]>
Commit | Line | Data |
---|---|---|
f1a9d64e PH |
1 | from __future__ import unicode_literals |
2 | ||
d6983cb4 | 3 | import base64 |
f4b1c7ad | 4 | import datetime |
3ec05685 | 5 | import hashlib |
3d3538e4 | 6 | import json |
4094b6e3 | 7 | import netrc |
d6983cb4 PH |
8 | import os |
9 | import re | |
10 | import socket | |
11 | import sys | |
4094b6e3 | 12 | import time |
267ed0c5 | 13 | import xml.etree.ElementTree |
d6983cb4 | 14 | |
8c25f81b | 15 | from ..compat import ( |
42939b61 | 16 | compat_cookiejar, |
799207e8 | 17 | compat_cookies, |
96a53167 | 18 | compat_HTTPError, |
d6983cb4 PH |
19 | compat_http_client, |
20 | compat_urllib_error, | |
c7deaa4c | 21 | compat_urllib_parse_urlparse, |
799207e8 | 22 | compat_urllib_request, |
f0b5d6af | 23 | compat_urlparse, |
d6983cb4 | 24 | compat_str, |
8c25f81b PH |
25 | ) |
26 | from ..utils import ( | |
c342041f | 27 | NO_DEFAULT, |
05900629 | 28 | age_restricted, |
08f2a92c | 29 | bug_reports_message, |
d6983cb4 PH |
30 | clean_html, |
31 | compiled_regex_type, | |
70f0f5a8 | 32 | determine_ext, |
d6983cb4 | 33 | ExtractorError, |
97f4aecf | 34 | fix_xml_ampersands, |
b14f3a4c | 35 | float_or_none, |
31bb8d3f | 36 | int_or_none, |
55b3e45b | 37 | RegexNotFoundError, |
d41e6efc | 38 | sanitize_filename, |
f38de77f | 39 | unescapeHTML, |
d6983cb4 | 40 | ) |
c342041f | 41 | |
d6983cb4 PH |
42 | |
43 | class InfoExtractor(object): | |
44 | """Information Extractor class. | |
45 | ||
46 | Information extractors are the classes that, given a URL, extract | |
47 | information about the video (or videos) the URL refers to. This | |
48 | information includes the real video URL, the video title, author and | |
49 | others. The information is stored in a dictionary which is then | |
5d380852 | 50 | passed to the YoutubeDL. The YoutubeDL processes this |
d6983cb4 PH |
51 | information possibly downloading the video to the file system, among |
52 | other possible outcomes. | |
53 | ||
cf0649f8 | 54 | The type field determines the type of the result. |
fed5d032 PH |
55 | By far the most common value (and the default if _type is missing) is |
56 | "video", which indicates a single video. | |
57 | ||
58 | For a video, the dictionaries must include the following fields: | |
d6983cb4 PH |
59 | |
60 | id: Video identifier. | |
d6983cb4 | 61 | title: Video title, unescaped. |
d67b0b15 | 62 | |
f49d89ee | 63 | Additionally, it must contain either a formats entry or a url one: |
d67b0b15 | 64 | |
f49d89ee PH |
65 | formats: A list of dictionaries for each format available, ordered |
66 | from worst to best quality. | |
67 | ||
68 | Potential fields: | |
d67b0b15 | 69 | * url Mandatory. The URL of the video file |
10952eb2 | 70 | * ext Will be calculated from URL if missing |
d67b0b15 PH |
71 | * format A human-readable description of the format |
72 | ("mp4 container with h264/opus"). | |
73 | Calculated from the format_id, width, height. | |
74 | and format_note fields if missing. | |
75 | * format_id A short description of the format | |
5d4f3985 PH |
76 | ("mp4_h264_opus" or "19"). |
77 | Technically optional, but strongly recommended. | |
d67b0b15 PH |
78 | * format_note Additional info about the format |
79 | ("3D" or "DASH video") | |
80 | * width Width of the video, if known | |
81 | * height Height of the video, if known | |
f49d89ee | 82 | * resolution Textual description of width and height |
7217e148 | 83 | * tbr Average bitrate of audio and video in KBit/s |
d67b0b15 PH |
84 | * abr Average audio bitrate in KBit/s |
85 | * acodec Name of the audio codec in use | |
dd27fd17 | 86 | * asr Audio sampling rate in Hertz |
d67b0b15 | 87 | * vbr Average video bitrate in KBit/s |
fbb21cf5 | 88 | * fps Frame rate |
d67b0b15 | 89 | * vcodec Name of the video codec in use |
1394ce65 | 90 | * container Name of the container format |
d67b0b15 | 91 | * filesize The number of bytes, if known in advance |
9732d77e | 92 | * filesize_approx An estimate for the number of bytes |
d67b0b15 | 93 | * player_url SWF Player URL (used for rtmpdump). |
c7deaa4c PH |
94 | * protocol The protocol that will be used for the actual |
95 | download, lower-case. | |
b04b8852 PH |
96 | "http", "https", "rtsp", "rtmp", "rtmpe", |
97 | "m3u8", or "m3u8_native". | |
f49d89ee | 98 | * preference Order number of this format. If this field is |
08d13955 | 99 | present and not None, the formats get sorted |
38d63d84 | 100 | by this field, regardless of all other values. |
f49d89ee PH |
101 | -1 for default (order by other properties), |
102 | -2 or smaller for less than default. | |
e65566a9 PH |
103 | < -1000 to hide the format (if there is |
104 | another one which is strictly better) | |
aff2f4f4 PH |
105 | * language_preference Is this in the correct requested |
106 | language? | |
107 | 10 if it's what the URL is about, | |
108 | -1 for default (don't know), | |
109 | -10 otherwise, other values reserved for now. | |
5d73273f PH |
110 | * quality Order number of the video quality of this |
111 | format, irrespective of the file format. | |
112 | -1 for default (order by other properties), | |
113 | -2 or smaller for less than default. | |
c64ed2a3 PH |
114 | * source_preference Order number for this video source |
115 | (quality takes higher priority) | |
116 | -1 for default (order by other properties), | |
117 | -2 or smaller for less than default. | |
d769be6c PH |
118 | * http_headers A dictionary of additional HTTP headers |
119 | to add to the request. | |
6271f1ca | 120 | * stretched_ratio If given and not 1, indicates that the |
3dee7826 PH |
121 | video's pixels are not square. |
122 | width : height ratio as float. | |
123 | * no_resume The server does not support resuming the | |
124 | (HTTP or RTMP) download. Boolean. | |
125 | ||
c0ba0f48 | 126 | url: Final video URL. |
d6983cb4 | 127 | ext: Video filename extension. |
d67b0b15 PH |
128 | format: The video format, defaults to ext (used for --get-format) |
129 | player_url: SWF Player URL (used for rtmpdump). | |
2f5865cc | 130 | |
d6983cb4 PH |
131 | The following fields are optional: |
132 | ||
f5e43bc6 | 133 | alt_title: A secondary title of the video. |
0afef30b PH |
134 | display_id An alternative identifier for the video, not necessarily |
135 | unique, but available before title. Typically, id is | |
136 | something like "4234987", title "Dancing naked mole rats", | |
137 | and display_id "dancing-naked-mole-rats" | |
d5519808 | 138 | thumbnails: A list of dictionaries, with the following entries: |
cfb56d1a | 139 | * "id" (optional, string) - Thumbnail format ID |
d5519808 | 140 | * "url" |
cfb56d1a | 141 | * "preference" (optional, int) - quality of the image |
d5519808 PH |
142 | * "width" (optional, int) |
143 | * "height" (optional, int) | |
