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 if os.name == 'nt':
  17     import ctypes
  18
  19 from .utils import (
  20     compat_http_client,
  21     compat_print,
  22     compat_str,
  23     compat_urllib_error,
  24     compat_urllib_request,
  25     ContentTooShortError,
  26     date_from_str,
  27     DateRange,
  28     determine_ext,
  29     DownloadError,
  30     encodeFilename,
  31     ExtractorError,
  32     locked_file,
  33     MaxDownloadsReached,
  34     PostProcessingError,
  35     preferredencoding,
  36     SameFileError,
  37     sanitize_filename,
  38     subtitles_filename,
  39     takewhile_inclusive,
  40     UnavailableVideoError,
  41     write_json_file,
  42     write_string,
  43 )
  44 from .extractor import get_info_extractor, gen_extractors
  45 from .FileDownloader import FileDownloader
  46
  47
  48 class YoutubeDL(object):
  49     """YoutubeDL class.
  50
  51     YoutubeDL objects are the ones responsible of downloading the
  52     actual video file and writing it to disk if the user has requested
  53     it, among some other tasks. In most cases there should be one per
  54     program. As, given a video URL, the downloader doesn't know how to
  55     extract all the needed information, task that InfoExtractors do, it
  56     has to pass the URL to one of them.
  57
  58     For this, YoutubeDL objects have a method that allows
  59     InfoExtractors to be registered in a given order. When it is passed
  60     a URL, the YoutubeDL object handles it to the first InfoExtractor it
  61     finds that reports being able to handle it. The InfoExtractor extracts
  62     all the information about the video or videos the URL refers to, and
  63     YoutubeDL process the extracted information, possibly using a File
  64     Downloader to download the video.
  65
  66     YoutubeDL objects accept a lot of parameters. In order not to saturate
  67     the object constructor with arguments, it receives a dictionary of
  68     options instead. These options are available through the params
  69     attribute for the InfoExtractors to use. The YoutubeDL also
  70     registers itself as the downloader in charge for the InfoExtractors
  71     that are added to it, so this is a "mutual registration".
  72
  73     Available options:
  74
  75     username:          Username for authentication purposes.
  76     password:          Password for authentication purposes.
  77     videopassword:     Password for acces a video.
  78     usenetrc:          Use netrc for authentication instead.
  79     verbose:           Print additional info to stdout.
  80     quiet:             Do not print messages to stdout.
  81     forceurl:          Force printing final URL.
  82     forcetitle:        Force printing title.
  83     forceid:           Force printing ID.
  84     forcethumbnail:    Force printing thumbnail URL.
  85     forcedescription:  Force printing description.
  86     forcefilename:     Force printing final filename.
  87     simulate:          Do not download the video files.
  88     format:            Video format code.
  89     format_limit:      Highest quality format to try.
  90     outtmpl:           Template for output names.
  91     restrictfilenames: Do not allow "&" and spaces in file names
  92     ignoreerrors:      Do not stop on download errors.
  93     nooverwrites:      Prevent overwriting files.
  94     playliststart:     Playlist item to start at.
  95     playlistend:       Playlist item to end at.
  96     matchtitle:        Download only matching titles.
  97     rejecttitle:       Reject downloads for matching titles.
  98     logtostderr:       Log messages to stderr instead of stdout.
  99     writedescription:  Write the video description to a .description file
 100     writeinfojson:     Write the video description to a .info.json file
 101     writeannotations:  Write the video annotations to a .annotations.xml file
 102     writethumbnail:    Write the thumbnail image to a file
 103     writesubtitles:    Write the video subtitles to a file
 104     writeautomaticsub: Write the automatic subtitles to a file
 105     allsubtitles:      Downloads all the subtitles of the video
 106                        (requires writesubtitles or writeautomaticsub)
 107     listsubtitles:     Lists all available subtitles for the video
 108     subtitlesformat:   Subtitle format [srt/sbv/vtt] (default=srt)
 109     subtitleslangs:    List of languages of the subtitles to download
 110     keepvideo:         Keep the video file after post-processing
 111     daterange:         A DateRange object, download only if the upload_date is in the range.
 112     skip_download:     Skip the actual download of the video file
 113     cachedir:          Location of the cache files in the filesystem.
 114                        None to disable filesystem cache.
 115     noplaylist:        Download single video instead of a playlist if in doubt.
