youtube_dl/YoutubeDL.py

   1 #!/usr/bin/env python
   2 # -*- coding: utf-8 -*-
   3
   4 from __future__ import absolute_import
   5
   6 import errno
   7 import io
   8 import os
   9 import re
  10 import shutil
  11 import socket
  12 import sys
  13 import time
  14 import traceback
  15
  16 from .utils import *
  17 from .extractor import get_info_extractor, gen_extractors
  18 from .FileDownloader import FileDownloader
  19
  20
  21 class YoutubeDL(object):
  22     """YoutubeDL class.
  23
  24     YoutubeDL objects are the ones responsible of downloading the
  25     actual video file and writing it to disk if the user has requested
  26     it, among some other tasks. In most cases there should be one per
  27     program. As, given a video URL, the downloader doesn't know how to
  28     extract all the needed information, task that InfoExtractors do, it
  29     has to pass the URL to one of them.
  30
  31     For this, YoutubeDL objects have a method that allows
  32     InfoExtractors to be registered in a given order. When it is passed
  33     a URL, the YoutubeDL object handles it to the first InfoExtractor it
  34     finds that reports being able to handle it. The InfoExtractor extracts
  35     all the information about the video or videos the URL refers to, and
  36     YoutubeDL process the extracted information, possibly using a File
  37     Downloader to download the video.
  38
  39     YoutubeDL objects accept a lot of parameters. In order not to saturate
  40     the object constructor with arguments, it receives a dictionary of
  41     options instead. These options are available through the params
  42     attribute for the InfoExtractors to use. The YoutubeDL also
  43     registers itself as the downloader in charge for the InfoExtractors
  44     that are added to it, so this is a "mutual registration".
  45
  46     Available options:
  47
  48     username:          Username for authentication purposes.
  49     password:          Password for authentication purposes.
  50     videopassword:     Password for acces a video.
  51     usenetrc:          Use netrc for authentication instead.
  52     verbose:           Print additional info to stdout.
  53     quiet:             Do not print messages to stdout.
  54     forceurl:          Force printing final URL.
  55     forcetitle:        Force printing title.
  56     forceid:           Force printing ID.
  57     forcethumbnail:    Force printing thumbnail URL.
  58     forcedescription:  Force printing description.
  59     forcefilename:     Force printing final filename.
  60     simulate:          Do not download the video files.
  61     format:            Video format code.
  62     format_limit:      Highest quality format to try.
  63     outtmpl:           Template for output names.
  64     restrictfilenames: Do not allow "&" and spaces in file names
  65     ignoreerrors:      Do not stop on download errors.
  66     nooverwrites:      Prevent overwriting files.
  67     playliststart:     Playlist item to start at.
  68     playlistend:       Playlist item to end at.
  69     matchtitle:        Download only matching titles.
  70     rejecttitle:       Reject downloads for matching titles.
  71     logtostderr:       Log messages to stderr instead of stdout.
  72     writedescription:  Write the video description to a .description file
  73     writeinfojson:     Write the video description to a .info.json file
  74     writeannotations:  Write the video annotations to a .annotations.xml file
  75     writethumbnail:    Write the thumbnail image to a file
  76     writesubtitles:    Write the video subtitles to a file
  77     writeautomaticsub: Write the automatic subtitles to a file
  78     allsubtitles:      Downloads all the subtitles of the video
  79                        (requires writesubtitles or writeautomaticsub)
  80     listsubtitles:     Lists all available subtitles for the video
  81     subtitlesformat:   Subtitle format [srt/sbv/vtt] (default=srt)
  82     subtitleslangs:    List of languages of the subtitles to download
  83     keepvideo:         Keep the video file after post-processing
  84     daterange:         A DateRange object, download only if the upload_date is in the range.
  85     skip_download:     Skip the actual download of the video file
  86     cachedir:          Location of the cache files in the filesystem.
  87                        None to disable filesystem cache.
  88     noplaylist:        Download single video instead of a playlist if in doubt.
  89     age_limit:         An integer representing the user's age in years.