144 | * "resolution" (optional, string "{width}x{height"}, | |
145 | deprecated) | |
d6983cb4 | 146 | thumbnail: Full URL to a video thumbnail image. |
f5e43bc6 | 147 | description: Full video description. |
d6983cb4 | 148 | uploader: Full name of the video uploader. |
9bb8e0a3 | 149 | creator: The main artist who created the video. |
955c4514 | 150 | timestamp: UNIX timestamp of the moment the video became available. |
d6983cb4 | 151 | upload_date: Video upload date (YYYYMMDD). |
955c4514 | 152 | If not explicitly set, calculated from timestamp. |
d6983cb4 | 153 | uploader_id: Nickname or id of the video uploader. |
da9ec3b9 | 154 | location: Physical location where the video was filmed. |
a504ced0 JMF |
155 | subtitles: The available subtitles as a dictionary in the format |
156 | {language: subformats}. "subformats" is a list sorted from | |
157 | lower to higher preference, each element is a dictionary | |
158 | with the "ext" entry and one of: | |
159 | * "data": The subtitles file contents | |
10952eb2 | 160 | * "url": A URL pointing to the subtitles file |
360e1ca5 JMF |
161 | automatic_captions: Like 'subtitles', used by the YoutubeIE for |
162 | automatically generated captions | |
c0ba0f48 | 163 | duration: Length of the video in seconds, as an integer. |
f3d29461 | 164 | view_count: How many users have watched the video on the platform. |
19e3dfc9 PH |
165 | like_count: Number of positive ratings of the video |
166 | dislike_count: Number of negative ratings of the video | |
2d30521a | 167 | average_rating: Average rating give by users, the scale used depends on the webpage |
19e3dfc9 | 168 | comment_count: Number of comments on the video |
dd622d7c PH |
169 | comments: A list of comments, each with one or more of the following |
170 | properties (all but one of text or html optional): | |
171 | * "author" - human-readable name of the comment author | |
172 | * "author_id" - user ID of the comment author | |
173 | * "id" - Comment ID | |
174 | * "html" - Comment as HTML | |
175 | * "text" - Plain text of the comment | |
176 | * "timestamp" - UNIX timestamp of comment | |
177 | * "parent" - ID of the comment this one is replying to. | |
178 | Set to "root" to indicate that this is a | |
179 | comment to the original video. | |
8dbe9899 | 180 | age_limit: Age restriction for the video, as an integer (years) |
10952eb2 | 181 | webpage_url: The URL to the video webpage, if given to youtube-dl it |
9103bbc5 JMF |
182 | should allow to get the same result again. (It will be set |
183 | by YoutubeDL if it's missing) | |
ad3bc6ac PH |
184 | categories: A list of categories that the video falls in, for example |
185 | ["Sports", "Berlin"] | |
864f24bd | 186 | tags: A list of tags assigned to the video, e.g. ["sweden", "pop music"] |
7267bd53 PH |
187 | is_live: True, False, or None (=unknown). Whether this video is a |
188 | live stream that goes on instead of a fixed-length video. | |
7c80519c | 189 | start_time: Time in seconds where the reproduction should start, as |
10952eb2 | 190 | specified in the URL. |
297a564b | 191 | end_time: Time in seconds where the reproduction should end, as |
10952eb2 | 192 | specified in the URL. |
d6983cb4 | 193 | |
deefc05b | 194 | Unless mentioned otherwise, the fields should be Unicode strings. |
d6983cb4 | 195 | |
d838b1bd PH |
196 | Unless mentioned otherwise, None is equivalent to absence of information. |
197 | ||
fed5d032 PH |
198 | |
199 | _type "playlist" indicates multiple videos. | |
b82f815f PH |
200 | There must be a key "entries", which is a list, an iterable, or a PagedList |
201 | object, each element of which is a valid dictionary by this specification. | |
fed5d032 PH |
202 | |
203 | Additionally, playlists can have "title" and "id" attributes with the same | |
204 | semantics as videos (see above). | |
205 | ||
206 | ||
207 | _type "multi_video" indicates that there are multiple videos that | |
208 | form a single show, for examples multiple acts of an opera or TV episode. | |
209 | It must have an entries key like a playlist and contain all the keys | |
210 | required for a video at the same time. | |
211 | ||
212 | ||
213 | _type "url" indicates that the video must be extracted from another | |
214 | location, possibly by a different extractor. Its only required key is: | |
215 | "url" - the next URL to extract. | |
f58766ce PH |
216 | The key "ie_key" can be set to the class name (minus the trailing "IE", |
217 | e.g. "Youtube") if the extractor class is known in advance. | |
218 | Additionally, the dictionary may have any properties of the resolved entity | |
219 | known in advance, for example "title" if the title of the referred video is | |
fed5d032 PH |
220 | known ahead of time. |
221 | ||
222 | ||
223 | _type "url_transparent" entities have the same specification as "url", but | |
224 | indicate that the given additional information is more precise than the one | |
225 | associated with the resolved URL. | |
226 | This is useful when a site employs a video service that hosts the video and | |
227 | its technical metadata, but that video service does not embed a useful | |
228 | title, description etc. | |
229 | ||
230 | ||
d6983cb4 PH |
231 | Subclasses of this one should re-define the _real_initialize() and |
232 | _real_extract() methods and define a _VALID_URL regexp. | |
233 | Probably, they should also be added to the list of extractors. | |
234 | ||
d6983cb4 PH |
235 | Finally, the _WORKING attribute should be set to False for broken IEs |
236 | in order to warn the users and skip the tests. | |
237 | """ | |
238 | ||
239 | _ready = False | |
240 | _downloader = None | |
241 | _WORKING = True | |
242 | ||
243 | def __init__(self, downloader=None): | |
244 | """Constructor. Receives an optional downloader.""" | |
245 | self._ready = False | |
246 | self.set_downloader(downloader) | |
247 | ||
248 | @classmethod | |
249 | def suitable(cls, url): | |
250 | """Receives a URL and returns True if suitable for this IE.""" | |
79cb2577 PH |
251 | |
252 | # This does not use has/getattr intentionally - we want to know whether | |
253 | # we have cached the regexp for *this* class, whereas getattr would also | |
254 | # match the superclass | |
255 | if '_VALID_URL_RE' not in cls.__dict__: | |
256 | cls._VALID_URL_RE = re.compile(cls._VALID_URL) | |
257 | return cls._VALID_URL_RE.match(url) is not None | |
d6983cb4 | 258 | |
ed9266db PH |
259 | @classmethod |
260 | def _match_id(cls, url): | |
261 | if '_VALID_URL_RE' not in cls.__dict__: | |
262 | cls._VALID_URL_RE = re.compile(cls._VALID_URL) | |
263 | m = cls._VALID_URL_RE.match(url) | |
264 | assert m | |
265 | return m.group('id') | |
266 | ||
d6983cb4 PH |
267 | @classmethod |
268 | def working(cls): | |
