf.get('BANDWIDTH'), scale=1000) | |
30d0b549 | 2125 | manifest_url = format_url(line.strip()) |
5ef62fc4 | 2126 | |
60755938 | 2127 | for idx in _extract_m3u8_playlist_indices(manifest_url): |
2128 | format_id = [m3u8_id, None, idx] | |
310c2ed2 | 2129 | # Bandwidth of live streams may differ over time thus making |
2130 | # format_id unpredictable. So it's better to keep provided | |
2131 | # format_id intact. | |
2132 | if not live: | |
60755938 | 2133 | stream_name = build_stream_name() |
2134 | format_id[1] = stream_name if stream_name else '%d' % (tbr if tbr else len(formats)) | |
310c2ed2 | 2135 | f = { |
60755938 | 2136 | 'format_id': '-'.join(map(str, filter(None, format_id))), |
2137 | 'format_index': idx, | |
310c2ed2 | 2138 | 'url': manifest_url, |
2139 | 'manifest_url': m3u8_url, | |
2140 | 'tbr': tbr, | |
2141 | 'ext': ext, | |
2142 | 'fps': float_or_none(last_stream_inf.get('FRAME-RATE')), | |
2143 | 'protocol': entry_protocol, | |
2144 | 'preference': preference, | |
2145 | 'quality': quality, | |
2146 | } | |
2147 | resolution = last_stream_inf.get('RESOLUTION') | |
2148 | if resolution: | |
2149 | mobj = re.search(r'(?P<width>\d+)[xX](?P<height>\d+)', resolution) | |
2150 | if mobj: | |
2151 | f['width'] = int(mobj.group('width')) | |
2152 | f['height'] = int(mobj.group('height')) | |
2153 | # Unified Streaming Platform | |
2154 | mobj = re.search( | |
2155 | r'audio.*?(?:%3D|=)(\d+)(?:-video.*?(?:%3D|=)(\d+))?', f['url']) | |
2156 | if mobj: | |
2157 | abr, vbr = mobj.groups() | |
2158 | abr, vbr = float_or_none(abr, 1000), float_or_none(vbr, 1000) | |
2159 | f.update({ | |
2160 | 'vbr': vbr, | |
2161 | 'abr': abr, | |
2162 | }) | |
2163 | codecs = parse_codecs(last_stream_inf.get('CODECS')) | |
2164 | f.update(codecs) | |
2165 | audio_group_id = last_stream_inf.get('AUDIO') | |
2166 | # As per [1, 4.3.4.1.1] any EXT-X-STREAM-INF tag which | |
2167 | # references a rendition group MUST have a CODECS attribute. | |
2168 | # However, this is not always respected, for example, [2] | |
2169 | # contains EXT-X-STREAM-INF tag which references AUDIO | |
2170 | # rendition group but does not have CODECS and despite | |
2171 | # referencing an audio group it represents a complete | |
2172 | # (with audio and video) format. So, for such cases we will | |
2173 | # ignore references to rendition groups and treat them | |
2174 | # as complete formats. | |
2175 | if audio_group_id and codecs and f.get('vcodec') != 'none': | |
2176 | audio_group = groups.get(audio_group_id) | |
2177 | if audio_group and audio_group[0].get('URI'): | |
2178 | # TODO: update acodec for audio only formats with | |
2179 | # the same GROUP-ID | |
2180 | f['acodec'] = 'none' | |
fc21af50 | 2181 | if not f.get('ext'): |
2182 | f['ext'] = 'm4a' if f.get('vcodec') == 'none' else 'mp4' | |
310c2ed2 | 2183 | formats.append(f) |
2184 | ||
2185 | # for DailyMotion | |
2186 | progressive_uri = last_stream_inf.get('PROGRESSIVE-URI') | |
2187 | if progressive_uri: | |
2188 | http_f = f.copy() | |
2189 | del http_f['manifest_url'] | |
2190 | http_f.update({ | |
2191 | 'format_id': f['format_id'].replace('hls-', 'http-'), | |
2192 | 'protocol': 'http', | |
2193 | 'url': progressive_uri, | |
2194 | }) | |
2195 | formats.append(http_f) | |
5ef62fc4 | 2196 | |
cb252080 | 2197 | last_stream_inf = {} |
a0c3b2d5 | 2198 | return formats, subtitles |
704df56d | 2199 | |
a107193e S |
2200 | @staticmethod |
2201 | def _xpath_ns(path, namespace=None): | |
2202 | if not namespace: | |
2203 | return path | |
2204 | out = [] | |
2205 | for c in path.split('/'): | |
2206 | if not c or c == '.': | |
2207 | out.append(c) | |
2208 | else: | |
2209 | out.append('{%s}%s' % (namespace, c)) | |
2210 | return '/'.join(out) | |
2211 | ||
da1c94ee | 2212 | def _extract_smil_formats_and_subtitles(self, smil_url, video_id, fatal=True, f4m_params=None, transform_source=None): |
09f572fb | 2213 | smil = self._download_smil(smil_url, video_id, fatal=fatal, transform_source=transform_source) |
a107193e | 2214 | |
995029a1 PH |
2215 | if smil is False: |
2216 | assert not fatal | |
2217 | return [] | |
e89a2aab | 2218 | |
17712eeb | 2219 | namespace = self._parse_smil_namespace(smil) |
a107193e | 2220 | |
da1c94ee | 2221 | fmts = self._parse_smil_formats( |
a107193e | 2222 | smil, smil_url, video_id, namespace=namespace, f4m_params=f4m_params) |
da1c94ee F |
2223 | subs = self._parse_smil_subtitles( |
2224 | smil, namespace=namespace) | |
2225 | ||
2226 | return fmts, subs | |
2227 | ||
2228 | def _extract_smil_formats(self, *args, **kwargs): | |
2229 | fmts, subs = self._extract_smil_formats_and_subtitles(*args, **kwargs) | |
2230 | if subs: | |
2231 | self.report_warning(bug_reports_message( | |
2232 | "Ignoring subtitle tracks found in the SMIL manifest; " | |
2233 | "if any subtitle tracks are missing," | |
2234 | )) | |
2235 | return fmts | |
a107193e S |
2236 | |
2237 | def _extract_smil_info(self, smil_url, video_id, fatal=True, f4m_params=None): | |
2238 | smil = self._download_smil(smil_url, video_id, fatal=fatal) | |
2239 | if smil is False: | |
2240 | return {} | |
2241 | return self._parse_smil(smil, smil_url, video_id, f4m_params=f4m_params) | |
2242 | ||
09f572fb | 2243 | def _download_smil(self, smil_url, video_id, fatal=True, transform_source=None): |
a107193e S |
2244 | return self._download_xml( |
2245 | smil_url, video_id, 'Downloading SMIL file', | |
09f572fb | 2246 | 'Unable to download SMIL file', fatal=fatal, transform_source=transform_source) |
a107193e S |
2247 | |
2248 | def _parse_smil(self, smil, smil_url, video_id, f4m_params=None): | |
17712eeb | 2249 | namespace = self._parse_smil_namespace(smil) |
a107193e S |
2250 | |
2251 | formats = self._parse_smil_formats( | |
2252 | smil, smil_url, video_id, namespace=namespace, f4m_params=f4m_params) | |
2253 | subtitles = self._parse_smil_subtitles(smil, namespace=namespace) | |
2254 | ||
2255 | video_id = os.path.splitext(url_basename(smil_url))[0] | |
2256 | title = None | |
2257 | description = None | |
647eab45 | 2258 | upload_date = None |
a107193e S |
2259 | for meta in smil.findall(self._xpath_ns('./head/meta', namespace)): |
2260 | name = meta.attrib.get('name') | |
2261 | content = meta.attrib.get('content') | |
2262 | if not name or not content: | |
2263 | continue | |
2264 | if not title and name == 'title': | |
2265 | title = content | |
2266 | elif not description and name in ('description', 'abstract'): | |
2267 | description = content | |
647eab45 S |
2268 | elif not upload_date and name == 'date': |
2269 | upload_date = unified_strdate(content) | |
a107193e | 2270 | |
1e5bcdec S |
2271 | thumbnails = [{ |
2272 | 'id': image.get('type'), | |
2273 | 'url': image.get('src'), | |
2274 | 'width': int_or_none(image.get('width')), | |
2275 | 'height': int_or_none(image.get('height')), | |
2276 | } for image in smil.findall(self._xpath_ns('.//image', namespace)) if image.get('src')] | |
2277 | ||
a107193e S |
2278 | return { |
2279 | 'id': video_id, | |
2280 | 'title': title or video_id, | |
2281 | 'description': description, | |
647eab45 | 2282 | 'upload_date': upload_date, |
1e5bcdec | 2283 | 'thumbnails': thumbnails, |
a107193e S |
2284 | 'formats': formats, |
2285 | 'subtitles': subtitles, | |
2286 | } | |
2287 | ||
17712eeb S |
2288 | def _parse_smil_namespace(self, smil): |
2289 | return self._search_regex( | |
2290 | r'(?i)^{([^}]+)?}smil$', smil.tag, 'namespace', default=None) | |
2291 | ||
f877c6ae | 2292 | def _parse_smil_formats(self, smil, smil_url, video_id, namespace=None, f4m_params=None, transform_rtmp_url=None): |
a107193e S |
2293 | base = smil_url |
2294 | for meta in smil.findall(self._xpath_ns('./head/meta', namespace)): | |
2295 | b = meta.get('base') or meta.get('httpBase') | |
2296 | if b: | |
2297 | base = b | |
2298 | break | |
e89a2aab S |
2299 | |
2300 | formats = [] | |
2301 | rtmp_count = 0 | |
a107193e | 2302 | http_count = 0 |
7f32e5dc | 2303 | m3u8_count = 0 |
a107193e | 2304 | |
81e1c4e2 | 2305 | srcs = [] |
ad96b4c8 YCH |
2306 | media = smil.findall(self._xpath_ns('.//video', namespace)) + smil.findall(self._xpath_ns('.//audio', namespace)) |
2307 | for medium in media: | |
2308 | src = medium.get('src') | |
81e1c4e2 | 2309 | if not src or src in srcs: |
a107193e | 2310 | continue |
81e1c4e2 | 2311 | srcs.append(src) |
a107193e | 2312 | |
ad96b4c8 YCH |
2313 | bitrate = float_or_none(medium.get('system-bitrate') or medium.get('systemBitrate'), 1000) |
2314 | filesize = int_or_none(medium.get('size') or medium.get('fileSize')) | |
2315 | width = int_or_none(medium.get('width')) | |
2316 | height = int_or_none(medium.get('height')) | |
