4 from __future__
import absolute_import
, unicode_literals
29 from string
import ascii_letters
34 compat_get_terminal_size
,
40 compat_tokenize_tokenize
,
42 compat_urllib_request
,
43 compat_urllib_request_DataHandler
,
54 DOT_DESKTOP_LINK_TEMPLATE
,
55 DOT_URL_LINK_TEMPLATE
,
56 DOT_WEBLOC_LINK_TEMPLATE
,
76 PerRequestProxyHandler
,
81 register_socks_protocols
,
93 UnavailableVideoError
,
99 YoutubeDLCookieProcessor
,
101 YoutubeDLRedirectHandler
,
103 from .cache
import Cache
104 from .extractor
import get_info_extractor
, gen_extractor_classes
, _LAZY_LOADER
105 from .extractor
.openload
import PhantomJSwrapper
106 from .downloader
import get_suitable_downloader
107 from .downloader
.rtmp
import rtmpdump_version
108 from .postprocessor
import (
111 FFmpegFixupStretchedPP
,
114 FFmpegSubtitlesConvertorPP
,
117 from .version
import __version__
119 if compat_os_name
== 'nt':
123 class YoutubeDL(object):
126 YoutubeDL objects are the ones responsible of downloading the
127 actual video file and writing it to disk if the user has requested
128 it, among some other tasks. In most cases there should be one per
129 program. As, given a video URL, the downloader doesn't know how to
130 extract all the needed information, task that InfoExtractors do, it
131 has to pass the URL to one of them.
133 For this, YoutubeDL objects have a method that allows
134 InfoExtractors to be registered in a given order. When it is passed
135 a URL, the YoutubeDL object handles it to the first InfoExtractor it
136 finds that reports being able to handle it. The InfoExtractor extracts
137 all the information about the video or videos the URL refers to, and
138 YoutubeDL process the extracted information, possibly using a File
139 Downloader to download the video.
141 YoutubeDL objects accept a lot of parameters. In order not to saturate
142 the object constructor with arguments, it receives a dictionary of
143 options instead. These options are available through the params
144 attribute for the InfoExtractors to use. The YoutubeDL also
145 registers itself as the downloader in charge for the InfoExtractors
146 that are added to it, so this is a "mutual registration".
150 username: Username for authentication purposes.
151 password: Password for authentication purposes.
152 videopassword: Password for accessing a video.
153 ap_mso: Adobe Pass multiple-system operator identifier.
154 ap_username: Multiple-system operator account username.
155 ap_password: Multiple-system operator account password.
156 usenetrc: Use netrc for authentication instead.
157 verbose: Print additional info to stdout.
158 quiet: Do not print messages to stdout.
159 no_warnings: Do not print out anything for warnings.
160 forceurl: Force printing final URL.
161 forcetitle: Force printing title.
162 forceid: Force printing ID.
163 forcethumbnail: Force printing thumbnail URL.
164 forcedescription: Force printing description.
165 forcefilename: Force printing final filename.
166 forceduration: Force printing duration.
167 forcejson: Force printing info_dict as JSON.
168 dump_single_json: Force printing the info_dict of the whole playlist
169 (or video) as a single JSON line.
170 force_write_download_archive: Force writing download archive regardless of
171 'skip_download' or 'simulate'.
172 simulate: Do not download the video files.
173 format: Video format code. see "FORMAT SELECTION" for more details.
174 format_sort: How to sort the video formats. see "Sorting Formats" for more details.
175 format_sort_force: Force the given format_sort. see "Sorting Formats" for more details.
176 allow_multiple_video_streams: Allow multiple video streams to be merged into a single file
177 allow_multiple_audio_streams: Allow multiple audio streams to be merged into a single file
178 outtmpl: Template for output names.
179 restrictfilenames: Do not allow "&" and spaces in file names.
180 trim_file_name: Limit length of filename (extension excluded).
181 ignoreerrors: Do not stop on download errors.
182 force_generic_extractor: Force downloader to use the generic extractor
183 nooverwrites: Prevent overwriting files.
184 playliststart: Playlist item to start at.
185 playlistend: Playlist item to end at.
186 playlist_items: Specific indices of playlist to download.
187 playlistreverse: Download playlist items in reverse order.
188 playlistrandom: Download playlist items in random order.
189 matchtitle: Download only matching titles.
190 rejecttitle: Reject downloads for matching titles.
191 logger: Log messages to a logging.Logger instance.
192 logtostderr: Log messages to stderr instead of stdout.
193 writedescription: Write the video description to a .description file
194 writeinfojson: Write the video description to a .info.json file
195 writeannotations: Write the video annotations to a .annotations.xml file
196 writethumbnail: Write the thumbnail image to a file
197 write_all_thumbnails: Write all thumbnail formats to files
198 writelink: Write an internet shortcut file, depending on the
199 current platform (.url/.webloc/.desktop)
200 writeurllink: Write a Windows internet shortcut file (.url)
201 writewebloclink: Write a macOS internet shortcut file (.webloc)
202 writedesktoplink: Write a Linux internet shortcut file (.desktop)
203 writesubtitles: Write the video subtitles to a file
204 writeautomaticsub: Write the automatically generated subtitles to a file
205 allsubtitles: Downloads all the subtitles of the video
206 (requires writesubtitles or writeautomaticsub)
207 listsubtitles: Lists all available subtitles for the video
208 subtitlesformat: The format code for subtitles
209 subtitleslangs: List of languages of the subtitles to download
210 keepvideo: Keep the video file after post-processing
211 daterange: A DateRange object, download only if the upload_date is in the range.
212 skip_download: Skip the actual download of the video file
213 cachedir: Location of the cache files in the filesystem.
214 False to disable filesystem cache.
215 noplaylist: Download single video instead of a playlist if in doubt.
216 age_limit: An integer representing the user's age in years.
217 Unsuitable videos for the given age are skipped.
218 min_views: An integer representing the minimum view count the video
219 must have in order to not be skipped.
220 Videos without view count information are always
221 downloaded. None for no limit.
222 max_views: An integer representing the maximum view count.
223 Videos that are more popular than that are not
225 Videos without view count information are always
226 downloaded. None for no limit.
227 download_archive: File name of a file where all downloads are recorded.
228 Videos already present in the file are not downloaded
230 break_on_existing: Stop the download process after attempting to download a file that's
232 cookiefile: File name where cookies should be read from and dumped to.
233 nocheckcertificate:Do not verify SSL certificates
234 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
235 At the moment, this is only supported by YouTube.
236 proxy: URL of the proxy server to use
237 geo_verification_proxy: URL of the proxy to use for IP address verification
238 on geo-restricted sites.
239 socket_timeout: Time to wait for unresponsive hosts, in seconds
240 bidi_workaround: Work around buggy terminals without bidirectional text
241 support, using fridibi
242 debug_printtraffic:Print out sent and received HTTP traffic
243 include_ads: Download ads as well
244 default_search: Prepend this string if an input url is not valid.
245 'auto' for elaborate guessing
246 encoding: Use this encoding instead of the system-specified.
247 extract_flat: Do not resolve URLs, return the immediate result.
248 Pass in 'in_playlist' to only show this behavior for
250 postprocessors: A list of dictionaries, each with an entry
251 * key: The name of the postprocessor. See
252 youtube_dlc/postprocessor/__init__.py for a list.
253 as well as any further keyword arguments for the
255 progress_hooks: A list of functions that get called on download
256 progress, with a dictionary with the entries
257 * status: One of "downloading", "error", or "finished".