269 | """Getter method for _WORKING.""" | |
270 | return cls._WORKING | |
271 | ||
272 | def initialize(self): | |
273 | """Initializes an instance (authentication, etc).""" | |
274 | if not self._ready: | |
275 | self._real_initialize() | |
276 | self._ready = True | |
277 | ||
278 | def extract(self, url): | |
279 | """Extracts URL information and returns it in list of dicts.""" | |
3a5bcd03 PH |
280 | try: |
281 | self.initialize() | |
282 | return self._real_extract(url) | |
283 | except ExtractorError: | |
284 | raise | |
285 | except compat_http_client.IncompleteRead as e: | |
286 | raise ExtractorError('A network error has occured.', cause=e, expected=True) | |
9650885b | 287 | except (KeyError, StopIteration) as e: |
3a5bcd03 | 288 | raise ExtractorError('An extractor error has occured.', cause=e) |
d6983cb4 PH |
289 | |
290 | def set_downloader(self, downloader): | |
291 | """Sets the downloader for this IE.""" | |
292 | self._downloader = downloader | |
293 | ||
294 | def _real_initialize(self): | |
295 | """Real initialization process. Redefine in subclasses.""" | |
296 | pass | |
297 | ||
298 | def _real_extract(self, url): | |
299 | """Real extraction process. Redefine in subclasses.""" | |
300 | pass | |
301 | ||
56c73665 JMF |
302 | @classmethod |
303 | def ie_key(cls): | |
304 | """A string for getting the InfoExtractor with get_info_extractor""" | |
305 | return cls.__name__[:-2] | |
306 | ||
d6983cb4 PH |
307 | @property |
308 | def IE_NAME(self): | |
309 | return type(self).__name__[:-2] | |
310 | ||
7cc3570e | 311 | def _request_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True): |
d6983cb4 PH |
312 | """ Returns the response handle """ |
313 | if note is None: | |
314 | self.report_download_webpage(video_id) | |
315 | elif note is not False: | |
7cc3570e | 316 | if video_id is None: |
f1a9d64e | 317 | self.to_screen('%s' % (note,)) |
7cc3570e | 318 | else: |
f1a9d64e | 319 | self.to_screen('%s: %s' % (video_id, note)) |
d6983cb4 | 320 | try: |
dca08720 | 321 | return self._downloader.urlopen(url_or_request) |
d6983cb4 | 322 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: |
aa94a6d3 PH |
323 | if errnote is False: |
324 | return False | |
d6983cb4 | 325 | if errnote is None: |
f1a9d64e PH |
326 | errnote = 'Unable to download webpage' |
327 | errmsg = '%s: %s' % (errnote, compat_str(err)) | |
7cc3570e PH |
328 | if fatal: |
329 | raise ExtractorError(errmsg, sys.exc_info()[2], cause=err) | |
330 | else: | |
331 | self._downloader.report_warning(errmsg) | |
332 | return False | |
d6983cb4 | 333 | |
c9a77969 | 334 | def _download_webpage_handle(self, url_or_request, video_id, note=None, errnote=None, fatal=True, encoding=None): |
d6983cb4 | 335 | """ Returns a tuple (page content as string, URL handle) """ |
b9d3e163 PH |
336 | # Strip hashes from the URL (#1038) |
337 | if isinstance(url_or_request, (compat_str, str)): | |
338 | url_or_request = url_or_request.partition('#')[0] | |
339 | ||
7cc3570e PH |
340 | urlh = self._request_webpage(url_or_request, video_id, note, errnote, fatal) |
341 | if urlh is False: | |
342 | assert not fatal | |
343 | return False | |
c9a77969 | 344 | content = self._webpage_read_content(urlh, url_or_request, video_id, note, errnote, fatal, encoding=encoding) |
23be51d8 PH |
345 | return (content, urlh) |
346 | ||
c9a77969 YCH |
347 | @staticmethod |
348 | def _guess_encoding_from_content(content_type, webpage_bytes): | |
d6983cb4 PH |
349 | m = re.match(r'[a-zA-Z0-9_.-]+/[a-zA-Z0-9_.-]+\s*;\s*charset=(.+)', content_type) |
350 | if m: | |
351 | encoding = m.group(1) | |
352 | else: | |
0d75ae2c | 353 | m = re.search(br'<meta[^>]+charset=[\'"]?([^\'")]+)[ /\'">]', |
f143d86a PH |
354 | webpage_bytes[:1024]) |
355 | if m: | |
356 | encoding = m.group(1).decode('ascii') | |
b60016e8 PH |
357 | elif webpage_bytes.startswith(b'\xff\xfe'): |
358 | encoding = 'utf-16' | |
f143d86a PH |
359 | else: |
360 | encoding = 'utf-8' | |
c9a77969 YCH |
361 | |
362 | return encoding | |
363 | ||
364 | def _webpage_read_content(self, urlh, url_or_request, video_id, note=None, errnote=None, fatal=True, prefix=None, encoding=None): | |
365 | content_type = urlh.headers.get('Content-Type', '') | |
366 | webpage_bytes = urlh.read() | |
367 | if prefix is not None: | |
368 | webpage_bytes = prefix + webpage_bytes | |
369 | if not encoding: | |
370 | encoding = self._guess_encoding_from_content(content_type, webpage_bytes) | |
d6983cb4 PH |
371 | if self._downloader.params.get('dump_intermediate_pages', False): |
372 | try: | |
373 | url = url_or_request.get_full_url() | |
374 | except AttributeError: | |
375 | url = url_or_request | |
f1a9d64e | 376 | self.to_screen('Dumping request to ' + url) |
d6983cb4 PH |
377 | dump = base64.b64encode(webpage_bytes).decode('ascii') |
378 | self._downloader.to_screen(dump) | |
d41e6efc PH |
379 | if self._downloader.params.get('write_pages', False): |
380 | try: | |
381 | url = url_or_request.get_full_url() | |
382 | except AttributeError: | |
383 | url = url_or_request | |
5afa7f8b | 384 | basen = '%s_%s' % (video_id, url) |
c1bce22f | 385 | if len(basen) > 240: |
f1a9d64e | 386 | h = '___' + hashlib.md5(basen.encode('utf-8')).hexdigest() |
c1bce22f PH |
387 | basen = basen[:240 - len(h)] + h |
388 | raw_filename = basen + '.dump' | |
d41e6efc | 389 | filename = sanitize_filename(raw_filename, restricted=True) |
f1a9d64e | 390 | self.to_screen('Saving request to ' + filename) |
5f58165d S |
391 | # Working around MAX_PATH limitation on Windows (see |
392 | # http://msdn.microsoft.com/en-us/library/windows/desktop/aa365247(v=vs.85).aspx) | |
393 | if os.name == 'nt': | |
394 | absfilepath = os.path.abspath(filename) | |
395 | if len(absfilepath) > 259: | |
396 | filename = '\\\\?\\' + absfilepath | |
d41e6efc PH |
397 | with open(filename, 'wb') as outf: |
398 | outf.write(webpage_bytes) | |
399 | ||
ec0fafbb AA |
400 | try: |
401 | content = webpage_bytes.decode(encoding, 'replace') | |
402 | except LookupError: | |
403 | content = webpage_bytes.decode('utf-8', 'replace') | |
2410c43d | 404 | |
f1a9d64e PH |
405 | if ('<title>Access to this site is blocked</title>' in content and |
406 | 'Websense' in content[:512]): | |
407 | msg = 'Access to this webpage has been blocked by Websense filtering software in your network.' | |
2410c43d PH |
408 | blocked_iframe = self._html_search_regex( |
409 | r'<iframe src="([^"]+)"', content, | |
f1a9d64e | 410 | 'Websense information URL', default=None) |
2410c43d | 411 | if blocked_iframe: |
f1a9d64e | 412 | msg += ' Visit %s for more details' % blocked_iframe |
2410c43d | 413 | raise ExtractorError(msg, expected=True) |
77b2986b PH |
414 | if '<title>The URL you requested has been blocked</title>' in content[:512]: |
