]>
Commit | Line | Data |
---|---|---|
d6983cb4 PH |
1 | import base64 |
2 | import os | |
3 | import re | |
4 | import socket | |
5 | import sys | |
fc79158d | 6 | import netrc |
267ed0c5 | 7 | import xml.etree.ElementTree |
d6983cb4 PH |
8 | |
9 | from ..utils import ( | |
10 | compat_http_client, | |
11 | compat_urllib_error, | |
d6983cb4 PH |
12 | compat_str, |
13 | ||
14 | clean_html, | |
15 | compiled_regex_type, | |
16 | ExtractorError, | |
55b3e45b | 17 | RegexNotFoundError, |
d41e6efc | 18 | sanitize_filename, |
f38de77f | 19 | unescapeHTML, |
d6983cb4 PH |
20 | ) |
21 | ||
dca08720 | 22 | |
d6983cb4 PH |
23 | class InfoExtractor(object): |
24 | """Information Extractor class. | |
25 | ||
26 | Information extractors are the classes that, given a URL, extract | |
27 | information about the video (or videos) the URL refers to. This | |
28 | information includes the real video URL, the video title, author and | |
29 | others. The information is stored in a dictionary which is then | |
30 | passed to the FileDownloader. The FileDownloader processes this | |
31 | information possibly downloading the video to the file system, among | |
32 | other possible outcomes. | |
33 | ||
34 | The dictionaries must include the following fields: | |
35 | ||
36 | id: Video identifier. | |
37 | url: Final video URL. | |
38 | title: Video title, unescaped. | |
39 | ext: Video filename extension. | |
40 | ||
2f5865cc PH |
41 | Instead of url and ext, formats can also specified. |
42 | ||
d6983cb4 PH |
43 | The following fields are optional: |
44 | ||
45 | format: The video format, defaults to ext (used for --get-format) | |
73e79f2a PH |
46 | thumbnails: A list of dictionaries (with the entries "resolution" and |
47 | "url") for the varying thumbnails | |
d6983cb4 PH |
48 | thumbnail: Full URL to a video thumbnail image. |
49 | description: One-line video description. | |
50 | uploader: Full name of the video uploader. | |
51 | upload_date: Video upload date (YYYYMMDD). | |
52 | uploader_id: Nickname or id of the video uploader. | |
53 | location: Physical location of the video. | |
54 | player_url: SWF Player URL (used for rtmpdump). | |
5d51a883 JMF |
55 | subtitles: The subtitle file contents as a dictionary in the format |
56 | {language: subtitles}. | |
f3d29461 | 57 | view_count: How many users have watched the video on the platform. |
19e3dfc9 PH |
58 | like_count: Number of positive ratings of the video |
59 | dislike_count: Number of negative ratings of the video | |
60 | comment_count: Number of comments on the video | |
d6983cb4 PH |
61 | urlhandle: [internal] The urlHandle to be used to download the file, |
62 | like returned by urllib.request.urlopen | |
8dbe9899 | 63 | age_limit: Age restriction for the video, as an integer (years) |
deefc05b PH |
64 | formats: A list of dictionaries for each format available, it must |
65 | be ordered from worst to best quality. Potential fields: | |
66 | * url Mandatory. The URL of the video file | |
67 | * ext Will be calculated from url if missing | |
68 | * format A human-readable description of the format | |
69 | ("mp4 container with h264/opus"). | |
b5d0d817 | 70 | Calculated from the format_id, width, height. |
8c51aa65 | 71 | and format_note fields if missing. |
deefc05b PH |
72 | * format_id A short description of the format |
73 | ("mp4_h264_opus" or "19") | |
8c51aa65 JMF |
74 | * format_note Additional info about the format |
75 | ("3D" or "DASH video") | |
deefc05b PH |
76 | * width Width of the video, if known |
77 | * height Height of the video, if known | |
91c7271a PH |
78 | * abr Average audio bitrate in KBit/s |
79 | * acodec Name of the audio codec in use | |
80 | * vbr Average video bitrate in KBit/s | |
81 | * vcodec Name of the video codec in use | |
02dbf93f | 82 | * filesize The number of bytes, if known in advance |
9103bbc5 JMF |
83 | webpage_url: The url to the video webpage, if given to youtube-dl it |
84 | should allow to get the same result again. (It will be set | |
