]>
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, |
e64b7569 | 18 | compat_getpass, |
96a53167 | 19 | compat_HTTPError, |
d6983cb4 PH |
20 | compat_http_client, |
21 | compat_urllib_error, | |
a107193e | 22 | compat_urllib_parse, |
c7deaa4c | 23 | compat_urllib_parse_urlparse, |
799207e8 | 24 | compat_urllib_request, |
f0b5d6af | 25 | compat_urlparse, |
d6983cb4 | 26 | compat_str, |
8c25f81b PH |
27 | ) |
28 | from ..utils import ( | |
c342041f | 29 | NO_DEFAULT, |
05900629 | 30 | age_restricted, |
08f2a92c | 31 | bug_reports_message, |
d6983cb4 PH |
32 | clean_html, |
33 | compiled_regex_type, | |
70f0f5a8 | 34 | determine_ext, |
d6983cb4 | 35 | ExtractorError, |
97f4aecf | 36 | fix_xml_ampersands, |
b14f3a4c | 37 | float_or_none, |
31bb8d3f | 38 | int_or_none, |
55b3e45b | 39 | RegexNotFoundError, |
d41e6efc | 40 | sanitize_filename, |
f38de77f | 41 | unescapeHTML, |
647eab45 | 42 | unified_strdate, |
a107193e | 43 | url_basename, |
8d6765cf S |
44 | xpath_text, |
45 | xpath_with_ns, | |
d6983cb4 | 46 | ) |
c342041f | 47 | |
d6983cb4 PH |
48 | |
49 | class InfoExtractor(object): | |
50 | """Information Extractor class. | |
51 | ||
52 | Information extractors are the classes that, given a URL, extract | |
53 | information about the video (or videos) the URL refers to. This | |
54 | information includes the real video URL, the video title, author and | |
55 | others. The information is stored in a dictionary which is then | |
5d380852 | 56 | passed to the YoutubeDL. The YoutubeDL processes this |
d6983cb4 PH |
57 | information possibly downloading the video to the file system, among |
58 | other possible outcomes. | |
59 | ||
cf0649f8 | 60 | The type field determines the type of the result. |
fed5d032 PH |
61 | By far the most common value (and the default if _type is missing) is |
62 | "video", which indicates a single video. | |
63 | ||
64 | For a video, the dictionaries must include the following fields: | |
d6983cb4 PH |
65 | |
66 | id: Video identifier. | |
d6983cb4 | 67 | title: Video title, unescaped. |
d67b0b15 | 68 | |
f49d89ee | 69 | Additionally, it must contain either a formats entry or a url one: |
d67b0b15 | 70 | |
f49d89ee PH |
71 | formats: A list of dictionaries for each format available, ordered |
72 | from worst to best quality. | |
73 | ||
74 | Potential fields: | |
d67b0b15 | 75 | * url Mandatory. The URL of the video file |
10952eb2 | 76 | * ext Will be calculated from URL if missing |
d67b0b15 PH |
77 | * format A human-readable description of the format |
78 | ("mp4 container with h264/opus"). | |
79 | Calculated from the format_id, width, height. | |
80 | and format_note fields if missing. | |
81 | * format_id A short description of the format | |
5d4f3985 PH |
82 | ("mp4_h264_opus" or "19"). |
83 | Technically optional, but strongly recommended. | |
d67b0b15 PH |
84 | * format_note Additional info about the format |
85 | ("3D" or "DASH video") | |
86 | * width Width of the video, if known | |
87 | * height Height of the video, if known | |
f49d89ee | 88 | * resolution Textual description of width and height |
7217e148 | 89 | * tbr Average bitrate of audio and video in KBit/s |
d67b0b15 PH |
90 | * abr Average audio bitrate in KBit/s |
91 | * acodec Name of the audio codec in use | |
dd27fd17 | 92 | * asr Audio sampling rate in Hertz |
d67b0b15 | 93 | * vbr Average video bitrate in KBit/s |
fbb21cf5 | 94 | * fps Frame rate |
d67b0b15 | 95 | * vcodec Name of the video codec in use |
1394ce65 | 96 | * container Name of the container format |
d67b0b15 | 97 | * filesize The number of bytes, if known in advance |
9732d77e | 98 | * filesize_approx An estimate for the number of bytes |
d67b0b15 | 99 | * player_url SWF Player URL (used for rtmpdump). |
c7deaa4c PH |
100 | * protocol The protocol that will be used for the actual |
101 | download, lower-case. | |
b04b8852 PH |
102 | "http", "https", "rtsp", "rtmp", "rtmpe", |
103 | "m3u8", or "m3u8_native". | |
f49d89ee | 104 | * preference Order number of this format. If this field is |
08d13955 | 105 | present and not None, the formats get sorted |
38d63d84 | 106 | by this field, regardless of all other values. |
f49d89ee PH |
107 | -1 for default (order by other properties), |
108 | -2 or smaller for less than default. | |
e65566a9 PH |
109 | < -1000 to hide the format (if there is |
110 | another one which is strictly better) | |
aff2f4f4 PH |
111 | * language_preference Is this in the correct requested |
112 | language? | |
113 | 10 if it's what the URL is about, | |
114 | -1 for default (don't know), | |
115 | -10 otherwise, other values reserved for now. | |
5d73273f PH |
116 | * quality Order number of the video quality of this |
117 | format, irrespective of the file format. | |
118 | -1 for default (order by other properties), | |
119 | -2 or smaller for less than default. | |
c64ed2a3 PH |
120 | * source_preference Order number for this video source |
121 | (quality takes higher priority) | |
122 | -1 for default (order by other properties), | |
123 | -2 or smaller for less than default. | |
d769be6c PH |
124 | * http_headers A dictionary of additional HTTP headers |
125 | to add to the request. | |
6271f1ca | 126 | * stretched_ratio If given and not 1, indicates that the |
3dee7826 PH |
127 | video's pixels are not square. |
128 | width : height ratio as float. | |
129 | * no_resume The server does not support resuming the | |
130 | (HTTP or RTMP) download. Boolean. | |
131 | ||
c0ba0f48 | 132 | url: Final video URL. |
d6983cb4 | 133 | ext: Video filename extension. |
d67b0b15 PH |
134 | format: The video format, defaults to ext (used for --get-format) |
135 | player_url: SWF Player URL (used for rtmpdump). | |
2f5865cc | 136 | |
d6983cb4 PH |
137 | The following fields are optional: |
138 | ||
f5e43bc6 | 139 | alt_title: A secondary title of the video. |
0afef30b PH |
140 | display_id An alternative identifier for the video, not necessarily |
141 | unique, but available before title. Typically, id is | |
142 | something like "4234987", title "Dancing naked mole rats", | |
143 | and display_id "dancing-naked-mole-rats" | |
d5519808 | 144 | thumbnails: A list of dictionaries, with the following entries: |
cfb56d1a | 145 | * "id" (optional, string) - Thumbnail format ID |
d5519808 | 146 | * "url" |
cfb56d1a | 147 | * "preference" (optional, int) - quality of the image |
d5519808 PH |
148 | * "width" (optional, int) |
149 | * "height" (optional, int) | |
150 | * "resolution" (optional, string "{width}x{height"}, | |
151 | deprecated) | |
d6983cb4 | 152 | thumbnail: Full URL to a video thumbnail image. |
f5e43bc6 | 153 | description: Full video description. |
d6983cb4 | 154 | uploader: Full name of the video uploader. |
9bb8e0a3 | 155 | creator: The main artist who created the video. |
8aab976b | 156 | release_date: The date (YYYYMMDD) when the video was released. |
955c4514 | 157 | timestamp: UNIX timestamp of the moment the video became available. |
d6983cb4 | 158 | upload_date: Video upload date (YYYYMMDD). |
955c4514 | 159 | If not explicitly set, calculated from timestamp. |
d6983cb4 | 160 | uploader_id: Nickname or id of the video uploader. |
da9ec3b9 | 161 | location: Physical location where the video was filmed. |
a504ced0 JMF |
162 | subtitles: The available subtitles as a dictionary in the format |
163 | {language: subformats}. "subformats" is a list sorted from | |
164 | lower to higher preference, each element is a dictionary | |
165 | with the "ext" entry and one of: | |
166 | * "data": The subtitles file contents | |
10952eb2 | 167 | * "url": A URL pointing to the subtitles file |
4bba3716 | 168 | "ext" will be calculated from URL if missing |
360e1ca5 JMF |
169 | automatic_captions: Like 'subtitles', used by the YoutubeIE for |
170 | automatically generated captions | |
c0ba0f48 | 171 | duration: Length of the video in seconds, as an integer. |