258 Check this first and ignore unknown values.
260 If status is one of "downloading", or "finished", the
261 following properties may also be present:
262 * filename: The final filename (always present)
263 * tmpfilename: The filename we're currently writing to
264 * downloaded_bytes: Bytes on disk
265 * total_bytes: Size of the whole file, None if unknown
266 * total_bytes_estimate: Guess of the eventual file size,
268 * elapsed: The number of seconds since download started.
269 * eta: The estimated time in seconds, None if unknown
270 * speed: The download speed in bytes/second, None if
272 * fragment_index: The counter of the currently
273 downloaded video fragment.
274 * fragment_count: The number of fragments (= individual
275 files that will be merged)
277 Progress hooks are guaranteed to be called at least once
278 (with status "finished") if the download is successful.
279 merge_output_format: Extension to use when merging formats.
280 fixup: Automatically correct known faults of the file.
282 - "never": do nothing
283 - "warn": only emit a warning
284 - "detect_or_warn": check whether we can do anything
285 about it, warn otherwise (default)
286 source_address: Client-side IP address to bind to.
287 call_home: Boolean, true iff we are allowed to contact the
288 youtube-dlc servers for debugging.
289 sleep_interval: Number of seconds to sleep before each download when
290 used alone or a lower bound of a range for randomized
291 sleep before each download (minimum possible number
292 of seconds to sleep) when used along with
294 max_sleep_interval:Upper bound of a range for randomized sleep before each
295 download (maximum possible number of seconds to sleep).
296 Must only be used along with sleep_interval.
297 Actual sleep time will be a random float from range
298 [sleep_interval; max_sleep_interval].
299 listformats: Print an overview of available video formats and exit.
300 list_thumbnails: Print a table of all thumbnails and exit.
301 match_filter: A function that gets called with the info_dict of
303 If it returns a message, the video is ignored.
304 If it returns None, the video is downloaded.
305 match_filter_func in utils.py is one example for this.
306 no_color: Do not emit color codes in output.
307 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
310 Two-letter ISO 3166-2 country code that will be used for
311 explicit geographic restriction bypassing via faking
312 X-Forwarded-For HTTP header
314 IP range in CIDR notation that will be used similarly to
317 The following options determine which downloader is picked:
318 external_downloader: Executable of the external downloader to call.
319 None or unset for standard (built-in) downloader.
320 hls_prefer_native: Use the native HLS downloader instead of ffmpeg/avconv
321 if True, otherwise use ffmpeg/avconv if False, otherwise
322 use downloader suggested by extractor if None.
324 The following parameters are not used by YoutubeDL itself, they are used by
325 the downloader (see youtube_dlc/downloader/common.py):
326 nopart, updatetime, buffersize, ratelimit, min_filesize, max_filesize, test,
327 noresizebuffer, retries, continuedl, noprogress, consoletitle,
328 xattr_set_filesize, external_downloader_args, hls_use_mpegts,
331 The following options are used by the post processors:
332 prefer_ffmpeg: If False, use avconv instead of ffmpeg if both are available,
333 otherwise prefer ffmpeg.
334 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
335 to the binary or its containing directory.
336 postprocessor_args: A list of additional command-line arguments for the
339 The following options are used by the Youtube extractor:
340 youtube_include_dash_manifest: If True (default), DASH manifests and related
341 data will be downloaded and processed by extractor.
342 You can reduce network I/O by disabling it if you don't
346 _NUMERIC_FIELDS
= set((
347 'width', 'height', 'tbr', 'abr', 'asr', 'vbr', 'fps', 'filesize', 'filesize_approx',
348 'timestamp', 'upload_year', 'upload_month', 'upload_day',
349 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
350 'average_rating', 'comment_count', 'age_limit',
351 'start_time', 'end_time',
352 'chapter_number', 'season_number', 'episode_number',
353 'track_number', 'disc_number', 'release_year',
360 _download_retcode
= None
361 _num_downloads
= None
364 def __init__(self
, params
=None, auto_init
=True):
365 """Create a FileDownloader object with the given options."""
369 self
._ies
_instances
= {}
371 self
._progress
_hooks
= []
372 self
._download
_retcode
= 0
373 self
._num
_downloads
= 0
374 self
._screen
_file
= [sys
.stdout
, sys
.stderr
][params
.get('logtostderr', False)]
375 self
._err
_file
= sys
.stderr
378 'nocheckcertificate': False,
380 self
.params
.update(params
)
381 self
.cache
= Cache(self
)
384 """Preload the archive, if any is specified"""
385 def preload_download_archive(self
):
386 fn
= self
.params
.get('download_archive')
390 with locked_file(fn
, 'r', encoding
='utf-8') as archive_file
:
391 for line
in archive_file
:
392 self
.archive
.add(line
.strip())
393 except IOError as ioe
:
394 if ioe
.errno
!= errno
.ENOENT
:
399 def check_deprecated(param
, option
, suggestion
):
400 if self
.params
.get(param
) is not None:
402 '%s is deprecated. Use %s instead.' % (option
, suggestion
))
406 if self
.params
.get('verbose'):
407 self
.to_stdout('[debug] Loading archive file %r' % self
.params
.get('download_archive'))
409 preload_download_archive(self
)
411 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
412 if self
.params
.get('geo_verification_proxy') is None:
413 self
.params
['geo_verification_proxy'] = self
.params
['cn_verification_proxy']
415 check_deprecated('autonumber_size', '--autonumber-size', 'output template with %(autonumber)0Nd, where N in the number of digits')
416 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
417 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
419 if params
.get('bidi_workaround', False):
422 master
, slave
= pty
.openpty()
423 width
= compat_get_terminal_size().columns
427 width_args
= ['-w', str(width
)]
429 stdin
=subprocess
.PIPE
,
431 stderr
=self
._err
_file
)
433 self
._output
_process
= subprocess
.Popen(
434 ['bidiv'] + width_args
, **sp_kwargs
437 self
._output
_process
= subprocess
.Popen(
438 ['fribidi', '-c', 'UTF-8'] + width_args
, **sp_kwargs
)
439 self
._output
_channel
= os
.fdopen(master
, 'rb')
440 except OSError as ose
:
441 if ose
.errno
== errno
.ENOENT
:
442 self
.report_warning('Could not find fribidi executable, ignoring --bidi-workaround . Make sure that fribidi is an executable file in one of the directories in your $PATH.')
446 if (sys
.platform
!= 'win32'
447 and sys
.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
448 and not params
.get('restrictfilenames', False)):
449 # Unicode filesystem API will throw errors (#1474, #13027)
451 'Assuming --restrict-filenames since file system encoding '
452 'cannot encode all characters. '
453 'Set the LC_ALL environment variable to fix this.')
454 self
.params
['restrictfilenames'] = True
456 if isinstance(params
.get('outtmpl'), bytes):
458 'Parameter outtmpl is bytes, but should be a unicode string. '
459 'Put from __future__ import unicode_literals at the top of your code file or consider switching to Python 3.x.')
464 self
.print_debug_header()
465 self
.add_default_info_extractors()
467 for pp_def_raw
in self
.params
.get('postprocessors', []):
468 pp_class
= get_postprocessor(pp_def_raw
['key'])
469 pp_def
= dict(pp_def_raw
)
471 pp
= pp_class(self
, **compat_kwargs(pp_def
))
472 self
.add_post_processor(pp
)
474 for ph
in self
.params
.get('progress_hooks', []):
475 self
.add_progress_hook(ph
)
477 register_socks_protocols()
479 def warn_if_short_id(self
, argv
):
480 # short YouTube ID starting with dash?