85 | by YoutubeDL if it's missing) | |
d6983cb4 | 86 | |
deefc05b | 87 | Unless mentioned otherwise, the fields should be Unicode strings. |
d6983cb4 PH |
88 | |
89 | Subclasses of this one should re-define the _real_initialize() and | |
90 | _real_extract() methods and define a _VALID_URL regexp. | |
91 | Probably, they should also be added to the list of extractors. | |
92 | ||
93 | _real_extract() must return a *list* of information dictionaries as | |
94 | described above. | |
95 | ||
96 | Finally, the _WORKING attribute should be set to False for broken IEs | |
97 | in order to warn the users and skip the tests. | |
98 | """ | |
99 | ||
100 | _ready = False | |
101 | _downloader = None | |
102 | _WORKING = True | |
103 | ||
104 | def __init__(self, downloader=None): | |
105 | """Constructor. Receives an optional downloader.""" | |
106 | self._ready = False | |
107 | self.set_downloader(downloader) | |
108 | ||
109 | @classmethod | |
110 | def suitable(cls, url): | |
111 | """Receives a URL and returns True if suitable for this IE.""" | |
79cb2577 PH |
112 | |
113 | # This does not use has/getattr intentionally - we want to know whether | |
114 | # we have cached the regexp for *this* class, whereas getattr would also | |
115 | # match the superclass | |
116 | if '_VALID_URL_RE' not in cls.__dict__: | |
117 | cls._VALID_URL_RE = re.compile(cls._VALID_URL) | |
118 | return cls._VALID_URL_RE.match(url) is not None | |
d6983cb4 PH |
119 | |
120 | @classmethod | |
121 | def working(cls): | |
122 | """Getter method for _WORKING.""" | |
123 | return cls._WORKING | |
124 | ||
125 | def initialize(self): | |
126 | """Initializes an instance (authentication, etc).""" | |
127 | if not self._ready: | |
128 | self._real_initialize() | |
129 | self._ready = True | |
130 | ||
131 | def extract(self, url): | |
132 | """Extracts URL information and returns it in list of dicts.""" | |
133 | self.initialize() | |
134 | return self._real_extract(url) | |
135 | ||
136 | def set_downloader(self, downloader): | |
137 | """Sets the downloader for this IE.""" | |
138 | self._downloader = downloader | |
139 | ||
140 | def _real_initialize(self): | |
141 | """Real initialization process. Redefine in subclasses.""" | |
142 | pass | |
143 | ||
144 | def _real_extract(self, url): | |
145 | """Real extraction process. Redefine in subclasses.""" | |
146 | pass | |
147 | ||
56c73665 JMF |
148 | @classmethod |
149 | def ie_key(cls): | |
150 | """A string for getting the InfoExtractor with get_info_extractor""" | |
151 | return cls.__name__[:-2] | |
152 | ||
d6983cb4 PH |
153 | @property |
154 | def IE_NAME(self): | |
155 | return type(self).__name__[:-2] | |
156 | ||
7cc3570e | 157 | def _request_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True): |
d6983cb4 PH |
158 | """ Returns the response handle """ |
159 | if note is None: | |
160 | self.report_download_webpage(video_id) | |
161 | elif note is not False: | |
7cc3570e PH |
162 | if video_id is None: |
163 | self.to_screen(u'%s' % (note,)) | |
164 | else: | |
165 | self.to_screen(u'%s: %s' % (video_id, note)) | |
d6983cb4 | 166 | try: |
dca08720 | 167 | return self._downloader.urlopen(url_or_request) |
d6983cb4 PH |
168 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: |
169 | if errnote is None: | |
170 | errnote = u'Unable to download webpage' | |
7cc3570e PH |
171 | errmsg = u'%s: %s' % (errnote, compat_str(err)) |
172 | if fatal: | |
173 | raise ExtractorError(errmsg, sys.exc_info()[2], cause=err) | |
174 | else: | |
175 | self._downloader.report_warning(errmsg) | |
176 | return False | |
d6983cb4 | 177 | |
7cc3570e | 178 | def _download_webpage_handle(self, url_or_request, video_id, note=None, errnote=None, fatal=True): |
d6983cb4 | 179 | """ Returns a tuple (page content as string, URL handle) """ |
b9d3e163 PH |
180 | |
181 | # Strip hashes from the URL (#1038) | |
182 | if isinstance(url_or_request, (compat_str, str)): | |
183 | url_or_request = url_or_request.partition('#')[0] | |
184 | ||
7cc3570e PH |