  90                        Unsuitable videos for the given age are skipped.
  91     downloadarchive:   File name of a file where all downloads are recorded.
  92                        Videos already present in the file are not downloaded
  93                        again.
  94
  95     The following parameters are not used by YoutubeDL itself, they are used by
  96     the FileDownloader:
  97     nopart, updatetime, buffersize, ratelimit, min_filesize, max_filesize, test,
  98     noresizebuffer, retries, continuedl, noprogress, consoletitle
  99     """
 100
 101     params = None
 102     _ies = []
 103     _pps = []
 104     _download_retcode = None
 105     _num_downloads = None
 106     _screen_file = None
 107
 108     def __init__(self, params):
 109         """Create a FileDownloader object with the given options."""
 110         self._ies = []
 111         self._ies_instances = {}
 112         self._pps = []
 113         self._progress_hooks = []
 114         self._download_retcode = 0
 115         self._num_downloads = 0
 116         self._screen_file = [sys.stdout, sys.stderr][params.get('logtostderr', False)]
 117
 118         if (sys.version_info >= (3,) and sys.platform != 'win32' and
 119                 sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
 120                 and not params['restrictfilenames']):
 121             # On Python 3, the Unicode filesystem API will throw errors (#1474)
 122             self.report_warning(
 123                 u'Assuming --restrict-filenames since file system encoding '
 124                 u'cannot encode all charactes. '
 125                 u'Set the LC_ALL environment variable to fix this.')
 126             params['restrictfilenames'] = True
 127
 128         self.params = params
 129         self.fd = FileDownloader(self, self.params)
 130
 131         if '%(stitle)s' in self.params['outtmpl']:
 132             self.report_warning(u'%(stitle)s is deprecated. Use the %(title)s and the --restrict-filenames flag(which also secures %(uploader)s et al) instead.')
 133
 134     def add_info_extractor(self, ie):
 135         """Add an InfoExtractor object to the end of the list."""
 136         self._ies.append(ie)
 137         self._ies_instances[ie.ie_key()] = ie
 138         ie.set_downloader(self)
 139
 140     def get_info_extractor(self, ie_key):
 141         """
 142         Get an instance of an IE with name ie_key, it will try to get one from
 143         the _ies list, if there's no instance it will create a new one and add
 144         it to the extractor list.
 145         """
 146         ie = self._ies_instances.get(ie_key)
 147         if ie is None:
 148             ie = get_info_extractor(ie_key)()
 149             self.add_info_extractor(ie)
 150         return ie
 151
 152     def add_default_info_extractors(self):
 153         """
 154         Add the InfoExtractors returned by gen_extractors to the end of the list
 155         """
 156         for ie in gen_extractors():
 157             self.add_info_extractor(ie)
 158
 159     def add_post_processor(self, pp):
 160         """Add a PostProcessor object to the end of the chain."""
 161         self._pps.append(pp)
 162         pp.set_downloader(self)
 163
 164     def to_screen(self, message, skip_eol=False):
 165         """Print message to stdout if not in quiet mode."""
 166         if not self.params.get('quiet', False):
 167             terminator = [u'\n', u''][skip_eol]
 168             output = message + terminator
 169             write_string(output, self._screen_file)
 170
 171     def to_stderr(self, message):
 172         """Print message to stderr."""
 173         assert type(message) == type(u'')
 174         output = message + u'\n'
 175         if 'b' in getattr(self._screen_file, 'mode', '') or sys.version_info[0] < 3: # Python 2 lies about the mode of sys.stdout/sys.stderr
 176             output = output.encode(preferredencoding())
 177         sys.stderr.write(output)
 178
 179     def fixed_template(self):
 180         """Checks if the output template is fixed."""
 181         return (re.search(u'(?u)%\\(.+?\\)s', self.params['outtmpl']) is None)
 182
 183     def trouble(self, message=None, tb=None):
 184         """Determine action to take when a download problem appears.
 185
 186         Depending on if the downloader has been configured to ignore
 187         download errors or not, this method may throw an exception or
 188         not when errors are found, after printing the message.