482 i
for i
, a
in enumerate(argv
)
483 if re
.match(r
'^-[0-9A-Za-z_-]{10}$', a
)]
487 + [a
for i
, a
in enumerate(argv
) if i
not in idxs
]
488 + ['--'] + [argv
[i
] for i
in idxs
]
491 'Long argument string detected. '
492 'Use -- to separate parameters and URLs, like this:\n%s\n' %
493 args_to_str(correct_argv
))
495 def add_info_extractor(self
, ie
):
496 """Add an InfoExtractor object to the end of the list."""
498 if not isinstance(ie
, type):
499 self
._ies
_instances
[ie
.ie_key()] = ie
500 ie
.set_downloader(self
)
502 def get_info_extractor(self
, ie_key
):
504 Get an instance of an IE with name ie_key, it will try to get one from
505 the _ies list, if there's no instance it will create a new one and add
506 it to the extractor list.
508 ie
= self
._ies
_instances
.get(ie_key
)
510 ie
= get_info_extractor(ie_key
)()
511 self
.add_info_extractor(ie
)
514 def add_default_info_extractors(self
):
516 Add the InfoExtractors returned by gen_extractors to the end of the list
518 for ie
in gen_extractor_classes():
519 self
.add_info_extractor(ie
)
521 def add_post_processor(self
, pp
):
522 """Add a PostProcessor object to the end of the chain."""
524 pp
.set_downloader(self
)
526 def add_progress_hook(self
, ph
):
527 """Add the progress hook (currently only for the file downloader)"""
528 self
._progress
_hooks
.append(ph
)
530 def _bidi_workaround(self
, message
):
531 if not hasattr(self
, '_output_channel'):
534 assert hasattr(self
, '_output_process')
535 assert isinstance(message
, compat_str
)
536 line_count
= message
.count('\n') + 1
537 self
._output
_process
.stdin
.write((message
+ '\n').encode('utf-8'))
538 self
._output
_process
.stdin
.flush()
539 res
= ''.join(self
._output
_channel
.readline().decode('utf-8')
540 for _
in range(line_count
))
541 return res
[:-len('\n')]
543 def to_screen(self
, message
, skip_eol
=False):
544 """Print message to stdout if not in quiet mode."""
545 return self
.to_stdout(message
, skip_eol
, check_quiet
=True)
547 def _write_string(self
, s
, out
=None):
548 write_string(s
, out
=out
, encoding
=self
.params
.get('encoding'))
550 def to_stdout(self
, message
, skip_eol
=False, check_quiet
=False):
551 """Print message to stdout if not in quiet mode."""
552 if self
.params
.get('logger'):
553 self
.params
['logger'].debug(message
)
554 elif not check_quiet
or not self
.params
.get('quiet', False):
555 message
= self
._bidi
_workaround
(message
)
556 terminator
= ['\n', ''][skip_eol
]
557 output
= message
+ terminator
559 self
._write
_string
(output
, self
._screen
_file
)
561 def to_stderr(self
, message
):
562 """Print message to stderr."""
563 assert isinstance(message
, compat_str
)
564 if self
.params
.get('logger'):
565 self
.params
['logger'].error(message
)
567 message
= self
._bidi
_workaround
(message
)
568 output
= message
+ '\n'
569 self
._write
_string
(output
, self
._err
_file
)
571 def to_console_title(self
, message
):
572 if not self
.params
.get('consoletitle', False):
574 if compat_os_name
== 'nt':
575 if ctypes
.windll
.kernel32
.GetConsoleWindow():
576 # c_wchar_p() might not be necessary if `message` is
577 # already of type unicode()
578 ctypes
.windll
.kernel32
.SetConsoleTitleW(ctypes
.c_wchar_p(message
))
579 elif 'TERM' in os
.environ
:
580 self
._write
_string
('\033]0;%s\007' % message
, self
._screen
_file
)
582 def save_console_title(self
):
583 if not self
.params
.get('consoletitle', False):
585 if self
.params
.get('simulate', False):
587 if compat_os_name
!= 'nt' and 'TERM' in os
.environ
:
588 # Save the title on stack
589 self
._write
_string
('\033[22;0t', self
._screen
_file
)
591 def restore_console_title(self
):
592 if not self
.params
.get('consoletitle', False):
594 if self
.params
.get('simulate', False):
596 if compat_os_name
!= 'nt' and 'TERM' in os
.environ
:
597 # Restore the title from stack
598 self
._write
_string
('\033[23;0t', self
._screen
_file
)
601 self
.save_console_title()
604 def __exit__(self
, *args
):
605 self
.restore_console_title()
607 if self
.params
.get('cookiefile') is not None:
608 self
.cookiejar
.save(ignore_discard
=True, ignore_expires
=True)
610 def trouble(self
, message
=None, tb
=None):
611 """Determine action to take when a download problem appears.
613 Depending on if the downloader has been configured to ignore
614 download errors or not, this method may throw an exception or
615 not when errors are found, after printing the message.
617 tb, if given, is additional traceback information.
619 if message
is not None:
620 self
.to_stderr(message
)
621 if self
.params
.get('verbose'):
623 if sys
.exc_info()[0]: # if .trouble has been called from an except block
625 if hasattr(sys
.exc_info()[1], 'exc_info') and sys
.exc_info()[1].exc_info
[0]:
626 tb
+= ''.join(traceback
.format_exception(*sys
.exc_info()[1].exc_info
))
627 tb
+= encode_compat_str(traceback
.format_exc())
629 tb_data
= traceback
.format_list(traceback
.extract_stack())
630 tb
= ''.join(tb_data
)
632 if not self
.params
.get('ignoreerrors', False):
633 if sys
.exc_info()[0] and hasattr(sys
.exc_info()[1], 'exc_info') and sys
.exc_info()[1].exc_info
[0]:
634 exc_info
= sys
.exc_info()[1].exc_info
636 exc_info
= sys
.exc_info()
637 raise DownloadError(message
, exc_info
)
638 self
._download
_retcode
= 1
640 def report_warning(self
, message
):
642 Print the message to stderr, it will be prefixed with 'WARNING:'
643 If stderr is a tty file the 'WARNING:' will be colored
645 if self
.params
.get('logger') is not None:
646 self
.params
['logger'].warning(message
)
648 if self
.params
.get('no_warnings'):
650 if not self
.params
.get('no_color') and self
._err
_file
.isatty() and compat_os_name
!= 'nt':
651 _msg_header
= '\033[0;33mWARNING:\033[0m'
653 _msg_header
= 'WARNING:'
654 warning_message
= '%s %s' % (_msg_header
, message
)
655 self
.to_stderr(warning_message
)
657 def report_error(self
, message
, tb
=None):
659 Do the same as trouble, but prefixes the message with 'ERROR:', colored
660 in red if stderr is a tty file.
662 if not self
.params
.get('no_color') and self
._err
_file
.isatty() and compat_os_name
!= 'nt':
663 _msg_header
= '\033[0;31mERROR:\033[0m'
665 _msg_header
= 'ERROR:'
666 error_message
= '%s %s' % (_msg_header
, message
)
667 self
.trouble(error_message
, tb
)
669 def report_file_already_downloaded(self
, file_name
):
670 """Report file has already been fully downloaded."""
672 self
.to_screen('[download] %s has already been downloaded' % file_name
)
673 except UnicodeEncodeError:
674 self
.to_screen('[download] The file has already been downloaded')
676 def prepare_filename(self
, info_dict
):
677 """Generate the output filename."""