2317 | proto = medium.get('proto') | |
2318 | ext = medium.get('ext') | |
a107193e | 2319 | src_ext = determine_ext(src) |
ad96b4c8 | 2320 | streamer = medium.get('streamer') or base |
a107193e S |
2321 | |
2322 | if proto == 'rtmp' or streamer.startswith('rtmp'): | |
2323 | rtmp_count += 1 | |
2324 | formats.append({ | |
2325 | 'url': streamer, | |
2326 | 'play_path': src, | |
2327 | 'ext': 'flv', | |
2328 | 'format_id': 'rtmp-%d' % (rtmp_count if bitrate is None else bitrate), | |
2329 | 'tbr': bitrate, | |
2330 | 'filesize': filesize, | |
2331 | 'width': width, | |
2332 | 'height': height, | |
2333 | }) | |
f877c6ae YCH |
2334 | if transform_rtmp_url: |
2335 | streamer, src = transform_rtmp_url(streamer, src) | |
2336 | formats[-1].update({ | |
2337 | 'url': streamer, | |
2338 | 'play_path': src, | |
2339 | }) | |
a107193e S |
2340 | continue |
2341 | ||
2342 | src_url = src if src.startswith('http') else compat_urlparse.urljoin(base, src) | |
c349456e | 2343 | src_url = src_url.strip() |
a107193e S |
2344 | |
2345 | if proto == 'm3u8' or src_ext == 'm3u8': | |
7f32e5dc | 2346 | m3u8_formats = self._extract_m3u8_formats( |
2347 | src_url, video_id, ext or 'mp4', m3u8_id='hls', fatal=False) | |
2348 | if len(m3u8_formats) == 1: | |
2349 | m3u8_count += 1 | |
2350 | m3u8_formats[0].update({ | |
2351 | 'format_id': 'hls-%d' % (m3u8_count if bitrate is None else bitrate), | |
2352 | 'tbr': bitrate, | |
2353 | 'width': width, | |
2354 | 'height': height, | |
2355 | }) | |
2356 | formats.extend(m3u8_formats) | |
bd21ead2 | 2357 | elif src_ext == 'f4m': |
a107193e S |
2358 | f4m_url = src_url |
2359 | if not f4m_params: | |
2360 | f4m_params = { | |
2361 | 'hdcore': '3.2.0', | |
2362 | 'plugin': 'flowplayer-3.2.0.1', | |
2363 | } | |
2364 | f4m_url += '&' if '?' in f4m_url else '?' | |
15707c7e | 2365 | f4m_url += compat_urllib_parse_urlencode(f4m_params) |
7e5edcfd | 2366 | formats.extend(self._extract_f4m_formats(f4m_url, video_id, f4m_id='hds', fatal=False)) |
bd21ead2 RA |
2367 | elif src_ext == 'mpd': |
2368 | formats.extend(self._extract_mpd_formats( | |
2369 | src_url, video_id, mpd_id='dash', fatal=False)) | |
2370 | elif re.search(r'\.ism/[Mm]anifest', src_url): | |
2371 | formats.extend(self._extract_ism_formats( | |
2372 | src_url, video_id, ism_id='mss', fatal=False)) | |
2373 | elif src_url.startswith('http') and self._is_valid_url(src, video_id): | |
a107193e S |
2374 | http_count += 1 |
2375 | formats.append({ | |
2376 | 'url': src_url, | |
2377 | 'ext': ext or src_ext or 'flv', | |
2378 | 'format_id': 'http-%d' % (bitrate or http_count), | |
2379 | 'tbr': bitrate, | |
2380 | 'filesize': filesize, | |
2381 | 'width': width, | |
2382 | 'height': height, | |
2383 | }) | |
63757032 | 2384 | |
e89a2aab S |
2385 | return formats |
2386 | ||
ce00af87 | 2387 | def _parse_smil_subtitles(self, smil, namespace=None, subtitles_lang='en'): |
d413095f | 2388 | urls = [] |
a107193e S |
2389 | subtitles = {} |
2390 | for num, textstream in enumerate(smil.findall(self._xpath_ns('.//textstream', namespace))): | |
2391 | src = textstream.get('src') | |
d413095f | 2392 | if not src or src in urls: |
a107193e | 2393 | continue |
d413095f | 2394 | urls.append(src) |
df634be2 | 2395 | ext = textstream.get('ext') or mimetype2ext(textstream.get('type')) or determine_ext(src) |
03bc7237 | 2396 | lang = textstream.get('systemLanguage') or textstream.get('systemLanguageName') or textstream.get('lang') or subtitles_lang |
a107193e S |
2397 | subtitles.setdefault(lang, []).append({ |
2398 | 'url': src, | |
2399 | 'ext': ext, | |
2400 | }) | |
2401 | return subtitles | |
63757032 | 2402 | |
47a5cb77 | 2403 | def _extract_xspf_playlist(self, xspf_url, playlist_id, fatal=True): |
942acef5 | 2404 | xspf = self._download_xml( |
47a5cb77 | 2405 | xspf_url, playlist_id, 'Downloading xpsf playlist', |
942acef5 S |
2406 | 'Unable to download xspf manifest', fatal=fatal) |
2407 | if xspf is False: | |
2408 | return [] | |
47a5cb77 S |
2409 | return self._parse_xspf( |
2410 | xspf, playlist_id, xspf_url=xspf_url, | |
2411 | xspf_base_url=base_url(xspf_url)) | |
8d6765cf | 2412 | |
47a5cb77 | 2413 | def _parse_xspf(self, xspf_doc, playlist_id, xspf_url=None, xspf_base_url=None): |
8d6765cf S |
2414 | NS_MAP = { |
2415 | 'xspf': 'http://xspf.org/ns/0/', | |
2416 | 's1': 'http://static.streamone.nl/player/ns/0', | |
2417 | } | |
2418 | ||
2419 | entries = [] | |
47a5cb77 | 2420 | for track in xspf_doc.findall(xpath_with_ns('./xspf:trackList/xspf:track', NS_MAP)): |
8d6765cf | 2421 | title = xpath_text( |
98044462 | 2422 | track, xpath_with_ns('./xspf:title', NS_MAP), 'title', default=playlist_id) |
8d6765cf S |
2423 | description = xpath_text( |
2424 | track, xpath_with_ns('./xspf:annotation', NS_MAP), 'description') | |
2425 | thumbnail = xpath_text( | |
2426 | track, xpath_with_ns('./xspf:image', NS_MAP), 'thumbnail') | |
2427 | duration = float_or_none( | |
2428 | xpath_text(track, xpath_with_ns('./xspf:duration', NS_MAP), 'duration'), 1000) | |
2429 | ||
47a5cb77 S |
2430 | formats = [] |
2431 | for location in track.findall(xpath_with_ns('./xspf:location', NS_MAP)): | |
2432 | format_url = urljoin(xspf_base_url, location.text) | |
2433 | if not format_url: | |
2434 | continue | |
2435 | formats.append({ | |
2436 | 'url': format_url, | |
2437 | 'manifest_url': xspf_url, | |
2438 | 'format_id': location.get(xpath_with_ns('s1:label', NS_MAP)), | |
2439 | 'width': int_or_none(location.get(xpath_with_ns('s1:width', NS_MAP))), | |
2440 | 'height': int_or_none(location.get(xpath_with_ns('s1:height', NS_MAP))), | |
2441 | }) | |
8d6765cf S |
2442 | self._sort_formats(formats) |
2443 | ||
2444 | entries.append({ | |
2445 | 'id': playlist_id, | |
2446 | 'title': title, | |
2447 | 'description': description, | |
2448 | 'thumbnail': thumbnail, | |
2449 | 'duration': duration, | |
2450 | 'formats': formats, | |
2451 | }) | |
2452 | return entries | |
2453 | ||
171e59ed F |
2454 | def _extract_mpd_formats(self, *args, **kwargs): |
2455 | fmts, subs = self._extract_mpd_formats_and_subtitles(*args, **kwargs) | |
2456 | if subs: | |
2457 | self.report_warning(bug_reports_message( | |
2458 | "Ignoring subtitle tracks found in the DASH manifest; " | |
2459 | "if any subtitle tracks are missing," | |
2460 | )) | |
2461 | return fmts | |
2462 | ||
2463 | def _extract_mpd_formats_and_subtitles( | |
2464 | self, mpd_url, video_id, mpd_id=None, note=None, errnote=None, | |
2465 | fatal=True, data=None, headers={}, query={}): | |
47a5cb77 | 2466 | res = self._download_xml_handle( |
1bac3455 | 2467 | mpd_url, video_id, |
37a3bb66 | 2468 | note='Downloading MPD manifest' if note is None else note, |
2469 | errnote='Failed to download MPD manifest' if errnote is None else errnote, | |
7360c06f | 2470 | fatal=fatal, data=data, headers=headers, query=query) |
1bac3455 | 2471 | if res is False: |
171e59ed | 2472 | return [], {} |
47a5cb77 | 2473 | mpd_doc, urlh = res |
c25720ef | 2474 | if mpd_doc is None: |
171e59ed | 2475 | return [], {} |
02dc0a36 | 2476 | mpd_base_url = base_url(urlh.geturl()) |
1bac3455 | 2477 | |
171e59ed | 2478 | return self._parse_mpd_formats_and_subtitles( |
545cc85d | 2479 | mpd_doc, mpd_id, mpd_base_url, mpd_url) |
2d2fa82d | 2480 | |
171e59ed F |
2481 | def _parse_mpd_formats(self, *args, **kwargs): |
2482 | fmts, subs = self._parse_mpd_formats_and_subtitles(*args, **kwargs) | |
2483 | if subs: | |
2484 | self.report_warning(bug_reports_message( | |
2485 | "Ignoring subtitle tracks found in the DASH manifest; " | |
2486 | "if any subtitle tracks are missing," | |
2487 | )) | |
2488 | return fmts | |
2489 | ||
2490 | def _parse_mpd_formats_and_subtitles( | |
2491 | self, mpd_doc, mpd_id=None, mpd_base_url='', mpd_url=None): | |
f0948348 S |
2492 | """ |
2493 | Parse formats from MPD manifest. | |
2494 | References: | |
2495 | 1. MPEG-DASH Standard, ISO/IEC 23009-1:2014(E), | |
2496 | http://standards.iso.org/ittf/PubliclyAvailableStandards/c065274_ISO_IEC_23009-1_2014.zip | |
2497 | 2. https://en.wikipedia.org/wiki/Dynamic_Adaptive_Streaming_over_HTTP | |
2498 | """ | |
a06916d9 | 2499 | if not self.get_param('dynamic_mpd', True): |
78895bd3 | 2500 | if mpd_doc.get('type') == 'dynamic': |
171e59ed | 2501 | return [], {} |
2d2fa82d | 2502 | |
91cb6b50 | 2503 | namespace = self._search_regex(r'(?i)^{([^}]+)?}MPD$', mpd_doc.tag, 'namespace', default=None) |
f14be228 | 2504 | |
2505 | def _add_ns(path): | |
2506 | return self._xpath_ns(path, namespace) | |
2507 | ||
675d0016 | 2508 | def is_drm_protected(element): |
2509 | return element.find(_add_ns('ContentProtection')) is not None | |
2510 | ||
1bac3455 | 2511 | def extract_multisegment_info(element, ms_parent_info): |
