Raphaël G. Git Repositories - youtubedl/blob - 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))
 283                                  for k, v in template_dict.items())
 284
 285             tmpl = os.path.expanduser(self.params['outtmpl'])
 286             filename = tmpl % template_dict
 287             return filename
 288         except KeyError as err:
 289             self.report_error(u'Erroneous output template')
 290             return None
 291         except ValueError as err:
 292             self.report_error(u'Error in output template: ' + str(err) + u' (encoding: ' + repr(preferredencoding()) + ')')
 293             return None
 294
 295     def _match_entry(self, info_dict):
 296         """ Returns None iff the file should be downloaded """
 297
 298         title = info_dict['title']
 299         matchtitle = self.params.get('matchtitle', False)
 300         if matchtitle:
 301             if not re.search(matchtitle, title, re.IGNORECASE):
 302                 return u'[download] "' + title + '" title did not match pattern "' + matchtitle + '"'
 303         rejecttitle = self.params.get('rejecttitle', False)
 304         if rejecttitle:
 305             if re.search(rejecttitle, title, re.IGNORECASE):
 306                 return u'"' + title + '" title matched reject pattern "' + rejecttitle + '"'
 307         date = info_dict.get('upload_date', None)
 308         if date is not None:
 309             dateRange = self.params.get('daterange', DateRange())
 310             if date not in dateRange:
 311                 return u'[download] %s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange)
 312         age_limit = self.params.get('age_limit')
 313         if age_limit is not None:
 314             if age_limit < info_dict.get('age_limit', 0):
 315                 return u'Skipping "' + title + '" because it is age restricted'
 316         if self.in_download_archive(info_dict):
 317             return (u'%(title)s has already been recorded in archive'
 318                     % info_dict)
 319         return None
 320
 321     def extract_info(self, url, download=True, ie_key=None, extra_info={}):
 322         '''
 323         Returns a list with a dictionary for each video we find.
 324         If 'download', also downloads the videos.
 325         extra_info is a dict containing the extra values to add to each result
 326          '''
 327
 328         if ie_key:
 329             ies = [self.get_info_extractor(ie_key)]
 330         else:
 331             ies = self._ies
 332
 333         for ie in ies:
 334             if not ie.suitable(url):
 335                 continue
 336
 337             if not ie.working():
 338                 self.report_warning(u'The program functionality for this site has been marked as broken, '
 339                                     u'and will probably not work.')
 340
 341             try:
 342                 ie_result = ie.extract(url)
 343                 if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
 344                     break
 345                 if isinstance(ie_result, list):
 346                     # Backwards compatibility: old IE result format
 347                     for result in ie_result:
 348                         result.update(extra_info)
 349                     ie_result = {
 350                         '_type': 'compat_list',
 351                         'entries': ie_result,
 352                     }
 353                 else:
 354                     ie_result.update(extra_info)
 355                 if 'extractor' not in ie_result:
 356                     ie_result['extractor'] = ie.IE_NAME
 357                 return self.process_ie_result(ie_result, download=download)
 358             except ExtractorError as de: # An error we somewhat expected
 359                 self.report_error(compat_str(de), de.format_traceback())
 360                 break
 361             except Exception as e:
 362                 if self.params.get('ignoreerrors', False):
 363                     self.report_error(compat_str(e), tb=compat_str(traceback.format_exc()))
 364                     break
 365                 else:
 366                     raise
 367         else:
 368             self.report_error(u'no suitable InfoExtractor: %s' % url)
 369
 370     def process_ie_result(self, ie_result, download=True, extra_info={}):
 371         """
 372         Take the result of the ie(may be modified) and resolve all unresolved
 373         references (URLs, playlist items).