679 template_dict
= dict(info_dict
)
681 template_dict
['epoch'] = int(time
.time())
682 autonumber_size
= self
.params
.get('autonumber_size')
683 if autonumber_size
is None:
685 template_dict
['autonumber'] = self
.params
.get('autonumber_start', 1) - 1 + self
._num
_downloads
686 if template_dict
.get('resolution') is None:
687 if template_dict
.get('width') and template_dict
.get('height'):
688 template_dict
['resolution'] = '%dx%d' % (template_dict
['width'], template_dict
['height'])
689 elif template_dict
.get('height'):
690 template_dict
['resolution'] = '%sp' % template_dict
['height']
691 elif template_dict
.get('width'):
692 template_dict
['resolution'] = '%dx?' % template_dict
['width']
694 sanitize
= lambda k
, v
: sanitize_filename(
696 restricted
=self
.params
.get('restrictfilenames'),
697 is_id
=(k
== 'id' or k
.endswith('_id')))
698 template_dict
= dict((k
, v
if isinstance(v
, compat_numeric_types
) else sanitize(k
, v
))
699 for k
, v
in template_dict
.items()
700 if v
is not None and not isinstance(v
, (list, tuple, dict)))
701 template_dict
= collections
.defaultdict(lambda: 'NA', template_dict
)
703 outtmpl
= self
.params
.get('outtmpl', DEFAULT_OUTTMPL
)
705 # For fields playlist_index and autonumber convert all occurrences
706 # of %(field)s to %(field)0Nd for backward compatibility
707 field_size_compat_map
= {
708 'playlist_index': len(str(template_dict
['n_entries'])),
709 'autonumber': autonumber_size
,
711 FIELD_SIZE_COMPAT_RE
= r
'(?<!%)%\((?P<field>autonumber|playlist_index)\)s'
712 mobj
= re
.search(FIELD_SIZE_COMPAT_RE
, outtmpl
)
715 FIELD_SIZE_COMPAT_RE
,
716 r
'%%(\1)0%dd' % field_size_compat_map
[mobj
.group('field')],
719 # Missing numeric fields used together with integer presentation types
720 # in format specification will break the argument substitution since
721 # string 'NA' is returned for missing fields. We will patch output
722 # template for missing fields to meet string presentation type.
723 for numeric_field
in self
._NUMERIC
_FIELDS
:
724 if numeric_field
not in template_dict
:
725 # As of [1] format syntax is:
726 # %[mapping_key][conversion_flags][minimum_width][.precision][length_modifier]type
727 # 1. https://docs.python.org/2/library/stdtypes.html#string-formatting
731 \({0}\) # mapping key
732 (?:[#0\-+ ]+)? # conversion flags (optional)
733 (?:\d+)? # minimum field width (optional)
734 (?:\.\d+)? # precision (optional)
735 [hlL]? # length modifier (optional)
736 [diouxXeEfFgGcrs%] # conversion type
739 FORMAT_RE
.format(numeric_field
),
740 r
'%({0})s'.format(numeric_field
), outtmpl
)
742 # expand_path translates '%%' into '%' and '$$' into '$'
743 # correspondingly that is not what we want since we need to keep
744 # '%%' intact for template dict substitution step. Working around
745 # with boundary-alike separator hack.
746 sep
= ''.join([random
.choice(ascii_letters
) for _
in range(32)])
747 outtmpl
= outtmpl
.replace('%%', '%{0}%'.format(sep
)).replace('$$', '${0}$'.format(sep
))
749 # outtmpl should be expand_path'ed before template dict substitution
750 # because meta fields may contain env variables we don't want to
751 # be expanded. For example, for outtmpl "%(title)s.%(ext)s" and
752 # title "Hello $PATH", we don't want `$PATH` to be expanded.
753 filename
= expand_path(outtmpl
).replace(sep
, '') % template_dict
755 # https://github.com/blackjack4494/youtube-dlc/issues/85
756 trim_file_name
= self
.params
.get('trim_file_name', False)
758 fn_groups
= filename
.rsplit('.')
761 if len(fn_groups
) > 2:
762 sub_ext
= fn_groups
[-2]
763 filename
= '.'.join(filter(None, [fn_groups
[0][:trim_file_name
], sub_ext
, ext
]))
765 # Temporary fix for #4787
766 # 'Treat' all problem characters by passing filename through preferredencoding
767 # to workaround encoding issues with subprocess on python2 @ Windows
768 if sys
.version_info
< (3, 0) and sys
.platform
== 'win32':
769 filename
= encodeFilename(filename
, True).decode(preferredencoding())
770 return sanitize_path(filename
)
771 except ValueError as err
:
772 self
.report_error('Error in output template: ' + str(err
) + ' (encoding: ' + repr(preferredencoding()) + ')')
775 def _match_entry(self
, info_dict
, incomplete
):
776 """ Returns None if the file should be downloaded """
778 video_title
= info_dict
.get('title', info_dict
.get('id', 'video'))
779 if 'title' in info_dict
:
780 # This can happen when we're just evaluating the playlist
781 title
= info_dict
['title']
782 matchtitle
= self
.params
.get('matchtitle', False)
784 if not re
.search(matchtitle
, title
, re
.IGNORECASE
):
785 return '"' + title
+ '" title did not match pattern "' + matchtitle
+ '"'
786 rejecttitle
= self
.params
.get('rejecttitle', False)
788 if re
.search(rejecttitle
, title
, re
.IGNORECASE
):
789 return '"' + title
+ '" title matched reject pattern "' + rejecttitle
+ '"'
790 date
= info_dict
.get('upload_date')
792 dateRange
= self
.params
.get('daterange', DateRange())
793 if date
not in dateRange
:
794 return '%s upload date is not in range %s' % (date_from_str(date
).isoformat(), dateRange
)
795 view_count
= info_dict
.get('view_count')
796 if view_count
is not None:
797 min_views
= self
.params
.get('min_views')
798 if min_views
is not None and view_count
< min_views
:
799 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title
, view_count
, min_views
)
800 max_views
= self
.params
.get('max_views')
801 if max_views
is not None and view_count
> max_views
:
802 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title
, view_count
, max_views
)
803 if age_restricted(info_dict
.get('age_limit'), self
.params
.get('age_limit')):
804 return 'Skipping "%s" because it is age restricted' % video_title
805 if self
.in_download_archive(info_dict
):
806 return '%s has already been recorded in archive' % video_title
809 match_filter
= self
.params
.get('match_filter')
810 if match_filter
is not None:
811 ret
= match_filter(info_dict
)
818 def add_extra_info(info_dict
, extra_info
):
819 '''Set the keys from extra_info in info dict if they are missing'''
820 for key
, value
in extra_info
.items():
821 info_dict
.setdefault(key
, value
)
823 def extract_info(self
, url
, download
=True, ie_key
=None, info_dict
=None, extra_info
={},
824 process
=True, force_generic_extractor
=False):
826 Returns a list with a dictionary for each video we find.
827 If 'download', also downloads the videos.
828 extra_info is a dict containing the extra values to add to each result
831 if not ie_key
and force_generic_extractor
:
835 ies
= [self
.get_info_extractor(ie_key
)]
840 if not ie
.suitable(url
):
844 ie
= self
.get_info_extractor(ie_key
)
846 self
.report_warning('The program functionality for this site has been marked as broken, '
847 'and will probably not work.')