 116     age_limit:         An integer representing the user's age in years.
 117                        Unsuitable videos for the given age are skipped.
 118     downloadarchive:   File name of a file where all downloads are recorded.
 119                        Videos already present in the file are not downloaded
 120                        again.
 121
 122     The following parameters are not used by YoutubeDL itself, they are used by
 123     the FileDownloader:
 124     nopart, updatetime, buffersize, ratelimit, min_filesize, max_filesize, test,
 125     noresizebuffer, retries, continuedl, noprogress, consoletitle
 126     """
 127
 128     params = None
 129     _ies = []
 130     _pps = []
 131     _download_retcode = None
 132     _num_downloads = None
 133     _screen_file = None
 134
 135     def __init__(self, params):
 136         """Create a FileDownloader object with the given options."""
 137         self._ies = []
 138         self._ies_instances = {}
 139         self._pps = []
 140         self._progress_hooks = []
 141         self._download_retcode = 0
 142         self._num_downloads = 0
 143         self._screen_file = [sys.stdout, sys.stderr][params.get('logtostderr', False)]
 144
 145         if (sys.version_info >= (3,) and sys.platform != 'win32' and
 146                 sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
 147                 and not params['restrictfilenames']):
 148             # On Python 3, the Unicode filesystem API will throw errors (#1474)
 149             self.report_warning(
 150                 u'Assuming --restrict-filenames since file system encoding '
 151                 u'cannot encode all charactes. '
 152                 u'Set the LC_ALL environment variable to fix this.')
 153             params['restrictfilenames'] = True
 154
 155         self.params = params
 156         self.fd = FileDownloader(self, self.params)
 157
 158         if '%(stitle)s' in self.params['outtmpl']:
 159             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.')
 160
 161     def add_info_extractor(self, ie):
 162         """Add an InfoExtractor object to the end of the list."""
 163         self._ies.append(ie)
 164         self._ies_instances[ie.ie_key()] = ie
 165         ie.set_downloader(self)
 166
 167     def get_info_extractor(self, ie_key):
 168         """
 169         Get an instance of an IE with name ie_key, it will try to get one from
 170         the _ies list, if there's no instance it will create a new one and add
 171         it to the extractor list.
 172         """
 173         ie = self._ies_instances.get(ie_key)
 174         if ie is None:
 175             ie = get_info_extractor(ie_key)()
 176             self.add_info_extractor(ie)
 177         return ie
 178
 179     def add_default_info_extractors(self):
 180         """
 181         Add the InfoExtractors returned by gen_extractors to the end of the list
 182         """
 183         for ie in gen_extractors():
 184             self.add_info_extractor(ie)
 185
 186     def add_post_processor(self, pp):
 187         """Add a PostProcessor object to the end of the chain."""
 188         self._pps.append(pp)
 189         pp.set_downloader(self)
 190
 191     def to_screen(self, message, skip_eol=False):
 192         """Print message to stdout if not in quiet mode."""
 193         if not self.params.get('quiet', False):
 194             terminator = [u'\n', u''][skip_eol]
 195             output = message + terminator
 196             write_string(output, self._screen_file)
 197
 198     def to_stderr(self, message):
 199         """Print message to stderr."""
 200         assert type(message) == type(u'')
 201         output = message + u'\n'
 202         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
 203             output = output.encode(preferredencoding())
 204         sys.stderr.write(output)
 205
 206     def to_console_title(self, message):
 207         if not self.params.get('consoletitle', False):
 208             return
 209         if os.name == 'nt' and ctypes.windll.kernel32.GetConsoleWindow():
 210             # c_wchar_p() might not be necessary if `message` is
 211             # already of type unicode()
 212             ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
 213         elif 'TERM' in os.environ:
 214             write_string(u'\033]0;%s\007' % message, self._screen_file)
 215
 216     def save_console_title(self):
 217         if not self.params.get('consoletitle', False):
 218             return
 219         if 'TERM' in os.environ:
 220             write_string(u'\033[22t', self._screen_file)
 221
 222     def restore_console_title(self):
 223         if not self.params.get('consoletitle', False):
 224             return
 225         if 'TERM' in os.environ:
 226             write_string(u'\033[23t', self._screen_file)
 227
 228     def __enter__(self):
 229         self.save_console_title()
 230         return self
 231
 232     def __exit__(self, *args):
 233         self.restore_console_title()
 234
 235     def fixed_template(self):
 236         """Checks if the output template is fixed."""