 189
 190         tb, if given, is additional traceback information.
 191         """
 192         if message is not None:
 193             self.to_stderr(message)
 194         if self.params.get('verbose'):
 195             if tb is None:
 196                 if sys.exc_info()[0]:  # if .trouble has been called from an except block
 197                     tb = u''
 198                     if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
 199                         tb += u''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
 200                     tb += compat_str(traceback.format_exc())
 201                 else:
 202                     tb_data = traceback.format_list(traceback.extract_stack())
 203                     tb = u''.join(tb_data)
 204             self.to_stderr(tb)
 205         if not self.params.get('ignoreerrors', False):
 206             if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
 207                 exc_info = sys.exc_info()[1].exc_info
 208             else:
 209                 exc_info = sys.exc_info()
 210             raise DownloadError(message, exc_info)
 211         self._download_retcode = 1
 212
 213     def report_warning(self, message):
 214         '''
 215         Print the message to stderr, it will be prefixed with 'WARNING:'
 216         If stderr is a tty file the 'WARNING:' will be colored
 217         '''
 218         if sys.stderr.isatty() and os.name != 'nt':
 219             _msg_header = u'\033[0;33mWARNING:\033[0m'
 220         else:
 221             _msg_header = u'WARNING:'
 222         warning_message = u'%s %s' % (_msg_header, message)
 223         self.to_stderr(warning_message)
 224
 225     def report_error(self, message, tb=None):
 226         '''
 227         Do the same as trouble, but prefixes the message with 'ERROR:', colored
 228         in red if stderr is a tty file.
 229         '''
 230         if sys.stderr.isatty() and os.name != 'nt':
 231             _msg_header = u'\033[0;31mERROR:\033[0m'
 232         else:
 233             _msg_header = u'ERROR:'
 234         error_message = u'%s %s' % (_msg_header, message)
 235         self.trouble(error_message, tb)
 236
 237     def report_writedescription(self, descfn):
 238         """ Report that the description file is being written """
 239         self.to_screen(u'[info] Writing video description to: ' + descfn)
 240
 241     def report_writesubtitles(self, sub_filename):
 242         """ Report that the subtitles file is being written """
 243         self.to_screen(u'[info] Writing video subtitles to: ' + sub_filename)
 244
 245     def report_writeinfojson(self, infofn):
 246         """ Report that the metadata file has been written """
 247         self.to_screen(u'[info] Video description metadata as JSON to: ' + infofn)
 248
 249     def report_writeannotations(self, annofn):
 250         """ Report that the annotations file has been written. """
 251         self.to_screen(u'[info] Writing video annotations to: ' + annofn)
 252
 253     def report_file_already_downloaded(self, file_name):
 254         """Report file has already been fully downloaded."""
 255         try:
 256             self.to_screen(u'[download] %s has already been downloaded' % file_name)
 257         except (UnicodeEncodeError) as err:
 258             self.to_screen(u'[download] The file has already been downloaded')
 259
 260     def increment_downloads(self):
 261         """Increment the ordinal that assigns a number to each file."""
 262         self._num_downloads += 1
 263
 264     def prepare_filename(self, info_dict):
 265         """Generate the output filename."""