2512 | ms_info = ms_parent_info.copy() | |
b4c1d6e8 S |
2513 | |
2514 | # As per [1, 5.3.9.2.2] SegmentList and SegmentTemplate share some | |
2515 | # common attributes and elements. We will only extract relevant | |
2516 | # for us. | |
2517 | def extract_common(source): | |
2518 | segment_timeline = source.find(_add_ns('SegmentTimeline')) | |
2519 | if segment_timeline is not None: | |
2520 | s_e = segment_timeline.findall(_add_ns('S')) | |
2521 | if s_e: | |
2522 | ms_info['total_number'] = 0 | |
2523 | ms_info['s'] = [] | |
2524 | for s in s_e: | |
2525 | r = int(s.get('r', 0)) | |
2526 | ms_info['total_number'] += 1 + r | |
2527 | ms_info['s'].append({ | |
2528 | 't': int(s.get('t', 0)), | |
2529 | # @d is mandatory (see [1, 5.3.9.6.2, Table 17, page 60]) | |
2530 | 'd': int(s.attrib['d']), | |
2531 | 'r': r, | |
2532 | }) | |
2533 | start_number = source.get('startNumber') | |
2534 | if start_number: | |
2535 | ms_info['start_number'] = int(start_number) | |
2536 | timescale = source.get('timescale') | |
2537 | if timescale: | |
2538 | ms_info['timescale'] = int(timescale) | |
2539 | segment_duration = source.get('duration') | |
2540 | if segment_duration: | |
48504785 | 2541 | ms_info['segment_duration'] = float(segment_duration) |
b4c1d6e8 S |
2542 | |
2543 | def extract_Initialization(source): | |
2544 | initialization = source.find(_add_ns('Initialization')) | |
2545 | if initialization is not None: | |
2546 | ms_info['initialization_url'] = initialization.attrib['sourceURL'] | |
2547 | ||
f14be228 | 2548 | segment_list = element.find(_add_ns('SegmentList')) |
1bac3455 | 2549 | if segment_list is not None: |
b4c1d6e8 S |
2550 | extract_common(segment_list) |
2551 | extract_Initialization(segment_list) | |
f14be228 | 2552 | segment_urls_e = segment_list.findall(_add_ns('SegmentURL')) |
1bac3455 | 2553 | if segment_urls_e: |
2554 | ms_info['segment_urls'] = [segment.attrib['media'] for segment in segment_urls_e] | |
1bac3455 | 2555 | else: |
f14be228 | 2556 | segment_template = element.find(_add_ns('SegmentTemplate')) |
1bac3455 | 2557 | if segment_template is not None: |
b4c1d6e8 | 2558 | extract_common(segment_template) |
e228616c S |
2559 | media = segment_template.get('media') |
2560 | if media: | |
2561 | ms_info['media'] = media | |
1bac3455 | 2562 | initialization = segment_template.get('initialization') |
2563 | if initialization: | |
e228616c | 2564 | ms_info['initialization'] = initialization |
1bac3455 | 2565 | else: |
b4c1d6e8 | 2566 | extract_Initialization(segment_template) |
1bac3455 | 2567 | return ms_info |
b323e170 | 2568 | |
a06916d9 | 2569 | skip_unplayable = not self.get_param('allow_unplayable_formats') |
63ad4d43 | 2570 | |
1bac3455 | 2571 | mpd_duration = parse_duration(mpd_doc.get('mediaPresentationDuration')) |
17b598d3 | 2572 | formats = [] |
171e59ed | 2573 | subtitles = {} |
f14be228 | 2574 | for period in mpd_doc.findall(_add_ns('Period')): |
1bac3455 | 2575 | period_duration = parse_duration(period.get('duration')) or mpd_duration |
2576 | period_ms_info = extract_multisegment_info(period, { | |
2577 | 'start_number': 1, | |
2578 | 'timescale': 1, | |
2579 | }) | |
f14be228 | 2580 | for adaptation_set in period.findall(_add_ns('AdaptationSet')): |
06869367 | 2581 | if skip_unplayable and is_drm_protected(adaptation_set): |
675d0016 | 2582 | continue |
1bac3455 | 2583 | adaption_set_ms_info = extract_multisegment_info(adaptation_set, period_ms_info) |
f14be228 | 2584 | for representation in adaptation_set.findall(_add_ns('Representation')): |
06869367 | 2585 | if skip_unplayable and is_drm_protected(representation): |
675d0016 | 2586 | continue |
1bac3455 | 2587 | representation_attrib = adaptation_set.attrib.copy() |
2588 | representation_attrib.update(representation.attrib) | |
f0948348 | 2589 | # According to [1, 5.3.7.2, Table 9, page 41], @mimeType is mandatory |
a6c8b759 | 2590 | mime_type = representation_attrib['mimeType'] |
171e59ed F |
2591 | content_type = representation_attrib.get('contentType', mime_type.split('/')[0]) |
2592 | ||
cdb19aa4 | 2593 | if content_type in ('video', 'audio', 'text') or mime_type == 'image/jpeg': |
1bac3455 | 2594 | base_url = '' |
2595 | for element in (representation, adaptation_set, period, mpd_doc): | |
f14be228 | 2596 | base_url_e = element.find(_add_ns('BaseURL')) |
1bac3455 | 2597 | if base_url_e is not None: |
2598 | base_url = base_url_e.text + base_url | |
2599 | if re.match(r'^https?://', base_url): | |
2600 | break | |
bb20526b S |
2601 | if mpd_base_url and not re.match(r'^https?://', base_url): |
2602 | if not mpd_base_url.endswith('/') and not base_url.startswith('/'): | |
2603 | mpd_base_url += '/' | |
1bac3455 | 2604 | base_url = mpd_base_url + base_url |
2605 | representation_id = representation_attrib.get('id') | |
d577c796 | 2606 | lang = representation_attrib.get('lang') |
51e9094f | 2607 | url_el = representation.find(_add_ns('BaseURL')) |
2608 | filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength') if url_el is not None else None) | |
e228616c | 2609 | bandwidth = int_or_none(representation_attrib.get('bandwidth')) |
cdb19aa4 | 2610 | if representation_id is not None: |
2611 | format_id = representation_id | |
2612 | else: | |
2613 | format_id = content_type | |
2614 | if mpd_id: | |
2615 | format_id = mpd_id + '-' + format_id | |
171e59ed F |
2616 | if content_type in ('video', 'audio'): |
2617 | f = { | |
cdb19aa4 | 2618 | 'format_id': format_id, |
171e59ed F |
2619 | 'manifest_url': mpd_url, |
2620 | 'ext': mimetype2ext(mime_type), | |
2621 | 'width': int_or_none(representation_attrib.get('width')), | |
2622 | 'height': int_or_none(representation_attrib.get('height')), | |
2623 | 'tbr': float_or_none(bandwidth, 1000), | |
2624 | 'asr': int_or_none(representation_attrib.get('audioSamplingRate')), | |
2625 | 'fps': int_or_none(representation_attrib.get('frameRate')), | |
2626 | 'language': lang if lang not in ('mul', 'und', 'zxx', 'mis') else None, | |
2627 | 'format_note': 'DASH %s' % content_type, | |
2628 | 'filesize': filesize, | |
2629 | 'container': mimetype2ext(mime_type) + '_dash', | |
2630 | } | |
2631 | f.update(parse_codecs(representation_attrib.get('codecs'))) | |
2632 | elif content_type == 'text': | |
2633 | f = { | |
2634 | 'ext': mimetype2ext(mime_type), | |
2635 | 'manifest_url': mpd_url, | |
2636 | 'filesize': filesize, | |
2637 | } | |
cdb19aa4 | 2638 | elif mime_type == 'image/jpeg': |
2639 | # See test case in VikiIE | |
2640 | # https://www.viki.com/videos/1175236v-choosing-spouse-by-lottery-episode-1 | |
2641 | f = { | |
2642 | 'format_id': format_id, | |
2643 | 'ext': 'mhtml', | |
2644 | 'manifest_url': mpd_url, | |
2645 | 'format_note': 'DASH storyboards (jpeg)', | |
2646 | 'acodec': 'none', | |
2647 | 'vcodec': 'none', | |
2648 | } | |
1bac3455 | 2649 | representation_ms_info = extract_multisegment_info(representation, adaption_set_ms_info) |
b4c1d6e8 | 2650 | |
e228616c | 2651 | def prepare_template(template_name, identifiers): |
eca1f0d1 S |
2652 | tmpl = representation_ms_info[template_name] |
2653 | # First of, % characters outside $...$ templates | |
2654 | # must be escaped by doubling for proper processing | |
2655 | # by % operator string formatting used further (see | |
067aa17e | 2656 | # https://github.com/ytdl-org/youtube-dl/issues/16867). |
eca1f0d1 S |
2657 | t = '' |
2658 | in_template = False | |
2659 | for c in tmpl: | |
2660 | t += c | |
2661 | if c == '$': | |
2662 | in_template = not in_template | |
2663 | elif c == '%' and not in_template: | |
2664 | t += c | |
2665 | # Next, $...$ templates are translated to their | |
2666 | # %(...) counterparts to be used with % operator | |
cdb19aa4 | 2667 | if representation_id is not None: |
2668 | t = t.replace('$RepresentationID$', representation_id) | |
e228616c S |
2669 | t = re.sub(r'\$(%s)\$' % '|'.join(identifiers), r'%(\1)d', t) |
2670 | t = re.sub(r'\$(%s)%%([^$]+)\$' % '|'.join(identifiers), r'%(\1)\2', t) | |
2671 | t.replace('$$', '$') | |
2672 | return t | |
2673 | ||
2674 | # @initialization is a regular template like @media one | |
2675 | # so it should be handled just the same way (see | |
067aa17e | 2676 | # https://github.com/ytdl-org/youtube-dl/issues/11605) |
e228616c S |
2677 | if 'initialization' in representation_ms_info: |
2678 | initialization_template = prepare_template( | |
2679 | 'initialization', | |
2680 | # As per [1, 5.3.9.4.2, Table 15, page 54] $Number$ and | |
2681 | # $Time$ shall not be included for @initialization thus | |
2682 | # only $Bandwidth$ remains | |
2683 | ('Bandwidth', )) | |
2684 | representation_ms_info['initialization_url'] = initialization_template % { | |