415 | msg = ( | |
416 | 'Access to this webpage has been blocked by Indian censorship. ' | |
417 | 'Use a VPN or proxy server (with --proxy) to route around it.') | |
418 | block_msg = self._html_search_regex( | |
419 | r'</h1><p>(.*?)</p>', | |
420 | content, 'block message', default=None) | |
421 | if block_msg: | |
422 | msg += ' (Message: "%s")' % block_msg.replace('\n', ' ') | |
423 | raise ExtractorError(msg, expected=True) | |
2410c43d | 424 | |
23be51d8 | 425 | return content |
d6983cb4 | 426 | |
c9a77969 | 427 | def _download_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True, tries=1, timeout=5, encoding=None): |
d6983cb4 | 428 | """ Returns the data of the page as a string """ |
995ad69c TF |
429 | success = False |
430 | try_count = 0 | |
431 | while success is False: | |
432 | try: | |
c9a77969 | 433 | res = self._download_webpage_handle(url_or_request, video_id, note, errnote, fatal, encoding=encoding) |
995ad69c TF |
434 | success = True |
435 | except compat_http_client.IncompleteRead as e: | |
436 | try_count += 1 | |
437 | if try_count >= tries: | |
438 | raise e | |
439 | self._sleep(timeout, video_id) | |
7cc3570e PH |
440 | if res is False: |
441 | return res | |
442 | else: | |
443 | content, _ = res | |
444 | return content | |
d6983cb4 | 445 | |
2a275ab0 | 446 | def _download_xml(self, url_or_request, video_id, |
f1a9d64e | 447 | note='Downloading XML', errnote='Unable to download XML', |
c9a77969 | 448 | transform_source=None, fatal=True, encoding=None): |
267ed0c5 | 449 | """Return the xml as an xml.etree.ElementTree.Element""" |
28746fbd | 450 | xml_string = self._download_webpage( |
c9a77969 | 451 | url_or_request, video_id, note, errnote, fatal=fatal, encoding=encoding) |
28746fbd PH |
452 | if xml_string is False: |
453 | return xml_string | |
e2b38da9 PH |
454 | if transform_source: |
455 | xml_string = transform_source(xml_string) | |
267ed0c5 JMF |
456 | return xml.etree.ElementTree.fromstring(xml_string.encode('utf-8')) |
457 | ||
3d3538e4 | 458 | def _download_json(self, url_or_request, video_id, |
f1a9d64e PH |
459 | note='Downloading JSON metadata', |
460 | errnote='Unable to download JSON metadata', | |
b090af59 | 461 | transform_source=None, |
c9a77969 | 462 | fatal=True, encoding=None): |
b090af59 | 463 | json_string = self._download_webpage( |
c9a77969 YCH |
464 | url_or_request, video_id, note, errnote, fatal=fatal, |
465 | encoding=encoding) | |
b090af59 PH |
466 | if (not fatal) and json_string is False: |
467 | return None | |
ebb64199 TF |
468 | return self._parse_json( |
469 | json_string, video_id, transform_source=transform_source, fatal=fatal) | |
470 | ||
471 | def _parse_json(self, json_string, video_id, transform_source=None, fatal=True): | |
81c2f20b PH |
472 | if transform_source: |
473 | json_string = transform_source(json_string) | |
3d3538e4 PH |
474 | try: |
475 | return json.loads(json_string) | |
476 | except ValueError as ve: | |
e7b6d122 PH |
477 | errmsg = '%s: Failed to parse JSON ' % video_id |
478 | if fatal: | |
479 | raise ExtractorError(errmsg, cause=ve) | |
480 | else: | |
481 | self.report_warning(errmsg + str(ve)) | |
3d3538e4 | 482 | |
f45f96f8 | 483 | def report_warning(self, msg, video_id=None): |
f1a9d64e | 484 | idstr = '' if video_id is None else '%s: ' % video_id |
f45f96f8 | 485 | self._downloader.report_warning( |
f1a9d64e | 486 | '[%s] %s%s' % (self.IE_NAME, idstr, msg)) |
f45f96f8 | 487 | |
d6983cb4 PH |
488 | def to_screen(self, msg): |
489 | """Print msg to screen, prefixing it with '[ie_name]'""" | |
f1a9d64e | 490 | self._downloader.to_screen('[%s] %s' % (self.IE_NAME, msg)) |
d6983cb4 PH |
491 | |
492 | def report_extraction(self, id_or_name): | |
493 | """Report information extraction.""" | |
f1a9d64e | 494 | self.to_screen('%s: Extracting information' % id_or_name) |
d6983cb4 PH |
495 | |
496 | def report_download_webpage(self, video_id): | |
497 | """Report webpage download.""" | |
f1a9d64e | 498 | self.to_screen('%s: Downloading webpage' % video_id) |
d6983cb4 PH |
499 | |
500 | def report_age_confirmation(self): | |
501 | """Report attempt to confirm age.""" | |
f1a9d64e | 502 | self.to_screen('Confirming age') |
d6983cb4 | 503 | |
fc79158d JMF |
504 | def report_login(self): |
505 | """Report attempt to log in.""" | |
f1a9d64e | 506 | self.to_screen('Logging in') |
fc79158d | 507 | |
5f6a1245 | 508 | # Methods for following #608 |
c0d0b01f | 509 | @staticmethod |
830d53bf | 510 | def url_result(url, ie=None, video_id=None, video_title=None): |
10952eb2 | 511 | """Returns a URL that points to a page that should be processed""" |
5f6a1245 | 512 | # TODO: ie should be the class used for getting the info |
d6983cb4 PH |
513 | video_info = {'_type': 'url', |
514 | 'url': url, | |
515 | 'ie_key': ie} | |
7012b23c PH |
516 | if video_id is not None: |
517 | video_info['id'] = video_id | |
830d53bf S |
518 | if video_title is not None: |
519 | video_info['title'] = video_title | |
d6983cb4 | 520 | return video_info |
5f6a1245 | 521 | |
c0d0b01f | 522 | @staticmethod |
acf5cbfe | 523 | def playlist_result(entries, playlist_id=None, playlist_title=None, playlist_description=None): |
d6983cb4 PH |
524 | """Returns a playlist""" |
525 | video_info = {'_type': 'playlist', | |
526 | 'entries': entries} | |
527 | if playlist_id: | |
528 | video_info['id'] = playlist_id | |
529 | if playlist_title: | |
530 | video_info['title'] = playlist_title | |
acf5cbfe S |
531 | if playlist_description: |
532 | video_info['description'] = playlist_description | |
d6983cb4 PH |
533 | return video_info |
534 | ||
c342041f | 535 | def _search_regex(self, pattern, string, name, default=NO_DEFAULT, fatal=True, flags=0, group=None): |
d6983cb4 PH |
536 | """ |
537 | Perform a regex search on the given string, using a single or a list of | |
538 | patterns returning the first matching group. | |
539 | In case of failure return a default value or raise a WARNING or a | |
55b3e45b | 540 | RegexNotFoundError, depending on fatal, specifying the field name. |
d6983cb4 PH |
541 | """ |
542 | if isinstance(pattern, (str, compat_str, compiled_regex_type)): | |
543 | mobj = re.search(pattern, string, flags) | |
544 | else: | |
545 | for p in pattern: | |
546 | mobj = re.search(p, string, flags) | |
c3415d1b PH |
547 | if mobj: |
548 | break | |
d6983cb4 | 549 | |
7e5db8c9 | 550 | if not self._downloader.params.get('no_color') and os.name != 'nt' and sys.stderr.isatty(): |
f1a9d64e | 551 | _name = '\033[0;34m%s\033[0m' % name |
d6983cb4 PH |
552 | else: |
553 | _name = name | |
554 | ||
555 | if mobj: | |
711ede6e PH |
556 | if group is None: |
557 | # return the first matching group | |
558 | return next(g for g in mobj.groups() if g is not None) | |