 374
 375         It will also download the videos if 'download'.
 376         Returns the resolved ie_result.
 377         """
 378
 379         result_type = ie_result.get('_type', 'video') # If not given we suppose it's a video, support the default old system
 380         if result_type == 'video':
 381             ie_result.update(extra_info)
 382             return self.process_video_result(ie_result)
 383         elif result_type == 'url':
 384             # We have to add extra_info to the results because it may be
 385             # contained in a playlist
 386             return self.extract_info(ie_result['url'],
 387                                      download,
 388                                      ie_key=ie_result.get('ie_key'),
 389                                      extra_info=extra_info)
 390         elif result_type == 'playlist':
 391             # We process each entry in the playlist
 392             playlist = ie_result.get('title', None) or ie_result.get('id', None)
 393             self.to_screen(u'[download] Downloading playlist: %s' % playlist)
 394
 395             playlist_results = []
 396
 397             n_all_entries = len(ie_result['entries'])
 398             playliststart = self.params.get('playliststart', 1) - 1
 399             playlistend = self.params.get('playlistend', -1)
 400
 401             if playlistend == -1:
 402                 entries = ie_result['entries'][playliststart:]
 403             else:
 404                 entries = ie_result['entries'][playliststart:playlistend]
 405
 406             n_entries = len(entries)
 407
 408             self.to_screen(u"[%s] playlist '%s': Collected %d video ids (downloading %d of them)" %
 409                 (ie_result['extractor'], playlist, n_all_entries, n_entries))
 410
 411             for i, entry in enumerate(entries, 1):
 412                 self.to_screen(u'[download] Downloading video #%s of %s' % (i, n_entries))
 413                 extra = {
 414                     'playlist': playlist,
 415                     'playlist_index': i + playliststart,
 416                 }
 417                 if not 'extractor' in entry:
 418                     # We set the extractor, if it's an url it will be set then to
 419                     # the new extractor, but if it's already a video we must make
 420                     # sure it's present: see issue #877
 421                     entry['extractor'] = ie_result['extractor']
 422                 entry_result = self.process_ie_result(entry,
 423                                                       download=download,
 424                                                       extra_info=extra)
 425                 playlist_results.append(entry_result)
 426             ie_result['entries'] = playlist_results
 427             return ie_result
 428         elif result_type == 'compat_list':
 429             def _fixup(r):
 430                 r.setdefault('extractor', ie_result['extractor'])
 431                 return r
 432             ie_result['entries'] = [
 433                 self.process_ie_result(_fixup(r), download=download)
 434                 for r in ie_result['entries']
 435             ]
 436             return ie_result
 437         else:
 438             raise Exception('Invalid result type: %s' % result_type)
 439
 440     def select_format(self, format_spec, available_formats):
 441         if format_spec == 'best' or format_spec is None:
 442             return available_formats[-1]
 443         elif format_spec == 'worst':
 444             return available_formats[0]
 445         else:
 446             extensions = [u'mp4', u'flv', u'webm', u'3gp']
 447             if format_spec in extensions:
 448                 filter_f = lambda f: f['ext'] == format_spec
 449             else:
 450                 filter_f = lambda f: f['format_id'] == format_spec
 451             matches = list(filter(filter_f, available_formats))
 452             if matches:
 453                 return matches[-1]
 454         return None
 455
 456     def process_video_result(self, info_dict, download=True):
 457         assert info_dict.get('_type', 'video') == 'video'
 458
 459         if 'playlist' not in info_dict:
 460             # It isn't part of a playlist
 461             info_dict['playlist'] = None
 462             info_dict['playlist_index'] = None
 463
 464         # This extractors handle format selection themselves
 465         if info_dict['extractor'] in [u'youtube', u'Youku', u'YouPorn', u'mixcloud']:
 466             if download:
 467                 self.process_info(info_dict)
 468             return info_dict
 469
 470         # We now pick which formats have to be downloaded
 471         if info_dict.get('formats') is None:
 472             # There's only one format available
 473             formats = [info_dict]
 474         else:
 475             formats = info_dict['formats']
 476
 477         # We check that all the formats have the format and format_id fields
 478         for (i, format) in enumerate(formats):
 479             if format.get('format_id') is None:
 480                 format['format_id'] = compat_str(i)
 481             if format.get('format') is None:
 482                 format['format'] = u'{id} - {res}{note}'.format(
 483                     id=format['format_id'],
 484                     res=self.format_resolution(format),
 485                     note=u' ({})'.format(format['format_note']) if format.get('format_note') is not None else '',
