]>
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 ( |
d6983cb4 PH |
16 | compat_http_client, |
17 | compat_urllib_error, | |
c7deaa4c | 18 | compat_urllib_parse_urlparse, |
f0b5d6af | 19 | compat_urlparse, |
d6983cb4 | 20 | compat_str, |
8c25f81b PH |
21 | ) |
22 | from ..utils import ( | |
d6983cb4 PH |
23 | clean_html, |
24 | compiled_regex_type, | |
25 | ExtractorError, | |
b14f3a4c | 26 | float_or_none, |
31bb8d3f | 27 | int_or_none, |
55b3e45b | 28 | RegexNotFoundError, |
d41e6efc | 29 | sanitize_filename, |
f38de77f | 30 | unescapeHTML, |
d6983cb4 | 31 | ) |
46374a56 | 32 | _NO_DEFAULT = object() |
d6983cb4 | 33 | |
dca08720 | 34 | |
d6983cb4 PH |
35 | class InfoExtractor(object): |
36 | """Information Extractor class. | |
37 | ||
38 | Information extractors are the classes that, given a URL, extract | |
39 | information about the video (or videos) the URL refers to. This | |
40 | information includes the real video URL, the video title, author and | |
41 | others. The information is stored in a dictionary which is then | |
42 | passed to the FileDownloader. The FileDownloader processes this | |
43 | information possibly downloading the video to the file system, among | |
44 | other possible outcomes. | |
45 | ||
46 | The dictionaries must include the following fields: | |
47 | ||
48 | id: Video identifier. | |
d6983cb4 | 49 | title: Video title, unescaped. |
d67b0b15 | 50 | |
f49d89ee | 51 | Additionally, it must contain either a formats entry or a url one: |
d67b0b15 | 52 | |
f49d89ee PH |
53 | formats: A list of dictionaries for each format available, ordered |
54 | from worst to best quality. | |
55 | ||
56 | Potential fields: | |
d67b0b15 PH |
57 | * url Mandatory. The URL of the video file |
58 | * ext Will be calculated from url if missing | |
59 | * format A human-readable description of the format | |
60 | ("mp4 container with h264/opus"). | |
61 | Calculated from the format_id, width, height. | |
62 | and format_note fields if missing. | |
63 | * format_id A short description of the format | |
5d4f3985 PH |
64 | ("mp4_h264_opus" or "19"). |
65 | Technically optional, but strongly recommended. | |
d67b0b15 PH |
66 | * format_note Additional info about the format |
67 | ("3D" or "DASH video") | |
68 | * width Width of the video, if known | |
69 | * height Height of the video, if known | |
f49d89ee | 70 | * resolution Textual description of width and height |
7217e148 | 71 | * tbr Average bitrate of audio and video in KBit/s |
d67b0b15 PH |
72 | * abr Average audio bitrate in KBit/s |
73 | * acodec Name of the audio codec in use | |
dd27fd17 | 74 | * asr Audio sampling rate in Hertz |
d67b0b15 | 75 | * vbr Average video bitrate in KBit/s |
fbb21cf5 | 76 | * fps Frame rate |
d67b0b15 | 77 | * vcodec Name of the video codec in use |
1394ce65 | 78 | * container Name of the container format |
d67b0b15 | 79 | * filesize The number of bytes, if known in advance |
9732d77e | 80 | * filesize_approx An estimate for the number of bytes |
d67b0b15 | 81 | * player_url SWF Player URL (used for rtmpdump). |
c7deaa4c PH |
82 | * protocol The protocol that will be used for the actual |
83 | download, lower-case. | |
db1f3888 | 84 | "http", "https", "rtsp", "rtmp", "m3u8" or so. |
f49d89ee | 85 | * preference Order number of this format. If this field is |
08d13955 | 86 | present and not None, the formats get sorted |
38d63d84 | 87 | by this field, regardless of all other values. |
f49d89ee PH |
88 | -1 for default (order by other properties), |
89 | -2 or smaller for less than default. | |
5d73273f PH |
90 | * quality Order number of the video quality of this |
91 | format, irrespective of the file format. | |
92 | -1 for default (order by other properties), | |
93 | -2 or smaller for less than default. | |
c64ed2a3 PH |
94 | * source_preference Order number for this video source |
95 | (quality takes higher priority) | |
96 | -1 for default (order by other properties), | |
97 | -2 or smaller for less than default. | |
d769be6c PH |
98 | * http_referer HTTP Referer header value to set. |
99 | * http_method HTTP method to use for the download. | |
100 | * http_headers A dictionary of additional HTTP headers | |
101 | to add to the request. | |
102 | * http_post_data Additional data to send with a POST | |
103 | request. | |
c0ba0f48 | 104 | url: Final video URL. |
d6983cb4 | 105 | ext: Video filename extension. |
d67b0b15 PH |
106 | format: The video format, defaults to ext (used for --get-format) |