559 | else: | |
560 | return mobj.group(group) | |
c342041f | 561 | elif default is not NO_DEFAULT: |
d6983cb4 PH |
562 | return default |
563 | elif fatal: | |
f1a9d64e | 564 | raise RegexNotFoundError('Unable to extract %s' % _name) |
d6983cb4 | 565 | else: |
08f2a92c | 566 | self._downloader.report_warning('unable to extract %s' % _name + bug_reports_message()) |
d6983cb4 PH |
567 | return None |
568 | ||
c342041f | 569 | def _html_search_regex(self, pattern, string, name, default=NO_DEFAULT, fatal=True, flags=0, group=None): |
d6983cb4 PH |
570 | """ |
571 | Like _search_regex, but strips HTML tags and unescapes entities. | |
572 | """ | |
711ede6e | 573 | res = self._search_regex(pattern, string, name, default, fatal, flags, group) |
d6983cb4 PH |
574 | if res: |
575 | return clean_html(res).strip() | |
576 | else: | |
577 | return res | |
578 | ||
fc79158d JMF |
579 | def _get_login_info(self): |
580 | """ | |
cf0649f8 | 581 | Get the login info as (username, password) |
fc79158d JMF |
582 | It will look in the netrc file using the _NETRC_MACHINE value |
583 | If there's no info available, return (None, None) | |
584 | """ | |
585 | if self._downloader is None: | |
586 | return (None, None) | |
587 | ||
588 | username = None | |
589 | password = None | |
590 | downloader_params = self._downloader.params | |
591 | ||
592 | # Attempt to use provided username and password or .netrc data | |
593 | if downloader_params.get('username', None) is not None: | |
594 | username = downloader_params['username'] | |
595 | password = downloader_params['password'] | |
596 | elif downloader_params.get('usenetrc', False): | |
597 | try: | |
598 | info = netrc.netrc().authenticators(self._NETRC_MACHINE) | |
599 | if info is not None: | |
600 | username = info[0] | |
601 | password = info[2] | |
602 | else: | |
603 | raise netrc.NetrcParseError('No authenticators for %s' % self._NETRC_MACHINE) | |
604 | except (IOError, netrc.NetrcParseError) as err: | |
f1a9d64e | 605 | self._downloader.report_warning('parsing .netrc: %s' % compat_str(err)) |
5f6a1245 | 606 | |
fc79158d JMF |
607 | return (username, password) |
608 | ||
83317f69 | 609 | def _get_tfa_info(self): |
610 | """ | |
611 | Get the two-factor authentication info | |
612 | TODO - asking the user will be required for sms/phone verify | |
613 | currently just uses the command line option | |
614 | If there's no info available, return None | |
615 | """ | |
616 | if self._downloader is None: | |
83317f69 | 617 | return None |
618 | downloader_params = self._downloader.params | |
619 | ||
620 | if downloader_params.get('twofactor', None) is not None: | |
621 | return downloader_params['twofactor'] | |
622 | ||
83317f69 | 623 | return None |
624 | ||
46720279 JMF |
625 | # Helper functions for extracting OpenGraph info |
626 | @staticmethod | |
ab2d5247 | 627 | def _og_regexes(prop): |
c1206423 | 628 | content_re = r'content=(?:"([^>]+?)"|\'([^>]+?)\')' |
9887c9b2 | 629 | property_re = r'(?:name|property)=[\'"]og:%s[\'"]' % re.escape(prop) |
78fb87b2 | 630 | template = r'<meta[^>]+?%s[^>]+?%s' |
ab2d5247 | 631 | return [ |
78fb87b2 JMF |
632 | template % (property_re, content_re), |
633 | template % (content_re, property_re), | |
ab2d5247 | 634 | ] |
46720279 | 635 | |
864f24bd S |
636 | @staticmethod |
637 | def _meta_regex(prop): | |
638 | return r'''(?isx)<meta | |
639 | (?=[^>]+(?:itemprop|name|property)=(["\']?)%s\1) | |
640 | [^>]+?content=(["\'])(?P<content>.*?)\2''' % re.escape(prop) | |
641 | ||
3c4e6d83 | 642 | def _og_search_property(self, prop, html, name=None, **kargs): |
46720279 | 643 | if name is None: |
3c4e6d83 | 644 | name = 'OpenGraph %s' % prop |
ab2d5247 | 645 | escaped = self._search_regex(self._og_regexes(prop), html, name, flags=re.DOTALL, **kargs) |
eb0a8398 PH |
646 | if escaped is None: |
647 | return None | |
648 | return unescapeHTML(escaped) | |
46720279 JMF |
649 | |
650 | def _og_search_thumbnail(self, html, **kargs): | |
10952eb2 | 651 | return self._og_search_property('image', html, 'thumbnail URL', fatal=False, **kargs) |
46720279 JMF |
652 | |
653 | def _og_search_description(self, html, **kargs): | |
654 | return self._og_search_property('description', html, fatal=False, **kargs) | |
655 | ||
656 | def _og_search_title(self, html, **kargs): | |
657 | return self._og_search_property('title', html, **kargs) | |
658 | ||
8ffa13e0 | 659 | def _og_search_video_url(self, html, name='video url', secure=True, **kargs): |
a3681973 PH |
660 | regexes = self._og_regexes('video') + self._og_regexes('video:url') |
661 | if secure: | |
662 | regexes = self._og_regexes('video:secure_url') + regexes | |
8ffa13e0 | 663 | return self._html_search_regex(regexes, html, name, **kargs) |
46720279 | 664 | |
78338f71 JMF |
665 | def _og_search_url(self, html, **kargs): |
666 | return self._og_search_property('url', html, **kargs) | |
667 | ||
40c696e5 | 668 | def _html_search_meta(self, name, html, display_name=None, fatal=False, **kwargs): |
59040888 PH |
669 | if display_name is None: |
670 | display_name = name | |
671 | return self._html_search_regex( | |
864f24bd | 672 | self._meta_regex(name), |
711ede6e | 673 | html, display_name, fatal=fatal, group='content', **kwargs) |
59040888 PH |
674 | |
675 | def _dc_search_uploader(self, html): | |
676 | return self._html_search_meta('dc.creator', html, 'uploader') | |
677 | ||
8dbe9899 PH |
678 | def _rta_search(self, html): |
679 | # See http://www.rtalabel.org/index.php?content=howtofaq#single | |
680 | if re.search(r'(?ix)<meta\s+name="rating"\s+' | |
681 | r' content="RTA-5042-1996-1400-1577-RTA"', | |
682 | html): | |
683 | return 18 | |
684 | return 0 | |
685 | ||
59040888 PH |
686 | def _media_rating_search(self, html): |
687 | # See http://www.tjg-designs.com/WP/metadata-code-examples-adding-metadata-to-your-web-pages/ | |
688 | rating = self._html_search_meta('rating', html) | |
689 | ||
690 | if not rating: | |
691 | return None | |
692 | ||
693 | RATING_TABLE = { | |
694 | 'safe for kids': 0, | |
695 | 'general': 8, | |
696 | '14 years': 14, | |
697 | 'mature': 17, | |
698 | 'restricted': 19, | |
699 | } | |
700 | return RATING_TABLE.get(rating.lower(), None) | |
701 | ||
69319969 | 702 | def _family_friendly_search(self, html): |
6ca7732d | 703 | # See http://schema.org/VideoObject |
69319969 NJ |
704 | family_friendly = self._html_search_meta('isFamilyFriendly', html) |
705 | ||
706 | if not family_friendly: | |
707 | return None | |
708 | ||
709 | RATING_TABLE = { | |
710 | '1': 0, | |
711 | 'true': 0, | |
712 | '0': 18, | |
713 | 'false': 18, | |
714 | } | |
715 | return RATING_TABLE.get(family_friendly.lower(), None) | |
716 | ||
0c708f11 JMF |
717 | def _twitter_search_player(self, html): |
