Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/common.py

   1 import base64
   2 import os
   3 import re
   4 import socket
   5 import sys
   6 import netrc
   7
   8 from ..utils import (
   9     compat_http_client,
  10     compat_urllib_error,
  11     compat_urllib_request,
  12     compat_str,
  13
  14     clean_html,
  15     compiled_regex_type,
  16     ExtractorError,
  17     unescapeHTML,
  18 )
  19
  20 class InfoExtractor(object):
  21     """Information Extractor class.
  22
  23     Information extractors are the classes that, given a URL, extract
  24     information about the video (or videos) the URL refers to. This
  25     information includes the real video URL, the video title, author and
  26     others. The information is stored in a dictionary which is then
  27     passed to the FileDownloader. The FileDownloader processes this
  28     information possibly downloading the video to the file system, among
  29     other possible outcomes.
  30
  31     The dictionaries must include the following fields:
  32
  33     id:             Video identifier.
  34     url:            Final video URL.
  35     title:          Video title, unescaped.
  36     ext:            Video filename extension.
  37
  38     The following fields are optional:
  39
  40     format:         The video format, defaults to ext (used for --get-format)
  41     thumbnails:     A list of dictionaries (with the entries "resolution" and
  42                     "url") for the varying thumbnails
  43     thumbnail:      Full URL to a video thumbnail image.
  44     description:    One-line video description.
  45     uploader:       Full name of the video uploader.
  46     upload_date:    Video upload date (YYYYMMDD).
  47     uploader_id:    Nickname or id of the video uploader.
  48     location:       Physical location of the video.
  49     player_url:     SWF Player URL (used for rtmpdump).
  50     subtitles:      The subtitle file contents as a dictionary in the format
  51                     {language: subtitles}.
  52     view_count:     How many users have watched the video on the platform.
  53     urlhandle:      [internal] The urlHandle to be used to download the file,
  54                     like returned by urllib.request.urlopen
  55
  56     The fields should all be Unicode strings.
  57
  58     Subclasses of this one should re-define the _real_initialize() and
  59     _real_extract() methods and define a _VALID_URL regexp.
  60     Probably, they should also be added to the list of extractors.
  61
  62     _real_extract() must return a *list* of information dictionaries as
  63     described above.
  64
  65     Finally, the _WORKING attribute should be set to False for broken IEs
  66     in order to warn the users and skip the tests.
  67     """
  68
  69     _ready = False
  70     _downloader = None
  71     _WORKING = True
  72
  73     def __init__(self, downloader=None):
  74         """Constructor. Receives an optional downloader."""
  75         self._ready = False
  76         self.set_downloader(downloader)
  77
  78     @classmethod
  79     def suitable(cls, url):
  80         """Receives a URL and returns True if suitable for this IE."""
  81
  82         # This does not use has/getattr intentionally - we want to know whether
  83         # we have cached the regexp for *this* class, whereas getattr would also
  84         # match the superclass
  85         if '_VALID_URL_RE' not in cls.__dict__:
  86             cls._VALID_URL_RE = re.compile(cls._VALID_URL)
  87         return cls._VALID_URL_RE.match(url) is not None
  88
  89     @classmethod
  90     def working(cls):
  91         """Getter method for _WORKING."""
  92         return cls._WORKING
  93
  94     def initialize(self):
  95         """Initializes an instance (authentication, etc)."""
  96         if not self._ready:
  97             self._real_initialize()
  98             self._ready = True
  99
 100     def extract(self, url):
 101         """Extracts URL information and returns it in list of dicts."""
 102         self.initialize()
 103         return self._real_extract(url)
 104
 105     def set_downloader(self, downloader):
 106         """Sets the downloader for this IE."""
 107         self._downloader = downloader
 108
 109     def _real_initialize(self):
 110         """Real initialization process. Redefine in subclasses."""
 111         pass
 112
 113     def _real_extract(self, url):
 114         """Real extraction process. Redefine in subclasses."""