107 | player_url: SWF Player URL (used for rtmpdump). | |
2f5865cc | 108 | |
d6983cb4 PH |
109 | The following fields are optional: |
110 | ||
0afef30b PH |
111 | display_id An alternative identifier for the video, not necessarily |
112 | unique, but available before title. Typically, id is | |
113 | something like "4234987", title "Dancing naked mole rats", | |
114 | and display_id "dancing-naked-mole-rats" | |
d5519808 PH |
115 | thumbnails: A list of dictionaries, with the following entries: |
116 | * "url" | |
117 | * "width" (optional, int) | |
118 | * "height" (optional, int) | |
119 | * "resolution" (optional, string "{width}x{height"}, | |
120 | deprecated) | |
d6983cb4 PH |
121 | thumbnail: Full URL to a video thumbnail image. |
122 | description: One-line video description. | |
123 | uploader: Full name of the video uploader. | |
955c4514 | 124 | timestamp: UNIX timestamp of the moment the video became available. |
d6983cb4 | 125 | upload_date: Video upload date (YYYYMMDD). |
955c4514 | 126 | If not explicitly set, calculated from timestamp. |
d6983cb4 | 127 | uploader_id: Nickname or id of the video uploader. |
da9ec3b9 | 128 | location: Physical location where the video was filmed. |
5d51a883 JMF |
129 | subtitles: The subtitle file contents as a dictionary in the format |
130 | {language: subtitles}. | |
c0ba0f48 | 131 | duration: Length of the video in seconds, as an integer. |
f3d29461 | 132 | view_count: How many users have watched the video on the platform. |
19e3dfc9 PH |
133 | like_count: Number of positive ratings of the video |
134 | dislike_count: Number of negative ratings of the video | |
135 | comment_count: Number of comments on the video | |
8dbe9899 | 136 | age_limit: Age restriction for the video, as an integer (years) |
9103bbc5 JMF |
137 | webpage_url: The url to the video webpage, if given to youtube-dl it |
138 | should allow to get the same result again. (It will be set | |
139 | by YoutubeDL if it's missing) | |
ad3bc6ac PH |
140 | categories: A list of categories that the video falls in, for example |
141 | ["Sports", "Berlin"] | |
7267bd53 PH |
142 | is_live: True, False, or None (=unknown). Whether this video is a |
143 | live stream that goes on instead of a fixed-length video. | |
d6983cb4 | 144 | |
deefc05b | 145 | Unless mentioned otherwise, the fields should be Unicode strings. |
d6983cb4 | 146 | |
d838b1bd PH |
147 | Unless mentioned otherwise, None is equivalent to absence of information. |
148 | ||
d6983cb4 PH |
149 | Subclasses of this one should re-define the _real_initialize() and |
150 | _real_extract() methods and define a _VALID_URL regexp. | |
151 | Probably, they should also be added to the list of extractors. | |
152 | ||
d6983cb4 PH |
153 | Finally, the _WORKING attribute should be set to False for broken IEs |
154 | in order to warn the users and skip the tests. | |
155 | """ | |
156 | ||
157 | _ready = False | |
158 | _downloader = None | |
159 | _WORKING = True | |
160 | ||
161 | def __init__(self, downloader=None): | |
162 | """Constructor. Receives an optional downloader.""" | |
163 | self._ready = False | |
164 | self.set_downloader(downloader) | |
165 | ||
166 | @classmethod | |
167 | def suitable(cls, url): | |
168 | """Receives a URL and returns True if suitable for this IE.""" | |
79cb2577 PH |
169 | |
170 | # This does not use has/getattr intentionally - we want to know whether | |
171 | # we have cached the regexp for *this* class, whereas getattr would also | |
172 | # match the superclass | |
173 | if '_VALID_URL_RE' not in cls.__dict__: | |
174 | cls._VALID_URL_RE = re.compile(cls._VALID_URL) | |
175 | return cls._VALID_URL_RE.match(url) is not None | |
d6983cb4 | 176 | |
ed9266db PH |
177 | @classmethod |
178 | def _match_id(cls, url): | |
179 | if '_VALID_URL_RE' not in cls.__dict__: | |
180 | cls._VALID_URL_RE = re.compile(cls._VALID_URL) | |
181 | m = cls._VALID_URL_RE.match(url) | |
182 | assert m | |
183 | return m.group('id') | |
184 | ||
d6983cb4 PH |
185 | @classmethod |
186 | def working(cls): | |
187 | """Getter method for _WORKING.""" | |
188 | return cls._WORKING | |
189 | ||
190 | def initialize(self): | |
191 | """Initializes an instance (authentication, etc).""" | |
192 | if not self._ready: | |
193 | self._real_initialize() | |
194 | self._ready = True | |
195 | ||
196 | def extract(self, url): | |
197 | """Extracts URL information and returns it in list of dicts.""" | |