f3d29461 | 172 | view_count: How many users have watched the video on the platform. |
19e3dfc9 PH |
173 | like_count: Number of positive ratings of the video |
174 | dislike_count: Number of negative ratings of the video | |
2d30521a | 175 | average_rating: Average rating give by users, the scale used depends on the webpage |
19e3dfc9 | 176 | comment_count: Number of comments on the video |
dd622d7c PH |
177 | comments: A list of comments, each with one or more of the following |
178 | properties (all but one of text or html optional): | |
179 | * "author" - human-readable name of the comment author | |
180 | * "author_id" - user ID of the comment author | |
181 | * "id" - Comment ID | |
182 | * "html" - Comment as HTML | |
183 | * "text" - Plain text of the comment | |
184 | * "timestamp" - UNIX timestamp of comment | |
185 | * "parent" - ID of the comment this one is replying to. | |
186 | Set to "root" to indicate that this is a | |
187 | comment to the original video. | |
8dbe9899 | 188 | age_limit: Age restriction for the video, as an integer (years) |
10952eb2 | 189 | webpage_url: The URL to the video webpage, if given to youtube-dl it |
9103bbc5 JMF |
190 | should allow to get the same result again. (It will be set |
191 | by YoutubeDL if it's missing) | |
ad3bc6ac PH |
192 | categories: A list of categories that the video falls in, for example |
193 | ["Sports", "Berlin"] | |
864f24bd | 194 | tags: A list of tags assigned to the video, e.g. ["sweden", "pop music"] |
7267bd53 PH |
195 | is_live: True, False, or None (=unknown). Whether this video is a |
196 | live stream that goes on instead of a fixed-length video. | |
7c80519c | 197 | start_time: Time in seconds where the reproduction should start, as |
10952eb2 | 198 | specified in the URL. |
297a564b | 199 | end_time: Time in seconds where the reproduction should end, as |
10952eb2 | 200 | specified in the URL. |
d6983cb4 | 201 | |
deefc05b | 202 | Unless mentioned otherwise, the fields should be Unicode strings. |
d6983cb4 | 203 | |
d838b1bd PH |
204 | Unless mentioned otherwise, None is equivalent to absence of information. |
205 | ||
fed5d032 PH |
206 | |
207 | _type "playlist" indicates multiple videos. | |
b82f815f PH |
208 | There must be a key "entries", which is a list, an iterable, or a PagedList |
209 | object, each element of which is a valid dictionary by this specification. | |
fed5d032 | 210 | |
e0b9d78f S |
211 | Additionally, playlists can have "title", "description" and "id" attributes |
212 | with the same semantics as videos (see above). | |
fed5d032 PH |
213 | |
214 | ||
215 | _type "multi_video" indicates that there are multiple videos that | |
216 | form a single show, for examples multiple acts of an opera or TV episode. | |
217 | It must have an entries key like a playlist and contain all the keys | |
218 | required for a video at the same time. | |
219 | ||
220 | ||
221 | _type "url" indicates that the video must be extracted from another | |
222 | location, possibly by a different extractor. Its only required key is: | |
223 | "url" - the next URL to extract. | |
f58766ce PH |
224 | The key "ie_key" can be set to the class name (minus the trailing "IE", |
225 | e.g. "Youtube") if the extractor class is known in advance. | |
226 | Additionally, the dictionary may have any properties of the resolved entity | |
227 | known in advance, for example "title" if the title of the referred video is | |
fed5d032 PH |
228 | known ahead of time. |
229 | ||
230 | ||
231 | _type "url_transparent" entities have the same specification as "url", but | |
232 | indicate that the given additional information is more precise than the one | |
233 | associated with the resolved URL. | |
234 | This is useful when a site employs a video service that hosts the video and | |
235 | its technical metadata, but that video service does not embed a useful | |
236 | title, description etc. | |
237 | ||
238 | ||
d6983cb4 PH |
239 | Subclasses of this one should re-define the _real_initialize() and |
240 | _real_extract() methods and define a _VALID_URL regexp. | |
241 | Probably, they should also be added to the list of extractors. | |
242 | ||
d6983cb4 PH |
243 | Finally, the _WORKING attribute should be set to False for broken IEs |
244 | in order to warn the users and skip the tests. | |
245 | """ | |
246 | ||
247 | _ready = False | |
248 | _downloader = None | |
249 | _WORKING = True | |
250 | ||
251 | def __init__(self, downloader=None): | |
252 | """Constructor. Receives an optional downloader.""" | |
253 | self._ready = False | |
254 | self.set_downloader(downloader) | |
255 | ||
256 | @classmethod | |
257 | def suitable(cls, url): | |
258 | """Receives a URL and returns True if suitable for this IE.""" | |
79cb2577 PH |
259 | |
260 | # This does not use has/getattr intentionally - we want to know whether | |
261 | # we have cached the regexp for *this* class, whereas getattr would also | |
262 | # match the superclass | |
263 | if '_VALID_URL_RE' not in cls.__dict__: | |
264 | cls._VALID_URL_RE = re.compile(cls._VALID_URL) | |
265 | return cls._VALID_URL_RE.match(url) is not None | |
d6983cb4 | 266 | |
ed9266db PH |
267 | @classmethod |
268 | def _match_id(cls, url): | |
269 | if '_VALID_URL_RE' not in cls.__dict__: | |
270 | cls._VALID_URL_RE = re.compile(cls._VALID_URL) | |
271 | m = cls._VALID_URL_RE.match(url) | |
272 | assert m | |
273 | return m.group('id') | |
274 | ||
d6983cb4 PH |
275 | @classmethod |
276 | def working(cls): | |
277 | """Getter method for _WORKING.""" | |
278 | return cls._WORKING | |
279 | ||
280 | def initialize(self): | |
281 | """Initializes an instance (authentication, etc).""" | |
282 | if not self._ready: | |
283 | self._real_initialize() | |
284 | self._ready = True | |
285 | ||
286 | def extract(self, url): | |
287 | """Extracts URL information and returns it in list of dicts.""" | |
3a5bcd03 PH |
288 | try: |
289 | self.initialize() | |
290 | return self._real_extract(url) | |
291 | except ExtractorError: | |
292 | raise | |
293 | except compat_http_client.IncompleteRead as e: | |
294 | raise ExtractorError('A network error has occured.', cause=e, expected=True) | |
9650885b | 295 | except (KeyError, StopIteration) as e: |
3a5bcd03 | 296 | raise ExtractorError('An extractor error has occured.', cause=e) |
d6983cb4 PH |
297 | |
298 | def set_downloader(self, downloader): | |
299 | """Sets the downloader for this IE.""" | |
300 | self._downloader = downloader | |
301 | ||
302 | def _real_initialize(self): | |
303 | """Real initialization process. Redefine in subclasses.""" | |
304 | pass | |
305 | ||
306 | def _real_extract(self, url): | |
307 | """Real extraction process. Redefine in subclasses.""" | |
308 | pass | |
309 | ||
56c73665 JMF |
310 | @classmethod |
311 | def ie_key(cls): | |
312 | """A string for getting the InfoExtractor with get_info_extractor""" | |
313 | return cls.__name__[:-2] | |
314 | ||
d6983cb4 PH |
315 | @property |
316 | def IE_NAME(self): | |
317 | return type(self).__name__[:-2] | |
318 | ||
7cc3570e | 319 | def _request_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True): |
d6983cb4 PH |
320 | """ Returns the response handle """ |
321 | if note is None: | |
322 | self.report_download_webpage(video_id) | |
323 | elif note is not False: | |
7cc3570e | 324 | if video_id is None: |
f1a9d64e | 325 | self.to_screen('%s' % (note,)) |
7cc3570e | 326 | else: |
f1a9d64e | 327 | self.to_screen('%s: %s' % (video_id, note)) |
d6983cb4 | 328 | try: |
dca08720 | 329 | return self._downloader.urlopen(url_or_request) |
d6983cb4 | 330 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: |
aa94a6d3 PH |
331 | if errnote is False: |
332 | return False | |
d6983cb4 | 333 | if errnote is None: |
f1a9d64e PH |
334 | errnote = 'Unable to download webpage' |
335 | errmsg = '%s: %s' % (errnote, compat_str(err)) | |