 237         return (re.search(u'(?u)%\\(.+?\\)s', self.params['outtmpl']) is None)
 238
 239     def trouble(self, message=None, tb=None):
 240         """Determine action to take when a download problem appears.
 241
 242         Depending on if the downloader has been configured to ignore
 243         download errors or not, this method may throw an exception or
 244         not when errors are found, after printing the message.
 245
 246         tb, if given, is additional traceback information.
 247         """
 248         if message is not None:
 249             self.to_stderr(message)
 250         if self.params.get('verbose'):
 251             if tb is None:
 252                 if sys.exc_info()[0]:  # if .trouble has been called from an except block
 253                     tb = u''
 254                     if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
 255                         tb += u''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
 256                     tb += compat_str(traceback.format_exc())
 257                 else:
 258                     tb_data = traceback.format_list(traceback.extract_stack())
 259                     tb = u''.join(tb_data)
 260             self.to_stderr(tb)
 261         if not self.params.get('ignoreerrors', False):
 262             if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
 263                 exc_info = sys.exc_info()[1].exc_info
 264             else:
 265                 exc_info = sys.exc_info()
 266             raise DownloadError(message, exc_info)
 267         self._download_retcode = 1
 268
 269     def report_warning(self, message):
 270         '''
 271         Print the message to stderr, it will be prefixed with 'WARNING:'
 272         If stderr is a tty file the 'WARNING:' will be colored
 273         '''
 274         if sys.stderr.isatty() and os.name != 'nt':
 275             _msg_header = u'\033[0;33mWARNING:\033[0m'
 276         else:
 277             _msg_header = u'WARNING:'
 278         warning_message = u'%s %s' % (_msg_header, message)
 279         self.to_stderr(warning_message)
 280
 281     def report_error(self, message, tb=None):
 282         '''
 283         Do the same as trouble, but prefixes the message with 'ERROR:', colored
 284         in red if stderr is a tty file.
 285         '''
 286         if sys.stderr.isatty() and os.name != 'nt':
 287             _msg_header = u'\033[0;31mERROR:\033[0m'
 288         else:
 289             _msg_header = u'ERROR:'
 290         error_message = u'%s %s' % (_msg_header, message)
 291         self.trouble(error_message, tb)
 292
 293     def report_writedescription(self, descfn):
 294         """ Report that the description file is being written """
 295         self.to_screen(u'[info] Writing video description to: ' + descfn)
 296
 297     def report_writesubtitles(self, sub_filename):
 298         """ Report that the subtitles file is being written """
 299         self.to_screen(u'[info] Writing video subtitles to: ' + sub_filename)
 300
 301     def report_writeinfojson(self, infofn):
 302         """ Report that the metadata file has been written """
 303         self.to_screen(u'[info] Video description metadata as JSON to: ' + infofn)
 304
 305     def report_writeannotations(self, annofn):
 306         """ Report that the annotations file has been written. """
 307         self.to_screen(u'[info] Writing video annotations to: ' + annofn)
 308
 309     def report_file_already_downloaded(self, file_name):
 310         """Report file has already been fully downloaded."""
 311         try:
 312             self.to_screen(u'[download] %s has already been downloaded' % file_name)
 313         except UnicodeEncodeError:
 314             self.to_screen(u'[download] The file has already been downloaded')
 315
 316     def increment_downloads(self):
 317         """Increment the ordinal that assigns a number to each file."""
 318         self._num_downloads += 1
 319
 320     def prepare_filename(self, info_dict):
 321         """Generate the output filename."""