198 | self.initialize() | |
199 | return self._real_extract(url) | |
200 | ||
201 | def set_downloader(self, downloader): | |
202 | """Sets the downloader for this IE.""" | |
203 | self._downloader = downloader | |
204 | ||
205 | def _real_initialize(self): | |
206 | """Real initialization process. Redefine in subclasses.""" | |
207 | pass | |
208 | ||
209 | def _real_extract(self, url): | |
210 | """Real extraction process. Redefine in subclasses.""" | |
211 | pass | |
212 | ||
56c73665 JMF |
213 | @classmethod |
214 | def ie_key(cls): | |
215 | """A string for getting the InfoExtractor with get_info_extractor""" | |
216 | return cls.__name__[:-2] | |
217 | ||
d6983cb4 PH |
218 | @property |
219 | def IE_NAME(self): | |
220 | return type(self).__name__[:-2] | |
221 | ||
7cc3570e | 222 | def _request_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True): |
d6983cb4 PH |
223 | """ Returns the response handle """ |
224 | if note is None: | |
225 | self.report_download_webpage(video_id) | |
226 | elif note is not False: | |
7cc3570e | 227 | if video_id is None: |
f1a9d64e | 228 | self.to_screen('%s' % (note,)) |
7cc3570e | 229 | else: |
f1a9d64e | 230 | self.to_screen('%s: %s' % (video_id, note)) |
d6983cb4 | 231 | try: |
dca08720 | 232 | return self._downloader.urlopen(url_or_request) |
d6983cb4 | 233 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: |
aa94a6d3 PH |
234 | if errnote is False: |
235 | return False | |
d6983cb4 | 236 | if errnote is None: |
f1a9d64e PH |
237 | errnote = 'Unable to download webpage' |
238 | errmsg = '%s: %s' % (errnote, compat_str(err)) | |
7cc3570e PH |
239 | if fatal: |
240 | raise ExtractorError(errmsg, sys.exc_info()[2], cause=err) | |
241 | else: | |
242 | self._downloader.report_warning(errmsg) | |
243 | return False | |
d6983cb4 | 244 | |
7cc3570e | 245 | def _download_webpage_handle(self, url_or_request, video_id, note=None, errnote=None, fatal=True): |
d6983cb4 | 246 | """ Returns a tuple (page content as string, URL handle) """ |
b9d3e163 PH |
247 | # Strip hashes from the URL (#1038) |
248 | if isinstance(url_or_request, (compat_str, str)): | |
249 | url_or_request = url_or_request.partition('#')[0] | |
250 | ||
7cc3570e PH |
251 | urlh = self._request_webpage(url_or_request, video_id, note, errnote, fatal) |
252 | if urlh is False: | |
253 | assert not fatal | |
254 | return False | |
23be51d8 PH |
255 | content = self._webpage_read_content(urlh, url_or_request, video_id, note, errnote, fatal) |
256 | return (content, urlh) | |
257 | ||
258 | def _webpage_read_content(self, urlh, url_or_request, video_id, note=None, errnote=None, fatal=True): | |
d6983cb4 | 259 | content_type = urlh.headers.get('Content-Type', '') |
f143d86a | 260 | webpage_bytes = urlh.read() |
d6983cb4 PH |
261 | m = re.match(r'[a-zA-Z0-9_.-]+/[a-zA-Z0-9_.-]+\s*;\s*charset=(.+)', content_type) |
262 | if m: | |
263 | encoding = m.group(1) | |
264 | else: | |
0d75ae2c | 265 | m = re.search(br'<meta[^>]+charset=[\'"]?([^\'")]+)[ /\'">]', |
f143d86a PH |
266 | webpage_bytes[:1024]) |
267 | if m: | |
268 | encoding = m.group(1).decode('ascii') | |
b60016e8 PH |
269 | elif webpage_bytes.startswith(b'\xff\xfe'): |
270 | encoding = 'utf-16' | |
f143d86a PH |
271 | else: |
272 | encoding = 'utf-8' | |
d6983cb4 PH |
273 | if self._downloader.params.get('dump_intermediate_pages', False): |
274 | try: | |
275 | url = url_or_request.get_full_url() | |
276 | except AttributeError: | |
277 | url = url_or_request | |
f1a9d64e | 278 | self.to_screen('Dumping request to ' + url) |
d6983cb4 PH |
279 | dump = base64.b64encode(webpage_bytes).decode('ascii') |
280 | self._downloader.to_screen(dump) | |
d41e6efc PH |
281 | if self._downloader.params.get('write_pages', False): |
282 | try: | |
283 | url = url_or_request.get_full_url() | |
284 | except AttributeError: | |
285 | url = url_or_request | |
5afa7f8b | 286 | basen = '%s_%s' % (video_id, url) |
c1bce22f | 287 | if len(basen) > 240: |
f1a9d64e | 288 | h = '___' + hashlib.md5(basen.encode('utf-8')).hexdigest() |
c1bce22f PH |
289 | basen = basen[:240 - len(h)] + h |
290 | raw_filename = basen + '.dump' | |
d41e6efc | 291 | filename = sanitize_filename(raw_filename, restricted=True) |
f1a9d64e | 292 | self.to_screen('Saving request to ' + filename) |
5f58165d S |
293 | # Working around MAX_PATH limitation on Windows (see |
294 | # http://msdn.microsoft.com/en-us/library/windows/desktop/aa365247(v=vs.85).aspx) | |