718 | return self._html_search_meta('twitter:player', html, | |
9e1a5b84 | 719 | 'twitter card player') |
0c708f11 | 720 | |
27713812 | 721 | @staticmethod |
f8da79f8 S |
722 | def _hidden_inputs(html): |
723 | return dict([ | |
724 | (input.group('name'), input.group('value')) for input in re.finditer( | |
725 | r'''(?x) | |
726 | <input\s+ | |
727 | type=(?P<q_hidden>["\'])hidden(?P=q_hidden)\s+ | |
728 | name=(?P<q_name>["\'])(?P<name>.+?)(?P=q_name)\s+ | |
729 | (?:id=(?P<q_id>["\']).+?(?P=q_id)\s+)? | |
730 | value=(?P<q_value>["\'])(?P<value>.*?)(?P=q_value) | |
731 | ''', html) | |
732 | ]) | |
27713812 | 733 | |
cf61d96d S |
734 | def _form_hidden_inputs(self, form_id, html): |
735 | form = self._search_regex( | |
736 | r'(?s)<form[^>]+?id=(["\'])%s\1[^>]*>(?P<form>.+?)</form>' % form_id, | |
737 | html, '%s form' % form_id, group='form') | |
738 | return self._hidden_inputs(form) | |
739 | ||
3ded7bac | 740 | def _sort_formats(self, formats, field_preference=None): |
7e8caf30 | 741 | if not formats: |
f1a9d64e | 742 | raise ExtractorError('No video formats found') |
7e8caf30 | 743 | |
4bcc7bd1 | 744 | def _formats_key(f): |
e6812ac9 PH |
745 | # TODO remove the following workaround |
746 | from ..utils import determine_ext | |
747 | if not f.get('ext') and 'url' in f: | |
748 | f['ext'] = determine_ext(f['url']) | |
749 | ||
3ded7bac S |
750 | if isinstance(field_preference, (list, tuple)): |
751 | return tuple(f.get(field) if f.get(field) is not None else -1 for field in field_preference) | |
752 | ||
4bcc7bd1 PH |
753 | preference = f.get('preference') |
754 | if preference is None: | |
c7deaa4c PH |
755 | proto = f.get('protocol') |
756 | if proto is None: | |
757 | proto = compat_urllib_parse_urlparse(f.get('url', '')).scheme | |
758 | ||
759 | preference = 0 if proto in ['http', 'https'] else -0.1 | |
4bcc7bd1 PH |
760 | if f.get('ext') in ['f4f', 'f4m']: # Not yet supported |
761 | preference -= 0.5 | |
762 | ||
763 | if f.get('vcodec') == 'none': # audio only | |
764 | if self._downloader.params.get('prefer_free_formats'): | |
f1a9d64e | 765 | ORDER = ['aac', 'mp3', 'm4a', 'webm', 'ogg', 'opus'] |
4bcc7bd1 | 766 | else: |
f1a9d64e | 767 | ORDER = ['webm', 'opus', 'ogg', 'mp3', 'aac', 'm4a'] |
4bcc7bd1 PH |
768 | ext_preference = 0 |
769 | try: | |
770 | audio_ext_preference = ORDER.index(f['ext']) | |
771 | except ValueError: | |
772 | audio_ext_preference = -1 | |
773 | else: | |
774 | if self._downloader.params.get('prefer_free_formats'): | |
f1a9d64e | 775 | ORDER = ['flv', 'mp4', 'webm'] |
4bcc7bd1 | 776 | else: |
f1a9d64e | 777 | ORDER = ['webm', 'flv', 'mp4'] |
4bcc7bd1 PH |
778 | try: |
779 | ext_preference = ORDER.index(f['ext']) | |
780 | except ValueError: | |
781 | ext_preference = -1 | |
782 | audio_ext_preference = 0 | |
783 | ||
784 | return ( | |
785 | preference, | |
aff2f4f4 | 786 | f.get('language_preference') if f.get('language_preference') is not None else -1, |
5d73273f | 787 | f.get('quality') if f.get('quality') is not None else -1, |
9933b574 | 788 | f.get('tbr') if f.get('tbr') is not None else -1, |
03cd72b0 | 789 | f.get('filesize') if f.get('filesize') is not None else -1, |
4bcc7bd1 | 790 | f.get('vbr') if f.get('vbr') is not None else -1, |
1a6373ef PH |
791 | f.get('height') if f.get('height') is not None else -1, |
792 | f.get('width') if f.get('width') is not None else -1, | |
1e1896f2 | 793 | ext_preference, |
4bcc7bd1 PH |
794 | f.get('abr') if f.get('abr') is not None else -1, |
795 | audio_ext_preference, | |
2c8e03d9 | 796 | f.get('fps') if f.get('fps') is not None else -1, |
9732d77e | 797 | f.get('filesize_approx') if f.get('filesize_approx') is not None else -1, |
c64ed2a3 | 798 | f.get('source_preference') if f.get('source_preference') is not None else -1, |
74f72824 | 799 | f.get('format_id') if f.get('format_id') is not None else '', |
4bcc7bd1 PH |
800 | ) |
801 | formats.sort(key=_formats_key) | |
59040888 | 802 | |
96a53167 S |
803 | def _check_formats(self, formats, video_id): |
804 | if formats: | |
805 | formats[:] = filter( | |
806 | lambda f: self._is_valid_url( | |
807 | f['url'], video_id, | |
808 | item='%s video format' % f.get('format_id') if f.get('format_id') else 'video'), | |
809 | formats) | |
810 | ||
811 | def _is_valid_url(self, url, video_id, item='video'): | |
2f0f6578 S |
812 | url = self._proto_relative_url(url, scheme='http:') |
813 | # For now assume non HTTP(S) URLs always valid | |
814 | if not (url.startswith('http://') or url.startswith('https://')): | |
815 | return True | |
96a53167 | 816 | try: |
4069766c | 817 | self._request_webpage(url, video_id, 'Checking %s URL' % item) |
96a53167 S |
818 | return True |
819 | except ExtractorError as e: | |
820 | if isinstance(e.cause, compat_HTTPError): | |
baa43cba S |
821 | self.to_screen( |
822 | '%s: %s URL is invalid, skipping' % (video_id, item)) | |
96a53167 S |
823 | return False |
824 | raise | |
825 | ||
20991253 | 826 | def http_scheme(self): |
1ede5b24 | 827 | """ Either "http:" or "https:", depending on the user's preferences """ |
20991253 PH |
828 | return ( |
829 | 'http:' | |
830 | if self._downloader.params.get('prefer_insecure', False) | |
831 | else 'https:') | |
832 | ||
57c7411f PH |
833 | def _proto_relative_url(self, url, scheme=None): |
834 | if url is None: | |
835 | return url | |
836 | if url.startswith('//'): | |
837 | if scheme is None: | |
838 | scheme = self.http_scheme() | |
839 | return scheme + url | |
840 | else: | |
841 | return url | |
842 | ||
4094b6e3 PH |
843 | def _sleep(self, timeout, video_id, msg_template=None): |
844 | if msg_template is None: | |
f1a9d64e | 845 | msg_template = '%(video_id)s: Waiting for %(timeout)s seconds' |
4094b6e3 PH |
846 | msg = msg_template % {'video_id': video_id, 'timeout': timeout} |
847 | self.to_screen(msg) | |
848 | time.sleep(timeout) | |
849 | ||
a38436e8 YCH |
850 | def _extract_f4m_formats(self, manifest_url, video_id, preference=None, f4m_id=None, |
851 | transform_source=lambda s: fix_xml_ampersands(s).strip()): | |
f036a632 JMF |
852 | manifest = self._download_xml( |
853 | manifest_url, video_id, 'Downloading f4m manifest', | |
97f4aecf S |
854 | 'Unable to download f4m manifest', |
855 | # Some manifests may be malformed, e.g. prosiebensat1 generated manifests | |
856 | # (see https://github.com/rg3/youtube-dl/issues/6215#issuecomment-121704244) | |
a38436e8 | 857 | transform_source=transform_source) |
31bb8d3f JMF |
858 | |
859 | formats = [] | |
7a47d07c | 860 | manifest_version = '1.0' |
b2527359 | 861 | media_nodes = manifest.findall('{http://ns.adobe.com/f4m/1.0}media') |
34e48bed | 862 | if not media_nodes: |
7a47d07c | 863 | manifest_version = '2.0' |
34e48bed | 864 | media_nodes = manifest.findall('{http://ns.adobe.com/f4m/2.0}media') |