850 temp_id
= ie
.extract_id(url
) if callable(getattr(ie
, 'extract_id', None)) else ie
._match
_id
(url
)
851 except (AssertionError, IndexError, AttributeError):
853 if temp_id
is not None and self
.in_download_archive({'id': temp_id, 'ie_key': ie_key}
):
854 self
.to_screen("[%s] %s: has already been recorded in archive" % (
858 return self
.__extract
_info
(url
, ie
, download
, extra_info
, process
, info_dict
)
861 self
.report_error('no suitable InfoExtractor for URL %s' % url
)
863 def __handle_extraction_exceptions(func
):
864 def wrapper(self
, *args
, **kwargs
):
866 return func(self
, *args
, **kwargs
)
867 except GeoRestrictedError
as e
:
870 msg
+= '\nThis video is available in %s.' % ', '.join(
871 map(ISO3166Utils
.short2full
, e
.countries
))
872 msg
+= '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
873 self
.report_error(msg
)
874 except ExtractorError
as e
: # An error we somewhat expected
875 self
.report_error(compat_str(e
), e
.format_traceback())
876 except MaxDownloadsReached
:
878 except Exception as e
:
879 if self
.params
.get('ignoreerrors', False):
880 self
.report_error(error_to_compat_str(e
), tb
=encode_compat_str(traceback
.format_exc()))
885 @__handle_extraction_exceptions
886 def __extract_info(self
, url
, ie
, download
, extra_info
, process
, info_dict
):
887 ie_result
= ie
.extract(url
)
888 if ie_result
is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
890 if isinstance(ie_result
, list):
891 # Backwards compatibility: old IE result format
893 '_type': 'compat_list',
894 'entries': ie_result
,
897 if info_dict
.get('id'):
898 ie_result
['id'] = info_dict
['id']
899 if info_dict
.get('title'):
900 ie_result
['title'] = info_dict
['title']
901 self
.add_default_extra_info(ie_result
, ie
, url
)
903 return self
.process_ie_result(ie_result
, download
, extra_info
)
907 def add_default_extra_info(self
, ie_result
, ie
, url
):
908 self
.add_extra_info(ie_result
, {
909 'extractor': ie
.IE_NAME
,
911 'webpage_url_basename': url_basename(url
),
912 'extractor_key': ie
.ie_key(),
915 def process_ie_result(self
, ie_result
, download
=True, extra_info
={}):
917 Take the result of the ie(may be modified) and resolve all unresolved
918 references (URLs, playlist items).
920 It will also download the videos if 'download'.
921 Returns the resolved ie_result.
923 result_type
= ie_result
.get('_type', 'video')
925 if result_type
in ('url', 'url_transparent'):
926 ie_result
['url'] = sanitize_url(ie_result
['url'])
927 extract_flat
= self
.params
.get('extract_flat', False)
928 if ((extract_flat
== 'in_playlist' and 'playlist' in extra_info
)
929 or extract_flat
is True):
930 self
.__forced
_printings
(
931 ie_result
, self
.prepare_filename(ie_result
),
935 if result_type
== 'video':
936 self
.add_extra_info(ie_result
, extra_info
)
937 return self
.process_video_result(ie_result
, download
=download
)
938 elif result_type
== 'url':
939 # We have to add extra_info to the results because it may be
940 # contained in a playlist
941 return self
.extract_info(ie_result
['url'],
942 download
, info_dict
=ie_result
,
943 ie_key
=ie_result
.get('ie_key'),
944 extra_info
=extra_info
)
945 elif result_type
== 'url_transparent':
946 # Use the information from the embedding page
947 info
= self
.extract_info(
948 ie_result
['url'], ie_key
=ie_result
.get('ie_key'),
949 extra_info
=extra_info
, download
=False, process
=False)
951 # extract_info may return None when ignoreerrors is enabled and
952 # extraction failed with an error, don't crash and return early
957 force_properties
= dict(
958 (k
, v
) for k
, v
in ie_result
.items() if v
is not None)
959 for f
in ('_type', 'url', 'id', 'extractor', 'extractor_key', 'ie_key'):
960 if f
in force_properties
:
961 del force_properties
[f
]
962 new_result
= info
.copy()
963 new_result
.update(force_properties
)
965 # Extracted info may not be a video result (i.e.
966 # info.get('_type', 'video') != video) but rather an url or
967 # url_transparent. In such cases outer metadata (from ie_result)
968 # should be propagated to inner one (info). For this to happen
969 # _type of info should be overridden with url_transparent. This
970 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
971 if new_result
.get('_type') == 'url':
972 new_result
['_type'] = 'url_transparent'
974 return self
.process_ie_result(
975 new_result
, download
=download
, extra_info
=extra_info
)
976 elif result_type
in ('playlist', 'multi_video'):
977 # We process each entry in the playlist
978 playlist
= ie_result
.get('title') or ie_result
.get('id')
979 self
.to_screen('[download] Downloading playlist: %s' % playlist
)
981 playlist_results
= []
983 playliststart
= self
.params
.get('playliststart', 1) - 1
984 playlistend
= self
.params
.get('playlistend')
985 # For backwards compatibility, interpret -1 as whole list
986 if playlistend
== -1:
989 playlistitems_str
= self
.params
.get('playlist_items')
991 if playlistitems_str
is not None:
992 def iter_playlistitems(format
):
993 for string_segment
in format
.split(','):
994 if '-' in string_segment
:
995 start
, end
= string_segment
.split('-')
996 for item
in range(int(start
), int(end
) + 1):
999 yield int(string_segment
)
1000 playlistitems
= orderedSet(iter_playlistitems(playlistitems_str
))
1002 ie_entries
= ie_result
['entries']
1004 def make_playlistitems_entries(list_ie_entries
):
1005 num_entries
= len(list_ie_entries
)
1007 list_ie_entries
[i
- 1] for i
in playlistitems
1008 if -num_entries
<= i
- 1 < num_entries
]
1010 def report_download(num_entries
):
1012 '[%s] playlist %s: Downloading %d videos' %
1013 (ie_result
['extractor'], playlist
, num_entries
))
1015 if isinstance(ie_entries
, list):
1016 n_all_entries
= len(ie_entries
)
1018 entries
= make_playlistitems_entries(ie_entries
)
1020 entries
= ie_entries
[playliststart
:playlistend
]
1021 n_entries
= len(entries
)
1023 '[%s] playlist %s: Collected %d video ids (downloading %d of them)' %
1024 (ie_result
['extractor'], playlist
, n_all_entries
, n_entries
))
1025 elif isinstance(ie_entries
, PagedList
):
1028 for item
in playlistitems
:
1029 entries
.extend(ie_entries
.getslice(
1033 entries
= ie_entries
.getslice(
1034 playliststart
, playlistend
)
1035 n_entries
= len(entries
)
1036 report_download(n_entries
)
1039 entries
= make_playlistitems_entries(list(itertools
.islice(
1040 ie_entries
, 0, max(playlistitems
))))
1042 entries
= list(itertools
.islice(
1043 ie_entries
, playliststart
, playlistend
))
1044 n_entries
= len(entries
)
1045 report_download(n_entries
)
1047 if self
.params
.get('playlistreverse', False):
1048 entries
= entries
[::-1]
1050 if self
.params
.get('playlistrandom', False):
1051 random
.shuffle(entries
)
1053 x_forwarded_for
= ie_result
.get('__x_forwarded_for_ip')
1055 for i
, entry
in enumerate(entries
, 1):
1056 self
.to_screen('[download] Downloading video %s of %s' % (i
, n_entries
))
1057 # This __x_forwarded_for_ip thing is a bit ugly but requires
1060 entry
['__x_forwarded_for_ip'] = x_forwarded_for
1062 'n_entries': n_entries
,
1063 'playlist': playlist
,
1064 'playlist_id': ie_result
.get('id'),
1065 'playlist_title': ie_result
.get('title'),
1066 'playlist_uploader': ie_result
.get('uploader'),
1067 'playlist_uploader_id': ie_result
.get('uploader_id'),
1068 'playlist_index': playlistitems
[i
- 1] if playlistitems
else i
+ playliststart
,
1069 'extractor': ie_result
['extractor'],
1070 'webpage_url': ie_result
['webpage_url'],
1071 'webpage_url_basename': url_basename(ie_result
['webpage_url']),
1072 'extractor_key': ie_result
['extractor_key'],
1075 reason
= self
._match
_entry
(entry
, incomplete
=True)
1076 if reason
is not None:
1077 if reason
.endswith('has already been recorded in the archive') and self
.params
.get('break_on_existing'):
1078 print('[download] tried downloading a file that\'s already in the archive, stopping since --break-on-existing is set.')