295 | if os.name == 'nt': | |
296 | absfilepath = os.path.abspath(filename) | |
297 | if len(absfilepath) > 259: | |
298 | filename = '\\\\?\\' + absfilepath | |
d41e6efc PH |
299 | with open(filename, 'wb') as outf: |
300 | outf.write(webpage_bytes) | |
301 | ||
ec0fafbb AA |
302 | try: |
303 | content = webpage_bytes.decode(encoding, 'replace') | |
304 | except LookupError: | |
305 | content = webpage_bytes.decode('utf-8', 'replace') | |
2410c43d | 306 | |
f1a9d64e PH |
307 | if ('<title>Access to this site is blocked</title>' in content and |
308 | 'Websense' in content[:512]): | |
309 | msg = 'Access to this webpage has been blocked by Websense filtering software in your network.' | |
2410c43d PH |
310 | blocked_iframe = self._html_search_regex( |
311 | r'<iframe src="([^"]+)"', content, | |
f1a9d64e | 312 | 'Websense information URL', default=None) |
2410c43d | 313 | if blocked_iframe: |
f1a9d64e | 314 | msg += ' Visit %s for more details' % blocked_iframe |
2410c43d PH |
315 | raise ExtractorError(msg, expected=True) |
316 | ||
23be51d8 | 317 | return content |
d6983cb4 | 318 | |
7cc3570e | 319 | def _download_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True): |
d6983cb4 | 320 | """ Returns the data of the page as a string """ |
7cc3570e PH |
321 | res = self._download_webpage_handle(url_or_request, video_id, note, errnote, fatal) |
322 | if res is False: | |
323 | return res | |
324 | else: | |
325 | content, _ = res | |
326 | return content | |
d6983cb4 | 327 | |
2a275ab0 | 328 | def _download_xml(self, url_or_request, video_id, |
f1a9d64e | 329 | note='Downloading XML', errnote='Unable to download XML', |
28746fbd | 330 | transform_source=None, fatal=True): |
267ed0c5 | 331 | """Return the xml as an xml.etree.ElementTree.Element""" |
28746fbd PH |
332 | xml_string = self._download_webpage( |
333 | url_or_request, video_id, note, errnote, fatal=fatal) | |
334 | if xml_string is False: | |
335 | return xml_string | |
e2b38da9 PH |
336 | if transform_source: |
337 | xml_string = transform_source(xml_string) | |
267ed0c5 JMF |
338 | return xml.etree.ElementTree.fromstring(xml_string.encode('utf-8')) |
339 | ||
3d3538e4 | 340 | def _download_json(self, url_or_request, video_id, |
f1a9d64e PH |
341 | note='Downloading JSON metadata', |
342 | errnote='Unable to download JSON metadata', | |
b090af59 PH |
343 | transform_source=None, |
344 | fatal=True): | |
345 | json_string = self._download_webpage( | |
346 | url_or_request, video_id, note, errnote, fatal=fatal) | |
347 | if (not fatal) and json_string is False: | |
348 | return None | |
81c2f20b PH |
349 | if transform_source: |
350 | json_string = transform_source(json_string) | |
3d3538e4 PH |
351 | try: |
352 | return json.loads(json_string) | |
353 | except ValueError as ve: | |
e7b6d122 PH |
354 | errmsg = '%s: Failed to parse JSON ' % video_id |
355 | if fatal: | |
356 | raise ExtractorError(errmsg, cause=ve) | |
357 | else: | |
358 | self.report_warning(errmsg + str(ve)) | |
3d3538e4 | 359 | |
f45f96f8 | 360 | def report_warning(self, msg, video_id=None): |
f1a9d64e | 361 | idstr = '' if video_id is None else '%s: ' % video_id |
f45f96f8 | 362 | self._downloader.report_warning( |
f1a9d64e | 363 | '[%s] %s%s' % (self.IE_NAME, idstr, msg)) |
f45f96f8 | 364 | |
d6983cb4 PH |
365 | def to_screen(self, msg): |
366 | """Print msg to screen, prefixing it with '[ie_name]'""" | |
f1a9d64e | 367 | self._downloader.to_screen('[%s] %s' % (self.IE_NAME, msg)) |
d6983cb4 PH |
368 | |
369 | def report_extraction(self, id_or_name): | |
370 | """Report information extraction.""" | |
f1a9d64e | 371 | self.to_screen('%s: Extracting information' % id_or_name) |
d6983cb4 PH |
372 | |
373 | def report_download_webpage(self, video_id): | |
374 | """Report webpage download.""" | |
f1a9d64e | 375 | self.to_screen('%s: Downloading webpage' % video_id) |
d6983cb4 PH |
376 | |
377 | def report_age_confirmation(self): | |
378 | """Report attempt to confirm age.""" | |
f1a9d64e | 379 | self.to_screen('Confirming age') |
d6983cb4 | 380 | |
fc79158d JMF |
381 | def report_login(self): |
382 | """Report attempt to log in.""" | |
f1a9d64e | 383 | self.to_screen('Logging in') |
fc79158d | 384 | |
d6983cb4 | 385 | #Methods for following #608 |
c0d0b01f JMF |
386 | @staticmethod |
387 | def url_result(url, ie=None, video_id=None): | |
d6983cb4 PH |
388 | """Returns a url that points to a page that should be processed""" |
389 | #TODO: ie should be the class used for getting the info | |