 486                 )
 487
 488         if self.params.get('listformats', None):
 489             self.list_formats(info_dict)
 490             return
 491
 492         format_limit = self.params.get('format_limit', None)
 493         if format_limit:
 494             formats = list(takewhile_inclusive(
 495                 lambda f: f['format_id'] != format_limit, formats
 496             ))
 497         if self.params.get('prefer_free_formats'):
 498             def _free_formats_key(f):
 499                 try:
 500                     ext_ord = [u'flv', u'mp4', u'webm'].index(f['ext'])
 501                 except ValueError:
 502                     ext_ord = -1
 503                 # We only compare the extension if they have the same height and width
 504                 return (f.get('height'), f.get('width'), ext_ord)
 505             formats = sorted(formats, key=_free_formats_key)
 506
 507         req_format = self.params.get('format', 'best')
 508         if req_format is None:
 509             req_format = 'best'
 510         formats_to_download = []
 511         # The -1 is for supporting YoutubeIE
 512         if req_format in ('-1', 'all'):
 513             formats_to_download = formats
 514         else:
 515             # We can accept formats requestd in the format: 34/5/best, we pick
 516             # the first that is available, starting from left
 517             req_formats = req_format.split('/')
 518             for rf in req_formats:
 519                 selected_format = self.select_format(rf, formats)
 520                 if selected_format is not None:
 521                     formats_to_download = [selected_format]
 522                     break
 523         if not formats_to_download:
 524             raise ExtractorError(u'requested format not available')
 525
 526         if download:
 527             if len(formats_to_download) > 1:
 528                 self.to_screen(u'[info] %s: downloading video in %s formats' % (info_dict['id'], len(formats_to_download)))
 529             for format in formats_to_download:
 530                 new_info = dict(info_dict)
 531                 new_info.update(format)
 532                 self.process_info(new_info)
 533         # We update the info dict with the best quality format (backwards compatibility)
 534         info_dict.update(formats_to_download[-1])
 535         return info_dict
 536
 537     def process_info(self, info_dict):
 538         """Process a single resolved IE result."""
 539
 540         assert info_dict.get('_type', 'video') == 'video'
 541         #We increment the download the download count here to match the previous behaviour.
 542         self.increment_downloads()
 543
 544         info_dict['fulltitle'] = info_dict['title']
 545         if len(info_dict['title']) > 200:
 546             info_dict['title'] = info_dict['title'][:197] + u'...'
 547
 548         # Keep for backwards compatibility
 549         info_dict['stitle'] = info_dict['title']
 550
 551         if not 'format' in info_dict:
 552             info_dict['format'] = info_dict['ext']
 553
 554         reason = self._match_entry(info_dict)
 555         if reason is not None:
 556             self.to_screen(u'[download] ' + reason)
 557             return
 558
 559         max_downloads = self.params.get('max_downloads')
 560         if max_downloads is not None:
 561             if self._num_downloads > int(max_downloads):
 562                 raise MaxDownloadsReached()
 563
 564         filename = self.prepare_filename(info_dict)
 565
 566         # Forced printings
 567         if self.params.get('forcetitle', False):
 568             compat_print(info_dict['title'])
 569         if self.params.get('forceid', False):
 570             compat_print(info_dict['id'])
 571         if self.params.get('forceurl', False):
 572             # For RTMP URLs, also include the playpath
 573             compat_print(info_dict['url'] + info_dict.get('play_path', u''))
 574         if self.params.get('forcethumbnail', False) and 'thumbnail' in info_dict:
 575             compat_print(info_dict['thumbnail'])
 576         if self.params.get('forcedescription', False) and 'description' in info_dict:
 577             compat_print(info_dict['description'])
 578         if self.params.get('forcefilename', False) and filename is not None:
 579             compat_print(filename)
 580         if self.params.get('forceformat', False):
 581             compat_print(info_dict['format'])
 582
 583         # Do nothing else if in simulate mode
 584         if self.params.get('simulate', False):
 585             return
 586
 587         if filename is None:
 588             return
 589
 590         try:
 591             dn = os.path.dirname(encodeFilename(filename))
 592             if dn != '' and not os.path.exists(dn):
 593                 os.makedirs(dn)
 594         except (OSError, IOError) as err:
 595             self.report_error(u'unable to create directory ' + compat_str(err))
 596             return
 597
 598         if self.params.get('writedescription', False):
 599             try:
 600                 descfn = filename + u'.description'
 601                 self.report_writedescription(descfn)
 602                 with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
 603                     descfile.write(info_dict['description'])
 604             except (KeyError, TypeError):
 605                 self.report_warning(u'There\'s no description to write.')