1081 self
.to_screen('[download] ' + reason
)
1084 entry_result
= self
.__process
_iterable
_entry
(entry
, download
, extra
)
1085 # TODO: skip failed (empty) entries?
1086 playlist_results
.append(entry_result
)
1087 ie_result
['entries'] = playlist_results
1088 self
.to_screen('[download] Finished downloading playlist: %s' % playlist
)
1090 elif result_type
== 'compat_list':
1091 self
.report_warning(
1092 'Extractor %s returned a compat_list result. '
1093 'It needs to be updated.' % ie_result
.get('extractor'))
1096 self
.add_extra_info(
1099 'extractor': ie_result
['extractor'],
1100 'webpage_url': ie_result
['webpage_url'],
1101 'webpage_url_basename': url_basename(ie_result
['webpage_url']),
1102 'extractor_key': ie_result
['extractor_key'],
1106 ie_result
['entries'] = [
1107 self
.process_ie_result(_fixup(r
), download
, extra_info
)
1108 for r
in ie_result
['entries']
1112 raise Exception('Invalid result type: %s' % result_type
)
1114 @__handle_extraction_exceptions
1115 def __process_iterable_entry(self
, entry
, download
, extra_info
):
1116 return self
.process_ie_result(
1117 entry
, download
=download
, extra_info
=extra_info
)
1119 def _build_format_filter(self
, filter_spec
):
1120 " Returns a function to filter the formats according to the filter_spec "
1130 operator_rex
= re
.compile(r
'''(?x)\s*
1131 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)
1132 \s*(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1133 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)
1135 ''' % '|'.join(map(re
.escape
, OPERATORS
.keys())))
1136 m
= operator_rex
.search(filter_spec
)
1139 comparison_value
= int(m
.group('value'))
1141 comparison_value
= parse_filesize(m
.group('value'))
1142 if comparison_value
is None:
1143 comparison_value
= parse_filesize(m
.group('value') + 'B')
1144 if comparison_value
is None:
1146 'Invalid value %r in format specification %r' % (
1147 m
.group('value'), filter_spec
))
1148 op
= OPERATORS
[m
.group('op')]
1153 '^=': lambda attr
, value
: attr
.startswith(value
),
1154 '$=': lambda attr
, value
: attr
.endswith(value
),
1155 '*=': lambda attr
, value
: value
in attr
,
1157 str_operator_rex
= re
.compile(r
'''(?x)
1158 \s*(?P<key>[a-zA-Z0-9._-]+)
1159 \s*(?P<negation>!\s*)?(?P<op>%s)(?P<none_inclusive>\s*\?)?
1160 \s*(?P<value>[a-zA-Z0-9._-]+)
1162 ''' % '|'.join(map(re
.escape
, STR_OPERATORS
.keys())))
1163 m
= str_operator_rex
.search(filter_spec
)
1165 comparison_value
= m
.group('value')
1166 str_op
= STR_OPERATORS
[m
.group('op')]
1167 if m
.group('negation'):
1168 op
= lambda attr
, value
: not str_op(attr
, value
)
1173 raise ValueError('Invalid filter specification %r' % filter_spec
)
1176 actual_value
= f
.get(m
.group('key'))
1177 if actual_value
is None:
1178 return m
.group('none_inclusive')
1179 return op(actual_value
, comparison_value
)
1182 def _default_format_spec(self
, info_dict
, download
=True):
1185 merger
= FFmpegMergerPP(self
)
1186 return merger
.available
and merger
.can_merge()
1189 if self
.params
.get('simulate', False):
1193 if self
.params
.get('outtmpl', DEFAULT_OUTTMPL
) == '-':
1195 if info_dict
.get('is_live'):
1201 req_format_list
= ['bestvideo+bestaudio', 'best']
1203 req_format_list
.reverse()
1204 return '/'.join(req_format_list
)
1206 def build_format_selector(self
, format_spec
):
1207 def syntax_error(note
, start
):
1209 'Invalid format specification: '
1210 '{0}\n\t{1}\n\t{2}^'.format(note
, format_spec
, ' ' * start
[1]))
1211 return SyntaxError(message
)
1213 PICKFIRST
= 'PICKFIRST'
1217 FormatSelector
= collections
.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
1219 allow_multiple_streams
= {'audio': self
.params
.get('allow_multiple_audio_streams', True),
1220 'video': self
.params
.get('allow_multiple_video_streams', True)}
1222 def _parse_filter(tokens
):
1224 for type, string
, start
, _
, _
in tokens
:
1225 if type == tokenize
.OP
and string
== ']':
1226 return ''.join(filter_parts
)
1228 filter_parts
.append(string
)
1230 def _remove_unused_ops(tokens
):
1231 # Remove operators that we don't use and join them with the surrounding strings
1232 # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
1233 ALLOWED_OPS
= ('/', '+', ',', '(', ')')
1234 last_string
, last_start
, last_end
, last_line
= None, None, None, None
1235 for type, string
, start
, end
, line
in tokens
:
1236 if type == tokenize
.OP
and string
== '[':
1238 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
1240 yield type, string
, start
, end
, line
1241 # everything inside brackets will be handled by _parse_filter
1242 for type, string
, start
, end
, line
in tokens
:
1243 yield type, string
, start
, end
, line
1244 if type == tokenize
.OP
and string
== ']':
1246 elif type == tokenize
.OP
and string
in ALLOWED_OPS
:
1248 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
1250 yield type, string
, start
, end
, line
1251 elif type in [tokenize
.NAME
, tokenize
.NUMBER
, tokenize
.OP
]:
1253 last_string
= string
1257 last_string
+= string
1259 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
1261 def _parse_format_selection(tokens
, inside_merge
=False, inside_choice
=False, inside_group
=False):
1263 current_selector
= None
1264 for type, string
, start
, _
, _
in tokens
:
1265 # ENCODING is only defined in python 3.x
1266 if type == getattr(tokenize
, 'ENCODING', None):
1268 elif type in [tokenize
.NAME
, tokenize
.NUMBER
]:
1269 current_selector
= FormatSelector(SINGLE
, string
, [])
1270 elif type == tokenize
.OP
:
1272 if not inside_group
:
1273 # ')' will be handled by the parentheses group
1274 tokens
.restore_last_token()
1276 elif inside_merge
and string
in ['/', ',']:
1277 tokens
.restore_last_token()
1279 elif inside_choice
and string
== ',':
1280 tokens
.restore_last_token()
1283 if not current_selector
:
1284 raise syntax_error('"," must follow a format selector', start
)
1285 selectors
.append(current_selector
)
1286 current_selector
= None
1288 if not current_selector
:
1289 raise syntax_error('"/" must follow a format selector', start
)
1290 first_choice
= current_selector
1291 second_choice
= _parse_format_selection(tokens
, inside_choice
=True)
1292 current_selector
= FormatSelector(PICKFIRST
, (first_choice
, second_choice
), [])
1294 if not current_selector
:
1295 current_selector
= FormatSelector(SINGLE
, 'best', [])
1296 format_filter
= _parse_filter(tokens
)
1297 current_selector
.filters
.append(format_filter
)
1299 if current_selector
:
1300 raise syntax_error('Unexpected "("', start
)
1301 group
= _parse_format_selection(tokens
, inside_group
=True)
1302 current_selector
= FormatSelector(GROUP
, group
, [])
1304 if not current_selector
:
1305 raise syntax_error('Unexpected "+"', start
)
1306 selector_1
= current_selector
1307 selector_2
= _parse_format_selection(tokens
, inside_merge
=True)
1309 raise syntax_error('Expected a selector', start
)
1310 current_selector
= FormatSelector(MERGE
, (selector_1
, selector_2
), [])
1312 raise syntax_error('Operator not recognized: "{0}"'.format(string
), start
)
1313 elif type == tokenize
.ENDMARKER
:
1315 if current_selector
:
1316 selectors
.append(current_selector
)
1319 def _build_selector_function(selector
):
1320 if isinstance(selector
, list): # ,
1321 fs
= [_build_selector_function(s
) for s
in selector
]
1323 def selector_function(ctx
):
1325 for format
in f(ctx
):
1327 return selector_function
1329 elif selector
.type == GROUP
: # ()
1330 selector_function
= _build_selector_function(selector
.selector
)
1332 elif selector
.type == PICKFIRST
: # /
1333 fs
= [_build_selector_function(s
) for s
in selector
.selector
]
1335 def selector_function(ctx
):
1337 picked_formats
= list(f(ctx
))
1339 return picked_formats
1342 elif selector
.type == SINGLE
: # atom
1343 format_spec
= selector
.selector
if selector
.selector
is not None else 'best'
1345 if format_spec
== 'all':
1346 def selector_function(ctx
):
1347 formats
= list(ctx
['formats'])
1353 format_fallback
= False
1354 format_spec_obj
= re
.match(r
'(best|worst|b|w)(video|audio|v|a)?(\*)?$', format_spec
)
1355 if format_spec_obj
is not None:
1356 format_idx
= 0 if format_spec_obj
.group(1)[0] == 'w' else -1
1357 format_type
= format_spec_obj
.group(2)[0] if format_spec_obj
.group(2) else False
1358 not_format_type
= 'v' if format_type
== 'a' else 'a'
1359 format_modified
= format_spec_obj
.group(3) is not None
1361 format_fallback
= not format_type
and not format_modified
# for b, w
1362 filter_f
= ((lambda f
: f
.get(format_type
+ 'codec') != 'none')
1363 if format_type
and format_modified
# bv*, ba*, wv*, wa*
1364 else (lambda f
: f
.get(not_format_type
+ 'codec') == 'none')
1365 if format_type
# bv, ba, wv, wa
1366 else (lambda f
: f
.get('vcodec') != 'none' and f
.get('acodec') != 'none')
1367 if not format_modified
# b, w
1371 filter_f
= ((lambda f
: f
.get('ext') == format_spec
)
1372 if format_spec
in ['mp4', 'flv', 'webm', '3gp', 'm4a', 'mp3', 'ogg', 'aac', 'wav'] # extension
1373 else (lambda f
: f
.get('format_id') == format_spec
)) # id
1375 def selector_function(ctx
):
1376 formats
= list(ctx
['formats'])
1379 matches
= list(filter(filter_f
, formats
)) if filter_f
is not None else formats
1381 yield matches
[format_idx
]
1382 elif format_fallback
== 'force' or (format_fallback
and ctx
['incomplete_formats']):
1383 # for extractors with incomplete formats (audio only (soundcloud)
1384 # or video only (imgur)) best/worst will fallback to
1385 # best/worst {video,audio}-only format
1386 yield formats
[format_idx
]
1388 elif selector
.type == MERGE
: # +
1389 def _merge(formats_pair
):
1390 format_1
, format_2
= formats_pair
1393 formats_info
.extend(format_1
.get('requested_formats', (format_1
,)))
1394 formats_info
.extend(format_2
.get('requested_formats', (format_2
,)))
1396 if not allow_multiple_streams
['video'] or not allow_multiple_streams
['audio']:
1397 get_no_more
= {"video": False, "audio": False}
1398 for (i
, fmt_info
) in enumerate(formats_info
):
1399 for aud_vid
in ["audio", "video"]:
1400 if not allow_multiple_streams
[aud_vid
] and fmt_info
.get(aud_vid
[0] + 'codec') != 'none':
1401 if get_no_more
[aud_vid
]:
1403 get_no_more
[aud_vid
] = True
1405 if len(formats_info
) == 1:
1406 return formats_info
[0]
1408 video_fmts
= [fmt_info
for fmt_info
in formats_info
if fmt_info
.get('vcodec') != 'none']
1409 audio_fmts
= [fmt_info
for fmt_info
in formats_info
if fmt_info
.get('acodec') != 'none']
1411 the_only_video
= video_fmts
[0] if len(video_fmts
) == 1 else None
1412 the_only_audio
= audio_fmts
[0] if len(audio_fmts
) == 1 else None
1414 output_ext
= self
.params
.get('merge_output_format')
1417 output_ext
= the_only_video
['ext']
1418 elif the_only_audio
and not video_fmts
:
1419 output_ext
= the_only_audio
['ext']
1424 'requested_formats': formats_info
,
1425 'format': '+'.join(fmt_info
.get('format') for fmt_info
in formats_info
),
1426 'format_id': '+'.join(fmt_info
.get('format_id') for fmt_info
in formats_info
),
1432 'width': the_only_video
.get('width'),
1433 'height': the_only_video
.get('height'),
1434 'resolution': the_only_video
.get('resolution'),
1435 'fps': the_only_video
.get('fps'),
1436 'vcodec': the_only_video
.get('vcodec'),
1437 'vbr': the_only_video
.get('vbr'),
1438 'stretched_ratio': the_only_video
.get('stretched_ratio'),
1443 'acodec': the_only_audio
.get('acodec'),
1444 'abr': the_only_audio
.get('abr'),
1449 selector_1
, selector_2
= map(_build_selector_function
, selector
.selector
)
1451 def selector_function(ctx
):
1452 for pair
in itertools
.product(
1453 selector_1(copy
.deepcopy(ctx
)), selector_2(copy
.deepcopy(ctx
))):
1456 filters
= [self
._build
_format
_filter
(f
) for f
in selector
.filters
]
1458 def final_selector(ctx
):
1459 ctx_copy
= copy
.deepcopy(ctx
)
1460 for _filter
in filters
:
1461 ctx_copy
['formats'] = list(filter(_filter
, ctx_copy
['formats']))
1462 return selector_function(ctx_copy
)
1463 return final_selector
1465 stream
= io
.BytesIO(format_spec
.encode('utf-8'))
1467 tokens
= list(_remove_unused_ops(compat_tokenize_tokenize(stream
.readline
)))
1468 except tokenize
.TokenError
:
1469 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec
)))
1471 class TokenIterator(object):
1472 def __init__(self
, tokens
):
1473 self
.tokens
= tokens
1480 if self
.counter
>= len(self
.tokens
):
1481 raise StopIteration()
1482 value
= self
.tokens
[self
.counter
]
1488 def restore_last_token(self
):
1491 parsed_selector
= _parse_format_selection(iter(TokenIterator(tokens
)))
1492 return _build_selector_function(parsed_selector
)
1494 def _calc_headers(self
, info_dict
):
1495 res
= std_headers
.copy()
1497 add_headers
= info_dict
.get('http_headers')
1499 res
.update(add_headers
)
1501 cookies
= self
._calc
_cookies
(info_dict
)
1503 res
['Cookie'] = cookies
1505 if 'X-Forwarded-For' not in res
:
1506 x_forwarded_for_ip
= info_dict
.get('__x_forwarded_for_ip')
1507 if x_forwarded_for_ip
:
1508 res
['X-Forwarded-For'] = x_forwarded_for_ip
1512 def _calc_cookies(self
, info_dict
):
1513 pr
= sanitized_Request(info_dict
['url'])
1514 self
.cookiejar
.add_cookie_header(pr
)
1515 return pr
.get_header('Cookie')
1517 def process_video_result(self
, info_dict
, download
=True):
1518 assert info_dict
.get('_type', 'video') == 'video'
1520 if 'id' not in info_dict
:
1521 raise ExtractorError('Missing "id" field in extractor result')
1522 if 'title' not in info_dict