390 | video_info = {'_type': 'url', | |
391 | 'url': url, | |
392 | 'ie_key': ie} | |
7012b23c PH |
393 | if video_id is not None: |
394 | video_info['id'] = video_id | |
d6983cb4 | 395 | return video_info |
c0d0b01f JMF |
396 | @staticmethod |
397 | def playlist_result(entries, playlist_id=None, playlist_title=None): | |
d6983cb4 PH |
398 | """Returns a playlist""" |
399 | video_info = {'_type': 'playlist', | |
400 | 'entries': entries} | |
401 | if playlist_id: | |
402 | video_info['id'] = playlist_id | |
403 | if playlist_title: | |
404 | video_info['title'] = playlist_title | |
405 | return video_info | |
406 | ||
46374a56 | 407 | def _search_regex(self, pattern, string, name, default=_NO_DEFAULT, fatal=True, flags=0): |
d6983cb4 PH |
408 | """ |
409 | Perform a regex search on the given string, using a single or a list of | |
410 | patterns returning the first matching group. | |
411 | In case of failure return a default value or raise a WARNING or a | |
55b3e45b | 412 | RegexNotFoundError, depending on fatal, specifying the field name. |
d6983cb4 PH |
413 | """ |
414 | if isinstance(pattern, (str, compat_str, compiled_regex_type)): | |
415 | mobj = re.search(pattern, string, flags) | |
416 | else: | |
417 | for p in pattern: | |
418 | mobj = re.search(p, string, flags) | |
c3415d1b PH |
419 | if mobj: |
420 | break | |
d6983cb4 | 421 | |
87a28127 | 422 | if os.name != 'nt' and sys.stderr.isatty(): |
f1a9d64e | 423 | _name = '\033[0;34m%s\033[0m' % name |
d6983cb4 PH |
424 | else: |
425 | _name = name | |
426 | ||
427 | if mobj: | |
428 | # return the first matching group | |
429 | return next(g for g in mobj.groups() if g is not None) | |
46374a56 | 430 | elif default is not _NO_DEFAULT: |
d6983cb4 PH |
431 | return default |
432 | elif fatal: | |
f1a9d64e | 433 | raise RegexNotFoundError('Unable to extract %s' % _name) |
d6983cb4 | 434 | else: |
f1a9d64e PH |
435 | self._downloader.report_warning('unable to extract %s; ' |
436 | 'please report this issue on http://yt-dl.org/bug' % _name) | |
d6983cb4 PH |
437 | return None |
438 | ||
46374a56 | 439 | def _html_search_regex(self, pattern, string, name, default=_NO_DEFAULT, fatal=True, flags=0): |
d6983cb4 PH |
440 | """ |
441 | Like _search_regex, but strips HTML tags and unescapes entities. | |
442 | """ | |
443 | res = self._search_regex(pattern, string, name, default, fatal, flags) | |
444 | if res: | |
445 | return clean_html(res).strip() | |
446 | else: | |
447 | return res | |
448 | ||
fc79158d JMF |
449 | def _get_login_info(self): |
450 | """ | |
451 | Get the the login info as (username, password) | |
452 | It will look in the netrc file using the _NETRC_MACHINE value | |
453 | If there's no info available, return (None, None) | |
454 | """ | |
455 | if self._downloader is None: | |
456 | return (None, None) | |
457 | ||
458 | username = None | |
459 | password = None | |
460 | downloader_params = self._downloader.params | |
461 | ||
462 | # Attempt to use provided username and password or .netrc data | |
463 | if downloader_params.get('username', None) is not None: | |
464 | username = downloader_params['username'] | |
465 | password = downloader_params['password'] | |
466 | elif downloader_params.get('usenetrc', False): | |
467 | try: | |
468 | info = netrc.netrc().authenticators(self._NETRC_MACHINE) | |
469 | if info is not None: | |
470 | username = info[0] | |
471 | password = info[2] | |
472 | else: | |
473 | raise netrc.NetrcParseError('No authenticators for %s' % self._NETRC_MACHINE) | |
474 | except (IOError, netrc.NetrcParseError) as err: | |
f1a9d64e | 475 | self._downloader.report_warning('parsing .netrc: %s' % compat_str(err)) |
fc79158d JMF |
476 | |
477 | return (username, password) | |
478 | ||
83317f69 | 479 | def _get_tfa_info(self): |
480 | """ | |
481 | Get the two-factor authentication info | |
482 | TODO - asking the user will be required for sms/phone verify | |
483 | currently just uses the command line option | |
484 | If there's no info available, return None | |
485 | """ | |
486 | if self._downloader is None: | |
83317f69 | 487 | return None |
488 | downloader_params = self._downloader.params | |
489 | ||
490 | if downloader_params.get('twofactor', None) is not None: | |
491 | return downloader_params['twofactor'] | |
492 | ||
83317f69 | 493 | return None |
494 | ||
46720279 JMF |
495 | # Helper functions for extracting OpenGraph info |
496 | @staticmethod | |