185 | urlh = self._request_webpage(url_or_request, video_id, note, errnote, fatal) |
186 | if urlh is False: | |
187 | assert not fatal | |
188 | return False | |
d6983cb4 | 189 | content_type = urlh.headers.get('Content-Type', '') |
f143d86a | 190 | webpage_bytes = urlh.read() |
d6983cb4 PH |
191 | m = re.match(r'[a-zA-Z0-9_.-]+/[a-zA-Z0-9_.-]+\s*;\s*charset=(.+)', content_type) |
192 | if m: | |
193 | encoding = m.group(1) | |
194 | else: | |
0d75ae2c | 195 | m = re.search(br'<meta[^>]+charset=[\'"]?([^\'")]+)[ /\'">]', |
f143d86a PH |
196 | webpage_bytes[:1024]) |
197 | if m: | |
198 | encoding = m.group(1).decode('ascii') | |
199 | else: | |
200 | encoding = 'utf-8' | |
d6983cb4 PH |
201 | if self._downloader.params.get('dump_intermediate_pages', False): |
202 | try: | |
203 | url = url_or_request.get_full_url() | |
204 | except AttributeError: | |
205 | url = url_or_request | |
206 | self.to_screen(u'Dumping request to ' + url) | |
207 | dump = base64.b64encode(webpage_bytes).decode('ascii') | |
208 | self._downloader.to_screen(dump) | |
d41e6efc PH |
209 | if self._downloader.params.get('write_pages', False): |
210 | try: | |
211 | url = url_or_request.get_full_url() | |
212 | except AttributeError: | |
213 | url = url_or_request | |
214 | raw_filename = ('%s_%s.dump' % (video_id, url)) | |
215 | filename = sanitize_filename(raw_filename, restricted=True) | |
216 | self.to_screen(u'Saving request to ' + filename) | |
217 | with open(filename, 'wb') as outf: | |
218 | outf.write(webpage_bytes) | |
219 | ||
d6983cb4 PH |
220 | content = webpage_bytes.decode(encoding, 'replace') |
221 | return (content, urlh) | |
222 | ||
7cc3570e | 223 | def _download_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True): |
d6983cb4 | 224 | """ Returns the data of the page as a string """ |
7cc3570e PH |
225 | res = self._download_webpage_handle(url_or_request, video_id, note, errnote, fatal) |
226 | if res is False: | |
227 | return res | |
228 | else: | |
229 | content, _ = res | |
230 | return content | |
d6983cb4 | 231 | |
2a275ab0 | 232 | def _download_xml(self, url_or_request, video_id, |
e2b38da9 PH |
233 | note=u'Downloading XML', errnote=u'Unable to download XML', |
234 | transform_source=None): | |
267ed0c5 JMF |
235 | """Return the xml as an xml.etree.ElementTree.Element""" |
236 | xml_string = self._download_webpage(url_or_request, video_id, note, errnote) | |
e2b38da9 PH |
237 | if transform_source: |
238 | xml_string = transform_source(xml_string) | |
267ed0c5 JMF |
239 | return xml.etree.ElementTree.fromstring(xml_string.encode('utf-8')) |
240 | ||
d6983cb4 PH |
241 | def to_screen(self, msg): |
242 | """Print msg to screen, prefixing it with '[ie_name]'""" | |
243 | self._downloader.to_screen(u'[%s] %s' % (self.IE_NAME, msg)) | |
244 | ||
245 | def report_extraction(self, id_or_name): | |
246 | """Report information extraction.""" | |
247 | self.to_screen(u'%s: Extracting information' % id_or_name) | |
248 | ||
249 | def report_download_webpage(self, video_id): | |
250 | """Report webpage download.""" | |
251 | self.to_screen(u'%s: Downloading webpage' % video_id) | |
252 | ||
253 | def report_age_confirmation(self): | |
254 | """Report attempt to confirm age.""" | |
255 | self.to_screen(u'Confirming age') | |
256 | ||
fc79158d JMF |
257 | def report_login(self): |
258 | """Report attempt to log in.""" | |
259 | self.to_screen(u'Logging in') | |
260 | ||
d6983cb4 | 261 | #Methods for following #608 |
7012b23c | 262 | def url_result(self, url, ie=None, video_id=None): |
d6983cb4 PH |
263 | """Returns a url that points to a page that should be processed""" |
264 | #TODO: ie should be the class used for getting the info | |
265 | video_info = {'_type': 'url', | |
266 | 'url': url, | |
267 | 'ie_key': ie} | |
7012b23c PH |
268 | if video_id is not None: |
269 | video_info['id'] = video_id | |
d6983cb4 PH |
270 | return video_info |
271 | def playlist_result(self, entries, playlist_id=None, playlist_title=None): | |
272 | """Returns a playlist""" | |