 322         try:
 323             template_dict = dict(info_dict)
 324
 325             template_dict['epoch'] = int(time.time())
 326             autonumber_size = self.params.get('autonumber_size')
 327             if autonumber_size is None:
 328                 autonumber_size = 5
 329             autonumber_templ = u'%0' + str(autonumber_size) + u'd'
 330             template_dict['autonumber'] = autonumber_templ % self._num_downloads
 331             if template_dict.get('playlist_index') is not None:
 332                 template_dict['playlist_index'] = u'%05d' % template_dict['playlist_index']
 333
 334             sanitize = lambda k, v: sanitize_filename(
 335                 u'NA' if v is None else compat_str(v),
 336                 restricted=self.params.get('restrictfilenames'),
 337                 is_id=(k == u'id'))
 338             template_dict = dict((k, sanitize(k, v))
 339                                  for k, v in template_dict.items())
 340
 341             tmpl = os.path.expanduser(self.params['outtmpl'])
 342             filename = tmpl % template_dict
 343             return filename
 344         except KeyError as err:
 345             self.report_error(u'Erroneous output template')
 346             return None
 347         except ValueError as err:
 348             self.report_error(u'Error in output template: ' + str(err) + u' (encoding: ' + repr(preferredencoding()) + ')')
 349             return None
 350
 351     def _match_entry(self, info_dict):
 352         """ Returns None iff the file should be downloaded """
 353
 354         title = info_dict['title']
 355         matchtitle = self.params.get('matchtitle', False)
 356         if matchtitle:
 357             if not re.search(matchtitle, title, re.IGNORECASE):
 358                 return u'[download] "' + title + '" title did not match pattern "' + matchtitle + '"'
 359         rejecttitle = self.params.get('rejecttitle', False)
 360         if rejecttitle:
 361             if re.search(rejecttitle, title, re.IGNORECASE):
 362                 return u'"' + title + '" title matched reject pattern "' + rejecttitle + '"'
 363         date = info_dict.get('upload_date', None)
 364         if date is not None:
 365             dateRange = self.params.get('daterange', DateRange())
 366             if date not in dateRange:
 367                 return u'[download] %s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange)
 368         age_limit = self.params.get('age_limit')
 369         if age_limit is not None:
 370             if age_limit < info_dict.get('age_limit', 0):
 371                 return u'Skipping "' + title + '" because it is age restricted'
 372         if self.in_download_archive(info_dict):
 373             return (u'%(title)s has already been recorded in archive'
 374                     % info_dict)
 375         return None
 376
 377     @staticmethod
 378     def add_extra_info(info_dict, extra_info):
 379         '''Set the keys from extra_info in info dict if they are missing'''
 380         for key, value in extra_info.items():
 381             info_dict.setdefault(key, value)
 382
 383     def extract_info(self, url, download=True, ie_key=None, extra_info={}):
 384         '''
 385         Returns a list with a dictionary for each video we find.
 386         If 'download', also downloads the videos.
 387         extra_info is a dict containing the extra values to add to each result
 388          '''
 389
 390         if ie_key:
 391             ies = [self.get_info_extractor(ie_key)]
 392         else:
 393             ies = self._ies
 394
 395         for ie in ies:
 396             if not ie.suitable(url):
 397                 continue
 398
 399             if not ie.working():
 400                 self.report_warning(u'The program functionality for this site has been marked as broken, '
 401                                     u'and will probably not work.')
 402
 403             try:
 404                 ie_result = ie.extract(url)
 405                 if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
 406                     break
 407                 if isinstance(ie_result, list):
 408                     # Backwards compatibility: old IE result format
 409                     ie_result = {
 410                         '_type': 'compat_list',
 411                         'entries': ie_result,
 412                     }
 413                 self.add_extra_info(ie_result,
 414                     {
 415                         'extractor': ie.IE_NAME,
 416                         'webpage_url': url,
 417                         'extractor_key': ie.ie_key(),
 418                     })
 419                 return self.process_ie_result(ie_result, download, extra_info)
 420             except ExtractorError as de: # An error we somewhat expected
 421                 self.report_error(compat_str(de), de.format_traceback())
 422                 break
 423             except Exception as e:
 424                 if self.params.get('ignoreerrors', False):
 425                     self.report_error(compat_str(e), tb=compat_str(traceback.format_exc()))
 426                     break
 427                 else:
 428                     raise
 429         else:
 430             self.report_error(u'no suitable InfoExtractor: %s' % url)
 431
 432     def process_ie_result(self, ie_result, download=True, extra_info={}):
 433         """
 434         Take the result of the ie(may be modified) and resolve all unresolved
 435         references (URLs, playlist items).