ab2d5247 | 497 | def _og_regexes(prop): |
c1206423 | 498 | content_re = r'content=(?:"([^>]+?)"|\'([^>]+?)\')' |
9887c9b2 | 499 | property_re = r'(?:name|property)=[\'"]og:%s[\'"]' % re.escape(prop) |
78fb87b2 | 500 | template = r'<meta[^>]+?%s[^>]+?%s' |
ab2d5247 | 501 | return [ |
78fb87b2 JMF |
502 | template % (property_re, content_re), |
503 | template % (content_re, property_re), | |
ab2d5247 | 504 | ] |
46720279 | 505 | |
3c4e6d83 | 506 | def _og_search_property(self, prop, html, name=None, **kargs): |
46720279 | 507 | if name is None: |
3c4e6d83 | 508 | name = 'OpenGraph %s' % prop |
ab2d5247 | 509 | escaped = self._search_regex(self._og_regexes(prop), html, name, flags=re.DOTALL, **kargs) |
eb0a8398 PH |
510 | if escaped is None: |
511 | return None | |
512 | return unescapeHTML(escaped) | |
46720279 JMF |
513 | |
514 | def _og_search_thumbnail(self, html, **kargs): | |
f1a9d64e | 515 | return self._og_search_property('image', html, 'thumbnail url', fatal=False, **kargs) |
46720279 JMF |
516 | |
517 | def _og_search_description(self, html, **kargs): | |
518 | return self._og_search_property('description', html, fatal=False, **kargs) | |
519 | ||
520 | def _og_search_title(self, html, **kargs): | |
521 | return self._og_search_property('title', html, **kargs) | |
522 | ||
8ffa13e0 | 523 | def _og_search_video_url(self, html, name='video url', secure=True, **kargs): |
a3681973 PH |
524 | regexes = self._og_regexes('video') + self._og_regexes('video:url') |
525 | if secure: | |
526 | regexes = self._og_regexes('video:secure_url') + regexes | |
8ffa13e0 | 527 | return self._html_search_regex(regexes, html, name, **kargs) |
46720279 | 528 | |
78338f71 JMF |
529 | def _og_search_url(self, html, **kargs): |
530 | return self._og_search_property('url', html, **kargs) | |
531 | ||
40c696e5 | 532 | def _html_search_meta(self, name, html, display_name=None, fatal=False, **kwargs): |
59040888 PH |
533 | if display_name is None: |
534 | display_name = name | |
535 | return self._html_search_regex( | |
aaebed13 | 536 | r'''(?ix)<meta |
1a30deca | 537 | (?=[^>]+(?:itemprop|name|property)=["\']?%s["\']?) |
59040888 | 538 | [^>]+content=["\']([^"\']+)["\']''' % re.escape(name), |
40c696e5 | 539 | html, display_name, fatal=fatal, **kwargs) |
59040888 PH |
540 | |
541 | def _dc_search_uploader(self, html): | |
542 | return self._html_search_meta('dc.creator', html, 'uploader') | |
543 | ||
8dbe9899 PH |
544 | def _rta_search(self, html): |
545 | # See http://www.rtalabel.org/index.php?content=howtofaq#single | |
546 | if re.search(r'(?ix)<meta\s+name="rating"\s+' | |
547 | r' content="RTA-5042-1996-1400-1577-RTA"', | |
548 | html): | |
549 | return 18 | |
550 | return 0 | |
551 | ||
59040888 PH |
552 | def _media_rating_search(self, html): |
553 | # See http://www.tjg-designs.com/WP/metadata-code-examples-adding-metadata-to-your-web-pages/ | |
554 | rating = self._html_search_meta('rating', html) | |
555 | ||
556 | if not rating: | |
557 | return None | |
558 | ||
559 | RATING_TABLE = { | |
560 | 'safe for kids': 0, | |
561 | 'general': 8, | |
562 | '14 years': 14, | |
563 | 'mature': 17, | |
564 | 'restricted': 19, | |
565 | } | |
566 | return RATING_TABLE.get(rating.lower(), None) | |
567 | ||
0c708f11 JMF |
568 | def _twitter_search_player(self, html): |
569 | return self._html_search_meta('twitter:player', html, | |
570 | 'twitter card player') | |
571 | ||
4bcc7bd1 | 572 | def _sort_formats(self, formats): |
7e8caf30 | 573 | if not formats: |
f1a9d64e | 574 | raise ExtractorError('No video formats found') |
7e8caf30 | 575 | |
4bcc7bd1 | 576 | def _formats_key(f): |
e6812ac9 PH |
577 | # TODO remove the following workaround |
578 | from ..utils import determine_ext | |
579 | if not f.get('ext') and 'url' in f: | |
580 | f['ext'] = determine_ext(f['url']) | |
581 | ||
4bcc7bd1 PH |
582 | preference = f.get('preference') |
583 | if preference is None: | |
c7deaa4c PH |
584 | proto = f.get('protocol') |
585 | if proto is None: | |
586 | proto = compat_urllib_parse_urlparse(f.get('url', '')).scheme | |
587 | ||
588 | preference = 0 if proto in ['http', 'https'] else -0.1 | |
4bcc7bd1 PH |
589 | if f.get('ext') in ['f4f', 'f4m']: # Not yet supported |
590 | preference -= 0.5 | |
591 | ||
592 | if f.get('vcodec') == 'none': # audio only | |
593 | if self._downloader.params.get('prefer_free_formats'): | |
f1a9d64e | 594 | ORDER = ['aac', 'mp3', 'm4a', 'webm', 'ogg', 'opus'] |