7cc3570e PH |
336 | if fatal: |
337 | raise ExtractorError(errmsg, sys.exc_info()[2], cause=err) | |
338 | else: | |
339 | self._downloader.report_warning(errmsg) | |
340 | return False | |
d6983cb4 | 341 | |
c9a77969 | 342 | def _download_webpage_handle(self, url_or_request, video_id, note=None, errnote=None, fatal=True, encoding=None): |
d6983cb4 | 343 | """ Returns a tuple (page content as string, URL handle) """ |
b9d3e163 PH |
344 | # Strip hashes from the URL (#1038) |
345 | if isinstance(url_or_request, (compat_str, str)): | |
346 | url_or_request = url_or_request.partition('#')[0] | |
347 | ||
7cc3570e PH |
348 | urlh = self._request_webpage(url_or_request, video_id, note, errnote, fatal) |
349 | if urlh is False: | |
350 | assert not fatal | |
351 | return False | |
c9a77969 | 352 | content = self._webpage_read_content(urlh, url_or_request, video_id, note, errnote, fatal, encoding=encoding) |
23be51d8 PH |
353 | return (content, urlh) |
354 | ||
c9a77969 YCH |
355 | @staticmethod |
356 | def _guess_encoding_from_content(content_type, webpage_bytes): | |
d6983cb4 PH |
357 | m = re.match(r'[a-zA-Z0-9_.-]+/[a-zA-Z0-9_.-]+\s*;\s*charset=(.+)', content_type) |
358 | if m: | |
359 | encoding = m.group(1) | |
360 | else: | |
0d75ae2c | 361 | m = re.search(br'<meta[^>]+charset=[\'"]?([^\'")]+)[ /\'">]', |
f143d86a PH |
362 | webpage_bytes[:1024]) |
363 | if m: | |
364 | encoding = m.group(1).decode('ascii') | |
b60016e8 PH |
365 | elif webpage_bytes.startswith(b'\xff\xfe'): |
366 | encoding = 'utf-16' | |
f143d86a PH |
367 | else: |
368 | encoding = 'utf-8' | |
c9a77969 YCH |
369 | |
370 | return encoding | |
371 | ||
372 | def _webpage_read_content(self, urlh, url_or_request, video_id, note=None, errnote=None, fatal=True, prefix=None, encoding=None): | |
373 | content_type = urlh.headers.get('Content-Type', '') | |
374 | webpage_bytes = urlh.read() | |
375 | if prefix is not None: | |
376 | webpage_bytes = prefix + webpage_bytes | |
377 | if not encoding: | |
378 | encoding = self._guess_encoding_from_content(content_type, webpage_bytes) | |
d6983cb4 PH |
379 | if self._downloader.params.get('dump_intermediate_pages', False): |
380 | try: | |
381 | url = url_or_request.get_full_url() | |
382 | except AttributeError: | |
383 | url = url_or_request | |
f1a9d64e | 384 | self.to_screen('Dumping request to ' + url) |
d6983cb4 PH |
385 | dump = base64.b64encode(webpage_bytes).decode('ascii') |
386 | self._downloader.to_screen(dump) | |
d41e6efc PH |
387 | if self._downloader.params.get('write_pages', False): |
388 | try: | |
389 | url = url_or_request.get_full_url() | |
390 | except AttributeError: | |
391 | url = url_or_request | |
5afa7f8b | 392 | basen = '%s_%s' % (video_id, url) |
c1bce22f | 393 | if len(basen) > 240: |
f1a9d64e | 394 | h = '___' + hashlib.md5(basen.encode('utf-8')).hexdigest() |
c1bce22f PH |
395 | basen = basen[:240 - len(h)] + h |
396 | raw_filename = basen + '.dump' | |
d41e6efc | 397 | filename = sanitize_filename(raw_filename, restricted=True) |
f1a9d64e | 398 | self.to_screen('Saving request to ' + filename) |
5f58165d S |
399 | # Working around MAX_PATH limitation on Windows (see |
400 | # http://msdn.microsoft.com/en-us/library/windows/desktop/aa365247(v=vs.85).aspx) | |
401 | if os.name == 'nt': | |
402 | absfilepath = os.path.abspath(filename) | |
403 | if len(absfilepath) > 259: | |
404 | filename = '\\\\?\\' + absfilepath | |
d41e6efc PH |
405 | with open(filename, 'wb') as outf: |
406 | outf.write(webpage_bytes) | |
407 | ||
ec0fafbb AA |
408 | try: |
409 | content = webpage_bytes.decode(encoding, 'replace') | |
410 | except LookupError: | |
411 | content = webpage_bytes.decode('utf-8', 'replace') | |
2410c43d | 412 | |
f1a9d64e PH |
413 | if ('<title>Access to this site is blocked</title>' in content and |
414 | 'Websense' in content[:512]): | |
415 | msg = 'Access to this webpage has been blocked by Websense filtering software in your network.' | |
2410c43d PH |
416 | blocked_iframe = self._html_search_regex( |
417 | r'<iframe src="([^"]+)"', content, | |
f1a9d64e | 418 | 'Websense information URL', default=None) |
2410c43d | 419 | if blocked_iframe: |
f1a9d64e | 420 | msg += ' Visit %s for more details' % blocked_iframe |
2410c43d | 421 | raise ExtractorError(msg, expected=True) |
77b2986b PH |
422 | if '<title>The URL you requested has been blocked</title>' in content[:512]: |
423 | msg = ( | |
424 | 'Access to this webpage has been blocked by Indian censorship. ' | |
425 | 'Use a VPN or proxy server (with --proxy) to route around it.') | |
426 | block_msg = self._html_search_regex( | |
427 | r'</h1><p>(.*?)</p>', | |
428 | content, 'block message', default=None) | |
429 | if block_msg: | |
430 | msg += ' (Message: "%s")' % block_msg.replace('\n', ' ') | |
431 | raise ExtractorError(msg, expected=True) | |
2410c43d | 432 | |
23be51d8 | 433 | return content |
d6983cb4 | 434 | |
c9a77969 | 435 | def _download_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True, tries=1, timeout=5, encoding=None): |
d6983cb4 | 436 | """ Returns the data of the page as a string """ |
995ad69c TF |
437 | success = False |
438 | try_count = 0 | |
439 | while success is False: | |
440 | try: | |
c9a77969 | 441 | res = self._download_webpage_handle(url_or_request, video_id, note, errnote, fatal, encoding=encoding) |
995ad69c TF |
442 | success = True |
443 | except compat_http_client.IncompleteRead as e: | |
444 | try_count += 1 | |
445 | if try_count >= tries: | |
446 | raise e | |
447 | self._sleep(timeout, video_id) | |
7cc3570e PH |
448 | if res is False: |
449 | return res | |
450 | else: | |
451 | content, _ = res | |
452 | return content | |
d6983cb4 | 453 | |
2a275ab0 | 454 | def _download_xml(self, url_or_request, video_id, |
f1a9d64e | 455 | note='Downloading XML', errnote='Unable to download XML', |
c9a77969 | 456 | transform_source=None, fatal=True, encoding=None): |
267ed0c5 | 457 | """Return the xml as an xml.etree.ElementTree.Element""" |
28746fbd | 458 | xml_string = self._download_webpage( |
c9a77969 | 459 | url_or_request, video_id, note, errnote, fatal=fatal, encoding=encoding) |
28746fbd PH |
460 | if xml_string is False: |
461 | return xml_string | |
e2b38da9 PH |
462 | if transform_source: |
463 | xml_string = transform_source(xml_string) | |
267ed0c5 JMF |
464 | return xml.etree.ElementTree.fromstring(xml_string.encode('utf-8')) |
465 | ||
3d3538e4 | 466 | def _download_json(self, url_or_request, video_id, |
f1a9d64e PH |
467 | note='Downloading JSON metadata', |
468 | errnote='Unable to download JSON metadata', | |
b090af59 | 469 | transform_source=None, |
c9a77969 | 470 | fatal=True, encoding=None): |
b090af59 | 471 | json_string = self._download_webpage( |
c9a77969 YCH |
472 | url_or_request, video_id, note, errnote, fatal=fatal, |
473 | encoding=encoding) | |
b090af59 PH |
474 | if (not fatal) and json_string is False: |
475 | return None | |
ebb64199 TF |
476 | return self._parse_json( |
477 | json_string, video_id, transform_source=transform_source, fatal=fatal) | |
478 | ||
479 | def _parse_json(self, json_string, video_id, transform_source=None, fatal=True): | |
81c2f20b PH |
480 | if transform_source: |
481 | json_string = transform_source(json_string) | |
3d3538e4 PH |
482 | try: |
483 | return json.loads(json_string) | |
484 | except ValueError as ve: | |
e7b6d122 PH |
485 | errmsg = '%s: Failed to parse JSON ' % video_id |
486 | if fatal: | |
487 | raise ExtractorError(errmsg, cause=ve) | |
488 | else: | |
489 | self.report_warning(errmsg + str(ve)) | |
3d3538e4 | 490 | |
f45f96f8 | 491 | def report_warning(self, msg, video_id=None): |
f1a9d64e | 492 | idstr = '' if video_id is None else '%s: ' % video_id |
f45f96f8 | 493 | self._downloader.report_warning( |
f1a9d64e | 494 | '[%s] %s%s' % (self.IE_NAME, idstr, msg)) |
f45f96f8 | 495 | |
d6983cb4 PH |