2685 | 'Bandwidth': bandwidth, | |
2686 | } | |
2687 | ||
1141e910 S |
2688 | def location_key(location): |
2689 | return 'url' if re.match(r'^https?://', location) else 'path' | |
2690 | ||
e228616c S |
2691 | if 'segment_urls' not in representation_ms_info and 'media' in representation_ms_info: |
2692 | ||
2693 | media_template = prepare_template('media', ('Number', 'Bandwidth', 'Time')) | |
1141e910 | 2694 | media_location_key = location_key(media_template) |
f0948348 S |
2695 | |
2696 | # As per [1, 5.3.9.4.4, Table 16, page 55] $Number$ and $Time$ | |
2697 | # can't be used at the same time | |
b4c1d6e8 S |
2698 | if '%(Number' in media_template and 's' not in representation_ms_info: |
2699 | segment_duration = None | |
c110944f | 2700 | if 'total_number' not in representation_ms_info and 'segment_duration' in representation_ms_info: |
b4c1d6e8 S |
2701 | segment_duration = float_or_none(representation_ms_info['segment_duration'], representation_ms_info['timescale']) |
2702 | representation_ms_info['total_number'] = int(math.ceil(float(period_duration) / segment_duration)) | |
b4c1d6e8 | 2703 | representation_ms_info['fragments'] = [{ |
1141e910 | 2704 | media_location_key: media_template % { |
b4c1d6e8 | 2705 | 'Number': segment_number, |
e228616c | 2706 | 'Bandwidth': bandwidth, |
b4c1d6e8 S |
2707 | }, |
2708 | 'duration': segment_duration, | |
2709 | } for segment_number in range( | |
2710 | representation_ms_info['start_number'], | |
2711 | representation_ms_info['total_number'] + representation_ms_info['start_number'])] | |
f0948348 | 2712 | else: |
b4c1d6e8 S |
2713 | # $Number*$ or $Time$ in media template with S list available |
2714 | # Example $Number*$: http://www.svtplay.se/klipp/9023742/stopptid-om-bjorn-borg | |
2715 | # Example $Time$: https://play.arkena.com/embed/avp/v2/player/media/b41dda37-d8e7-4d3f-b1b5-9a9db578bdfe/1/129411 | |
b4c1d6e8 | 2716 | representation_ms_info['fragments'] = [] |
f0948348 | 2717 | segment_time = 0 |
b4c1d6e8 S |
2718 | segment_d = None |
2719 | segment_number = representation_ms_info['start_number'] | |
f0948348 S |
2720 | |
2721 | def add_segment_url(): | |
b4c1d6e8 S |
2722 | segment_url = media_template % { |
2723 | 'Time': segment_time, | |
e228616c | 2724 | 'Bandwidth': bandwidth, |
b4c1d6e8 S |
2725 | 'Number': segment_number, |
2726 | } | |
b4c1d6e8 | 2727 | representation_ms_info['fragments'].append({ |
1141e910 | 2728 | media_location_key: segment_url, |
b4c1d6e8 S |
2729 | 'duration': float_or_none(segment_d, representation_ms_info['timescale']), |
2730 | }) | |
f0948348 S |
2731 | |
2732 | for num, s in enumerate(representation_ms_info['s']): | |
2733 | segment_time = s.get('t') or segment_time | |
b4c1d6e8 | 2734 | segment_d = s['d'] |
f0948348 | 2735 | add_segment_url() |
b4c1d6e8 | 2736 | segment_number += 1 |
f0948348 | 2737 | for r in range(s.get('r', 0)): |
b4c1d6e8 | 2738 | segment_time += segment_d |
f0948348 | 2739 | add_segment_url() |
b4c1d6e8 S |
2740 | segment_number += 1 |
2741 | segment_time += segment_d | |
2742 | elif 'segment_urls' in representation_ms_info and 's' in representation_ms_info: | |
2743 | # No media template | |
2744 | # Example: https://www.youtube.com/watch?v=iXZV5uAYMJI | |
2745 | # or any YouTube dashsegments video | |
2746 | fragments = [] | |
d04621da S |
2747 | segment_index = 0 |
2748 | timescale = representation_ms_info['timescale'] | |
2749 | for s in representation_ms_info['s']: | |
2750 | duration = float_or_none(s['d'], timescale) | |
b4c1d6e8 | 2751 | for r in range(s.get('r', 0) + 1): |
1141e910 | 2752 | segment_uri = representation_ms_info['segment_urls'][segment_index] |
b4c1d6e8 | 2753 | fragments.append({ |
1141e910 | 2754 | location_key(segment_uri): segment_uri, |
d04621da | 2755 | 'duration': duration, |
b4c1d6e8 | 2756 | }) |
d04621da | 2757 | segment_index += 1 |
b4c1d6e8 | 2758 | representation_ms_info['fragments'] = fragments |
41bf647e PN |
2759 | elif 'segment_urls' in representation_ms_info: |
2760 | # Segment URLs with no SegmentTimeline | |
2761 | # Example: https://www.seznam.cz/zpravy/clanek/cesko-zasahne-vitr-o-sile-vichrice-muze-byt-i-zivotu-nebezpecny-39091 | |
067aa17e | 2762 | # https://github.com/ytdl-org/youtube-dl/pull/14844 |
41bf647e | 2763 | fragments = [] |
603fc4e0 S |
2764 | segment_duration = float_or_none( |
2765 | representation_ms_info['segment_duration'], | |
2766 | representation_ms_info['timescale']) if 'segment_duration' in representation_ms_info else None | |
41bf647e | 2767 | for segment_url in representation_ms_info['segment_urls']: |
603fc4e0 | 2768 | fragment = { |
41bf647e | 2769 | location_key(segment_url): segment_url, |
603fc4e0 S |
2770 | } |
2771 | if segment_duration: | |
2772 | fragment['duration'] = segment_duration | |
2773 | fragments.append(fragment) | |
41bf647e | 2774 | representation_ms_info['fragments'] = fragments |
79d2077e S |
2775 | # If there is a fragments key available then we correctly recognized fragmented media. |
2776 | # Otherwise we will assume unfragmented media with direct access. Technically, such | |
2777 | # assumption is not necessarily correct since we may simply have no support for | |
2778 | # some forms of fragmented media renditions yet, but for now we'll use this fallback. | |
86f4d14f | 2779 | if 'fragments' in representation_ms_info: |
1bac3455 | 2780 | f.update({ |
79d2077e S |
2781 | # NB: mpd_url may be empty when MPD manifest is parsed from a string |
2782 | 'url': mpd_url or base_url, | |
1141e910 | 2783 | 'fragment_base_url': base_url, |
b4c1d6e8 | 2784 | 'fragments': [], |
cdb19aa4 | 2785 | 'protocol': 'http_dash_segments' if mime_type != 'image/jpeg' else 'mhtml', |
df374b52 | 2786 | }) |
1bac3455 | 2787 | if 'initialization_url' in representation_ms_info: |
e228616c | 2788 | initialization_url = representation_ms_info['initialization_url'] |
1bac3455 | 2789 | if not f.get('url'): |
2790 | f['url'] = initialization_url | |
1141e910 | 2791 | f['fragments'].append({location_key(initialization_url): initialization_url}) |
b4c1d6e8 | 2792 | f['fragments'].extend(representation_ms_info['fragments']) |
79d2077e S |
2793 | else: |
2794 | # Assuming direct URL to unfragmented media. | |
2795 | f['url'] = base_url | |
cdb19aa4 | 2796 | if content_type in ('video', 'audio') or mime_type == 'image/jpeg': |
fd76a142 F |
2797 | formats.append(f) |
2798 | elif content_type == 'text': | |
2799 | subtitles.setdefault(lang or 'und', []).append(f) | |
17b598d3 | 2800 | else: |
1bac3455 | 2801 | self.report_warning('Unknown MIME type %s in DASH manifest' % mime_type) |
171e59ed | 2802 | return formats, subtitles |
17b598d3 | 2803 | |
fd76a142 F |
2804 | def _extract_ism_formats(self, *args, **kwargs): |
2805 | fmts, subs = self._extract_ism_formats_and_subtitles(*args, **kwargs) | |
2806 | if subs: | |
2807 | self.report_warning(bug_reports_message( | |
2808 | "Ignoring subtitle tracks found in the ISM manifest; " | |
2809 | "if any subtitle tracks are missing," | |
2810 | )) | |
2811 | return fmts | |
2812 | ||
2813 | def _extract_ism_formats_and_subtitles(self, ism_url, video_id, ism_id=None, note=None, errnote=None, fatal=True, data=None, headers={}, query={}): | |
47a5cb77 | 2814 | res = self._download_xml_handle( |
b2758123 | 2815 | ism_url, video_id, |
37a3bb66 | 2816 | note='Downloading ISM manifest' if note is None else note, |
2817 | errnote='Failed to download ISM manifest' if errnote is None else errnote, | |
7360c06f | 2818 | fatal=fatal, data=data, headers=headers, query=query) |
b2758123 | 2819 | if res is False: |
fd76a142 | 2820 | return [], {} |
47a5cb77 | 2821 | ism_doc, urlh = res |
13b08034 | 2822 | if ism_doc is None: |
fd76a142 | 2823 | return [], {} |
b2758123 | 2824 | |
fd76a142 | 2825 | return self._parse_ism_formats_and_subtitles(ism_doc, urlh.geturl(), ism_id) |
b2758123 | 2826 | |
fd76a142 | 2827 | def _parse_ism_formats_and_subtitles(self, ism_doc, ism_url, ism_id=None): |
76d5a363 S |
2828 | """ |
2829 | Parse formats from ISM manifest. | |
2830 | References: | |
2831 | 1. [MS-SSTR]: Smooth Streaming Protocol, | |
2832 | https://msdn.microsoft.com/en-us/library/ff469518.aspx | |
2833 | """ | |
06869367 | 2834 | if ism_doc.get('IsLive') == 'TRUE': |
fd76a142 | 2835 | return [], {} |
a06916d9 | 2836 | if (not self.get_param('allow_unplayable_formats') |
06869367 | 2837 | and ism_doc.find('Protection') is not None): |
fd76a142 | 2838 | return [], {} |
b2758123 | 2839 | |
b2758123 RA |
2840 | duration = int(ism_doc.attrib['Duration']) |
2841 | timescale = int_or_none(ism_doc.get('TimeScale')) or 10000000 | |
2842 | ||
2843 | formats = [] | |