 266         try:
 267             template_dict = dict(info_dict)
 268
 269             template_dict['epoch'] = int(time.time())
 270             autonumber_size = self.params.get('autonumber_size')
 271             if autonumber_size is None:
 272                 autonumber_size = 5
 273             autonumber_templ = u'%0' + str(autonumber_size) + u'd'
 274             template_dict['autonumber'] = autonumber_templ % self._num_downloads
 275             if template_dict['playlist_index'] is not None:
 276                 template_dict['playlist_index'] = u'%05d' % template_dict['playlist_index']
 277
 278             sanitize = lambda k, v: sanitize_filename(
 279                 u'NA' if v is None else compat_str(v),
 280                 restricted=self.params.get('restrictfilenames'),
 281                 is_id=k == u'id')
 282             template_dict = dict((k, sanitize(k, v)) for k, v in template_dict.items())
 283
 284             filename = self.params['outtmpl'] % template_dict
 285             return filename
 286         except KeyError as err:
 287             self.report_error(u'Erroneous output template')
 288             return None
 289         except ValueError as err:
 290             self.report_error(u'Error in output template: ' + str(err) + u' (encoding: ' + repr(preferredencoding()) + ')')
 291             return None
 292
 293     def _match_entry(self, info_dict):
 294         """ Returns None iff the file should be downloaded """
 295
 296         title = info_dict['title']
 297         matchtitle = self.params.get('matchtitle', False)
 298         if matchtitle:
 299             if not re.search(matchtitle, title, re.IGNORECASE):
 300                 return u'[download] "' + title + '" title did not match pattern "' + matchtitle + '"'
 301         rejecttitle = self.params.get('rejecttitle', False)
 302         if rejecttitle:
 303             if re.search(rejecttitle, title, re.IGNORECASE):
 304                 return u'"' + title + '" title matched reject pattern "' + rejecttitle + '"'
 305         date = info_dict.get('upload_date', None)
 306         if date is not None:
 307             dateRange = self.params.get('daterange', DateRange())
 308             if date not in dateRange:
 309                 return u'[download] %s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange)
 310         age_limit = self.params.get('age_limit')
 311         if age_limit is not None:
 312             if age_limit < info_dict.get('age_limit', 0):
 313                 return u'Skipping "' + title + '" because it is age restricted'
 314         if self.in_download_archive(info_dict):
 315             return (u'%(title)s has already been recorded in archive'
 316                     % info_dict)
 317         return None
 318
 319     def extract_info(self, url, download=True, ie_key=None, extra_info={}):
 320         '''
 321         Returns a list with a dictionary for each video we find.
 322         If 'download', also downloads the videos.
 323         extra_info is a dict containing the extra values to add to each result
 324          '''
 325
 326         if ie_key:
 327             ies = [self.get_info_extractor(ie_key)]
 328         else:
 329             ies = self._ies
 330
 331         for ie in ies:
 332             if not ie.suitable(url):
 333                 continue
 334
 335             if not ie.working():
 336                 self.report_warning(u'The program functionality for this site has been marked as broken, '
 337                                     u'and will probably not work.')
 338
 339             try:
 340                 ie_result = ie.extract(url)
 341                 if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
 342                     break
 343                 if isinstance(ie_result, list):
 344                     # Backwards compatibility: old IE result format
 345                     for result in ie_result:
 346                         result.update(extra_info)
 347                     ie_result = {
 348                         '_type': 'compat_list',
 349                         'entries': ie_result,
 350                     }
 351                 else:
 352                     ie_result.update(extra_info)
 353                 if 'extractor' not in ie_result:
 354                     ie_result['extractor'] = ie.IE_NAME
 355                 return self.process_ie_result(ie_result, download=download)
 356             except ExtractorError as de: # An error we somewhat expected
 357                 self.report_error(compat_str(de), de.format_traceback())
 358                 break
 359             except Exception as e:
 360                 if self.params.get('ignoreerrors', False):
 361                     self.report_error(compat_str(e), tb=compat_str(traceback.format_exc()))
 362                     break
 363                 else:
 364                     raise
 365         else:
 366             self.report_error(u'no suitable InfoExtractor: %s' % url)
 367
 368     def process_ie_result(self, ie_result, download=True, extra_info={}):
 369         """
 370         Take the result of the ie(may be modified) and resolve all unresolved
 371         references (URLs, playlist items).