 436
 437         It will also download the videos if 'download'.
 438         Returns the resolved ie_result.
 439         """
 440
 441         result_type = ie_result.get('_type', 'video') # If not given we suppose it's a video, support the default old system
 442         if result_type == 'video':
 443             self.add_extra_info(ie_result, extra_info)
 444             return self.process_video_result(ie_result, download=download)
 445         elif result_type == 'url':
 446             # We have to add extra_info to the results because it may be
 447             # contained in a playlist
 448             return self.extract_info(ie_result['url'],
 449                                      download,
 450                                      ie_key=ie_result.get('ie_key'),
 451                                      extra_info=extra_info)
 452         elif result_type == 'playlist':
 453             self.add_extra_info(ie_result, extra_info)
 454             # We process each entry in the playlist
 455             playlist = ie_result.get('title', None) or ie_result.get('id', None)
 456             self.to_screen(u'[download] Downloading playlist: %s' % playlist)
 457
 458             playlist_results = []
 459
 460             n_all_entries = len(ie_result['entries'])
 461             playliststart = self.params.get('playliststart', 1) - 1
 462             playlistend = self.params.get('playlistend', -1)
 463
 464             if playlistend == -1:
 465                 entries = ie_result['entries'][playliststart:]
 466             else:
 467                 entries = ie_result['entries'][playliststart:playlistend]
 468
 469             n_entries = len(entries)
 470
 471             self.to_screen(u"[%s] playlist '%s': Collected %d video ids (downloading %d of them)" %
 472                 (ie_result['extractor'], playlist, n_all_entries, n_entries))
 473
 474             for i, entry in enumerate(entries, 1):
 475                 self.to_screen(u'[download] Downloading video #%s of %s' % (i, n_entries))
 476                 extra = {
 477                     'playlist': playlist,
 478                     'playlist_index': i + playliststart,
 479                     'extractor': ie_result['extractor'],
 480                     'webpage_url': ie_result['webpage_url'],
 481                     'extractor_key': ie_result['extractor_key'],
 482                 }
 483                 entry_result = self.process_ie_result(entry,
 484                                                       download=download,
 485                                                       extra_info=extra)
 486                 playlist_results.append(entry_result)
 487             ie_result['entries'] = playlist_results
 488             return ie_result
 489         elif result_type == 'compat_list':
 490             def _fixup(r):
 491                 self.add_extra_info(r,
 492                     {
 493                         'extractor': ie_result['extractor'],
 494                         'webpage_url': ie_result['webpage_url'],
 495                         'extractor_key': ie_result['extractor_key'],
 496                     })
 497                 return r
 498             ie_result['entries'] = [
 499                 self.process_ie_result(_fixup(r), download, extra_info)
 500                 for r in ie_result['entries']
 501             ]
 502             return ie_result
 503         else:
 504             raise Exception('Invalid result type: %s' % result_type)
 505
 506     def select_format(self, format_spec, available_formats):
 507         if format_spec == 'best' or format_spec is None:
 508             return available_formats[-1]
 509         elif format_spec == 'worst':
 510             return available_formats[0]
 511         else:
 512             extensions = [u'mp4', u'flv', u'webm', u'3gp']
 513             if format_spec in extensions:
 514                 filter_f = lambda f: f['ext'] == format_spec
 515             else:
 516                 filter_f = lambda f: f['format_id'] == format_spec
 517             matches = list(filter(filter_f, available_formats))
 518             if matches:
 519                 return matches[-1]
 520         return None
 521
 522     def process_video_result(self, info_dict, download=True):
 523         assert info_dict.get('_type', 'video') == 'video'
 524
 525         if 'playlist' not in info_dict:
 526             # It isn't part of a playlist
 527             info_dict['playlist'] = None
 528             info_dict['playlist_index'] = None
 529
 530         # This extractors handle format selection themselves
 531         if info_dict['extractor'] in [u'youtube', u'Youku']:
 532             if download:
 533                 self.process_info(info_dict)
 534             return info_dict
 535
 536         # We now pick which formats have to be downloaded
 537         if info_dict.get('formats') is None:
 538             # There's only one format available
 539             formats = [info_dict]
 540         else:
 541             formats = info_dict['formats']
 542
 543         # We check that all the formats have the format and format_id fields
 544         for (i, format) in enumerate(formats):
 545             if format.get('format_id') is None:
 546                 format['format_id'] = compat_str(i)
 547             if format.get('format') is None:
 548                 format['format'] = u'{id} - {res}{note}'.format(
 549                     id=format['format_id'],
 550                     res=self.format_resolution(format),
 551                     note=u' ({0})'.format(format['format_note']) if format.get('format_note') is not None else '',
 552                 )