b2527359 | 865 | for i, media_el in enumerate(media_nodes): |
7a47d07c | 866 | if manifest_version == '2.0': |
31c746e5 S |
867 | media_url = media_el.attrib.get('href') or media_el.attrib.get('url') |
868 | if not media_url: | |
869 | continue | |
cc357c4d S |
870 | manifest_url = ( |
871 | media_url if media_url.startswith('http://') or media_url.startswith('https://') | |
872 | else ('/'.join(manifest_url.split('/')[:-1]) + '/' + media_url)) | |
70f0f5a8 S |
873 | # If media_url is itself a f4m manifest do the recursive extraction |
874 | # since bitrates in parent manifest (this one) and media_url manifest | |
875 | # may differ leading to inability to resolve the format by requested | |
876 | # bitrate in f4m downloader | |
877 | if determine_ext(manifest_url) == 'f4m': | |
878 | formats.extend(self._extract_f4m_formats(manifest_url, video_id, preference, f4m_id)) | |
879 | continue | |
b2527359 | 880 | tbr = int_or_none(media_el.attrib.get('bitrate')) |
31bb8d3f | 881 | formats.append({ |
e21a55ab | 882 | 'format_id': '-'.join(filter(None, [f4m_id, compat_str(i if tbr is None else tbr)])), |
31bb8d3f JMF |
883 | 'url': manifest_url, |
884 | 'ext': 'flv', | |
b2527359 | 885 | 'tbr': tbr, |
31bb8d3f JMF |
886 | 'width': int_or_none(media_el.attrib.get('width')), |
887 | 'height': int_or_none(media_el.attrib.get('height')), | |
60ca389c | 888 | 'preference': preference, |
31bb8d3f JMF |
889 | }) |
890 | self._sort_formats(formats) | |
891 | ||
892 | return formats | |
893 | ||
f0b5d6af | 894 | def _extract_m3u8_formats(self, m3u8_url, video_id, ext=None, |
60ca389c | 895 | entry_protocol='m3u8', preference=None, |
13af92fd YCH |
896 | m3u8_id=None, note=None, errnote=None, |
897 | fatal=True): | |
f0b5d6af | 898 | |
704df56d | 899 | formats = [{ |
f207019c | 900 | 'format_id': '-'.join(filter(None, [m3u8_id, 'meta'])), |
704df56d PH |
901 | 'url': m3u8_url, |
902 | 'ext': ext, | |
903 | 'protocol': 'm3u8', | |
9fe6ef7a | 904 | 'preference': preference - 1 if preference else -1, |
704df56d PH |
905 | 'resolution': 'multiple', |
906 | 'format_note': 'Quality selection URL', | |
907 | }] | |
908 | ||
f0b5d6af PH |
909 | format_url = lambda u: ( |
910 | u | |
911 | if re.match(r'^https?://', u) | |
912 | else compat_urlparse.urljoin(m3u8_url, u)) | |
913 | ||
81515ad9 PH |
914 | m3u8_doc = self._download_webpage( |
915 | m3u8_url, video_id, | |
621ed9f5 | 916 | note=note or 'Downloading m3u8 information', |
13af92fd YCH |
917 | errnote=errnote or 'Failed to download m3u8 information', |
918 | fatal=fatal) | |
919 | if m3u8_doc is False: | |
920 | return m3u8_doc | |
704df56d | 921 | last_info = None |
fa156077 | 922 | last_media = None |
704df56d PH |
923 | kv_rex = re.compile( |
924 | r'(?P<key>[a-zA-Z_-]+)=(?P<val>"[^"]+"|[^",]+)(?:,|$)') | |
925 | for line in m3u8_doc.splitlines(): | |
926 | if line.startswith('#EXT-X-STREAM-INF:'): | |
927 | last_info = {} | |
928 | for m in kv_rex.finditer(line): | |
929 | v = m.group('val') | |
930 | if v.startswith('"'): | |
931 | v = v[1:-1] | |
932 | last_info[m.group('key')] = v | |
4cd95bcb JMF |
933 | elif line.startswith('#EXT-X-MEDIA:'): |
934 | last_media = {} | |
935 | for m in kv_rex.finditer(line): | |
936 | v = m.group('val') | |
937 | if v.startswith('"'): | |
938 | v = v[1:-1] | |
939 | last_media[m.group('key')] = v | |
704df56d PH |
940 | elif line.startswith('#') or not line.strip(): |
941 | continue | |
942 | else: | |
daebaab6 | 943 | if last_info is None: |
f0b5d6af | 944 | formats.append({'url': format_url(line)}) |
3524cc25 | 945 | continue |
704df56d | 946 | tbr = int_or_none(last_info.get('BANDWIDTH'), scale=1000) |
8dc9d361 S |
947 | format_id = [] |
948 | if m3u8_id: | |
949 | format_id.append(m3u8_id) | |
05d5392c | 950 | last_media_name = last_media.get('NAME') if last_media and last_media.get('TYPE') != 'SUBTITLES' else None |
8dc9d361 | 951 | format_id.append(last_media_name if last_media_name else '%d' % (tbr if tbr else len(formats))) |
704df56d | 952 | f = { |
8dc9d361 | 953 | 'format_id': '-'.join(format_id), |
f0b5d6af | 954 | 'url': format_url(line.strip()), |
704df56d PH |
955 | 'tbr': tbr, |
956 | 'ext': ext, | |
f0b5d6af PH |
957 | 'protocol': entry_protocol, |
958 | 'preference': preference, | |
704df56d PH |
959 | } |
960 | codecs = last_info.get('CODECS') | |
961 | if codecs: | |
9ebf22b7 S |
962 | # TODO: looks like video codec is not always necessarily goes first |
963 | va_codecs = codecs.split(',') | |
964 | if va_codecs[0]: | |
965 | f['vcodec'] = va_codecs[0].partition('.')[0] | |
966 | if len(va_codecs) > 1 and va_codecs[1]: | |
967 | f['acodec'] = va_codecs[1].partition('.')[0] | |
704df56d PH |
968 | resolution = last_info.get('RESOLUTION') |
969 | if resolution: | |
970 | width_str, height_str = resolution.split('x') | |
971 | f['width'] = int(width_str) | |
972 | f['height'] = int(height_str) | |
4cd95bcb JMF |
973 | if last_media is not None: |
974 | f['m3u8_media'] = last_media | |
975 | last_media = None | |
704df56d PH |
976 | formats.append(f) |
977 | last_info = {} | |
978 | self._sort_formats(formats) | |
979 | return formats | |
980 | ||
e89a2aab | 981 | # TODO: improve extraction |
995029a1 | 982 | def _extract_smil_formats(self, smil_url, video_id, fatal=True): |
e89a2aab S |
983 | smil = self._download_xml( |
984 | smil_url, video_id, 'Downloading SMIL file', | |
995029a1 PH |
985 | 'Unable to download SMIL file', fatal=fatal) |
986 | if smil is False: | |
987 | assert not fatal | |
988 | return [] | |
e89a2aab S |
989 | |
990 | base = smil.find('./head/meta').get('base') | |
991 | ||
992 | formats = [] | |
993 | rtmp_count = 0 | |
63757032 AA |
994 | if smil.findall('./body/seq/video'): |
995 | video = smil.findall('./body/seq/video')[0] | |
72a406e7 | 996 | fmts, rtmp_count = self._parse_smil_video(video, video_id, base, rtmp_count) |
63757032 AA |
997 | formats.extend(fmts) |
998 | else: | |
999 | for video in smil.findall('./body/switch/video'): | |
72a406e7 | 1000 | fmts, rtmp_count = self._parse_smil_video(video, video_id, base, rtmp_count) |
63757032 AA |
1001 | formats.extend(fmts) |
1002 | ||
e89a2aab S |
1003 | self._sort_formats(formats) |
1004 | ||
1005 | return formats | |
1006 | ||
72a406e7 | 1007 | def _parse_smil_video(self, video, video_id, base, rtmp_count): |
63757032 AA |
1008 | src = video.get('src') |
1009 | if not src: | |
74fe23ec | 1010 | return [], rtmp_count |
63757032 AA |
1011 | bitrate = int_or_none(video.get('system-bitrate') or video.get('systemBitrate'), 1000) |
1012 | width = int_or_none(video.get('width')) | |
1013 | height = int_or_none(video.get('height')) | |
1014 | proto = video.get('proto') | |
1015 | if not proto: | |
1016 | if base: | |