fd76a142 | 2844 | subtitles = {} |
b2758123 RA |
2845 | for stream in ism_doc.findall('StreamIndex'): |
2846 | stream_type = stream.get('Type') | |
fd76a142 | 2847 | if stream_type not in ('video', 'audio', 'text'): |
b2758123 RA |
2848 | continue |
2849 | url_pattern = stream.attrib['Url'] | |
2850 | stream_timescale = int_or_none(stream.get('TimeScale')) or timescale | |
2851 | stream_name = stream.get('Name') | |
fd76a142 | 2852 | stream_language = stream.get('Language', 'und') |
b2758123 | 2853 | for track in stream.findall('QualityLevel'): |
e2efe599 | 2854 | fourcc = track.get('FourCC') or ('AACL' if track.get('AudioTag') == '255' else None) |
b2758123 | 2855 | # TODO: add support for WVC1 and WMAP |
66a1b864 | 2856 | if fourcc not in ('H264', 'AVC1', 'AACL', 'TTML'): |
b2758123 RA |
2857 | self.report_warning('%s is not a supported codec' % fourcc) |
2858 | continue | |
2859 | tbr = int(track.attrib['Bitrate']) // 1000 | |
76d5a363 S |
2860 | # [1] does not mention Width and Height attributes. However, |
2861 | # they're often present while MaxWidth and MaxHeight are | |
2862 | # missing, so should be used as fallbacks | |
2863 | width = int_or_none(track.get('MaxWidth') or track.get('Width')) | |
2864 | height = int_or_none(track.get('MaxHeight') or track.get('Height')) | |
b2758123 RA |
2865 | sampling_rate = int_or_none(track.get('SamplingRate')) |
2866 | ||
2867 | track_url_pattern = re.sub(r'{[Bb]itrate}', track.attrib['Bitrate'], url_pattern) | |
2868 | track_url_pattern = compat_urlparse.urljoin(ism_url, track_url_pattern) | |
2869 | ||
2870 | fragments = [] | |
2871 | fragment_ctx = { | |
2872 | 'time': 0, | |
2873 | } | |
2874 | stream_fragments = stream.findall('c') | |
2875 | for stream_fragment_index, stream_fragment in enumerate(stream_fragments): | |
2876 | fragment_ctx['time'] = int_or_none(stream_fragment.get('t')) or fragment_ctx['time'] | |
2877 | fragment_repeat = int_or_none(stream_fragment.get('r')) or 1 | |
2878 | fragment_ctx['duration'] = int_or_none(stream_fragment.get('d')) | |
2879 | if not fragment_ctx['duration']: | |
2880 | try: | |
2881 | next_fragment_time = int(stream_fragment[stream_fragment_index + 1].attrib['t']) | |
2882 | except IndexError: | |
2883 | next_fragment_time = duration | |
1616f9b4 | 2884 | fragment_ctx['duration'] = (next_fragment_time - fragment_ctx['time']) / fragment_repeat |
b2758123 RA |
2885 | for _ in range(fragment_repeat): |
2886 | fragments.append({ | |
1616f9b4 | 2887 | 'url': re.sub(r'{start[ _]time}', compat_str(fragment_ctx['time']), track_url_pattern), |
b2758123 RA |
2888 | 'duration': fragment_ctx['duration'] / stream_timescale, |
2889 | }) | |
2890 | fragment_ctx['time'] += fragment_ctx['duration'] | |
2891 | ||
2892 | format_id = [] | |
2893 | if ism_id: | |
2894 | format_id.append(ism_id) | |
2895 | if stream_name: | |
2896 | format_id.append(stream_name) | |
2897 | format_id.append(compat_str(tbr)) | |
2898 | ||
fd76a142 F |
2899 | if stream_type == 'text': |
2900 | subtitles.setdefault(stream_language, []).append({ | |
2901 | 'ext': 'ismt', | |
2902 | 'protocol': 'ism', | |
2903 | 'url': ism_url, | |
2904 | 'manifest_url': ism_url, | |
2905 | 'fragments': fragments, | |
2906 | '_download_params': { | |
2907 | 'stream_type': stream_type, | |
2908 | 'duration': duration, | |
2909 | 'timescale': stream_timescale, | |
2910 | 'fourcc': fourcc, | |
2911 | 'language': stream_language, | |
2912 | 'codec_private_data': track.get('CodecPrivateData'), | |
2913 | } | |
2914 | }) | |
2915 | elif stream_type in ('video', 'audio'): | |
2916 | formats.append({ | |
2917 | 'format_id': '-'.join(format_id), | |
2918 | 'url': ism_url, | |
2919 | 'manifest_url': ism_url, | |
2920 | 'ext': 'ismv' if stream_type == 'video' else 'isma', | |
2921 | 'width': width, | |
2922 | 'height': height, | |
2923 | 'tbr': tbr, | |
2924 | 'asr': sampling_rate, | |
2925 | 'vcodec': 'none' if stream_type == 'audio' else fourcc, | |
2926 | 'acodec': 'none' if stream_type == 'video' else fourcc, | |
2927 | 'protocol': 'ism', | |
2928 | 'fragments': fragments, | |
2929 | '_download_params': { | |
2930 | 'stream_type': stream_type, | |
2931 | 'duration': duration, | |
2932 | 'timescale': stream_timescale, | |
2933 | 'width': width or 0, | |
2934 | 'height': height or 0, | |
2935 | 'fourcc': fourcc, | |
2936 | 'language': stream_language, | |
2937 | 'codec_private_data': track.get('CodecPrivateData'), | |
2938 | 'sampling_rate': sampling_rate, | |
2939 | 'channels': int_or_none(track.get('Channels', 2)), | |
2940 | 'bits_per_sample': int_or_none(track.get('BitsPerSample', 16)), | |
2941 | 'nal_unit_length_field': int_or_none(track.get('NALUnitLengthField', 4)), | |
2942 | }, | |
2943 | }) | |
2944 | return formats, subtitles | |
b2758123 | 2945 | |
f983b875 | 2946 | def _parse_html5_media_entries(self, base_url, webpage, video_id, m3u8_id=None, m3u8_entry_protocol='m3u8', mpd_id=None, preference=None, quality=None): |
6780154e S |
2947 | def absolute_url(item_url): |
2948 | return urljoin(base_url, item_url) | |
59bbe491 | 2949 | |
2950 | def parse_content_type(content_type): | |
2951 | if not content_type: | |
2952 | return {} | |
2953 | ctr = re.search(r'(?P<mimetype>[^/]+/[^;]+)(?:;\s*codecs="?(?P<codecs>[^"]+))?', content_type) | |
2954 | if ctr: | |
2955 | mimetype, codecs = ctr.groups() | |
2956 | f = parse_codecs(codecs) | |
2957 | f['ext'] = mimetype2ext(mimetype) | |
2958 | return f | |
2959 | return {} | |
2960 | ||
868f79db | 2961 | def _media_formats(src, cur_media_type, type_info={}): |
520251c0 | 2962 | full_url = absolute_url(src) |
82889d4a | 2963 | ext = type_info.get('ext') or determine_ext(full_url) |
87a449c1 | 2964 | if ext == 'm3u8': |
520251c0 YCH |
2965 | is_plain_url = False |
2966 | formats = self._extract_m3u8_formats( | |
ad120ae1 | 2967 | full_url, video_id, ext='mp4', |
eeb0a956 | 2968 | entry_protocol=m3u8_entry_protocol, m3u8_id=m3u8_id, |
f983b875 | 2969 | preference=preference, quality=quality, fatal=False) |
87a449c1 S |
2970 | elif ext == 'mpd': |
2971 | is_plain_url = False | |
2972 | formats = self._extract_mpd_formats( | |
b359e977 | 2973 | full_url, video_id, mpd_id=mpd_id, fatal=False) |
520251c0 YCH |
2974 | else: |
2975 | is_plain_url = True | |
2976 | formats = [{ | |
2977 | 'url': full_url, | |
2978 | 'vcodec': 'none' if cur_media_type == 'audio' else None, | |
2979 | }] | |
2980 | return is_plain_url, formats | |
2981 | ||
59bbe491 | 2982 | entries = [] |
4328ddf8 S |
2983 | # amp-video and amp-audio are very similar to their HTML5 counterparts |
2984 | # so we wll include them right here (see | |
2985 | # https://www.ampproject.org/docs/reference/components/amp-video) | |
29f7c58a | 2986 | # For dl8-* tags see https://delight-vr.com/documentation/dl8-video/ |
2987 | _MEDIA_TAG_NAME_RE = r'(?:(?:amp|dl8(?:-live)?)-)?(video|audio)' | |
2988 | media_tags = [(media_tag, media_tag_name, media_type, '') | |
2989 | for media_tag, media_tag_name, media_type | |
2990 | in re.findall(r'(?s)(<(%s)[^>]*/>)' % _MEDIA_TAG_NAME_RE, webpage)] | |
2aec7256 S |
2991 | media_tags.extend(re.findall( |
2992 | # We only allow video|audio followed by a whitespace or '>'. | |
2993 | # Allowing more characters may end up in significant slow down (see | |
067aa17e | 2994 | # https://github.com/ytdl-org/youtube-dl/issues/11979, example URL: |
2aec7256 | 2995 | # http://www.porntrex.com/maps/videositemap.xml). |
29f7c58a | 2996 | r'(?s)(<(?P<tag>%s)(?:\s+[^>]*)?>)(.*?)</(?P=tag)>' % _MEDIA_TAG_NAME_RE, webpage)) |
2997 | for media_tag, _, media_type, media_content in media_tags: | |
59bbe491 | 2998 | media_info = { |
2999 | 'formats': [], | |
3000 | 'subtitles': {}, | |
3001 | } | |
3002 | media_attributes = extract_attributes(media_tag) | |
f856816b | 3003 | src = strip_or_none(media_attributes.get('src')) |
59bbe491 | 3004 | if src: |
dedb1770 | 3005 | _, formats = _media_formats(src, media_type) |
520251c0 | 3006 | media_info['formats'].extend(formats) |
6780154e | 3007 | media_info['thumbnail'] = absolute_url(media_attributes.get('poster')) |
59bbe491 | 3008 | if media_content: |
3009 | for source_tag in re.findall(r'<source[^>]+>', media_content): | |
d493f15c S |
3010 | s_attr = extract_attributes(source_tag) |
3011 | # data-video-src and data-src are non standard but seen | |
3012 | # several times in the wild | |
f856816b | 3013 | src = strip_or_none(dict_get(s_attr, ('src', 'data-video-src', 'data-src'))) |
59bbe491 | 3014 | if not src: |
3015 | continue | |
d493f15c | 3016 | f = parse_content_type(s_attr.get('type')) |
868f79db | 3017 | is_plain_url, formats = _media_formats(src, media_type, f) |
520251c0 | 3018 | if is_plain_url: |
d493f15c S |