 553             # Automatically determine file extension if missing
 554             if 'ext' not in format:
 555                 format['ext'] = determine_ext(format['url'])
 556
 557         if self.params.get('listformats', None):
 558             self.list_formats(info_dict)
 559             return
 560
 561         format_limit = self.params.get('format_limit', None)
 562         if format_limit:
 563             formats = list(takewhile_inclusive(
 564                 lambda f: f['format_id'] != format_limit, formats
 565             ))
 566         if self.params.get('prefer_free_formats'):
 567             def _free_formats_key(f):
 568                 try:
 569                     ext_ord = [u'flv', u'mp4', u'webm'].index(f['ext'])
 570                 except ValueError:
 571                     ext_ord = -1
 572                 # We only compare the extension if they have the same height and width
 573                 return (f.get('height'), f.get('width'), ext_ord)
 574             formats = sorted(formats, key=_free_formats_key)
 575
 576         req_format = self.params.get('format', 'best')
 577         if req_format is None:
 578             req_format = 'best'
 579         formats_to_download = []
 580         # The -1 is for supporting YoutubeIE
 581         if req_format in ('-1', 'all'):
 582             formats_to_download = formats
 583         else:
 584             # We can accept formats requestd in the format: 34/5/best, we pick
 585             # the first that is available, starting from left
 586             req_formats = req_format.split('/')
 587             for rf in req_formats:
 588                 selected_format = self.select_format(rf, formats)
 589                 if selected_format is not None:
 590                     formats_to_download = [selected_format]
 591                     break
 592         if not formats_to_download:
 593             raise ExtractorError(u'requested format not available',
 594                                  expected=True)
 595
 596         if download:
 597             if len(formats_to_download) > 1:
 598                 self.to_screen(u'[info] %s: downloading video in %s formats' % (info_dict['id'], len(formats_to_download)))
 599             for format in formats_to_download:
 600                 new_info = dict(info_dict)
 601                 new_info.update(format)
 602                 self.process_info(new_info)
 603         # We update the info dict with the best quality format (backwards compatibility)
 604         info_dict.update(formats_to_download[-1])
 605         return info_dict
 606
 607     def process_info(self, info_dict):
 608         """Process a single resolved IE result."""
 609
 610         assert info_dict.get('_type', 'video') == 'video'
 611         #We increment the download the download count here to match the previous behaviour.
 612         self.increment_downloads()
 613
 614         info_dict['fulltitle'] = info_dict['title']
 615         if len(info_dict['title']) > 200:
 616             info_dict['title'] = info_dict['title'][:197] + u'...'
 617
 618         # Keep for backwards compatibility
 619         info_dict['stitle'] = info_dict['title']
 620
 621         if not 'format' in info_dict:
 622             info_dict['format'] = info_dict['ext']
 623
 624         reason = self._match_entry(info_dict)
 625         if reason is not None:
 626             self.to_screen(u'[download] ' + reason)
 627             return
 628
 629         max_downloads = self.params.get('max_downloads')
 630         if max_downloads is not None:
 631             if self._num_downloads > int(max_downloads):
 632                 raise MaxDownloadsReached()
 633
 634         filename = self.prepare_filename(info_dict)
 635
 636         # Forced printings
 637         if self.params.get('forcetitle', False):
 638             compat_print(info_dict['title'])
 639         if self.params.get('forceid', False):
 640             compat_print(info_dict['id'])
 641         if self.params.get('forceurl', False):
 642             # For RTMP URLs, also include the playpath
 643             compat_print(info_dict['url'] + info_dict.get('play_path', u''))
 644         if self.params.get('forcethumbnail', False) and info_dict.get('thumbnail') is not None:
 645             compat_print(info_dict['thumbnail'])
 646         if self.params.get('forcedescription', False) and info_dict.get('description') is not None:
 647             compat_print(info_dict['description'])
 648         if self.params.get('forcefilename', False) and filename is not None:
 649             compat_print(filename)
 650         if self.params.get('forceformat', False):
 651             compat_print(info_dict['format'])
 652
 653         # Do nothing else if in simulate mode
 654         if self.params.get('simulate', False):
 655             return
 656
 657         if filename is None:
 658             return
 659
 660         try:
 661             dn = os.path.dirname(encodeFilename(filename))
 662             if dn != '' and not os.path.exists(dn):
 663                 os.makedirs(dn)
 664         except (OSError, IOError) as err:
 665             self.report_error(u'unable to create directory ' + compat_str(err))
 666             return
 667
 668         if self.params.get('writedescription', False):
 669             try:
 670                 descfn = filename + u'.description'
 671                 self.report_writedescription(descfn)
 672                 with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
 673                     descfile.write(info_dict['description'])
 674             except (KeyError, TypeError):
 675                 self.report_warning(u'There\'s no description to write.')