 372
 373         It will also download the videos if 'download'.
 374         Returns the resolved ie_result.
 375         """
 376
 377         result_type = ie_result.get('_type', 'video') # If not given we suppose it's a video, support the default old system
 378         if result_type == 'video':
 379             ie_result.update(extra_info)
 380             return self.process_video_result(ie_result)
 381         elif result_type == 'url':
 382             # We have to add extra_info to the results because it may be
 383             # contained in a playlist
 384             return self.extract_info(ie_result['url'],
 385                                      download,
 386                                      ie_key=ie_result.get('ie_key'),
 387                                      extra_info=extra_info)
 388         elif result_type == 'playlist':
 389             # We process each entry in the playlist
 390             playlist = ie_result.get('title', None) or ie_result.get('id', None)
 391             self.to_screen(u'[download] Downloading playlist: %s' % playlist)
 392
 393             playlist_results = []
 394
 395             n_all_entries = len(ie_result['entries'])
 396             playliststart = self.params.get('playliststart', 1) - 1
 397             playlistend = self.params.get('playlistend', -1)
 398
 399             if playlistend == -1:
 400                 entries = ie_result['entries'][playliststart:]
 401             else:
 402                 entries = ie_result['entries'][playliststart:playlistend]
 403
 404             n_entries = len(entries)
 405
 406             self.to_screen(u"[%s] playlist '%s': Collected %d video ids (downloading %d of them)" %
 407                 (ie_result['extractor'], playlist, n_all_entries, n_entries))
 408
 409             for i, entry in enumerate(entries, 1):
 410                 self.to_screen(u'[download] Downloading video #%s of %s' % (i, n_entries))
 411                 extra = {
 412                     'playlist': playlist,
 413                     'playlist_index': i + playliststart,
 414                 }
 415                 if not 'extractor' in entry:
 416                     # We set the extractor, if it's an url it will be set then to
 417                     # the new extractor, but if it's already a video we must make
 418                     # sure it's present: see issue #877
 419                     entry['extractor'] = ie_result['extractor']
 420                 entry_result = self.process_ie_result(entry,
 421                                                       download=download,
 422                                                       extra_info=extra)
 423                 playlist_results.append(entry_result)
 424             ie_result['entries'] = playlist_results
 425             return ie_result
 426         elif result_type == 'compat_list':
 427             def _fixup(r):
 428                 r.setdefault('extractor', ie_result['extractor'])
 429                 return r
 430             ie_result['entries'] = [
 431                 self.process_ie_result(_fixup(r), download=download)
 432                 for r in ie_result['entries']
 433             ]
 434             return ie_result
 435         else:
 436             raise Exception('Invalid result type: %s' % result_type)
 437
 438     def select_format(self, format_spec, available_formats):
 439         if format_spec == 'best' or format_spec is None:
 440             return available_formats[-1]
 441         elif format_spec == 'worst':
 442             return available_formats[0]
 443         else:
 444             extensions = [u'mp4', u'flv', u'webm', u'3gp']
 445             if format_spec in extensions:
 446                 filter_f = lambda f: f['ext'] == format_spec
 447             else:
 448                 filter_f = lambda f: f['format_id'] == format_spec
 449             matches = list(filter(filter_f, available_formats))
 450             if matches:
 451                 return matches[-1]
 452         return None
 453
 454     def process_video_result(self, info_dict, download=True):
 455         assert info_dict.get('_type', 'video') == 'video'
 456
 457         if 'playlist' not in info_dict:
 458             # It isn't part of a playlist
 459             info_dict['playlist'] = None
 460             info_dict['playlist_index'] = None
 461
 462         # This extractors handle format selection themselves
 463         if info_dict['extractor'] in [u'youtube', u'Youku', u'YouPorn', u'mixcloud']:
 464             if download:
 465                 self.process_info(info_dict)
 466             return info_dict
 467
 468         # We now pick which formats have to be downloaded
 469         if info_dict.get('formats') is None:
 470             # There's only one format available
 471             formats = [info_dict]
 472         else:
 473             formats = info_dict['formats']
 474
 475         # We check that all the formats have the format and format_id fields
 476         for (i, format) in enumerate(formats):
 477             if format.get('format_id') is None:
 478                 format['format_id'] = compat_str(i)
 479             if format.get('format') is None:
 480                 format['format'] = u'{id} - {res}{note}'.format(
 481                     id=format['format_id'],
 482                     res=self.format_resolution(format),
 483                     note=u' ({})'.format(format['format_note']) if format.get('format_note') is not None else '',