3019 | # width, height, res, label and title attributes are |
3020 | # all not standard but seen several times in the wild | |
3021 | labels = [ | |
3022 | s_attr.get(lbl) | |
3023 | for lbl in ('label', 'title') | |
3024 | if str_or_none(s_attr.get(lbl)) | |
3025 | ] | |
3026 | width = int_or_none(s_attr.get('width')) | |
3089bc74 S |
3027 | height = (int_or_none(s_attr.get('height')) |
3028 | or int_or_none(s_attr.get('res'))) | |
d493f15c S |
3029 | if not width or not height: |
3030 | for lbl in labels: | |
3031 | resolution = parse_resolution(lbl) | |
3032 | if not resolution: | |
3033 | continue | |
3034 | width = width or resolution.get('width') | |
3035 | height = height or resolution.get('height') | |
3036 | for lbl in labels: | |
3037 | tbr = parse_bitrate(lbl) | |
3038 | if tbr: | |
3039 | break | |
3040 | else: | |
3041 | tbr = None | |
1ed45499 | 3042 | f.update({ |
d493f15c S |
3043 | 'width': width, |
3044 | 'height': height, | |
3045 | 'tbr': tbr, | |
3046 | 'format_id': s_attr.get('label') or s_attr.get('title'), | |
1ed45499 | 3047 | }) |
520251c0 YCH |
3048 | f.update(formats[0]) |
3049 | media_info['formats'].append(f) | |
3050 | else: | |
3051 | media_info['formats'].extend(formats) | |
59bbe491 | 3052 | for track_tag in re.findall(r'<track[^>]+>', media_content): |
3053 | track_attributes = extract_attributes(track_tag) | |
3054 | kind = track_attributes.get('kind') | |
5968d7d2 | 3055 | if not kind or kind in ('subtitles', 'captions'): |
f856816b | 3056 | src = strip_or_none(track_attributes.get('src')) |
59bbe491 | 3057 | if not src: |
3058 | continue | |
3059 | lang = track_attributes.get('srclang') or track_attributes.get('lang') or track_attributes.get('label') | |
3060 | media_info['subtitles'].setdefault(lang, []).append({ | |
3061 | 'url': absolute_url(src), | |
3062 | }) | |
5e8e2fa5 S |
3063 | for f in media_info['formats']: |
3064 | f.setdefault('http_headers', {})['Referer'] = base_url | |
5968d7d2 | 3065 | if media_info['formats'] or media_info['subtitles']: |
59bbe491 | 3066 | entries.append(media_info) |
3067 | return entries | |
3068 | ||
f6a1d69a F |
3069 | def _extract_akamai_formats(self, *args, **kwargs): |
3070 | fmts, subs = self._extract_akamai_formats_and_subtitles(*args, **kwargs) | |
3071 | if subs: | |
3072 | self.report_warning(bug_reports_message( | |
3073 | "Ignoring subtitle tracks found in the manifests; " | |
3074 | "if any subtitle tracks are missing," | |
3075 | )) | |
3076 | return fmts | |
3077 | ||
3078 | def _extract_akamai_formats_and_subtitles(self, manifest_url, video_id, hosts={}): | |
29f7c58a | 3079 | signed = 'hdnea=' in manifest_url |
3080 | if not signed: | |
3081 | # https://learn.akamai.com/en-us/webhelp/media-services-on-demand/stream-packaging-user-guide/GUID-BE6C0F73-1E06-483B-B0EA-57984B91B7F9.html | |
3082 | manifest_url = re.sub( | |
3083 | r'(?:b=[\d,-]+|(?:__a__|attributes)=off|__b__=\d+)&?', | |
3084 | '', manifest_url).strip('?') | |
3085 | ||
c7c43a93 | 3086 | formats = [] |
f6a1d69a | 3087 | subtitles = {} |
70c5802b | 3088 | |
e71a4509 | 3089 | hdcore_sign = 'hdcore=3.7.0' |
ff6f9a67 | 3090 | f4m_url = re.sub(r'(https?://[^/]+)/i/', r'\1/z/', manifest_url).replace('/master.m3u8', '/manifest.f4m') |
c4251b9a RA |
3091 | hds_host = hosts.get('hds') |
3092 | if hds_host: | |
3093 | f4m_url = re.sub(r'(https?://)[^/]+', r'\1' + hds_host, f4m_url) | |
e71a4509 RA |
3094 | if 'hdcore=' not in f4m_url: |
3095 | f4m_url += ('&' if '?' in f4m_url else '?') + hdcore_sign | |
3096 | f4m_formats = self._extract_f4m_formats( | |
3097 | f4m_url, video_id, f4m_id='hds', fatal=False) | |
3098 | for entry in f4m_formats: | |
3099 | entry.update({'extra_param_to_segment_url': hdcore_sign}) | |
3100 | formats.extend(f4m_formats) | |
70c5802b | 3101 | |
c4251b9a RA |
3102 | m3u8_url = re.sub(r'(https?://[^/]+)/z/', r'\1/i/', manifest_url).replace('/manifest.f4m', '/master.m3u8') |
3103 | hls_host = hosts.get('hls') | |
3104 | if hls_host: | |
3105 | m3u8_url = re.sub(r'(https?://)[^/]+', r'\1' + hls_host, m3u8_url) | |
f6a1d69a | 3106 | m3u8_formats, m3u8_subtitles = self._extract_m3u8_formats_and_subtitles( |
c7c43a93 | 3107 | m3u8_url, video_id, 'mp4', 'm3u8_native', |
29f7c58a | 3108 | m3u8_id='hls', fatal=False) |
3109 | formats.extend(m3u8_formats) | |
f6a1d69a | 3110 | subtitles = self._merge_subtitles(subtitles, m3u8_subtitles) |
70c5802b | 3111 | |
3112 | http_host = hosts.get('http') | |
29f7c58a | 3113 | if http_host and m3u8_formats and not signed: |
3114 | REPL_REGEX = r'https?://[^/]+/i/([^,]+),([^/]+),([^/]+)\.csmil/.+' | |
70c5802b | 3115 | qualities = re.match(REPL_REGEX, m3u8_url).group(2).split(',') |
3116 | qualities_length = len(qualities) | |
29f7c58a | 3117 | if len(m3u8_formats) in (qualities_length, qualities_length + 1): |
70c5802b | 3118 | i = 0 |
29f7c58a | 3119 | for f in m3u8_formats: |
3120 | if f['vcodec'] != 'none': | |
70c5802b | 3121 | for protocol in ('http', 'https'): |
3122 | http_f = f.copy() | |
3123 | del http_f['manifest_url'] | |
3124 | http_url = re.sub( | |
29f7c58a | 3125 | REPL_REGEX, protocol + r'://%s/\g<1>%s\3' % (http_host, qualities[i]), f['url']) |
70c5802b | 3126 | http_f.update({ |
3127 | 'format_id': http_f['format_id'].replace('hls-', protocol + '-'), | |
3128 | 'url': http_url, | |
3129 | 'protocol': protocol, | |
3130 | }) | |
29f7c58a | 3131 | formats.append(http_f) |
70c5802b | 3132 | i += 1 |
70c5802b | 3133 | |
f6a1d69a | 3134 | return formats, subtitles |
c7c43a93 | 3135 | |
6ad02195 | 3136 | def _extract_wowza_formats(self, url, video_id, m3u8_entry_protocol='m3u8_native', skip_protocols=[]): |
044eeb14 | 3137 | query = compat_urlparse.urlparse(url).query |
6ad02195 | 3138 | url = re.sub(r'/(?:manifest|playlist|jwplayer)\.(?:m3u8|f4m|mpd|smil)', '', url) |
240f2622 S |
3139 | mobj = re.search( |
3140 | r'(?:(?:http|rtmp|rtsp)(?P<s>s)?:)?(?P<url>//[^?]+)', url) | |
3141 | url_base = mobj.group('url') | |
3142 | http_base_url = '%s%s:%s' % ('http', mobj.group('s') or '', url_base) | |
6ad02195 | 3143 | formats = [] |
044eeb14 S |
3144 | |
3145 | def manifest_url(manifest): | |
3146 | m_url = '%s/%s' % (http_base_url, manifest) | |
3147 | if query: | |
3148 | m_url += '?%s' % query | |
3149 | return m_url | |
3150 | ||
6ad02195 RA |
3151 | if 'm3u8' not in skip_protocols: |
3152 | formats.extend(self._extract_m3u8_formats( | |
044eeb14 | 3153 | manifest_url('playlist.m3u8'), video_id, 'mp4', |
6ad02195 RA |
3154 | m3u8_entry_protocol, m3u8_id='hls', fatal=False)) |
3155 | if 'f4m' not in skip_protocols: | |
3156 | formats.extend(self._extract_f4m_formats( | |
044eeb14 | 3157 | manifest_url('manifest.f4m'), |
6ad02195 | 3158 | video_id, f4m_id='hds', fatal=False)) |
0384932e RA |
3159 | if 'dash' not in skip_protocols: |
3160 | formats.extend(self._extract_mpd_formats( | |
044eeb14 | 3161 | manifest_url('manifest.mpd'), |
0384932e | 3162 | video_id, mpd_id='dash', fatal=False)) |
6ad02195 | 3163 | if re.search(r'(?:/smil:|\.smil)', url_base): |
6ad02195 RA |
3164 | if 'smil' not in skip_protocols: |
3165 | rtmp_formats = self._extract_smil_formats( | |
044eeb14 | 3166 | manifest_url('jwplayer.smil'), |
6ad02195 RA |
3167 | video_id, fatal=False) |
3168 | for rtmp_format in rtmp_formats: | |
3169 | rtsp_format = rtmp_format.copy() | |
3170 | rtsp_format['url'] = '%s/%s' % (rtmp_format['url'], rtmp_format['play_path']) | |
3171 | del rtsp_format['play_path'] | |
3172 | del rtsp_format['ext'] | |
3173 | rtsp_format.update({ | |
3174 | 'url': rtsp_format['url'].replace('rtmp://', 'rtsp://'), | |
3175 | 'format_id': rtmp_format['format_id'].replace('rtmp', 'rtsp'), | |
3176 | 'protocol': 'rtsp', | |
3177 | }) | |
3178 | formats.extend([rtmp_format, rtsp_format]) | |
3179 | else: | |
3180 | for protocol in ('rtmp', 'rtsp'): | |
3181 | if protocol not in skip_protocols: | |
3182 | formats.append({ | |
f2e2f0c7 | 3183 | 'url': '%s:%s' % (protocol, url_base), |
6ad02195 RA |
3184 | 'format_id': protocol, |
3185 | 'protocol': protocol, | |
3186 | }) | |
3187 | return formats | |
3188 | ||
c73e330e | 3189 | def _find_jwplayer_data(self, webpage, video_id=None, transform_source=js_to_json): |
a4a554a7 | 3190 | mobj = re.search( |
ac9c69ac | 3191 | r'(?s)jwplayer\((?P<quote>[\'"])[^\'" ]+(?P=quote)\)(?!</script>).*?\.setup\s*\((?P<options>[^)]+)\)', |
a4a554a7 YCH |
3192 | webpage) |
3193 | if mobj: | |
c73e330e RU |
3194 | try: |
3195 | jwplayer_data = self._parse_json(mobj.group('options'), | |
3196 | video_id=video_id, | |
3197 | transform_source=transform_source) | |
3198 | except ExtractorError: | |
3199 | pass | |
3200 | else: | |
3201 | if isinstance(jwplayer_data, dict): | |