 115         pass
 116
 117     @classmethod
 118     def ie_key(cls):
 119         """A string for getting the InfoExtractor with get_info_extractor"""
 120         return cls.__name__[:-2]
 121
 122     @property
 123     def IE_NAME(self):
 124         return type(self).__name__[:-2]
 125
 126     def _request_webpage(self, url_or_request, video_id, note=None, errnote=None):
 127         """ Returns the response handle """
 128         if note is None:
 129             self.report_download_webpage(video_id)
 130         elif note is not False:
 131             self.to_screen(u'%s: %s' % (video_id, note))
 132         try:
 133             return compat_urllib_request.urlopen(url_or_request)
 134         except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
 135             if errnote is None:
 136                 errnote = u'Unable to download webpage'
 137             raise ExtractorError(u'%s: %s' % (errnote, compat_str(err)), sys.exc_info()[2], cause=err)
 138
 139     def _download_webpage_handle(self, url_or_request, video_id, note=None, errnote=None):
 140         """ Returns a tuple (page content as string, URL handle) """
 141
 142         # Strip hashes from the URL (#1038)
 143         if isinstance(url_or_request, (compat_str, str)):
 144             url_or_request = url_or_request.partition('#')[0]
 145
 146         urlh = self._request_webpage(url_or_request, video_id, note, errnote)
 147         content_type = urlh.headers.get('Content-Type', '')
 148         webpage_bytes = urlh.read()
 149         m = re.match(r'[a-zA-Z0-9_.-]+/[a-zA-Z0-9_.-]+\s*;\s*charset=(.+)', content_type)
 150         if m:
 151             encoding = m.group(1)
 152         else:
 153             m = re.search(br'<meta[^>]+charset=[\'"]?([^\'")]+)[ /\'">]',
 154                           webpage_bytes[:1024])
 155             if m:
 156                 encoding = m.group(1).decode('ascii')
 157             else:
 158                 encoding = 'utf-8'
 159         if self._downloader.params.get('dump_intermediate_pages', False):
 160             try:
 161                 url = url_or_request.get_full_url()
 162             except AttributeError:
 163                 url = url_or_request
 164             self.to_screen(u'Dumping request to ' + url)
 165             dump = base64.b64encode(webpage_bytes).decode('ascii')
 166             self._downloader.to_screen(dump)
 167         content = webpage_bytes.decode(encoding, 'replace')
 168         return (content, urlh)
 169
 170     def _download_webpage(self, url_or_request, video_id, note=None, errnote=None):
 171         """ Returns the data of the page as a string """
 172         return self._download_webpage_handle(url_or_request, video_id, note, errnote)[0]
 173
 174     def to_screen(self, msg):
 175         """Print msg to screen, prefixing it with '[ie_name]'"""
 176         self._downloader.to_screen(u'[%s] %s' % (self.IE_NAME, msg))
 177
 178     def report_extraction(self, id_or_name):
 179         """Report information extraction."""
 180         self.to_screen(u'%s: Extracting information' % id_or_name)
 181
 182     def report_download_webpage(self, video_id):
 183         """Report webpage download."""
 184         self.to_screen(u'%s: Downloading webpage' % video_id)
 185
 186     def report_age_confirmation(self):
 187         """Report attempt to confirm age."""
 188         self.to_screen(u'Confirming age')
 189
 190     def report_login(self):
 191         """Report attempt to log in."""
 192         self.to_screen(u'Logging in')
 193
 194     #Methods for following #608
 195     def url_result(self, url, ie=None):
 196         """Returns a url that points to a page that should be processed"""
 197         #TODO: ie should be the class used for getting the info
 198         video_info = {'_type': 'url',
 199                       'url': url,
 200                       'ie_key': ie}
 201         return video_info
 202     def playlist_result(self, entries, playlist_id=None, playlist_title=None):
 203         """Returns a playlist"""
 204         video_info = {'_type': 'playlist',
 205                       'entries': entries}
 206         if playlist_id:
 207             video_info['id'] = playlist_id
 208         if playlist_title:
 209             video_info['title'] = playlist_title
 210         return video_info
 211
 212     def _search_regex(self, pattern, string, name, default=None, fatal=True, flags=0):
 213         """
 214         Perform a regex search on the given string, using a single or a list of
 215         patterns returning the first matching group.
 216         In case of failure return a default value or raise a WARNING or a
 217         ExtractorError, depending on fatal, specifying the field name.
 218         """
 219         if isinstance(pattern, (str, compat_str, compiled_regex_type)):
 220             mobj = re.search(pattern, string, flags)
 221         else:
 222             for p in pattern:
 223                 mobj = re.search(p, string, flags)
 224                 if mobj: break
 225
 226         if sys.stderr.isatty() and os.name != 'nt':
 227             _name = u'\033[0;34m%s\033[0m' % name
 228         else:
 229             _name = name
 230
 231         if mobj:
 232             # return the first matching group
 233             return next(g for g in mobj.groups() if g is not None)
 234         elif default is not None:
 235             return default
 236         elif fatal:
 237             raise ExtractorError(u'Unable to extract %s' % _name)
 238         else:
 239             self._downloader.report_warning(u'unable to extract %s; '
 240                 u'please report this issue on http://yt-dl.org/bug' % _name)
 241             return None
 242
 243     def _html_search_regex(self, pattern, string, name, default=None, fatal=True, flags=0):
 244         """