 606             except (OSError, IOError):
 607                 self.report_error(u'Cannot write description file ' + descfn)
 608                 return
 609
 610         if self.params.get('writeannotations', False):
 611             try:
 612                 annofn = filename + u'.annotations.xml'
 613                 self.report_writeannotations(annofn)
 614                 with io.open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
 615                     annofile.write(info_dict['annotations'])
 616             except (KeyError, TypeError):
 617                 self.report_warning(u'There are no annotations to write.')
 618             except (OSError, IOError):
 619                 self.report_error(u'Cannot write annotations file: ' + annofn)
 620                 return
 621
 622         subtitles_are_requested = any([self.params.get('writesubtitles', False),
 623                                        self.params.get('writeautomaticsub')])
 624
 625         if subtitles_are_requested and 'subtitles' in info_dict and info_dict['subtitles']:
 626             # subtitles download errors are already managed as troubles in relevant IE
 627             # that way it will silently go on when used with unsupporting IE
 628             subtitles = info_dict['subtitles']
 629             sub_format = self.params.get('subtitlesformat')
 630             for sub_lang in subtitles.keys():
 631                 sub = subtitles[sub_lang]
 632                 if sub is None:
 633                     continue
 634                 try:
 635                     sub_filename = subtitles_filename(filename, sub_lang, sub_format)
 636                     self.report_writesubtitles(sub_filename)
 637                     with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8') as subfile:
 638                             subfile.write(sub)
 639                 except (OSError, IOError):
 640                     self.report_error(u'Cannot write subtitles file ' + descfn)
 641                     return
 642
 643         if self.params.get('writeinfojson', False):
 644             infofn = filename + u'.info.json'
 645             self.report_writeinfojson(infofn)
 646             try:
 647                 json_info_dict = dict((k, v) for k, v in info_dict.items() if not k in ['urlhandle'])
 648                 write_json_file(json_info_dict, encodeFilename(infofn))
 649             except (OSError, IOError):
 650                 self.report_error(u'Cannot write metadata to JSON file ' + infofn)
 651                 return
 652
 653         if self.params.get('writethumbnail', False):
 654             if info_dict.get('thumbnail') is not None:
 655                 thumb_format = determine_ext(info_dict['thumbnail'], u'jpg')
 656                 thumb_filename = filename.rpartition('.')[0] + u'.' + thumb_format
 657                 self.to_screen(u'[%s] %s: Downloading thumbnail ...' %
 658                                (info_dict['extractor'], info_dict['id']))
 659                 try:
 660                     uf = compat_urllib_request.urlopen(info_dict['thumbnail'])
 661                     with open(thumb_filename, 'wb') as thumbf:
 662                         shutil.copyfileobj(uf, thumbf)
 663                     self.to_screen(u'[%s] %s: Writing thumbnail to: %s' %
 664                         (info_dict['extractor'], info_dict['id'], thumb_filename))
 665                 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
 666                     self.report_warning(u'Unable to download thumbnail "%s": %s' %
 667                         (info_dict['thumbnail'], compat_str(err)))
 668
 669         if not self.params.get('skip_download', False):
 670             if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(filename)):
 671                 success = True
 672             else:
 673                 try:
 674                     success = self.fd._do_download(filename, info_dict)
 675                 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
 676                     self.report_error(u'unable to download video data: %s' % str(err))
 677                     return
 678                 except (OSError, IOError) as err:
 679                     raise UnavailableVideoError(err)
 680                 except (ContentTooShortError, ) as err:
 681                     self.report_error(u'content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded))
 682                     return
 683
 684             if success:
 685                 try:
 686                     self.post_process(filename, info_dict)
 687                 except (PostProcessingError) as err:
 688                     self.report_error(u'postprocessing: %s' % str(err))
 689                     return
 690
 691         self.record_download_archive(info_dict)
 692
 693     def download(self, url_list):
 694         """Download a given list of URLs."""