3202 | return jwplayer_data | |
a4a554a7 YCH |
3203 | |
3204 | def _extract_jwplayer_data(self, webpage, video_id, *args, **kwargs): | |
c73e330e RU |
3205 | jwplayer_data = self._find_jwplayer_data( |
3206 | webpage, video_id, transform_source=js_to_json) | |
a4a554a7 YCH |
3207 | return self._parse_jwplayer_data( |
3208 | jwplayer_data, video_id, *args, **kwargs) | |
3209 | ||
3210 | def _parse_jwplayer_data(self, jwplayer_data, video_id=None, require_title=True, | |
3211 | m3u8_id=None, mpd_id=None, rtmp_params=None, base_url=None): | |
3212 | # JWPlayer backward compatibility: flattened playlists | |
3213 | # https://github.com/jwplayer/jwplayer/blob/v7.4.3/src/js/api/config.js#L81-L96 | |
3214 | if 'playlist' not in jwplayer_data: | |
3215 | jwplayer_data = {'playlist': [jwplayer_data]} | |
3216 | ||
3217 | entries = [] | |
3218 | ||
3219 | # JWPlayer backward compatibility: single playlist item | |
3220 | # https://github.com/jwplayer/jwplayer/blob/v7.7.0/src/js/playlist/playlist.js#L10 | |
3221 | if not isinstance(jwplayer_data['playlist'], list): | |
3222 | jwplayer_data['playlist'] = [jwplayer_data['playlist']] | |
3223 | ||
3224 | for video_data in jwplayer_data['playlist']: | |
3225 | # JWPlayer backward compatibility: flattened sources | |
3226 | # https://github.com/jwplayer/jwplayer/blob/v7.4.3/src/js/playlist/item.js#L29-L35 | |
3227 | if 'sources' not in video_data: | |
3228 | video_data['sources'] = [video_data] | |
3229 | ||
3230 | this_video_id = video_id or video_data['mediaid'] | |
3231 | ||
1a2192cb S |
3232 | formats = self._parse_jwplayer_formats( |
3233 | video_data['sources'], video_id=this_video_id, m3u8_id=m3u8_id, | |
3234 | mpd_id=mpd_id, rtmp_params=rtmp_params, base_url=base_url) | |
a4a554a7 YCH |
3235 | |
3236 | subtitles = {} | |
3237 | tracks = video_data.get('tracks') | |
3238 | if tracks and isinstance(tracks, list): | |
3239 | for track in tracks: | |
96a2daa1 S |
3240 | if not isinstance(track, dict): |
3241 | continue | |
f4b74272 S |
3242 | track_kind = track.get('kind') |
3243 | if not track_kind or not isinstance(track_kind, compat_str): | |
3244 | continue | |
3245 | if track_kind.lower() not in ('captions', 'subtitles'): | |
a4a554a7 YCH |
3246 | continue |
3247 | track_url = urljoin(base_url, track.get('file')) | |
3248 | if not track_url: | |
3249 | continue | |
3250 | subtitles.setdefault(track.get('label') or 'en', []).append({ | |
3251 | 'url': self._proto_relative_url(track_url) | |
3252 | }) | |
3253 | ||
50d808f5 | 3254 | entry = { |
a4a554a7 | 3255 | 'id': this_video_id, |
50d808f5 | 3256 | 'title': unescapeHTML(video_data['title'] if require_title else video_data.get('title')), |
f81dd65b | 3257 | 'description': clean_html(video_data.get('description')), |
6945b9e7 | 3258 | 'thumbnail': urljoin(base_url, self._proto_relative_url(video_data.get('image'))), |
a4a554a7 YCH |
3259 | 'timestamp': int_or_none(video_data.get('pubdate')), |
3260 | 'duration': float_or_none(jwplayer_data.get('duration') or video_data.get('duration')), | |
3261 | 'subtitles': subtitles, | |
50d808f5 RA |
3262 | } |
3263 | # https://github.com/jwplayer/jwplayer/blob/master/src/js/utils/validator.js#L32 | |
3264 | if len(formats) == 1 and re.search(r'^(?:http|//).*(?:youtube\.com|youtu\.be)/.+', formats[0]['url']): | |
3265 | entry.update({ | |
3266 | '_type': 'url_transparent', | |
3267 | 'url': formats[0]['url'], | |
3268 | }) | |
3269 | else: | |
3270 | self._sort_formats(formats) | |
3271 | entry['formats'] = formats | |
3272 | entries.append(entry) | |
a4a554a7 YCH |
3273 | if len(entries) == 1: |
3274 | return entries[0] | |
3275 | else: | |
3276 | return self.playlist_result(entries) | |
3277 | ||
ed0cf9b3 S |
3278 | def _parse_jwplayer_formats(self, jwplayer_sources_data, video_id=None, |
3279 | m3u8_id=None, mpd_id=None, rtmp_params=None, base_url=None): | |
bf1b87cd | 3280 | urls = [] |
ed0cf9b3 | 3281 | formats = [] |
1a2192cb | 3282 | for source in jwplayer_sources_data: |
0a268c6e S |
3283 | if not isinstance(source, dict): |
3284 | continue | |
6945b9e7 RA |
3285 | source_url = urljoin( |
3286 | base_url, self._proto_relative_url(source.get('file'))) | |
3287 | if not source_url or source_url in urls: | |
bf1b87cd RA |
3288 | continue |
3289 | urls.append(source_url) | |
ed0cf9b3 S |
3290 | source_type = source.get('type') or '' |
3291 | ext = mimetype2ext(source_type) or determine_ext(source_url) | |
3292 | if source_type == 'hls' or ext == 'm3u8': | |
3293 | formats.extend(self._extract_m3u8_formats( | |
0236cd0d S |
3294 | source_url, video_id, 'mp4', entry_protocol='m3u8_native', |
3295 | m3u8_id=m3u8_id, fatal=False)) | |
0d9c48de | 3296 | elif source_type == 'dash' or ext == 'mpd': |
ed0cf9b3 S |
3297 | formats.extend(self._extract_mpd_formats( |
3298 | source_url, video_id, mpd_id=mpd_id, fatal=False)) | |
b51dc9db S |
3299 | elif ext == 'smil': |
3300 | formats.extend(self._extract_smil_formats( | |
3301 | source_url, video_id, fatal=False)) | |
ed0cf9b3 | 3302 | # https://github.com/jwplayer/jwplayer/blob/master/src/js/providers/default.js#L67 |
0236cd0d S |
3303 | elif source_type.startswith('audio') or ext in ( |
3304 | 'oga', 'aac', 'mp3', 'mpeg', 'vorbis'): | |
ed0cf9b3 S |
3305 | formats.append({ |
3306 | 'url': source_url, | |
3307 | 'vcodec': 'none', | |
3308 | 'ext': ext, | |
3309 | }) | |
3310 | else: | |
3311 | height = int_or_none(source.get('height')) | |
3312 | if height is None: | |
3313 | # Often no height is provided but there is a label in | |
0236cd0d | 3314 | # format like "1080p", "720p SD", or 1080. |
ed0cf9b3 | 3315 | height = int_or_none(self._search_regex( |
0236cd0d | 3316 | r'^(\d{3,4})[pP]?(?:\b|$)', compat_str(source.get('label') or ''), |
ed0cf9b3 S |
3317 | 'height', default=None)) |
3318 | a_format = { | |
3319 | 'url': source_url, | |
3320 | 'width': int_or_none(source.get('width')), | |
3321 | 'height': height, | |
0236cd0d | 3322 | 'tbr': int_or_none(source.get('bitrate')), |
ed0cf9b3 S |
3323 | 'ext': ext, |
3324 | } | |
3325 | if source_url.startswith('rtmp'): | |
3326 | a_format['ext'] = 'flv' | |
ed0cf9b3 S |
3327 | # See com/longtailvideo/jwplayer/media/RTMPMediaProvider.as |
3328 | # of jwplayer.flash.swf | |
3329 | rtmp_url_parts = re.split( | |
3330 | r'((?:mp4|mp3|flv):)', source_url, 1) | |
3331 | if len(rtmp_url_parts) == 3: | |
3332 | rtmp_url, prefix, play_path = rtmp_url_parts | |
3333 | a_format.update({ | |
3334 | 'url': rtmp_url, | |
3335 | 'play_path': prefix + play_path, | |
3336 | }) | |
3337 | if rtmp_params: | |
3338 | a_format.update(rtmp_params) | |
3339 | formats.append(a_format) | |
3340 | return formats | |
3341 | ||
f4b1c7ad PH |
3342 | def _live_title(self, name): |
3343 | """ Generate the title for a live video """ | |
3344 | now = datetime.datetime.now() | |
611c1dd9 | 3345 | now_str = now.strftime('%Y-%m-%d %H:%M') |
f4b1c7ad PH |
3346 | return name + ' ' + now_str |
3347 | ||
b14f3a4c PH |
3348 | def _int(self, v, name, fatal=False, **kwargs): |
3349 | res = int_or_none(v, **kwargs) | |
3350 | if 'get_attr' in kwargs: | |
3351 | print(getattr(v, kwargs['get_attr'])) | |
3352 | if res is None: | |
3353 | msg = 'Failed to extract %s: Could not parse value %r' % (name, v) | |
3354 | if fatal: | |
3355 | raise ExtractorError(msg) | |
3356 | else: | |
6a39ee13 | 3357 | self.report_warning(msg) |
b14f3a4c PH |
3358 | return res |
3359 | ||
3360 | def _float(self, v, name, fatal=False, **kwargs): | |
3361 | res = float_or_none(v, **kwargs) | |
3362 | if res is None: | |
3363 | msg = 'Failed to extract %s: Could not parse value %r' % (name, v) | |
3364 | if fatal: | |
3365 | raise ExtractorError(msg) | |
3366 | else: | |
6a39ee13 | 3367 | self.report_warning(msg) |
b14f3a4c PH |
3368 | return res |
3369 | ||
40e41780 TF |
3370 | def _set_cookie(self, domain, name, value, expire_time=None, port=None, |
3371 | path='/', secure=False, discard=False, rest={}, **kwargs): | |
6c22cee6 | 3372 | cookie = compat_cookiejar_Cookie( |
4ed2d7b7 | 3373 | 0, name, value, port, port is not None, domain, True, |
40e41780 TF |
3374 | domain.startswith('.'), path, True, secure, expire_time, |
3375 | discard, None, None, rest) | |
42939b61 JMF |
3376 | self._downloader.cookiejar.set_cookie(cookie) |
3377 | ||
799207e8 | 3378 | def _get_cookies(self, url): |
f7ad7160 | 3379 | """ Return a compat_cookies_SimpleCookie with the cookies for the url """ |
5c2266df | 3380 | req = sanitized_Request(url) |
799207e8 | 3381 | self._downloader.cookiejar.add_cookie_header(req) |
f7ad7160 | 3382 | return compat_cookies_SimpleCookie(req.get_header('Cookie')) |
799207e8 | 3383 | |
e3c1266f | 3384 | def _apply_first_set_cookie_header(self, url_handle, cookie): |