496 | def to_screen(self, msg): |
497 | """Print msg to screen, prefixing it with '[ie_name]'""" | |
f1a9d64e | 498 | self._downloader.to_screen('[%s] %s' % (self.IE_NAME, msg)) |
d6983cb4 PH |
499 | |
500 | def report_extraction(self, id_or_name): | |
501 | """Report information extraction.""" | |
f1a9d64e | 502 | self.to_screen('%s: Extracting information' % id_or_name) |
d6983cb4 PH |
503 | |
504 | def report_download_webpage(self, video_id): | |
505 | """Report webpage download.""" | |
f1a9d64e | 506 | self.to_screen('%s: Downloading webpage' % video_id) |
d6983cb4 PH |
507 | |
508 | def report_age_confirmation(self): | |
509 | """Report attempt to confirm age.""" | |
f1a9d64e | 510 | self.to_screen('Confirming age') |
d6983cb4 | 511 | |
fc79158d JMF |
512 | def report_login(self): |
513 | """Report attempt to log in.""" | |
f1a9d64e | 514 | self.to_screen('Logging in') |
fc79158d | 515 | |
43e7d3c9 S |
516 | @staticmethod |
517 | def raise_login_required(msg='This video is only available for registered users'): | |
518 | raise ExtractorError( | |
519 | '%s. Use --username and --password or --netrc to provide account credentials.' % msg, | |
520 | expected=True) | |
521 | ||
c430802e S |
522 | @staticmethod |
523 | def raise_geo_restricted(msg='This video is not available from your location due to geo restriction'): | |
524 | raise ExtractorError( | |
525 | '%s. You might want to use --proxy to workaround.' % msg, | |
526 | expected=True) | |
527 | ||
5f6a1245 | 528 | # Methods for following #608 |
c0d0b01f | 529 | @staticmethod |
830d53bf | 530 | def url_result(url, ie=None, video_id=None, video_title=None): |
10952eb2 | 531 | """Returns a URL that points to a page that should be processed""" |
5f6a1245 | 532 | # TODO: ie should be the class used for getting the info |
d6983cb4 PH |
533 | video_info = {'_type': 'url', |
534 | 'url': url, | |
535 | 'ie_key': ie} | |
7012b23c PH |
536 | if video_id is not None: |
537 | video_info['id'] = video_id | |
830d53bf S |
538 | if video_title is not None: |
539 | video_info['title'] = video_title | |
d6983cb4 | 540 | return video_info |
5f6a1245 | 541 | |
c0d0b01f | 542 | @staticmethod |
acf5cbfe | 543 | def playlist_result(entries, playlist_id=None, playlist_title=None, playlist_description=None): |
d6983cb4 PH |
544 | """Returns a playlist""" |
545 | video_info = {'_type': 'playlist', | |
546 | 'entries': entries} | |
547 | if playlist_id: | |
548 | video_info['id'] = playlist_id | |
549 | if playlist_title: | |
550 | video_info['title'] = playlist_title | |
acf5cbfe S |
551 | if playlist_description: |
552 | video_info['description'] = playlist_description | |
d6983cb4 PH |
553 | return video_info |
554 | ||
c342041f | 555 | def _search_regex(self, pattern, string, name, default=NO_DEFAULT, fatal=True, flags=0, group=None): |
d6983cb4 PH |
556 | """ |
557 | Perform a regex search on the given string, using a single or a list of | |
558 | patterns returning the first matching group. | |
559 | In case of failure return a default value or raise a WARNING or a | |
55b3e45b | 560 | RegexNotFoundError, depending on fatal, specifying the field name. |
d6983cb4 PH |
561 | """ |
562 | if isinstance(pattern, (str, compat_str, compiled_regex_type)): | |
563 | mobj = re.search(pattern, string, flags) | |
564 | else: | |
565 | for p in pattern: | |
566 | mobj = re.search(p, string, flags) | |
c3415d1b PH |
567 | if mobj: |
568 | break | |
d6983cb4 | 569 | |
7e5db8c9 | 570 | if not self._downloader.params.get('no_color') and os.name != 'nt' and sys.stderr.isatty(): |
f1a9d64e | 571 | _name = '\033[0;34m%s\033[0m' % name |
d6983cb4 PH |
572 | else: |
573 | _name = name | |
574 | ||
575 | if mobj: | |
711ede6e PH |
576 | if group is None: |
577 | # return the first matching group | |
578 | return next(g for g in mobj.groups() if g is not None) | |
579 | else: | |
580 | return mobj.group(group) | |
c342041f | 581 | elif default is not NO_DEFAULT: |
d6983cb4 PH |
582 | return default |
583 | elif fatal: | |
f1a9d64e | 584 | raise RegexNotFoundError('Unable to extract %s' % _name) |
d6983cb4 | 585 | else: |
08f2a92c | 586 | self._downloader.report_warning('unable to extract %s' % _name + bug_reports_message()) |
d6983cb4 PH |
587 | return None |
588 | ||
c342041f | 589 | def _html_search_regex(self, pattern, string, name, default=NO_DEFAULT, fatal=True, flags=0, group=None): |
d6983cb4 PH |
590 | """ |
591 | Like _search_regex, but strips HTML tags and unescapes entities. | |
592 | """ | |
711ede6e | 593 | res = self._search_regex(pattern, string, name, default, fatal, flags, group) |
d6983cb4 PH |
594 | if res: |
595 | return clean_html(res).strip() | |
596 | else: | |
597 | return res | |
598 | ||
fc79158d JMF |
599 | def _get_login_info(self): |
600 | """ | |
cf0649f8 | 601 | Get the login info as (username, password) |
fc79158d JMF |
602 | It will look in the netrc file using the _NETRC_MACHINE value |
603 | If there's no info available, return (None, None) | |
604 | """ | |
605 | if self._downloader is None: | |
606 | return (None, None) | |
607 | ||
608 | username = None | |
609 | password = None | |
610 | downloader_params = self._downloader.params | |
611 | ||
612 | # Attempt to use provided username and password or .netrc data | |
613 | if downloader_params.get('username', None) is not None: | |
614 | username = downloader_params['username'] | |
615 | password = downloader_params['password'] | |
616 | elif downloader_params.get('usenetrc', False): | |
617 | try: | |
618 | info = netrc.netrc().authenticators(self._NETRC_MACHINE) | |
619 | if info is not None: | |
620 | username = info[0] | |
621 | password = info[2] | |
622 | else: | |
623 | raise netrc.NetrcParseError('No authenticators for %s' % self._NETRC_MACHINE) | |
624 | except (IOError, netrc.NetrcParseError) as err: | |
f1a9d64e | 625 | self._downloader.report_warning('parsing .netrc: %s' % compat_str(err)) |
5f6a1245 | 626 | |
fc79158d JMF |
627 | return (username, password) |
628 | ||
e64b7569 | 629 | def _get_tfa_info(self, note='two-factor verification code'): |
83317f69 | 630 | """ |
631 | Get the two-factor authentication info | |
632 | TODO - asking the user will be required for sms/phone verify | |
633 | currently just uses the command line option | |
634 | If there's no info available, return None | |
635 | """ | |
636 | if self._downloader is None: | |
83317f69 | 637 | return None |
638 | downloader_params = self._downloader.params | |
639 | ||
640 | if downloader_params.get('twofactor', None) is not None: | |
641 | return downloader_params['twofactor'] | |
642 | ||
e64b7569 | 643 | return compat_getpass('Type %s and press [Return]: ' % note) |
83317f69 | 644 | |
46720279 JMF |
645 | # Helper functions for extracting OpenGraph info |
646 | @staticmethod | |
ab2d5247 | 647 | def _og_regexes(prop): |
4180a3d8 | 648 | content_re = r'content=(?:"([^>]+?)"|\'([^>]+?)\'|\s*([^\s"\'=<>`]+?))' |
57935b25 | 649 | property_re = r'(?:name|property)=[\'"]?og:%s[\'"]?' % re.escape(prop) |
78fb87b2 | 650 | template = r'<meta[^>]+?%s[^>]+?%s' |
ab2d5247 | 651 | return [ |
78fb87b2 JMF |
652 | template % (property_re, content_re), |
653 | template % (content_re, property_re), | |
ab2d5247 | 654 | ] |
46720279 | 655 | |
864f24bd S |
656 | @staticmethod |
657 | def _meta_regex(prop): | |
658 | return r'''(?isx)<meta | |
8b9848ac | 659 | (?=[^>]+(?:itemprop|name|property|id|http-equiv)=(["\']?)%s\1) |
864f24bd S |
660 | [^>]+?content=(["\'])(?P<content>.*?)\2''' % re.escape(prop) |
661 | ||
3c4e6d83 | 662 | def _og_search_property(self, prop, html, name=None, **kargs): |
46720279 | 663 | if name is None: |
3c4e6d83 | 664 | name = 'OpenGraph %s' % prop |
ab2d5247 | 665 | escaped = self._search_regex(self._og_regexes(prop), html, name, flags=re.DOTALL, **kargs) |
eb0a8398 PH |
666 | if escaped is None: |
667 | return None | |
668 | return unescapeHTML(escaped) | |