 484                 )
 485
 486         if self.params.get('listformats', None):
 487             self.list_formats(info_dict)
 488             return
 489
 490         format_limit = self.params.get('format_limit', None)
 491         if format_limit:
 492             formats = list(takewhile_inclusive(
 493                 lambda f: f['format_id'] != format_limit, formats
 494             ))
 495         if self.params.get('prefer_free_formats'):
 496             def _free_formats_key(f):
 497                 try:
 498                     ext_ord = [u'flv', u'mp4', u'webm'].index(f['ext'])
 499                 except ValueError:
 500                     ext_ord = -1
 501                 # We only compare the extension if they have the same height and width
 502                 return (f.get('height'), f.get('width'), ext_ord)
 503             formats = sorted(formats, key=_free_formats_key)
 504
 505         req_format = self.params.get('format', 'best')
 506         if req_format is None:
 507             req_format = 'best'
 508         formats_to_download = []
 509         # The -1 is for supporting YoutubeIE
 510         if req_format in ('-1', 'all'):
 511             formats_to_download = formats
 512         else:
 513             # We can accept formats requestd in the format: 34/5/best, we pick
 514             # the first that is available, starting from left
 515             req_formats = req_format.split('/')
 516             for rf in req_formats:
 517                 selected_format = self.select_format(rf, formats)
 518                 if selected_format is not None:
 519                     formats_to_download = [selected_format]
 520                     break
 521         if not formats_to_download:
 522             raise ExtractorError(u'requested format not available')
 523
 524         if download:
 525             if len(formats_to_download) > 1:
 526                 self.to_screen(u'[info] %s: downloading video in %s formats' % (info_dict['id'], len(formats_to_download)))
 527             for format in formats_to_download:
 528                 new_info = dict(info_dict)
 529                 new_info.update(format)
 530                 self.process_info(new_info)
 531         # We update the info dict with the best quality format (backwards compatibility)
 532         info_dict.update(formats_to_download[-1])
 533         return info_dict
 534
 535     def process_info(self, info_dict):
 536         """Process a single resolved IE result."""
 537
 538         assert info_dict.get('_type', 'video') == 'video'
 539         #We increment the download the download count here to match the previous behaviour.
 540         self.increment_downloads()
 541
 542         info_dict['fulltitle'] = info_dict['title']
 543         if len(info_dict['title']) > 200:
 544             info_dict['title'] = info_dict['title'][:197] + u'...'
 545
 546         # Keep for backwards compatibility
 547         info_dict['stitle'] = info_dict['title']
 548
 549         if not 'format' in info_dict:
 550             info_dict['format'] = info_dict['ext']
 551
 552         reason = self._match_entry(info_dict)
 553         if reason is not None:
 554             self.to_screen(u'[download] ' + reason)
 555             return
 556
 557         max_downloads = self.params.get('max_downloads')
 558         if max_downloads is not None:
 559             if self._num_downloads > int(max_downloads):
 560                 raise MaxDownloadsReached()
 561
 562         filename = self.prepare_filename(info_dict)
 563
 564         # Forced printings
 565         if self.params.get('forcetitle', False):
 566             compat_print(info_dict['title'])
 567         if self.params.get('forceid', False):
 568             compat_print(info_dict['id'])
 569         if self.params.get('forceurl', False):
 570             # For RTMP URLs, also include the playpath
 571             compat_print(info_dict['url'] + info_dict.get('play_path', u''))
 572         if self.params.get('forcethumbnail', False) and 'thumbnail' in info_dict:
 573             compat_print(info_dict['thumbnail'])
 574         if self.params.get('forcedescription', False) and 'description' in info_dict:
 575             compat_print(info_dict['description'])
 576         if self.params.get('forcefilename', False) and filename is not None:
 577             compat_print(filename)
 578         if self.params.get('forceformat', False):
 579             compat_print(info_dict['format'])
 580
 581         # Do nothing else if in simulate mode
 582         if self.params.get('simulate', False):
 583             return
 584
 585         if filename is None:
 586             return
 587
 588         try:
 589             dn = os.path.dirname(encodeFilename(filename))
 590             if dn != '' and not os.path.exists(dn):
 591                 os.makedirs(dn)
 592         except (OSError, IOError) as err:
 593             self.report_error(u'unable to create directory ' + compat_str(err))
 594             return
 595
 596         if self.params.get('writedescription', False):
 597             try:
 598                 descfn = filename + u'.description'
 599                 self.report_writedescription(descfn)
 600                 with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
 601                     descfile.write(info_dict['description'])
 602             except (KeyError, TypeError):
 603                 self.report_warning(u'There\'s no description to write.')