1017 | if base.startswith('rtmp'): | |
1018 | proto = 'rtmp' | |
1019 | elif base.startswith('http'): | |
1020 | proto = 'http' | |
1021 | ext = video.get('ext') | |
1022 | if proto == 'm3u8': | |
74fe23ec | 1023 | return self._extract_m3u8_formats(src, video_id, ext), rtmp_count |
63757032 AA |
1024 | elif proto == 'rtmp': |
1025 | rtmp_count += 1 | |
1026 | streamer = video.get('streamer') or base | |
1027 | return ([{ | |
1028 | 'url': streamer, | |
1029 | 'play_path': src, | |
1030 | 'ext': 'flv', | |
1031 | 'format_id': 'rtmp-%d' % (rtmp_count if bitrate is None else bitrate), | |
1032 | 'tbr': bitrate, | |
1033 | 'width': width, | |
1034 | 'height': height, | |
1035 | }], rtmp_count) | |
6f4ba540 AA |
1036 | elif proto.startswith('http'): |
1037 | return ([{ | |
1038 | 'url': base + src, | |
1039 | 'ext': ext or 'flv', | |
1040 | 'tbr': bitrate, | |
1041 | 'width': width, | |
1042 | 'height': height, | |
1043 | }], rtmp_count) | |
63757032 | 1044 | |
f4b1c7ad PH |
1045 | def _live_title(self, name): |
1046 | """ Generate the title for a live video """ | |
1047 | now = datetime.datetime.now() | |
1048 | now_str = now.strftime("%Y-%m-%d %H:%M") | |
1049 | return name + ' ' + now_str | |
1050 | ||
b14f3a4c PH |
1051 | def _int(self, v, name, fatal=False, **kwargs): |
1052 | res = int_or_none(v, **kwargs) | |
1053 | if 'get_attr' in kwargs: | |
1054 | print(getattr(v, kwargs['get_attr'])) | |
1055 | if res is None: | |
1056 | msg = 'Failed to extract %s: Could not parse value %r' % (name, v) | |
1057 | if fatal: | |
1058 | raise ExtractorError(msg) | |
1059 | else: | |
1060 | self._downloader.report_warning(msg) | |
1061 | return res | |
1062 | ||
1063 | def _float(self, v, name, fatal=False, **kwargs): | |
1064 | res = float_or_none(v, **kwargs) | |
1065 | if res is None: | |
1066 | msg = 'Failed to extract %s: Could not parse value %r' % (name, v) | |
1067 | if fatal: | |
1068 | raise ExtractorError(msg) | |
1069 | else: | |
1070 | self._downloader.report_warning(msg) | |
1071 | return res | |
1072 | ||
42939b61 | 1073 | def _set_cookie(self, domain, name, value, expire_time=None): |
810fb84d PH |
1074 | cookie = compat_cookiejar.Cookie( |
1075 | 0, name, value, None, None, domain, None, | |
42939b61 JMF |
1076 | None, '/', True, False, expire_time, '', None, None, None) |
1077 | self._downloader.cookiejar.set_cookie(cookie) | |
1078 | ||
799207e8 | 1079 | def _get_cookies(self, url): |
1080 | """ Return a compat_cookies.SimpleCookie with the cookies for the url """ | |
1081 | req = compat_urllib_request.Request(url) | |
1082 | self._downloader.cookiejar.add_cookie_header(req) | |
1083 | return compat_cookies.SimpleCookie(req.get_header('Cookie')) | |
1084 | ||
05900629 PH |
1085 | def get_testcases(self, include_onlymatching=False): |
1086 | t = getattr(self, '_TEST', None) | |
1087 | if t: | |
1088 | assert not hasattr(self, '_TESTS'), \ | |
1089 | '%s has _TEST and _TESTS' % type(self).__name__ | |
1090 | tests = [t] | |
1091 | else: | |
1092 | tests = getattr(self, '_TESTS', []) | |
1093 | for t in tests: | |
1094 | if not include_onlymatching and t.get('only_matching', False): | |
1095 | continue | |
1096 | t['name'] = type(self).__name__[:-len('IE')] | |
1097 | yield t | |
1098 | ||
1099 | def is_suitable(self, age_limit): | |
1100 | """ Test whether the extractor is generally suitable for the given | |
1101 | age limit (i.e. pornographic sites are not, all others usually are) """ | |
1102 | ||
1103 | any_restricted = False | |
1104 | for tc in self.get_testcases(include_onlymatching=False): | |
1105 | if 'playlist' in tc: | |
1106 | tc = tc['playlist'][0] | |
1107 | is_restricted = age_restricted( | |
1108 | tc.get('info_dict', {}).get('age_limit'), age_limit) | |
1109 | if not is_restricted: | |
1110 | return True | |
1111 | any_restricted = any_restricted or is_restricted | |
1112 | return not any_restricted | |
1113 | ||
a504ced0 | 1114 | def extract_subtitles(self, *args, **kwargs): |
9868ea49 JMF |
1115 | if (self._downloader.params.get('writesubtitles', False) or |
1116 | self._downloader.params.get('listsubtitles')): | |
1117 | return self._get_subtitles(*args, **kwargs) | |
1118 | return {} | |
a504ced0 JMF |
1119 | |
1120 | def _get_subtitles(self, *args, **kwargs): | |
1121 | raise NotImplementedError("This method must be implemented by subclasses") | |
1122 | ||
360e1ca5 | 1123 | def extract_automatic_captions(self, *args, **kwargs): |
9868ea49 JMF |
1124 | if (self._downloader.params.get('writeautomaticsub', False) or |
1125 | self._downloader.params.get('listsubtitles')): | |
1126 | return self._get_automatic_captions(*args, **kwargs) | |
1127 | return {} | |
360e1ca5 JMF |
1128 | |
1129 | def _get_automatic_captions(self, *args, **kwargs): | |
1130 | raise NotImplementedError("This method must be implemented by subclasses") | |
1131 | ||
8dbe9899 | 1132 | |
d6983cb4 PH |
1133 | class SearchInfoExtractor(InfoExtractor): |
1134 | """ | |
1135 | Base class for paged search queries extractors. | |
10952eb2 | 1136 | They accept URLs in the format _SEARCH_KEY(|all|[0-9]):{query} |
d6983cb4 PH |
1137 | Instances should define _SEARCH_KEY and _MAX_RESULTS. |
1138 | """ | |
1139 | ||
1140 | @classmethod | |
1141 | def _make_valid_url(cls): | |
1142 | return r'%s(?P<prefix>|[1-9][0-9]*|all):(?P<query>[\s\S]+)' % cls._SEARCH_KEY | |
1143 | ||
1144 | @classmethod | |
1145 | def suitable(cls, url): | |
1146 | return re.match(cls._make_valid_url(), url) is not None | |
1147 | ||
1148 | def _real_extract(self, query): | |
1149 | mobj = re.match(self._make_valid_url(), query) | |
1150 | if mobj is None: | |
f1a9d64e | 1151 | raise ExtractorError('Invalid search query "%s"' % query) |
d6983cb4 PH |
1152 | |
1153 | prefix = mobj.group('prefix') | |
1154 | query = mobj.group('query') | |
1155 | if prefix == '': | |
1156 | return self._get_n_results(query, 1) | |
1157 | elif prefix == 'all': | |
1158 | return self._get_n_results(query, self._MAX_RESULTS) | |
1159 | else: | |
1160 | n = int(prefix) | |
1161 | if n <= 0: | |
f1a9d64e | 1162 | raise ExtractorError('invalid download number %s for query "%s"' % (n, query)) |
d6983cb4 | 1163 | elif n > self._MAX_RESULTS: |
f1a9d64e | 1164 | self._downloader.report_warning('%s returns max %i results (you requested %i)' % (self._SEARCH_KEY, self._MAX_RESULTS, n)) |
d6983cb4 PH |
1165 | n = self._MAX_RESULTS |
1166 | return self._get_n_results(query, n) | |
1167 | ||
1168 | def _get_n_results(self, query, n): | |
1169 | """Get a specified number of results for a query""" | |
416a5efc | 1170 | raise NotImplementedError("This method must be implemented by subclasses") |
0f818663 PH |
1171 | |
1172 | @property | |
1173 | def SEARCH_KEY(self): | |
1174 | return self._SEARCH_KEY |