46720279 JMF |
669 | |
670 | def _og_search_thumbnail(self, html, **kargs): | |
10952eb2 | 671 | return self._og_search_property('image', html, 'thumbnail URL', fatal=False, **kargs) |
46720279 JMF |
672 | |
673 | def _og_search_description(self, html, **kargs): | |
674 | return self._og_search_property('description', html, fatal=False, **kargs) | |
675 | ||
676 | def _og_search_title(self, html, **kargs): | |
677 | return self._og_search_property('title', html, **kargs) | |
678 | ||
8ffa13e0 | 679 | def _og_search_video_url(self, html, name='video url', secure=True, **kargs): |
a3681973 PH |
680 | regexes = self._og_regexes('video') + self._og_regexes('video:url') |
681 | if secure: | |
682 | regexes = self._og_regexes('video:secure_url') + regexes | |
8ffa13e0 | 683 | return self._html_search_regex(regexes, html, name, **kargs) |
46720279 | 684 | |
78338f71 JMF |
685 | def _og_search_url(self, html, **kargs): |
686 | return self._og_search_property('url', html, **kargs) | |
687 | ||
40c696e5 | 688 | def _html_search_meta(self, name, html, display_name=None, fatal=False, **kwargs): |
59040888 PH |
689 | if display_name is None: |
690 | display_name = name | |
691 | return self._html_search_regex( | |
864f24bd | 692 | self._meta_regex(name), |
711ede6e | 693 | html, display_name, fatal=fatal, group='content', **kwargs) |
59040888 PH |
694 | |
695 | def _dc_search_uploader(self, html): | |
696 | return self._html_search_meta('dc.creator', html, 'uploader') | |
697 | ||
8dbe9899 PH |
698 | def _rta_search(self, html): |
699 | # See http://www.rtalabel.org/index.php?content=howtofaq#single | |
700 | if re.search(r'(?ix)<meta\s+name="rating"\s+' | |
701 | r' content="RTA-5042-1996-1400-1577-RTA"', | |
702 | html): | |
703 | return 18 | |
704 | return 0 | |
705 | ||
59040888 PH |
706 | def _media_rating_search(self, html): |
707 | # See http://www.tjg-designs.com/WP/metadata-code-examples-adding-metadata-to-your-web-pages/ | |
708 | rating = self._html_search_meta('rating', html) | |
709 | ||
710 | if not rating: | |
711 | return None | |
712 | ||
713 | RATING_TABLE = { | |
714 | 'safe for kids': 0, | |
715 | 'general': 8, | |
716 | '14 years': 14, | |
717 | 'mature': 17, | |
718 | 'restricted': 19, | |
719 | } | |
720 | return RATING_TABLE.get(rating.lower(), None) | |
721 | ||
69319969 | 722 | def _family_friendly_search(self, html): |
6ca7732d | 723 | # See http://schema.org/VideoObject |
69319969 NJ |
724 | family_friendly = self._html_search_meta('isFamilyFriendly', html) |
725 | ||
726 | if not family_friendly: | |
727 | return None | |
728 | ||
729 | RATING_TABLE = { | |
730 | '1': 0, | |
731 | 'true': 0, | |
732 | '0': 18, | |
733 | 'false': 18, | |
734 | } | |
735 | return RATING_TABLE.get(family_friendly.lower(), None) | |
736 | ||
0c708f11 JMF |
737 | def _twitter_search_player(self, html): |
738 | return self._html_search_meta('twitter:player', html, | |
9e1a5b84 | 739 | 'twitter card player') |
0c708f11 | 740 | |
27713812 | 741 | @staticmethod |
f8da79f8 | 742 | def _hidden_inputs(html): |
586f1cc5 | 743 | html = re.sub(r'<!--(?:(?!<!--).)*-->', '', html) |
201ea3ee | 744 | hidden_inputs = {} |
73eb13df | 745 | for input in re.findall(r'(?i)<input([^>]+)>', html): |
be0e5dbd | 746 | if not re.search(r'type=(["\'])(?:hidden|submit)\1', input): |
201ea3ee S |
747 | continue |
748 | name = re.search(r'name=(["\'])(?P<value>.+?)\1', input) | |
749 | if not name: | |
750 | continue | |
751 | value = re.search(r'value=(["\'])(?P<value>.*?)\1', input) | |
752 | if not value: | |
753 | continue | |
754 | hidden_inputs[name.group('value')] = value.group('value') | |
755 | return hidden_inputs | |
27713812 | 756 | |
cf61d96d S |
757 | def _form_hidden_inputs(self, form_id, html): |
758 | form = self._search_regex( | |
73eb13df | 759 | r'(?is)<form[^>]+?id=(["\'])%s\1[^>]*>(?P<form>.+?)</form>' % form_id, |
cf61d96d S |
760 | html, '%s form' % form_id, group='form') |
761 | return self._hidden_inputs(form) | |
762 | ||
3ded7bac | 763 | def _sort_formats(self, formats, field_preference=None): |
7e8caf30 | 764 | if not formats: |
f1a9d64e | 765 | raise ExtractorError('No video formats found') |
7e8caf30 | 766 | |
4bcc7bd1 | 767 | def _formats_key(f): |
e6812ac9 PH |
768 | # TODO remove the following workaround |
769 | from ..utils import determine_ext | |
770 | if not f.get('ext') and 'url' in f: | |
771 | f['ext'] = determine_ext(f['url']) | |
772 | ||
3ded7bac S |
773 | if isinstance(field_preference, (list, tuple)): |
774 | return tuple(f.get(field) if f.get(field) is not None else -1 for field in field_preference) | |
775 | ||
4bcc7bd1 PH |
776 | preference = f.get('preference') |
777 | if preference is None: | |
c7deaa4c PH |
778 | proto = f.get('protocol') |
779 | if proto is None: | |
780 | proto = compat_urllib_parse_urlparse(f.get('url', '')).scheme | |
781 | ||
782 | preference = 0 if proto in ['http', 'https'] else -0.1 | |
4bcc7bd1 PH |
783 | if f.get('ext') in ['f4f', 'f4m']: # Not yet supported |
784 | preference -= 0.5 | |
785 | ||
786 | if f.get('vcodec') == 'none': # audio only | |
787 | if self._downloader.params.get('prefer_free_formats'): | |
f1a9d64e | 788 | ORDER = ['aac', 'mp3', 'm4a', 'webm', 'ogg', 'opus'] |
4bcc7bd1 | 789 | else: |
f1a9d64e | 790 | ORDER = ['webm', 'opus', 'ogg', 'mp3', 'aac', 'm4a'] |
4bcc7bd1 PH |
791 | ext_preference = 0 |
792 | try: | |
793 | audio_ext_preference = ORDER.index(f['ext']) | |
794 | except ValueError: | |
795 | audio_ext_preference = -1 | |
796 | else: | |
797 | if self._downloader.params.get('prefer_free_formats'): | |
f1a9d64e | 798 | ORDER = ['flv', 'mp4', 'webm'] |
4bcc7bd1 | 799 | else: |
f1a9d64e | 800 | ORDER = ['webm', 'flv', 'mp4'] |
4bcc7bd1 PH |
801 | try: |
802 | ext_preference = ORDER.index(f['ext']) | |
803 | except ValueError: | |
804 | ext_preference = -1 | |
805 | audio_ext_preference = 0 | |
806 | ||
807 | return ( | |
808 | preference, | |
aff2f4f4 | 809 | f.get('language_preference') if f.get('language_preference') is not None else -1, |
5d73273f | 810 | f.get('quality') if f.get('quality') is not None else -1, |
9933b574 | 811 | f.get('tbr') if f.get('tbr') is not None else -1, |
03cd72b0 | 812 | f.get('filesize') if f.get('filesize') is not None else -1, |
4bcc7bd1 | 813 | f.get('vbr') if f.get('vbr') is not None else -1, |
1a6373ef PH |
814 | f.get('height') if f.get('height') is not None else -1, |
815 | f.get('width') if f.get('width') is not None else -1, | |
1e1896f2 | 816 | ext_preference, |
4bcc7bd1 PH |
817 | f.get('abr') if f.get('abr') is not None else -1, |
818 | audio_ext_preference, | |
2c8e03d9 | 819 | f.get('fps') if f.get('fps') is not None else -1, |
9732d77e | 820 | f.get('filesize_approx') if f.get('filesize_approx') is not None else -1, |
c64ed2a3 | 821 | f.get('source_preference') if f.get('source_preference') is not None else -1, |
74f72824 | 822 | f.get('format_id') if f.get('format_id') is not None else '', |
4bcc7bd1 PH |
823 | ) |
824 | formats.sort(key=_formats_key) | |
59040888 | 825 | |
96a53167 S |
826 | def _check_formats(self, formats, video_id): |
827 | if formats: | |
828 | formats[:] = filter( | |
829 | lambda f: self._is_valid_url( | |
830 | f['url'], video_id, | |
831 | item='%s video format' % f.get('format_id') if f.get('format_id') else 'video'), | |
832 | formats) | |
833 | ||
834 | def _is_valid_url(self, url, video_id, item='video'): | |
2f0f6578 S |
835 | url = self._proto_relative_url(url, scheme='http:') |
836 | # For now assume non HTTP(S) URLs always valid | |
837 | if not (url.startswith('http://') or url.startswith('https://')): | |
838 | return True | |
96a53167 | 839 | try: |
4069766c | 840 | self._request_webpage(url, video_id, 'Checking %s URL' % item) |
96a53167 S |
841 | return True |
842 | except ExtractorError as e: | |
843 | if isinstance(e.cause, compat_HTTPError): | |