273 | video_info = {'_type': 'playlist', | |
274 | 'entries': entries} | |
275 | if playlist_id: | |
276 | video_info['id'] = playlist_id | |
277 | if playlist_title: | |
278 | video_info['title'] = playlist_title | |
279 | return video_info | |
280 | ||
281 | def _search_regex(self, pattern, string, name, default=None, fatal=True, flags=0): | |
282 | """ | |
283 | Perform a regex search on the given string, using a single or a list of | |
284 | patterns returning the first matching group. | |
285 | In case of failure return a default value or raise a WARNING or a | |
55b3e45b | 286 | RegexNotFoundError, depending on fatal, specifying the field name. |
d6983cb4 PH |
287 | """ |
288 | if isinstance(pattern, (str, compat_str, compiled_regex_type)): | |
289 | mobj = re.search(pattern, string, flags) | |
290 | else: | |
291 | for p in pattern: | |
292 | mobj = re.search(p, string, flags) | |
293 | if mobj: break | |
294 | ||
295 | if sys.stderr.isatty() and os.name != 'nt': | |
296 | _name = u'\033[0;34m%s\033[0m' % name | |
297 | else: | |
298 | _name = name | |
299 | ||
300 | if mobj: | |
301 | # return the first matching group | |
302 | return next(g for g in mobj.groups() if g is not None) | |
303 | elif default is not None: | |
304 | return default | |
305 | elif fatal: | |
55b3e45b | 306 | raise RegexNotFoundError(u'Unable to extract %s' % _name) |
d6983cb4 PH |
307 | else: |
308 | self._downloader.report_warning(u'unable to extract %s; ' | |
98bcd283 | 309 | u'please report this issue on http://yt-dl.org/bug' % _name) |
d6983cb4 PH |
310 | return None |
311 | ||
312 | def _html_search_regex(self, pattern, string, name, default=None, fatal=True, flags=0): | |
313 | """ | |
314 | Like _search_regex, but strips HTML tags and unescapes entities. | |
315 | """ | |
316 | res = self._search_regex(pattern, string, name, default, fatal, flags) | |
317 | if res: | |
318 | return clean_html(res).strip() | |
319 | else: | |
320 | return res | |
321 | ||
fc79158d JMF |
322 | def _get_login_info(self): |
323 | """ | |
324 | Get the the login info as (username, password) | |
325 | It will look in the netrc file using the _NETRC_MACHINE value | |
326 | If there's no info available, return (None, None) | |
327 | """ | |
328 | if self._downloader is None: | |
329 | return (None, None) | |
330 | ||
331 | username = None | |
332 | password = None | |
333 | downloader_params = self._downloader.params | |
334 | ||
335 | # Attempt to use provided username and password or .netrc data | |
336 | if downloader_params.get('username', None) is not None: | |
337 | username = downloader_params['username'] | |
338 | password = downloader_params['password'] | |
339 | elif downloader_params.get('usenetrc', False): | |
340 | try: | |
341 | info = netrc.netrc().authenticators(self._NETRC_MACHINE) | |
342 | if info is not None: | |
343 | username = info[0] | |
344 | password = info[2] | |
345 | else: | |
346 | raise netrc.NetrcParseError('No authenticators for %s' % self._NETRC_MACHINE) | |
347 | except (IOError, netrc.NetrcParseError) as err: | |
348 | self._downloader.report_warning(u'parsing .netrc: %s' % compat_str(err)) | |
349 | ||
350 | return (username, password) | |
351 | ||
46720279 JMF |
352 | # Helper functions for extracting OpenGraph info |
353 | @staticmethod | |
ab2d5247 | 354 | def _og_regexes(prop): |
78fb87b2 JMF |
355 | content_re = r'content=(?:"([^>]+?)"|\'(.+?)\')' |
356 | property_re = r'property=[\'"]og:%s[\'"]' % re.escape(prop) | |
357 | template = r'<meta[^>]+?%s[^>]+?%s' | |
ab2d5247 | 358 | return [ |
78fb87b2 JMF |
359 | template % (property_re, content_re), |
360 | template % (content_re, property_re), | |
ab2d5247 | 361 | ] |
46720279 | 362 | |
3c4e6d83 | 363 | def _og_search_property(self, prop, html, name=None, **kargs): |
46720279 | 364 | if name is None: |
3c4e6d83 | 365 | name = 'OpenGraph %s' % prop |
ab2d5247 | 366 | escaped = self._search_regex(self._og_regexes(prop), html, name, flags=re.DOTALL, **kargs) |
eb0a8398 PH |
367 | if escaped is None: |