 695         if len(url_list) > 1 and self.fixed_template():
 696             raise SameFileError(self.params['outtmpl'])
 697
 698         for url in url_list:
 699             try:
 700                 #It also downloads the videos
 701                 videos = self.extract_info(url)
 702             except UnavailableVideoError:
 703                 self.report_error(u'unable to download video')
 704             except MaxDownloadsReached:
 705                 self.to_screen(u'[info] Maximum number of downloaded files reached.')
 706                 raise
 707
 708         return self._download_retcode
 709
 710     def post_process(self, filename, ie_info):
 711         """Run all the postprocessors on the given file."""
 712         info = dict(ie_info)
 713         info['filepath'] = filename
 714         keep_video = None
 715         for pp in self._pps:
 716             try:
 717                 keep_video_wish, new_info = pp.run(info)
 718                 if keep_video_wish is not None:
 719                     if keep_video_wish:
 720                         keep_video = keep_video_wish
 721                     elif keep_video is None:
 722                         # No clear decision yet, let IE decide
 723                         keep_video = keep_video_wish
 724             except PostProcessingError as e:
 725                 self.report_error(e.msg)
 726         if keep_video is False and not self.params.get('keepvideo', False):
 727             try:
 728                 self.to_screen(u'Deleting original file %s (pass -k to keep)' % filename)
 729                 os.remove(encodeFilename(filename))
 730             except (IOError, OSError):
 731                 self.report_warning(u'Unable to remove downloaded video file')
 732
 733     def in_download_archive(self, info_dict):
 734         fn = self.params.get('download_archive')
 735         if fn is None:
 736             return False
 737         vid_id = info_dict['extractor'] + u' ' + info_dict['id']
 738         try:
 739             with locked_file(fn, 'r', encoding='utf-8') as archive_file:
 740                 for line in archive_file:
 741                     if line.strip() == vid_id:
 742                         return True
 743         except IOError as ioe:
 744             if ioe.errno != errno.ENOENT:
 745                 raise
 746         return False
 747
 748     def record_download_archive(self, info_dict):
 749         fn = self.params.get('download_archive')
 750         if fn is None:
 751             return
 752         vid_id = info_dict['extractor'] + u' ' + info_dict['id']
 753         with locked_file(fn, 'a', encoding='utf-8') as archive_file:
 754             archive_file.write(vid_id + u'\n')
 755
 756     @staticmethod
 757     def format_resolution(format):
 758         if format.get('height') is not None:
 759             if format.get('width') is not None:
 760                 res = u'%sx%s' % (format['width'], format['height'])
 761             else:
 762                 res = u'%sp' % format['height']
 763         else:
 764             res = '???'
 765         return res
 766
 767     def list_formats(self, info_dict):
 768         formats_s = []
 769         for format in info_dict.get('formats', [info_dict]):
 770             formats_s.append(u'%-15s: %-5s     %-15s[%s]' % (
 771                 format['format_id'],
 772                 format['ext'],
 773                 format.get('format_note') or '-',
 774                 self.format_resolution(format),
 775                 )
 776             )
 777         if len(formats_s) != 1:
 778             formats_s[0] += ' (worst)'
 779             formats_s[-1] += ' (best)'
 780         formats_s = "\n".join(formats_s)
 781         self.to_screen(u'[info] Available formats for %s:\n'
 782             u'format code    extension   note           resolution\n%s' % (
 783                 info_dict['id'], formats_s))