baa43cba S |
844 | self.to_screen( |
845 | '%s: %s URL is invalid, skipping' % (video_id, item)) | |
96a53167 S |
846 | return False |
847 | raise | |
848 | ||
20991253 | 849 | def http_scheme(self): |
1ede5b24 | 850 | """ Either "http:" or "https:", depending on the user's preferences """ |
20991253 PH |
851 | return ( |
852 | 'http:' | |
853 | if self._downloader.params.get('prefer_insecure', False) | |
854 | else 'https:') | |
855 | ||
57c7411f PH |
856 | def _proto_relative_url(self, url, scheme=None): |
857 | if url is None: | |
858 | return url | |
859 | if url.startswith('//'): | |
860 | if scheme is None: | |
861 | scheme = self.http_scheme() | |
862 | return scheme + url | |
863 | else: | |
864 | return url | |
865 | ||
4094b6e3 PH |
866 | def _sleep(self, timeout, video_id, msg_template=None): |
867 | if msg_template is None: | |
f1a9d64e | 868 | msg_template = '%(video_id)s: Waiting for %(timeout)s seconds' |
4094b6e3 PH |
869 | msg = msg_template % {'video_id': video_id, 'timeout': timeout} |
870 | self.to_screen(msg) | |
871 | time.sleep(timeout) | |
872 | ||
a38436e8 | 873 | def _extract_f4m_formats(self, manifest_url, video_id, preference=None, f4m_id=None, |
4de61310 S |
874 | transform_source=lambda s: fix_xml_ampersands(s).strip(), |
875 | fatal=True): | |
f036a632 JMF |
876 | manifest = self._download_xml( |
877 | manifest_url, video_id, 'Downloading f4m manifest', | |
97f4aecf S |
878 | 'Unable to download f4m manifest', |
879 | # Some manifests may be malformed, e.g. prosiebensat1 generated manifests | |
880 | # (see https://github.com/rg3/youtube-dl/issues/6215#issuecomment-121704244) | |
4de61310 S |
881 | transform_source=transform_source, |
882 | fatal=fatal) | |
883 | ||
884 | if manifest is False: | |
885 | return manifest | |
31bb8d3f JMF |
886 | |
887 | formats = [] | |
7a47d07c | 888 | manifest_version = '1.0' |
b2527359 | 889 | media_nodes = manifest.findall('{http://ns.adobe.com/f4m/1.0}media') |
34e48bed | 890 | if not media_nodes: |
7a47d07c | 891 | manifest_version = '2.0' |
34e48bed | 892 | media_nodes = manifest.findall('{http://ns.adobe.com/f4m/2.0}media') |
b2527359 | 893 | for i, media_el in enumerate(media_nodes): |
7a47d07c | 894 | if manifest_version == '2.0': |
31c746e5 S |
895 | media_url = media_el.attrib.get('href') or media_el.attrib.get('url') |
896 | if not media_url: | |
897 | continue | |
cc357c4d S |
898 | manifest_url = ( |
899 | media_url if media_url.startswith('http://') or media_url.startswith('https://') | |
900 | else ('/'.join(manifest_url.split('/')[:-1]) + '/' + media_url)) | |
70f0f5a8 S |
901 | # If media_url is itself a f4m manifest do the recursive extraction |
902 | # since bitrates in parent manifest (this one) and media_url manifest | |
903 | # may differ leading to inability to resolve the format by requested | |
904 | # bitrate in f4m downloader | |
905 | if determine_ext(manifest_url) == 'f4m': | |
4de61310 S |
906 | f4m_formats = self._extract_f4m_formats( |
907 | manifest_url, video_id, preference, f4m_id, fatal=fatal) | |
908 | if f4m_formats: | |
909 | formats.extend(f4m_formats) | |
70f0f5a8 | 910 | continue |
b2527359 | 911 | tbr = int_or_none(media_el.attrib.get('bitrate')) |
31bb8d3f | 912 | formats.append({ |
e21a55ab | 913 | 'format_id': '-'.join(filter(None, [f4m_id, compat_str(i if tbr is None else tbr)])), |
31bb8d3f JMF |
914 | 'url': manifest_url, |
915 | 'ext': 'flv', | |
b2527359 | 916 | 'tbr': tbr, |
31bb8d3f JMF |
917 | 'width': int_or_none(media_el.attrib.get('width')), |
918 | 'height': int_or_none(media_el.attrib.get('height')), | |
60ca389c | 919 | 'preference': preference, |
31bb8d3f JMF |
920 | }) |
921 | self._sort_formats(formats) | |
922 | ||
923 | return formats | |
924 | ||
f0b5d6af | 925 | def _extract_m3u8_formats(self, m3u8_url, video_id, ext=None, |
60ca389c | 926 | entry_protocol='m3u8', preference=None, |
13af92fd YCH |
927 | m3u8_id=None, note=None, errnote=None, |
928 | fatal=True): | |
f0b5d6af | 929 | |
704df56d | 930 | formats = [{ |
f207019c | 931 | 'format_id': '-'.join(filter(None, [m3u8_id, 'meta'])), |
704df56d PH |
932 | 'url': m3u8_url, |
933 | 'ext': ext, | |
934 | 'protocol': 'm3u8', | |
9fe6ef7a | 935 | 'preference': preference - 1 if preference else -1, |
704df56d PH |
936 | 'resolution': 'multiple', |
937 | 'format_note': 'Quality selection URL', | |
938 | }] | |
939 | ||
f0b5d6af PH |
940 | format_url = lambda u: ( |
941 | u | |
942 | if re.match(r'^https?://', u) | |
943 | else compat_urlparse.urljoin(m3u8_url, u)) | |
944 | ||
81515ad9 PH |
945 | m3u8_doc = self._download_webpage( |
946 | m3u8_url, video_id, | |
621ed9f5 | 947 | note=note or 'Downloading m3u8 information', |
13af92fd YCH |
948 | errnote=errnote or 'Failed to download m3u8 information', |
949 | fatal=fatal) | |
950 | if m3u8_doc is False: | |
951 | return m3u8_doc | |
704df56d | 952 | last_info = None |
fa156077 | 953 | last_media = None |
704df56d PH |
954 | kv_rex = re.compile( |
955 | r'(?P<key>[a-zA-Z_-]+)=(?P<val>"[^"]+"|[^",]+)(?:,|$)') | |
956 | for line in m3u8_doc.splitlines(): | |
957 | if line.startswith('#EXT-X-STREAM-INF:'): | |
958 | last_info = {} | |
959 | for m in kv_rex.finditer(line): | |
960 | v = m.group('val') | |
961 | if v.startswith('"'): | |
962 | v = v[1:-1] | |
963 | last_info[m.group('key')] = v | |
4cd95bcb JMF |
964 | elif line.startswith('#EXT-X-MEDIA:'): |
965 | last_media = {} | |
966 | for m in kv_rex.finditer(line): | |
967 | v = m.group('val') | |
968 | if v.startswith('"'): | |
969 | v = v[1:-1] | |
970 | last_media[m.group('key')] = v | |
704df56d PH |
971 | elif line.startswith('#') or not line.strip(): |
972 | continue | |
973 | else: | |
daebaab6 | 974 | if last_info is None: |
f0b5d6af | 975 | formats.append({'url': format_url(line)}) |
3524cc25 | 976 | continue |
704df56d | 977 | tbr = int_or_none(last_info.get('BANDWIDTH'), scale=1000) |
8dc9d361 S |
978 | format_id = [] |
979 | if m3u8_id: | |
980 | format_id.append(m3u8_id) | |
05d5392c | 981 | last_media_name = last_media.get('NAME') if last_media and last_media.get('TYPE') != 'SUBTITLES' else None |
8dc9d361 | 982 | format_id.append(last_media_name if last_media_name else '%d' % (tbr if tbr else len(formats))) |
704df56d | 983 | f = { |
8dc9d361 | 984 | 'format_id': '-'.join(format_id), |
f0b5d6af | 985 | 'url': format_url(line.strip()), |
704df56d PH |
986 | 'tbr': tbr, |
987 | 'ext': ext, | |
f0b5d6af PH |
988 | 'protocol': entry_protocol, |
989 | 'preference': preference, | |
704df56d PH |
990 | } |
991 | codecs = last_info.get('CODECS') | |
992 | if codecs: | |
9ebf22b7 S |
993 | # TODO: looks like video codec is not always necessarily goes first |
994 | va_codecs = codecs.split(',') | |
995 | if va_codecs[0]: | |
996 | f['vcodec'] = va_codecs[0].partition('.')[0] | |
997 | if len(va_codecs) > 1 and va_codecs[1]: | |
998 | f['acodec'] = va_codecs[1].partition('.')[0] | |
704df56d PH |
999 | resolution = last_info.get('RESOLUTION') |
1000 | if resolution: | |
1001 | width_str, height_str = resolution.split('x') | |
1002 | f['width'] = int(width_str) | |
1003 | f['height'] = int(height_str) | |
4cd95bcb JMF |
1004 | if last_media is not None: |
1005 | f['m3u8_media'] = last_media | |
1006 | last_media = None | |
704df56d PH |
1007 | formats.append(f) |
1008 | last_info = {} | |
1009 | self._sort_formats(formats) | |
1010 | return formats | |
1011 | ||
a107193e S |
1012 | @staticmethod |
1013 | def _xpath_ns(path, namespace=None): | |
1014 | if not namespace: | |
1015 | return path | |
1016 | out = [] | |
1017 | for c in path.split('/'): | |
1018 | if not c or c == '.': | |
1019 | out.append(c) | |
1020 | else: | |
1021 | out.append('{%s}%s' % (namespace, c)) | |
1022 | return '/'.join(out) | |
1023 | ||
1024 | def _extract_smil_formats(self, smil_url, video_id, fatal=True, f4m_params=None): | |
1025 | smil = self._download_smil(smil_url, video_id, fatal=fatal) | |