ce2fe4c0 S |
3385 | """ |
3386 | Apply first Set-Cookie header instead of the last. Experimental. | |
3387 | ||
3388 | Some sites (e.g. [1-3]) may serve two cookies under the same name | |
3389 | in Set-Cookie header and expect the first (old) one to be set rather | |
3390 | than second (new). However, as of RFC6265 the newer one cookie | |
3391 | should be set into cookie store what actually happens. | |
3392 | We will workaround this issue by resetting the cookie to | |
3393 | the first one manually. | |
3394 | 1. https://new.vk.com/ | |
3395 | 2. https://github.com/ytdl-org/youtube-dl/issues/9841#issuecomment-227871201 | |
3396 | 3. https://learning.oreilly.com/ | |
3397 | """ | |
e3c1266f S |
3398 | for header, cookies in url_handle.headers.items(): |
3399 | if header.lower() != 'set-cookie': | |
3400 | continue | |
3401 | if sys.version_info[0] >= 3: | |
3402 | cookies = cookies.encode('iso-8859-1') | |
3403 | cookies = cookies.decode('utf-8') | |
3404 | cookie_value = re.search( | |
3405 | r'%s=(.+?);.*?\b[Dd]omain=(.+?)(?:[,;]|$)' % cookie, cookies) | |
3406 | if cookie_value: | |
3407 | value, domain = cookie_value.groups() | |
3408 | self._set_cookie(domain, cookie, value) | |
3409 | break | |
3410 | ||
05900629 PH |
3411 | def get_testcases(self, include_onlymatching=False): |
3412 | t = getattr(self, '_TEST', None) | |
3413 | if t: | |
3414 | assert not hasattr(self, '_TESTS'), \ | |
3415 | '%s has _TEST and _TESTS' % type(self).__name__ | |
3416 | tests = [t] | |
3417 | else: | |
3418 | tests = getattr(self, '_TESTS', []) | |
3419 | for t in tests: | |
3420 | if not include_onlymatching and t.get('only_matching', False): | |
3421 | continue | |
3422 | t['name'] = type(self).__name__[:-len('IE')] | |
3423 | yield t | |
3424 | ||
3425 | def is_suitable(self, age_limit): | |
3426 | """ Test whether the extractor is generally suitable for the given | |
3427 | age limit (i.e. pornographic sites are not, all others usually are) """ | |
3428 | ||
3429 | any_restricted = False | |
3430 | for tc in self.get_testcases(include_onlymatching=False): | |
40090e8d | 3431 | if tc.get('playlist', []): |
05900629 PH |
3432 | tc = tc['playlist'][0] |
3433 | is_restricted = age_restricted( | |
3434 | tc.get('info_dict', {}).get('age_limit'), age_limit) | |
3435 | if not is_restricted: | |
3436 | return True | |
3437 | any_restricted = any_restricted or is_restricted | |
3438 | return not any_restricted | |
3439 | ||
a504ced0 | 3440 | def extract_subtitles(self, *args, **kwargs): |
a06916d9 | 3441 | if (self.get_param('writesubtitles', False) |
3442 | or self.get_param('listsubtitles')): | |
9868ea49 JMF |
3443 | return self._get_subtitles(*args, **kwargs) |
3444 | return {} | |
a504ced0 JMF |
3445 | |
3446 | def _get_subtitles(self, *args, **kwargs): | |
611c1dd9 | 3447 | raise NotImplementedError('This method must be implemented by subclasses') |
a504ced0 | 3448 | |
912e0b7e YCH |
3449 | @staticmethod |
3450 | def _merge_subtitle_items(subtitle_list1, subtitle_list2): | |
3451 | """ Merge subtitle items for one language. Items with duplicated URLs | |
3452 | will be dropped. """ | |
3453 | list1_urls = set([item['url'] for item in subtitle_list1]) | |
3454 | ret = list(subtitle_list1) | |
3455 | ret.extend([item for item in subtitle_list2 if item['url'] not in list1_urls]) | |
3456 | return ret | |
3457 | ||
3458 | @classmethod | |
46890374 | 3459 | def _merge_subtitles(cls, *dicts, target=None): |
19bb3920 | 3460 | """ Merge subtitle dictionaries, language by language. """ |
19bb3920 F |
3461 | if target is None: |
3462 | target = {} | |
3463 | for d in dicts: | |
3464 | for lang, subs in d.items(): | |
3465 | target[lang] = cls._merge_subtitle_items(target.get(lang, []), subs) | |
3466 | return target | |
912e0b7e | 3467 | |
360e1ca5 | 3468 | def extract_automatic_captions(self, *args, **kwargs): |
a06916d9 | 3469 | if (self.get_param('writeautomaticsub', False) |
3470 | or self.get_param('listsubtitles')): | |
9868ea49 JMF |
3471 | return self._get_automatic_captions(*args, **kwargs) |
3472 | return {} | |
360e1ca5 JMF |
3473 | |
3474 | def _get_automatic_captions(self, *args, **kwargs): | |
611c1dd9 | 3475 | raise NotImplementedError('This method must be implemented by subclasses') |
360e1ca5 | 3476 | |
d77ab8e2 | 3477 | def mark_watched(self, *args, **kwargs): |
a06916d9 | 3478 | if (self.get_param('mark_watched', False) |
3089bc74 | 3479 | and (self._get_login_info()[0] is not None |
a06916d9 | 3480 | or self.get_param('cookiefile') is not None)): |
d77ab8e2 S |
3481 | self._mark_watched(*args, **kwargs) |
3482 | ||
3483 | def _mark_watched(self, *args, **kwargs): | |
3484 | raise NotImplementedError('This method must be implemented by subclasses') | |
3485 | ||
38cce791 YCH |
3486 | def geo_verification_headers(self): |
3487 | headers = {} | |
a06916d9 | 3488 | geo_verification_proxy = self.get_param('geo_verification_proxy') |
38cce791 YCH |
3489 | if geo_verification_proxy: |
3490 | headers['Ytdl-request-proxy'] = geo_verification_proxy | |
3491 | return headers | |
3492 | ||
98763ee3 YCH |
3493 | def _generic_id(self, url): |
3494 | return compat_urllib_parse_unquote(os.path.splitext(url.rstrip('/').split('/')[-1])[0]) | |
3495 | ||
3496 | def _generic_title(self, url): | |
3497 | return compat_urllib_parse_unquote(os.path.splitext(url_basename(url))[0]) | |
3498 | ||
c224251a | 3499 | @staticmethod |
b0089e89 | 3500 | def _availability(is_private=None, needs_premium=None, needs_subscription=None, needs_auth=None, is_unlisted=None): |
c224251a M |
3501 | all_known = all(map( |
3502 | lambda x: x is not None, | |
3503 | (is_private, needs_premium, needs_subscription, needs_auth, is_unlisted))) | |
3504 | return ( | |
3505 | 'private' if is_private | |
3506 | else 'premium_only' if needs_premium | |
3507 | else 'subscriber_only' if needs_subscription | |
3508 | else 'needs_auth' if needs_auth | |
3509 | else 'unlisted' if is_unlisted | |
3510 | else 'public' if all_known | |
3511 | else None) | |
3512 | ||
4bb6b02f | 3513 | def _configuration_arg(self, key, default=NO_DEFAULT, casesense=False): |
3514 | ''' | |
3515 | @returns A list of values for the extractor argument given by "key" | |
3516 | or "default" if no such key is present | |
3517 | @param default The default value to return when the key is not present (default: []) | |
3518 | @param casesense When false, the values are converted to lower case | |
3519 | ''' | |
3520 | val = traverse_obj( | |
5d3a0e79 | 3521 | self._downloader.params, ('extractor_args', self.ie_key().lower(), key)) |
4bb6b02f | 3522 | if val is None: |
3523 | return [] if default is NO_DEFAULT else default | |
3524 | return list(val) if casesense else [x.lower() for x in val] | |
5d3a0e79 | 3525 | |
8dbe9899 | 3526 | |
d6983cb4 PH |
3527 | class SearchInfoExtractor(InfoExtractor): |
3528 | """ | |
3529 | Base class for paged search queries extractors. | |
10952eb2 | 3530 | They accept URLs in the format _SEARCH_KEY(|all|[0-9]):{query} |
d6983cb4 PH |
3531 | Instances should define _SEARCH_KEY and _MAX_RESULTS. |
3532 | """ | |
3533 | ||
3534 | @classmethod | |
3535 | def _make_valid_url(cls): | |
3536 | return r'%s(?P<prefix>|[1-9][0-9]*|all):(?P<query>[\s\S]+)' % cls._SEARCH_KEY | |
3537 | ||
3538 | @classmethod | |
3539 | def suitable(cls, url): | |
3540 | return re.match(cls._make_valid_url(), url) is not None | |
3541 | ||
3542 | def _real_extract(self, query): | |
3543 | mobj = re.match(self._make_valid_url(), query) | |
3544 | if mobj is None: | |
f1a9d64e | 3545 | raise ExtractorError('Invalid search query "%s"' % query) |
d6983cb4 PH |
3546 | |
3547 | prefix = mobj.group('prefix') | |
3548 | query = mobj.group('query') | |
3549 | if prefix == '': | |
3550 | return self._get_n_results(query, 1) | |
3551 | elif prefix == 'all': | |
3552 | return self._get_n_results(query, self._MAX_RESULTS) | |
3553 | else: | |
3554 | n = int(prefix) | |
3555 | if n <= 0: | |
f1a9d64e | 3556 | raise ExtractorError('invalid download number %s for query "%s"' % (n, query)) |
d6983cb4 | 3557 | elif n > self._MAX_RESULTS: |
6a39ee13 | 3558 | self.report_warning('%s returns max %i results (you requested %i)' % (self._SEARCH_KEY, self._MAX_RESULTS, n)) |
d6983cb4 PH |
3559 | n = self._MAX_RESULTS |
3560 | return self._get_n_results(query, n) | |
3561 | ||
3562 | def _get_n_results(self, query, n): | |
3563 | """Get a specified number of results for a query""" | |
611c1dd9 | 3564 | raise NotImplementedError('This method must be implemented by subclasses') |
0f818663 PH |
3565 | |
3566 | @property | |
3567 | def SEARCH_KEY(self): | |
3568 | return self._SEARCH_KEY |