4bcc7bd1 | 595 | else: |
f1a9d64e | 596 | ORDER = ['webm', 'opus', 'ogg', 'mp3', 'aac', 'm4a'] |
4bcc7bd1 PH |
597 | ext_preference = 0 |
598 | try: | |
599 | audio_ext_preference = ORDER.index(f['ext']) | |
600 | except ValueError: | |
601 | audio_ext_preference = -1 | |
602 | else: | |
603 | if self._downloader.params.get('prefer_free_formats'): | |
f1a9d64e | 604 | ORDER = ['flv', 'mp4', 'webm'] |
4bcc7bd1 | 605 | else: |
f1a9d64e | 606 | ORDER = ['webm', 'flv', 'mp4'] |
4bcc7bd1 PH |
607 | try: |
608 | ext_preference = ORDER.index(f['ext']) | |
609 | except ValueError: | |
610 | ext_preference = -1 | |
611 | audio_ext_preference = 0 | |
612 | ||
613 | return ( | |
614 | preference, | |
5d73273f | 615 | f.get('quality') if f.get('quality') is not None else -1, |
4bcc7bd1 PH |
616 | f.get('height') if f.get('height') is not None else -1, |
617 | f.get('width') if f.get('width') is not None else -1, | |
618 | ext_preference, | |
9933b574 | 619 | f.get('tbr') if f.get('tbr') is not None else -1, |
4bcc7bd1 PH |
620 | f.get('vbr') if f.get('vbr') is not None else -1, |
621 | f.get('abr') if f.get('abr') is not None else -1, | |
622 | audio_ext_preference, | |
2c8e03d9 | 623 | f.get('fps') if f.get('fps') is not None else -1, |
4bcc7bd1 | 624 | f.get('filesize') if f.get('filesize') is not None else -1, |
9732d77e | 625 | f.get('filesize_approx') if f.get('filesize_approx') is not None else -1, |
c64ed2a3 | 626 | f.get('source_preference') if f.get('source_preference') is not None else -1, |
4bcc7bd1 PH |
627 | f.get('format_id'), |
628 | ) | |
629 | formats.sort(key=_formats_key) | |
59040888 | 630 | |
20991253 | 631 | def http_scheme(self): |
1ede5b24 | 632 | """ Either "http:" or "https:", depending on the user's preferences """ |
20991253 PH |
633 | return ( |
634 | 'http:' | |
635 | if self._downloader.params.get('prefer_insecure', False) | |
636 | else 'https:') | |
637 | ||
57c7411f PH |
638 | def _proto_relative_url(self, url, scheme=None): |
639 | if url is None: | |
640 | return url | |
641 | if url.startswith('//'): | |
642 | if scheme is None: | |
643 | scheme = self.http_scheme() | |
644 | return scheme + url | |
645 | else: | |
646 | return url | |
647 | ||
4094b6e3 PH |
648 | def _sleep(self, timeout, video_id, msg_template=None): |
649 | if msg_template is None: | |
f1a9d64e | 650 | msg_template = '%(video_id)s: Waiting for %(timeout)s seconds' |
4094b6e3 PH |
651 | msg = msg_template % {'video_id': video_id, 'timeout': timeout} |
652 | self.to_screen(msg) | |
653 | time.sleep(timeout) | |
654 | ||
31bb8d3f | 655 | def _extract_f4m_formats(self, manifest_url, video_id): |
f036a632 JMF |
656 | manifest = self._download_xml( |
657 | manifest_url, video_id, 'Downloading f4m manifest', | |
658 | 'Unable to download f4m manifest') | |
31bb8d3f JMF |
659 | |
660 | formats = [] | |
b2527359 PH |
661 | media_nodes = manifest.findall('{http://ns.adobe.com/f4m/1.0}media') |
662 | for i, media_el in enumerate(media_nodes): | |
663 | tbr = int_or_none(media_el.attrib.get('bitrate')) | |
664 | format_id = 'f4m-%d' % (i if tbr is None else tbr) | |
31bb8d3f | 665 | formats.append({ |
b2527359 | 666 | 'format_id': format_id, |
31bb8d3f JMF |
667 | 'url': manifest_url, |
668 | 'ext': 'flv', | |
b2527359 | 669 | 'tbr': tbr, |
31bb8d3f JMF |
670 | 'width': int_or_none(media_el.attrib.get('width')), |
671 | 'height': int_or_none(media_el.attrib.get('height')), | |
672 | }) | |
673 | self._sort_formats(formats) | |
674 | ||
675 | return formats | |
676 | ||
f0b5d6af PH |
677 | def _extract_m3u8_formats(self, m3u8_url, video_id, ext=None, |
678 | entry_protocol='m3u8', preference=None): | |
679 | ||
704df56d PH |
680 | formats = [{ |
681 | 'format_id': 'm3u8-meta', | |
682 | 'url': m3u8_url, | |
683 | 'ext': ext, | |
684 | 'protocol': 'm3u8', | |
685 | 'preference': -1, | |
686 | 'resolution': 'multiple', | |
687 | 'format_note': 'Quality selection URL', | |
688 | }] | |
689 | ||
f0b5d6af PH |
690 | format_url = lambda u: ( |
691 | u | |
692 | if re.match(r'^https?://', u) | |
693 | else compat_urlparse.urljoin(m3u8_url, u)) | |
694 | ||
81515ad9 PH |
695 | m3u8_doc = self._download_webpage( |
696 | m3u8_url, video_id, | |
697 | note='Downloading m3u8 information', | |
698 | errnote='Failed to download m3u8 information') | |
704df56d PH |
699 | last_info = None |
700 | kv_rex = re.compile( | |
701 | r'(?P<key>[a-zA-Z_-]+)=(?P<val>"[^"]+"|[^",]+)(?:,|$)') | |
702 | for line in m3u8_doc.splitlines(): | |