368 | return None | |
369 | return unescapeHTML(escaped) | |
46720279 JMF |
370 | |
371 | def _og_search_thumbnail(self, html, **kargs): | |
3c4e6d83 | 372 | return self._og_search_property('image', html, u'thumbnail url', fatal=False, **kargs) |
46720279 JMF |
373 | |
374 | def _og_search_description(self, html, **kargs): | |
375 | return self._og_search_property('description', html, fatal=False, **kargs) | |
376 | ||
377 | def _og_search_title(self, html, **kargs): | |
378 | return self._og_search_property('title', html, **kargs) | |
379 | ||
8ffa13e0 | 380 | def _og_search_video_url(self, html, name='video url', secure=True, **kargs): |
ab2d5247 JMF |
381 | regexes = self._og_regexes('video') |
382 | if secure: regexes = self._og_regexes('video:secure_url') + regexes | |
8ffa13e0 | 383 | return self._html_search_regex(regexes, html, name, **kargs) |
46720279 | 384 | |
59040888 PH |
385 | def _html_search_meta(self, name, html, display_name=None): |
386 | if display_name is None: | |
387 | display_name = name | |
388 | return self._html_search_regex( | |
aaebed13 PH |
389 | r'''(?ix)<meta |
390 | (?=[^>]+(?:itemprop|name|property)=["\']%s["\']) | |
59040888 PH |
391 | [^>]+content=["\']([^"\']+)["\']''' % re.escape(name), |
392 | html, display_name, fatal=False) | |
393 | ||
394 | def _dc_search_uploader(self, html): | |
395 | return self._html_search_meta('dc.creator', html, 'uploader') | |
396 | ||
8dbe9899 PH |
397 | def _rta_search(self, html): |
398 | # See http://www.rtalabel.org/index.php?content=howtofaq#single | |
399 | if re.search(r'(?ix)<meta\s+name="rating"\s+' | |
400 | r' content="RTA-5042-1996-1400-1577-RTA"', | |
401 | html): | |
402 | return 18 | |
403 | return 0 | |
404 | ||
59040888 PH |
405 | def _media_rating_search(self, html): |
406 | # See http://www.tjg-designs.com/WP/metadata-code-examples-adding-metadata-to-your-web-pages/ | |
407 | rating = self._html_search_meta('rating', html) | |
408 | ||
409 | if not rating: | |
410 | return None | |
411 | ||
412 | RATING_TABLE = { | |
413 | 'safe for kids': 0, | |
414 | 'general': 8, | |
415 | '14 years': 14, | |
416 | 'mature': 17, | |
417 | 'restricted': 19, | |
418 | } | |
419 | return RATING_TABLE.get(rating.lower(), None) | |
420 | ||
421 | ||
8dbe9899 | 422 | |
d6983cb4 PH |
423 | class SearchInfoExtractor(InfoExtractor): |
424 | """ | |
425 | Base class for paged search queries extractors. | |
426 | They accept urls in the format _SEARCH_KEY(|all|[0-9]):{query} | |
427 | Instances should define _SEARCH_KEY and _MAX_RESULTS. | |
428 | """ | |
429 | ||
430 | @classmethod | |
431 | def _make_valid_url(cls): | |
432 | return r'%s(?P<prefix>|[1-9][0-9]*|all):(?P<query>[\s\S]+)' % cls._SEARCH_KEY | |
433 | ||
434 | @classmethod | |
435 | def suitable(cls, url): | |
436 | return re.match(cls._make_valid_url(), url) is not None | |
437 | ||
438 | def _real_extract(self, query): | |
439 | mobj = re.match(self._make_valid_url(), query) | |
440 | if mobj is None: | |
441 | raise ExtractorError(u'Invalid search query "%s"' % query) | |
442 | ||
443 | prefix = mobj.group('prefix') | |
444 | query = mobj.group('query') | |
445 | if prefix == '': | |
446 | return self._get_n_results(query, 1) | |
447 | elif prefix == 'all': | |
448 | return self._get_n_results(query, self._MAX_RESULTS) | |
449 | else: | |
450 | n = int(prefix) | |
451 | if n <= 0: | |
452 | raise ExtractorError(u'invalid download number %s for query "%s"' % (n, query)) | |
453 | elif n > self._MAX_RESULTS: | |
454 | self._downloader.report_warning(u'%s returns max %i results (you requested %i)' % (self._SEARCH_KEY, self._MAX_RESULTS, n)) | |
455 | n = self._MAX_RESULTS | |
456 | return self._get_n_results(query, n) | |
457 | ||
458 | def _get_n_results(self, query, n): | |
459 | """Get a specified number of results for a query""" | |
416a5efc | 460 | raise NotImplementedError("This method must be implemented by subclasses") |
0f818663 PH |
461 | |
462 | @property | |
463 | def SEARCH_KEY(self): | |
464 | return self._SEARCH_KEY |