 604             except (OSError, IOError):
 605                 self.report_error(u'Cannot write description file ' + descfn)
 606                 return
 607
 608         if self.params.get('writeannotations', False):
 609             try:
 610                 annofn = filename + u'.annotations.xml'
 611                 self.report_writeannotations(annofn)
 612                 with io.open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
 613                     annofile.write(info_dict['annotations'])
 614             except (KeyError, TypeError):
 615                 self.report_warning(u'There are no annotations to write.')
 616             except (OSError, IOError):
 617                 self.report_error(u'Cannot write annotations file: ' + annofn)
 618                 return
 619
 620         subtitles_are_requested = any([self.params.get('writesubtitles', False),
 621                                        self.params.get('writeautomaticsub')])
 622
 623         if subtitles_are_requested and 'subtitles' in info_dict and info_dict['subtitles']:
 624             # subtitles download errors are already managed as troubles in relevant IE
 625             # that way it will silently go on when used with unsupporting IE
 626             subtitles = info_dict['subtitles']
 627             sub_format = self.params.get('subtitlesformat')
 628             for sub_lang in subtitles.keys():
 629                 sub = subtitles[sub_lang]
 630                 if sub is None:
 631                     continue
 632                 try:
 633                     sub_filename = subtitles_filename(filename, sub_lang, sub_format)
 634                     self.report_writesubtitles(sub_filename)
 635                     with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8') as subfile:
 636                             subfile.write(sub)
 637                 except (OSError, IOError):
 638                     self.report_error(u'Cannot write subtitles file ' + descfn)
 639                     return
 640
 641         if self.params.get('writeinfojson', False):
 642             infofn = filename + u'.info.json'
 643             self.report_writeinfojson(infofn)
 644             try:
 645                 json_info_dict = dict((k, v) for k, v in info_dict.items() if not k in ['urlhandle'])
 646                 write_json_file(json_info_dict, encodeFilename(infofn))
 647             except (OSError, IOError):
 648                 self.report_error(u'Cannot write metadata to JSON file ' + infofn)
 649                 return
 650
 651         if self.params.get('writethumbnail', False):
 652             if info_dict.get('thumbnail') is not None:
 653                 thumb_format = determine_ext(info_dict['thumbnail'], u'jpg')
 654                 thumb_filename = filename.rpartition('.')[0] + u'.' + thumb_format
 655                 self.to_screen(u'[%s] %s: Downloading thumbnail ...' %
 656                                (info_dict['extractor'], info_dict['id']))
 657                 try:
 658                     uf = compat_urllib_request.urlopen(info_dict['thumbnail'])
 659                     with open(thumb_filename, 'wb') as thumbf:
 660                         shutil.copyfileobj(uf, thumbf)
 661                     self.to_screen(u'[%s] %s: Writing thumbnail to: %s' %
 662                         (info_dict['extractor'], info_dict['id'], thumb_filename))
 663                 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
 664                     self.report_warning(u'Unable to download thumbnail "%s": %s' %
 665                         (info_dict['thumbnail'], compat_str(err)))
 666
 667         if not self.params.get('skip_download', False):
 668             if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(filename)):
 669                 success = True
 670             else:
 671                 try:
 672                     success = self.fd._do_download(filename, info_dict)
 673                 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
 674                     self.report_error(u'unable to download video data: %s' % str(err))
 675                     return
 676                 except (OSError, IOError) as err:
 677                     raise UnavailableVideoError(err)
 678                 except (ContentTooShortError, ) as err:
 679                     self.report_error(u'content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded))
 680                     return
 681
 682             if success:
 683                 try:
 684                     self.post_process(filename, info_dict)
 685                 except (PostProcessingError) as err:
 686                     self.report_error(u'postprocessing: %s' % str(err))
 687                     return
 688
 689         self.record_download_archive(info_dict)
 690
 691     def download(self, url_list):
 692         """Download a given list of URLs."""