1026 | ||
995029a1 PH |
1027 | if smil is False: |
1028 | assert not fatal | |
1029 | return [] | |
e89a2aab | 1030 | |
17712eeb | 1031 | namespace = self._parse_smil_namespace(smil) |
a107193e S |
1032 | |
1033 | return self._parse_smil_formats( | |
1034 | smil, smil_url, video_id, namespace=namespace, f4m_params=f4m_params) | |
1035 | ||
1036 | def _extract_smil_info(self, smil_url, video_id, fatal=True, f4m_params=None): | |
1037 | smil = self._download_smil(smil_url, video_id, fatal=fatal) | |
1038 | if smil is False: | |
1039 | return {} | |
1040 | return self._parse_smil(smil, smil_url, video_id, f4m_params=f4m_params) | |
1041 | ||
1042 | def _download_smil(self, smil_url, video_id, fatal=True): | |
1043 | return self._download_xml( | |
1044 | smil_url, video_id, 'Downloading SMIL file', | |
1045 | 'Unable to download SMIL file', fatal=fatal) | |
1046 | ||
1047 | def _parse_smil(self, smil, smil_url, video_id, f4m_params=None): | |
17712eeb | 1048 | namespace = self._parse_smil_namespace(smil) |
a107193e S |
1049 | |
1050 | formats = self._parse_smil_formats( | |
1051 | smil, smil_url, video_id, namespace=namespace, f4m_params=f4m_params) | |
1052 | subtitles = self._parse_smil_subtitles(smil, namespace=namespace) | |
1053 | ||
1054 | video_id = os.path.splitext(url_basename(smil_url))[0] | |
1055 | title = None | |
1056 | description = None | |
647eab45 | 1057 | upload_date = None |
a107193e S |
1058 | for meta in smil.findall(self._xpath_ns('./head/meta', namespace)): |
1059 | name = meta.attrib.get('name') | |
1060 | content = meta.attrib.get('content') | |
1061 | if not name or not content: | |
1062 | continue | |
1063 | if not title and name == 'title': | |
1064 | title = content | |
1065 | elif not description and name in ('description', 'abstract'): | |
1066 | description = content | |
647eab45 S |
1067 | elif not upload_date and name == 'date': |
1068 | upload_date = unified_strdate(content) | |
a107193e | 1069 | |
1e5bcdec S |
1070 | thumbnails = [{ |
1071 | 'id': image.get('type'), | |
1072 | 'url': image.get('src'), | |
1073 | 'width': int_or_none(image.get('width')), | |
1074 | 'height': int_or_none(image.get('height')), | |
1075 | } for image in smil.findall(self._xpath_ns('.//image', namespace)) if image.get('src')] | |
1076 | ||
a107193e S |
1077 | return { |
1078 | 'id': video_id, | |
1079 | 'title': title or video_id, | |
1080 | 'description': description, | |
647eab45 | 1081 | 'upload_date': upload_date, |
1e5bcdec | 1082 | 'thumbnails': thumbnails, |
a107193e S |
1083 | 'formats': formats, |
1084 | 'subtitles': subtitles, | |
1085 | } | |
1086 | ||
17712eeb S |
1087 | def _parse_smil_namespace(self, smil): |
1088 | return self._search_regex( | |
1089 | r'(?i)^{([^}]+)?}smil$', smil.tag, 'namespace', default=None) | |
1090 | ||
f877c6ae | 1091 | def _parse_smil_formats(self, smil, smil_url, video_id, namespace=None, f4m_params=None, transform_rtmp_url=None): |
a107193e S |
1092 | base = smil_url |
1093 | for meta in smil.findall(self._xpath_ns('./head/meta', namespace)): | |
1094 | b = meta.get('base') or meta.get('httpBase') | |
1095 | if b: | |
1096 | base = b | |
1097 | break | |
e89a2aab S |
1098 | |
1099 | formats = [] | |
1100 | rtmp_count = 0 | |
a107193e S |
1101 | http_count = 0 |
1102 | ||
1103 | videos = smil.findall(self._xpath_ns('.//video', namespace)) | |
1104 | for video in videos: | |
1105 | src = video.get('src') | |
1106 | if not src: | |
1107 | continue | |
1108 | ||
e7d8e98a | 1109 | bitrate = float_or_none(video.get('system-bitrate') or video.get('systemBitrate'), 1000) |
a107193e S |
1110 | filesize = int_or_none(video.get('size') or video.get('fileSize')) |
1111 | width = int_or_none(video.get('width')) | |
1112 | height = int_or_none(video.get('height')) | |
1113 | proto = video.get('proto') | |
1114 | ext = video.get('ext') | |
1115 | src_ext = determine_ext(src) | |
1116 | streamer = video.get('streamer') or base | |
1117 | ||
1118 | if proto == 'rtmp' or streamer.startswith('rtmp'): | |
1119 | rtmp_count += 1 | |
1120 | formats.append({ | |
1121 | 'url': streamer, | |
1122 | 'play_path': src, | |
1123 | 'ext': 'flv', | |
1124 | 'format_id': 'rtmp-%d' % (rtmp_count if bitrate is None else bitrate), | |
1125 | 'tbr': bitrate, | |
1126 | 'filesize': filesize, | |
1127 | 'width': width, | |
1128 | 'height': height, | |
1129 | }) | |
f877c6ae YCH |
1130 | if transform_rtmp_url: |
1131 | streamer, src = transform_rtmp_url(streamer, src) | |
1132 | formats[-1].update({ | |
1133 | 'url': streamer, | |
1134 | 'play_path': src, | |
1135 | }) | |
a107193e S |
1136 | continue |
1137 | ||
1138 | src_url = src if src.startswith('http') else compat_urlparse.urljoin(base, src) | |
1139 | ||
1140 | if proto == 'm3u8' or src_ext == 'm3u8': | |
3a1341a7 S |
1141 | m3u8_formats = self._extract_m3u8_formats( |
1142 | src_url, video_id, ext or 'mp4', m3u8_id='hls', fatal=False) | |
1143 | if m3u8_formats: | |
1144 | formats.extend(m3u8_formats) | |
a107193e S |
1145 | continue |
1146 | ||
1147 | if src_ext == 'f4m': | |
1148 | f4m_url = src_url | |
1149 | if not f4m_params: | |
1150 | f4m_params = { | |
1151 | 'hdcore': '3.2.0', | |
1152 | 'plugin': 'flowplayer-3.2.0.1', | |
1153 | } | |
1154 | f4m_url += '&' if '?' in f4m_url else '?' | |
41c3a5a7 | 1155 | f4m_url += compat_urllib_parse.urlencode(f4m_params) |
e5851b96 S |
1156 | f4m_formats = self._extract_f4m_formats(f4m_url, video_id, f4m_id='hds', fatal=False) |
1157 | if f4m_formats: | |
1158 | formats.extend(f4m_formats) | |
a107193e S |
1159 | continue |
1160 | ||
c78e4817 | 1161 | if src_url.startswith('http') and self._is_valid_url(src, video_id): |
a107193e S |
1162 | http_count += 1 |
1163 | formats.append({ | |
1164 | 'url': src_url, | |
1165 | 'ext': ext or src_ext or 'flv', | |
1166 | 'format_id': 'http-%d' % (bitrate or http_count), | |
1167 | 'tbr': bitrate, | |
1168 | 'filesize': filesize, | |
1169 | 'width': width, | |
1170 | 'height': height, | |
1171 | }) | |
1172 | continue | |
63757032 | 1173 | |
e89a2aab S |
1174 | self._sort_formats(formats) |
1175 | ||
1176 | return formats | |
1177 | ||
ce00af87 | 1178 | def _parse_smil_subtitles(self, smil, namespace=None, subtitles_lang='en'): |
a107193e S |
1179 | subtitles = {} |
1180 | for num, textstream in enumerate(smil.findall(self._xpath_ns('.//textstream', namespace))): | |
1181 | src = textstream.get('src') | |
1182 | if not src: | |
1183 | continue | |
1184 | ext = textstream.get('ext') or determine_ext(src) | |
1185 | if not ext: | |
1186 | type_ = textstream.get('type') | |
5cdefc46 S |
1187 | SUBTITLES_TYPES = { |
1188 | 'text/vtt': 'vtt', | |
1189 | 'text/srt': 'srt', | |
1190 | 'application/smptett+xml': 'tt', | |
1191 | } | |
1192 | if type_ in SUBTITLES_TYPES: | |
1193 | ext = SUBTITLES_TYPES[type_] | |
03bc7237 | 1194 | lang = textstream.get('systemLanguage') or textstream.get('systemLanguageName') or textstream.get('lang') or subtitles_lang |
a107193e S |
1195 | subtitles.setdefault(lang, []).append({ |
1196 | 'url': src, | |
1197 | 'ext': ext, | |
1198 | }) | |
1199 | return subtitles | |
63757032 | 1200 | |
942acef5 S |
1201 | def _extract_xspf_playlist(self, playlist_url, playlist_id, fatal=True): |
1202 | xspf = self._download_xml( | |
8d6765cf | 1203 | playlist_url, playlist_id, 'Downloading xpsf playlist', |
942acef5 S |
1204 | 'Unable to download xspf manifest', fatal=fatal) |
1205 | if xspf is False: | |
1206 | return [] | |
1207 | return self._parse_xspf(xspf, playlist_id) | |
8d6765cf | 1208 | |
942acef5 | 1209 | def _parse_xspf(self, playlist, playlist_id): |
8d6765cf S |
1210 | NS_MAP = { |
1211 | 'xspf': 'http://xspf.org/ns/0/', | |
1212 | 's1': 'http://static.streamone.nl/player/ns/0', | |
1213 | } | |
1214 | ||
1215 | entries = [] | |
1216 | for track in playlist.findall(xpath_with_ns('./xspf:trackList/xspf:track', NS_MAP)): | |