 245         Like _search_regex, but strips HTML tags and unescapes entities.
 246         """
 247         res = self._search_regex(pattern, string, name, default, fatal, flags)
 248         if res:
 249             return clean_html(res).strip()
 250         else:
 251             return res
 252
 253     def _get_login_info(self):
 254         """
 255         Get the the login info as (username, password)
 256         It will look in the netrc file using the _NETRC_MACHINE value
 257         If there's no info available, return (None, None)
 258         """
 259         if self._downloader is None:
 260             return (None, None)
 261
 262         username = None
 263         password = None
 264         downloader_params = self._downloader.params
 265
 266         # Attempt to use provided username and password or .netrc data
 267         if downloader_params.get('username', None) is not None:
 268             username = downloader_params['username']
 269             password = downloader_params['password']
 270         elif downloader_params.get('usenetrc', False):
 271             try:
 272                 info = netrc.netrc().authenticators(self._NETRC_MACHINE)
 273                 if info is not None:
 274                     username = info[0]
 275                     password = info[2]
 276                 else:
 277                     raise netrc.NetrcParseError('No authenticators for %s' % self._NETRC_MACHINE)
 278             except (IOError, netrc.NetrcParseError) as err:
 279                 self._downloader.report_warning(u'parsing .netrc: %s' % compat_str(err))
 280
 281         return (username, password)
 282
 283     # Helper functions for extracting OpenGraph info
 284     @staticmethod
 285     def _og_regex(prop):
 286         return r'<meta.+?property=[\'"]og:%s[\'"].+?content=(?:"(.+?)"|\'(.+?)\')' % re.escape(prop)
 287
 288     def _og_search_property(self, prop, html, name=None, **kargs):
 289         if name is None:
 290             name = 'OpenGraph %s' % prop
 291         escaped = self._search_regex(self._og_regex(prop), html, name, flags=re.DOTALL, **kargs)
 292         return unescapeHTML(escaped)
 293
 294     def _og_search_thumbnail(self, html, **kargs):
 295         return self._og_search_property('image', html, u'thumbnail url', fatal=False, **kargs)
 296
 297     def _og_search_description(self, html, **kargs):
 298         return self._og_search_property('description', html, fatal=False, **kargs)
 299
 300     def _og_search_title(self, html, **kargs):
 301         return self._og_search_property('title', html, **kargs)
 302
 303     def _og_search_video_url(self, html, name='video url', **kargs):
 304         return self._html_search_regex([self._og_regex('video:secure_url'),
 305                                         self._og_regex('video')],
 306                                        html, name, **kargs)
 307
 308 class SearchInfoExtractor(InfoExtractor):
 309     """
 310     Base class for paged search queries extractors.
 311     They accept urls in the format _SEARCH_KEY(|all|[0-9]):{query}
 312     Instances should define _SEARCH_KEY and _MAX_RESULTS.
 313     """
 314
 315     @classmethod
 316     def _make_valid_url(cls):
 317         return r'%s(?P<prefix>|[1-9][0-9]*|all):(?P<query>[\s\S]+)' % cls._SEARCH_KEY
 318
 319     @classmethod
 320     def suitable(cls, url):
 321         return re.match(cls._make_valid_url(), url) is not None
 322
 323     def _real_extract(self, query):
 324         mobj = re.match(self._make_valid_url(), query)
 325         if mobj is None:
 326             raise ExtractorError(u'Invalid search query "%s"' % query)
 327
 328         prefix = mobj.group('prefix')
 329         query = mobj.group('query')
 330         if prefix == '':
 331             return self._get_n_results(query, 1)
 332         elif prefix == 'all':
 333             return self._get_n_results(query, self._MAX_RESULTS)
 334         else:
 335             n = int(prefix)
 336             if n <= 0:
 337                 raise ExtractorError(u'invalid download number %s for query "%s"' % (n, query))
 338             elif n > self._MAX_RESULTS:
 339                 self._downloader.report_warning(u'%s returns max %i results (you requested %i)' % (self._SEARCH_KEY, self._MAX_RESULTS, n))
 340                 n = self._MAX_RESULTS
 341             return self._get_n_results(query, n)
 342
 343     def _get_n_results(self, query, n):
 344         """Get a specified number of results for a query"""
 345         raise NotImplementedError("This method must be implemented by sublclasses")
 346
 347     @property
 348     def SEARCH_KEY(self):
 349         return self._SEARCH_KEY