703 | if line.startswith('#EXT-X-STREAM-INF:'): | |
704 | last_info = {} | |
705 | for m in kv_rex.finditer(line): | |
706 | v = m.group('val') | |
707 | if v.startswith('"'): | |
708 | v = v[1:-1] | |
709 | last_info[m.group('key')] = v | |
710 | elif line.startswith('#') or not line.strip(): | |
711 | continue | |
712 | else: | |
daebaab6 | 713 | if last_info is None: |
f0b5d6af | 714 | formats.append({'url': format_url(line)}) |
3524cc25 | 715 | continue |
704df56d PH |
716 | tbr = int_or_none(last_info.get('BANDWIDTH'), scale=1000) |
717 | ||
718 | f = { | |
719 | 'format_id': 'm3u8-%d' % (tbr if tbr else len(formats)), | |
f0b5d6af | 720 | 'url': format_url(line.strip()), |
704df56d PH |
721 | 'tbr': tbr, |
722 | 'ext': ext, | |
f0b5d6af PH |
723 | 'protocol': entry_protocol, |
724 | 'preference': preference, | |
704df56d PH |
725 | } |
726 | codecs = last_info.get('CODECS') | |
727 | if codecs: | |
9ebf22b7 S |
728 | # TODO: looks like video codec is not always necessarily goes first |
729 | va_codecs = codecs.split(',') | |
730 | if va_codecs[0]: | |
731 | f['vcodec'] = va_codecs[0].partition('.')[0] | |
732 | if len(va_codecs) > 1 and va_codecs[1]: | |
733 | f['acodec'] = va_codecs[1].partition('.')[0] | |
704df56d PH |
734 | resolution = last_info.get('RESOLUTION') |
735 | if resolution: | |
736 | width_str, height_str = resolution.split('x') | |
737 | f['width'] = int(width_str) | |
738 | f['height'] = int(height_str) | |
739 | formats.append(f) | |
740 | last_info = {} | |
741 | self._sort_formats(formats) | |
742 | return formats | |
743 | ||
f4b1c7ad PH |
744 | def _live_title(self, name): |
745 | """ Generate the title for a live video """ | |
746 | now = datetime.datetime.now() | |
747 | now_str = now.strftime("%Y-%m-%d %H:%M") | |
748 | return name + ' ' + now_str | |
749 | ||
b14f3a4c PH |
750 | def _int(self, v, name, fatal=False, **kwargs): |
751 | res = int_or_none(v, **kwargs) | |
752 | if 'get_attr' in kwargs: | |
753 | print(getattr(v, kwargs['get_attr'])) | |
754 | if res is None: | |
755 | msg = 'Failed to extract %s: Could not parse value %r' % (name, v) | |
756 | if fatal: | |
757 | raise ExtractorError(msg) | |
758 | else: | |
759 | self._downloader.report_warning(msg) | |
760 | return res | |
761 | ||
762 | def _float(self, v, name, fatal=False, **kwargs): | |
763 | res = float_or_none(v, **kwargs) | |
764 | if res is None: | |
765 | msg = 'Failed to extract %s: Could not parse value %r' % (name, v) | |
766 | if fatal: | |
767 | raise ExtractorError(msg) | |
768 | else: | |
769 | self._downloader.report_warning(msg) | |
770 | return res | |
771 | ||
8dbe9899 | 772 | |
d6983cb4 PH |
773 | class SearchInfoExtractor(InfoExtractor): |
774 | """ | |
775 | Base class for paged search queries extractors. | |
776 | They accept urls in the format _SEARCH_KEY(|all|[0-9]):{query} | |
777 | Instances should define _SEARCH_KEY and _MAX_RESULTS. | |
778 | """ | |
779 | ||
780 | @classmethod | |
781 | def _make_valid_url(cls): | |
782 | return r'%s(?P<prefix>|[1-9][0-9]*|all):(?P<query>[\s\S]+)' % cls._SEARCH_KEY | |
783 | ||
784 | @classmethod | |
785 | def suitable(cls, url): | |
786 | return re.match(cls._make_valid_url(), url) is not None | |
787 | ||
788 | def _real_extract(self, query): | |
789 | mobj = re.match(self._make_valid_url(), query) | |
790 | if mobj is None: | |
f1a9d64e | 791 | raise ExtractorError('Invalid search query "%s"' % query) |
d6983cb4 PH |
792 | |
793 | prefix = mobj.group('prefix') | |
794 | query = mobj.group('query') | |
795 | if prefix == '': | |
796 | return self._get_n_results(query, 1) | |
797 | elif prefix == 'all': | |
798 | return self._get_n_results(query, self._MAX_RESULTS) | |
799 | else: | |
800 | n = int(prefix) | |
801 | if n <= 0: | |
f1a9d64e | 802 | raise ExtractorError('invalid download number %s for query "%s"' % (n, query)) |
d6983cb4 | 803 | elif n > self._MAX_RESULTS: |
f1a9d64e | 804 | self._downloader.report_warning('%s returns max %i results (you requested %i)' % (self._SEARCH_KEY, self._MAX_RESULTS, n)) |
d6983cb4 PH |
805 | n = self._MAX_RESULTS |
806 | return self._get_n_results(query, n) | |
807 | ||
808 | def _get_n_results(self, query, n): | |
809 | """Get a specified number of results for a query""" | |
416a5efc | 810 | raise NotImplementedError("This method must be implemented by subclasses") |
0f818663 PH |
811 | |
812 | @property | |
813 | def SEARCH_KEY(self): | |
814 | return self._SEARCH_KEY |