 676             except (OSError, IOError):
 677                 self.report_error(u'Cannot write description file ' + descfn)
 678                 return
 679
 680         if self.params.get('writeannotations', False):
 681             try:
 682                 annofn = filename + u'.annotations.xml'
 683                 self.report_writeannotations(annofn)
 684                 with io.open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
 685                     annofile.write(info_dict['annotations'])
 686             except (KeyError, TypeError):
 687                 self.report_warning(u'There are no annotations to write.')
 688             except (OSError, IOError):
 689                 self.report_error(u'Cannot write annotations file: ' + annofn)
 690                 return
 691
 692         subtitles_are_requested = any([self.params.get('writesubtitles', False),
 693                                        self.params.get('writeautomaticsub')])
 694
 695         if subtitles_are_requested and 'subtitles' in info_dict and info_dict['subtitles']:
 696             # subtitles download errors are already managed as troubles in relevant IE
 697             # that way it will silently go on when used with unsupporting IE
 698             subtitles = info_dict['subtitles']
 699             sub_format = self.params.get('subtitlesformat', 'srt')
 700             for sub_lang in subtitles.keys():
 701                 sub = subtitles[sub_lang]
 702                 if sub is None:
 703                     continue
 704                 try:
 705                     sub_filename = subtitles_filename(filename, sub_lang, sub_format)
 706                     self.report_writesubtitles(sub_filename)
 707                     with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8') as subfile:
 708                             subfile.write(sub)
 709                 except (OSError, IOError):
 710                     self.report_error(u'Cannot write subtitles file ' + descfn)
 711                     return
 712
 713         if self.params.get('writeinfojson', False):
 714             infofn = filename + u'.info.json'
 715             self.report_writeinfojson(infofn)
 716             try:
 717                 json_info_dict = dict((k, v) for k, v in info_dict.items() if not k in ['urlhandle'])
 718                 write_json_file(json_info_dict, encodeFilename(infofn))
 719             except (OSError, IOError):
 720                 self.report_error(u'Cannot write metadata to JSON file ' + infofn)
 721                 return
 722
 723         if self.params.get('writethumbnail', False):
 724             if info_dict.get('thumbnail') is not None:
 725                 thumb_format = determine_ext(info_dict['thumbnail'], u'jpg')
 726                 thumb_filename = filename.rpartition('.')[0] + u'.' + thumb_format
 727                 self.to_screen(u'[%s] %s: Downloading thumbnail ...' %
 728                                (info_dict['extractor'], info_dict['id']))
 729                 try:
 730                     uf = compat_urllib_request.urlopen(info_dict['thumbnail'])
 731                     with open(thumb_filename, 'wb') as thumbf:
 732                         shutil.copyfileobj(uf, thumbf)
 733                     self.to_screen(u'[%s] %s: Writing thumbnail to: %s' %
 734                         (info_dict['extractor'], info_dict['id'], thumb_filename))
 735                 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
 736                     self.report_warning(u'Unable to download thumbnail "%s": %s' %
 737                         (info_dict['thumbnail'], compat_str(err)))
 738
 739         if not self.params.get('skip_download', False):
 740             if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(filename)):
 741                 success = True
 742             else:
 743                 try:
 744                     success = self.fd._do_download(filename, info_dict)
 745                 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
 746                     self.report_error(u'unable to download video data: %s' % str(err))
 747                     return
 748                 except (OSError, IOError) as err:
 749                     raise UnavailableVideoError(err)
 750                 except (ContentTooShortError, ) as err:
 751                     self.report_error(u'content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded))
 752                     return
 753
 754             if success:
 755                 try:
 756                     self.post_process(filename, info_dict)
 757                 except (PostProcessingError) as err:
 758                     self.report_error(u'postprocessing: %s' % str(err))
 759                     return
 760
 761         self.record_download_archive(info_dict)
 762
 763     def download(self, url_list):
 764         """Download a given list of URLs."""