1217 | title = xpath_text( | |
98044462 | 1218 | track, xpath_with_ns('./xspf:title', NS_MAP), 'title', default=playlist_id) |
8d6765cf S |
1219 | description = xpath_text( |
1220 | track, xpath_with_ns('./xspf:annotation', NS_MAP), 'description') | |
1221 | thumbnail = xpath_text( | |
1222 | track, xpath_with_ns('./xspf:image', NS_MAP), 'thumbnail') | |
1223 | duration = float_or_none( | |
1224 | xpath_text(track, xpath_with_ns('./xspf:duration', NS_MAP), 'duration'), 1000) | |
1225 | ||
1226 | formats = [{ | |
1227 | 'url': location.text, | |
1228 | 'format_id': location.get(xpath_with_ns('s1:label', NS_MAP)), | |
1229 | 'width': int_or_none(location.get(xpath_with_ns('s1:width', NS_MAP))), | |
1230 | 'height': int_or_none(location.get(xpath_with_ns('s1:height', NS_MAP))), | |
1231 | } for location in track.findall(xpath_with_ns('./xspf:location', NS_MAP))] | |
1232 | self._sort_formats(formats) | |
1233 | ||
1234 | entries.append({ | |
1235 | 'id': playlist_id, | |
1236 | 'title': title, | |
1237 | 'description': description, | |
1238 | 'thumbnail': thumbnail, | |
1239 | 'duration': duration, | |
1240 | 'formats': formats, | |
1241 | }) | |
1242 | return entries | |
1243 | ||
f4b1c7ad PH |
1244 | def _live_title(self, name): |
1245 | """ Generate the title for a live video """ | |
1246 | now = datetime.datetime.now() | |
1247 | now_str = now.strftime("%Y-%m-%d %H:%M") | |
1248 | return name + ' ' + now_str | |
1249 | ||
b14f3a4c PH |
1250 | def _int(self, v, name, fatal=False, **kwargs): |
1251 | res = int_or_none(v, **kwargs) | |
1252 | if 'get_attr' in kwargs: | |
1253 | print(getattr(v, kwargs['get_attr'])) | |
1254 | if res is None: | |
1255 | msg = 'Failed to extract %s: Could not parse value %r' % (name, v) | |
1256 | if fatal: | |
1257 | raise ExtractorError(msg) | |
1258 | else: | |
1259 | self._downloader.report_warning(msg) | |
1260 | return res | |
1261 | ||
1262 | def _float(self, v, name, fatal=False, **kwargs): | |
1263 | res = float_or_none(v, **kwargs) | |
1264 | if res is None: | |
1265 | msg = 'Failed to extract %s: Could not parse value %r' % (name, v) | |
1266 | if fatal: | |
1267 | raise ExtractorError(msg) | |
1268 | else: | |
1269 | self._downloader.report_warning(msg) | |
1270 | return res | |
1271 | ||
42939b61 | 1272 | def _set_cookie(self, domain, name, value, expire_time=None): |
810fb84d PH |
1273 | cookie = compat_cookiejar.Cookie( |
1274 | 0, name, value, None, None, domain, None, | |
42939b61 JMF |
1275 | None, '/', True, False, expire_time, '', None, None, None) |
1276 | self._downloader.cookiejar.set_cookie(cookie) | |
1277 | ||
799207e8 | 1278 | def _get_cookies(self, url): |
1279 | """ Return a compat_cookies.SimpleCookie with the cookies for the url """ | |
1280 | req = compat_urllib_request.Request(url) | |
1281 | self._downloader.cookiejar.add_cookie_header(req) | |
1282 | return compat_cookies.SimpleCookie(req.get_header('Cookie')) | |
1283 | ||
05900629 PH |
1284 | def get_testcases(self, include_onlymatching=False): |
1285 | t = getattr(self, '_TEST', None) | |
1286 | if t: | |
1287 | assert not hasattr(self, '_TESTS'), \ | |
1288 | '%s has _TEST and _TESTS' % type(self).__name__ | |
1289 | tests = [t] | |
1290 | else: | |
1291 | tests = getattr(self, '_TESTS', []) | |
1292 | for t in tests: | |
1293 | if not include_onlymatching and t.get('only_matching', False): | |
1294 | continue | |
1295 | t['name'] = type(self).__name__[:-len('IE')] | |
1296 | yield t | |
1297 | ||
1298 | def is_suitable(self, age_limit): | |
1299 | """ Test whether the extractor is generally suitable for the given | |
1300 | age limit (i.e. pornographic sites are not, all others usually are) """ | |
1301 | ||
1302 | any_restricted = False | |
1303 | for tc in self.get_testcases(include_onlymatching=False): | |
1304 | if 'playlist' in tc: | |
1305 | tc = tc['playlist'][0] | |
1306 | is_restricted = age_restricted( | |
1307 | tc.get('info_dict', {}).get('age_limit'), age_limit) | |
1308 | if not is_restricted: | |
1309 | return True | |
1310 | any_restricted = any_restricted or is_restricted | |
1311 | return not any_restricted | |
1312 | ||
a504ced0 | 1313 | def extract_subtitles(self, *args, **kwargs): |
9868ea49 JMF |
1314 | if (self._downloader.params.get('writesubtitles', False) or |
1315 | self._downloader.params.get('listsubtitles')): | |
1316 | return self._get_subtitles(*args, **kwargs) | |
1317 | return {} | |
a504ced0 JMF |
1318 | |
1319 | def _get_subtitles(self, *args, **kwargs): | |
1320 | raise NotImplementedError("This method must be implemented by subclasses") | |
1321 | ||
912e0b7e YCH |
1322 | @staticmethod |
1323 | def _merge_subtitle_items(subtitle_list1, subtitle_list2): | |
1324 | """ Merge subtitle items for one language. Items with duplicated URLs | |
1325 | will be dropped. """ | |
1326 | list1_urls = set([item['url'] for item in subtitle_list1]) | |
1327 | ret = list(subtitle_list1) | |
1328 | ret.extend([item for item in subtitle_list2 if item['url'] not in list1_urls]) | |
1329 | return ret | |
1330 | ||
1331 | @classmethod | |
8c97f819 | 1332 | def _merge_subtitles(cls, subtitle_dict1, subtitle_dict2): |
912e0b7e | 1333 | """ Merge two subtitle dictionaries, language by language. """ |
912e0b7e YCH |
1334 | ret = dict(subtitle_dict1) |
1335 | for lang in subtitle_dict2: | |
8c97f819 | 1336 | ret[lang] = cls._merge_subtitle_items(subtitle_dict1.get(lang, []), subtitle_dict2[lang]) |
912e0b7e YCH |
1337 | return ret |
1338 | ||
360e1ca5 | 1339 | def extract_automatic_captions(self, *args, **kwargs): |
9868ea49 JMF |
1340 | if (self._downloader.params.get('writeautomaticsub', False) or |
1341 | self._downloader.params.get('listsubtitles')): | |
1342 | return self._get_automatic_captions(*args, **kwargs) | |
1343 | return {} | |
360e1ca5 JMF |
1344 | |
1345 | def _get_automatic_captions(self, *args, **kwargs): | |
1346 | raise NotImplementedError("This method must be implemented by subclasses") | |
1347 | ||
8dbe9899 | 1348 | |
d6983cb4 PH |
1349 | class SearchInfoExtractor(InfoExtractor): |
1350 | """ | |
1351 | Base class for paged search queries extractors. | |
10952eb2 | 1352 | They accept URLs in the format _SEARCH_KEY(|all|[0-9]):{query} |
d6983cb4 PH |
1353 | Instances should define _SEARCH_KEY and _MAX_RESULTS. |
1354 | """ | |
1355 | ||
1356 | @classmethod | |
1357 | def _make_valid_url(cls): | |
1358 | return r'%s(?P<prefix>|[1-9][0-9]*|all):(?P<query>[\s\S]+)' % cls._SEARCH_KEY | |
1359 | ||
1360 | @classmethod | |
1361 | def suitable(cls, url): | |
1362 | return re.match(cls._make_valid_url(), url) is not None | |
1363 | ||
1364 | def _real_extract(self, query): | |
1365 | mobj = re.match(self._make_valid_url(), query) | |
1366 | if mobj is None: | |
f1a9d64e | 1367 | raise ExtractorError('Invalid search query "%s"' % query) |
d6983cb4 PH |
1368 | |
1369 | prefix = mobj.group('prefix') | |
1370 | query = mobj.group('query') | |
1371 | if prefix == '': | |
1372 | return self._get_n_results(query, 1) | |
1373 | elif prefix == 'all': | |
1374 | return self._get_n_results(query, self._MAX_RESULTS) | |
1375 | else: | |
1376 | n = int(prefix) | |
1377 | if n <= 0: | |
f1a9d64e | 1378 | raise ExtractorError('invalid download number %s for query "%s"' % (n, query)) |
d6983cb4 | 1379 | elif n > self._MAX_RESULTS: |
f1a9d64e | 1380 | self._downloader.report_warning('%s returns max %i results (you requested %i)' % (self._SEARCH_KEY, self._MAX_RESULTS, n)) |
d6983cb4 PH |
1381 | n = self._MAX_RESULTS |
1382 | return self._get_n_results(query, n) | |
1383 | ||
1384 | def _get_n_results(self, query, n): | |
1385 | """Get a specified number of results for a query""" | |
416a5efc | 1386 | raise NotImplementedError("This method must be implemented by subclasses") |
0f818663 PH |
1387 | |
1388 | @property | |
1389 | def SEARCH_KEY(self): | |
1390 | return self._SEARCH_KEY |