 693         if len(url_list) > 1 and self.fixed_template():
 694             raise SameFileError(self.params['outtmpl'])
 695
 696         for url in url_list:
 697             try:
 698                 #It also downloads the videos
 699                 videos = self.extract_info(url)
 700             except UnavailableVideoError:
 701                 self.report_error(u'unable to download video')
 702             except MaxDownloadsReached:
 703                 self.to_screen(u'[info] Maximum number of downloaded files reached.')
 704                 raise
 705
 706         return self._download_retcode
 707
 708     def post_process(self, filename, ie_info):
 709         """Run all the postprocessors on the given file."""
 710         info = dict(ie_info)
 711         info['filepath'] = filename
 712         keep_video = None
 713         for pp in self._pps:
 714             try:
 715                 keep_video_wish, new_info = pp.run(info)
 716                 if keep_video_wish is not None:
 717                     if keep_video_wish:
 718                         keep_video = keep_video_wish
 719                     elif keep_video is None:
 720                         # No clear decision yet, let IE decide
 721                         keep_video = keep_video_wish
 722             except PostProcessingError as e:
 723                 self.report_error(e.msg)
 724         if keep_video is False and not self.params.get('keepvideo', False):
 725             try:
 726                 self.to_screen(u'Deleting original file %s (pass -k to keep)' % filename)
 727                 os.remove(encodeFilename(filename))
 728             except (IOError, OSError):
 729                 self.report_warning(u'Unable to remove downloaded video file')
 730
 731     def in_download_archive(self, info_dict):
 732         fn = self.params.get('download_archive')
 733         if fn is None:
 734             return False
 735         vid_id = info_dict['extractor'] + u' ' + info_dict['id']
 736         try:
 737             with locked_file(fn, 'r', encoding='utf-8') as archive_file:
 738                 for line in archive_file:
 739                     if line.strip() == vid_id:
 740                         return True
 741         except IOError as ioe:
 742             if ioe.errno != errno.ENOENT:
 743                 raise
 744         return False
 745
 746     def record_download_archive(self, info_dict):
 747         fn = self.params.get('download_archive')
 748         if fn is None:
 749             return
 750         vid_id = info_dict['extractor'] + u' ' + info_dict['id']
 751         with locked_file(fn, 'a', encoding='utf-8') as archive_file:
 752             archive_file.write(vid_id + u'\n')
 753
 754     @staticmethod
 755     def format_resolution(format):
 756         if format.get('height') is not None:
 757             if format.get('width') is not None:
 758                 res = u'%sx%s' % (format['width'], format['height'])
 759             else:
 760                 res = u'%sp' % format['height']
 761         else:
 762             res = '???'
 763         return res
 764
 765     def list_formats(self, info_dict):
 766         formats_s = []
 767         for format in info_dict.get('formats', [info_dict]):
 768             formats_s.append(u'%-15s: %-5s     %-15s[%s]' % (
 769                 format['format_id'],
 770                 format['ext'],
 771                 format.get('format_note') or '-',
 772                 self.format_resolution(format),
 773                 )
 774             )
 775         if len(formats_s) != 1:
 776             formats_s[0] += ' (worst)'
 777             formats_s[-1] += ' (best)'
 778         formats_s = "\n".join(formats_s)
 779         self.to_screen(u'[info] Available formats for %s:\n'
 780             u'format code    extension   note           resolution\n%s' % (
 781                 info_dict['id'], formats_s))