 765         if len(url_list) > 1 and self.fixed_template():
 766             raise SameFileError(self.params['outtmpl'])
 767
 768         for url in url_list:
 769             try:
 770                 #It also downloads the videos
 771                 videos = self.extract_info(url)
 772             except UnavailableVideoError:
 773                 self.report_error(u'unable to download video')
 774             except MaxDownloadsReached:
 775                 self.to_screen(u'[info] Maximum number of downloaded files reached.')
 776                 raise
 777
 778         return self._download_retcode
 779
 780     def post_process(self, filename, ie_info):
 781         """Run all the postprocessors on the given file."""
 782         info = dict(ie_info)
 783         info['filepath'] = filename
 784         keep_video = None
 785         for pp in self._pps:
 786             try:
 787                 keep_video_wish, new_info = pp.run(info)
 788                 if keep_video_wish is not None:
 789                     if keep_video_wish:
 790                         keep_video = keep_video_wish
 791                     elif keep_video is None:
 792                         # No clear decision yet, let IE decide
 793                         keep_video = keep_video_wish
 794             except PostProcessingError as e:
 795                 self.report_error(e.msg)
 796         if keep_video is False and not self.params.get('keepvideo', False):
 797             try:
 798                 self.to_screen(u'Deleting original file %s (pass -k to keep)' % filename)
 799                 os.remove(encodeFilename(filename))
 800             except (IOError, OSError):
 801                 self.report_warning(u'Unable to remove downloaded video file')
 802
 803     def in_download_archive(self, info_dict):
 804         fn = self.params.get('download_archive')
 805         if fn is None:
 806             return False
 807         vid_id = info_dict['extractor'] + u' ' + info_dict['id']
 808         try:
 809             with locked_file(fn, 'r', encoding='utf-8') as archive_file:
 810                 for line in archive_file:
 811                     if line.strip() == vid_id:
 812                         return True
 813         except IOError as ioe:
 814             if ioe.errno != errno.ENOENT:
 815                 raise
 816         return False
 817
 818     def record_download_archive(self, info_dict):
 819         fn = self.params.get('download_archive')
 820         if fn is None:
 821             return
 822         vid_id = info_dict['extractor'] + u' ' + info_dict['id']
 823         with locked_file(fn, 'a', encoding='utf-8') as archive_file:
 824             archive_file.write(vid_id + u'\n')
 825
 826     @staticmethod
 827     def format_resolution(format, default='unknown'):
 828         if format.get('_resolution') is not None:
 829             return format['_resolution']
 830         if format.get('height') is not None:
 831             if format.get('width') is not None:
 832                 res = u'%sx%s' % (format['width'], format['height'])
 833             else:
 834                 res = u'%sp' % format['height']
 835         else:
 836             res = default
 837         return res
 838
 839     def list_formats(self, info_dict):
 840         def format_note(fdict):
 841             if fdict.get('format_note') is not None:
 842                 return fdict['format_note']
 843             res = u''
 844             if fdict.get('vcodec') is not None:
 845                 res += u'%-5s' % fdict['vcodec']
 846             elif fdict.get('vbr') is not None:
 847                 res += u'video'
 848             if fdict.get('vbr') is not None:
 849                 res += u'@%4dk' % fdict['vbr']
 850             if fdict.get('acodec') is not None:
 851                 if res:
 852                     res += u', '
 853                 res += u'%-5s' % fdict['acodec']
 854             elif fdict.get('abr') is not None:
 855                 if res:
 856                     res += u', '
 857                 res += 'audio'
 858             if fdict.get('abr') is not None:
 859                 res += u'@%3dk' % fdict['abr']
 860             return res
 861
 862         def line(format):
 863             return (u'%-20s%-10s%-12s%s' % (
 864                 format['format_id'],
 865                 format['ext'],
 866                 self.format_resolution(format),
 867                 format_note(format),
 868                 )
 869             )
 870
 871         formats = info_dict.get('formats', [info_dict])
 872         formats_s = list(map(line, formats))
 873         if len(formats) > 1:
 874             formats_s[0] += (' ' if format_note(formats[0]) else '') + '(worst)'
 875             formats_s[-1] += (' ' if format_note(formats[-1]) else '') + '(best)'
 876
 877         header_line = line({
 878             'format_id': u'format code', 'ext': u'extension',
 879             '_resolution': u'resolution', 'format_note': u'note'})
 880         self.to_screen(u'[info] Available formats for %s:\n%s\n%s' %
 881                        (info_dict['id'], header_line, u"\n".join(formats_s)))