:
1523 raise ExtractorError('Missing "title" field in extractor result')
1525 def report_force_conversion(field
, field_not
, conversion
):
1526 self
.report_warning(
1527 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
1528 % (field
, field_not
, conversion
))
1530 def sanitize_string_field(info
, string_field
):
1531 field
= info
.get(string_field
)
1532 if field
is None or isinstance(field
, compat_str
):
1534 report_force_conversion(string_field
, 'a string', 'string')
1535 info
[string_field
] = compat_str(field
)
1537 def sanitize_numeric_fields(info
):
1538 for numeric_field
in self
._NUMERIC
_FIELDS
:
1539 field
= info
.get(numeric_field
)
1540 if field
is None or isinstance(field
, compat_numeric_types
):
1542 report_force_conversion(numeric_field
, 'numeric', 'int')
1543 info
[numeric_field
] = int_or_none(field
)
1545 sanitize_string_field(info_dict
, 'id')
1546 sanitize_numeric_fields(info_dict
)
1548 if 'playlist' not in info_dict
:
1549 # It isn't part of a playlist
1550 info_dict
['playlist'] = None
1551 info_dict
['playlist_index'] = None
1553 thumbnails
= info_dict
.get('thumbnails')
1554 if thumbnails
is None:
1555 thumbnail
= info_dict
.get('thumbnail')
1557 info_dict
['thumbnails'] = thumbnails
= [{'url': thumbnail}
]
1559 thumbnails
.sort(key
=lambda t
: (
1560 t
.get('preference') if t
.get('preference') is not None else -1,
1561 t
.get('width') if t
.get('width') is not None else -1,
1562 t
.get('height') if t
.get('height') is not None else -1,
1563 t
.get('id') if t
.get('id') is not None else '', t
.get('url')))
1564 for i
, t
in enumerate(thumbnails
):
1565 t
['url'] = sanitize_url(t
['url'])
1566 if t
.get('width') and t
.get('height'):
1567 t
['resolution'] = '%dx%d' % (t
['width'], t
['height'])
1568 if t
.get('id') is None:
1571 if self
.params
.get('list_thumbnails'):
1572 self
.list_thumbnails(info_dict
)
1575 thumbnail
= info_dict
.get('thumbnail')
1577 info_dict
['thumbnail'] = sanitize_url(thumbnail
)
1579 info_dict
['thumbnail'] = thumbnails
[-1]['url']
1581 if 'display_id' not in info_dict
and 'id' in info_dict
:
1582 info_dict
['display_id'] = info_dict
['id']
1584 if info_dict
.get('upload_date') is None and info_dict
.get('timestamp') is not None:
1585 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
1586 # see http://bugs.python.org/issue1646728)
1588 upload_date
= datetime
.datetime
.utcfromtimestamp(info_dict
['timestamp'])
1589 info_dict
['upload_date'] = upload_date
.strftime('%Y%m%d')
1590 except (ValueError, OverflowError, OSError):
1593 # Auto generate title fields corresponding to the *_number fields when missing
1594 # in order to always have clean titles. This is very common for TV series.
1595 for field
in ('chapter', 'season', 'episode'):
1596 if info_dict
.get('%s_number' % field
) is not None and not info_dict
.get(field
):
1597 info_dict
[field
] = '%s %d' % (field
.capitalize(), info_dict
['%s_number' % field
])
1599 for cc_kind
in ('subtitles', 'automatic_captions'):
1600 cc
= info_dict
.get(cc_kind
)
1602 for _
, subtitle
in cc
.items():
1603 for subtitle_format
in subtitle
:
1604 if subtitle_format
.get('url'):
1605 subtitle_format
['url'] = sanitize_url(subtitle_format
['url'])
1606 if subtitle_format
.get('ext') is None:
1607 subtitle_format
['ext'] = determine_ext(subtitle_format
['url']).lower()
1609 automatic_captions
= info_dict
.get('automatic_captions')
1610 subtitles
= info_dict
.get('subtitles')
1612 if self
.params
.get('listsubtitles', False):
1613 if 'automatic_captions' in info_dict
:
1614 self
.list_subtitles(
1615 info_dict
['id'], automatic_captions
, 'automatic captions')
1616 self
.list_subtitles(info_dict
['id'], subtitles
, 'subtitles')
1619 info_dict
['requested_subtitles'] = self
.process_subtitles(
1620 info_dict
['id'], subtitles
, automatic_captions
)
1622 # We now pick which formats have to be downloaded
1623 if info_dict
.get('formats') is None:
1624 # There's only one format available
1625 formats
= [info_dict
]
1627 formats
= info_dict
['formats']
1630 raise ExtractorError('No video formats found!')
1632 def is_wellformed(f
):
1635 self
.report_warning(
1636 '"url" field is missing or empty - skipping format, '
1637 'there is an error in extractor')
1639 if isinstance(url
, bytes):
1640 sanitize_string_field(f
, 'url')
1643 # Filter out malformed formats for better extraction robustness
1644 formats
= list(filter(is_wellformed
, formats
))
1648 # We check that all the formats have the format and format_id fields
1649 for i
, format
in enumerate(formats
):
1650 sanitize_string_field(format
, 'format_id')
1651 sanitize_numeric_fields(format
)
1652 format
['url'] = sanitize_url(format
['url'])
1653 if not format
.get('format_id'):
1654 format
['format_id'] = compat_str(i
)
1656 # Sanitize format_id from characters used in format selector expression
1657 format
['format_id'] = re
.sub(r
'[\s,/+\[\]()]', '_', format
['format_id'])
1658 format_id
= format
['format_id']
1659 if format_id
not in formats_dict
:
1660 formats_dict
[format_id
] = []
1661 formats_dict
[format_id
].append(format
)
1663 # Make sure all formats have unique format_id
1664 for format_id
, ambiguous_formats
in formats_dict
.items():
1665 if len(ambiguous_formats
) > 1:
1666 for i
, format
in enumerate(ambiguous_formats
):
1667 format
['format_id'] = '%s-%d' % (format_id
, i
)
1669 for i
, format
in enumerate(formats
):
1670 if format
.get('format') is None:
1671 format
['format'] = '{id} - {res}{note}'.format(
1672 id=format
['format_id'],
1673 res
=self
.format_resolution(format
),
1674 note
=' ({0})'.format(format
['format_note']) if format
.get('format_note') is not None else '',
1676 # Automatically determine file extension if missing
1677 if format
.get('ext') is None:
1678 format
['ext'] = determine_ext(format
['url']).lower()
1679 # Automatically determine protocol if missing (useful for format
1680 # selection purposes)
1681 if format
.get('protocol') is None:
1682 format
['protocol'] = determine_protocol(format
)
1683 # Add HTTP headers, so that external programs can use them from the
1685 full_format_info
= info_dict
.copy()
1686 full_format_info
.update(format
)
1687 format
['http_headers'] = self
._calc
_headers
(full_format_info
)
1688 # Remove private housekeeping stuff
1689 if '__x_forwarded_for_ip' in info_dict
:
1690 del info_dict
['__x_forwarded_for_ip']
1692 # TODO Central sorting goes here
1694 if formats
[0] is not info_dict
:
1695 # only set the 'formats' fields if the original info_dict list them
1696 # otherwise we end up with a circular reference, the first (and unique)
1697 # element in the 'formats' field in info_dict is info_dict itself,
1698 # which can't be exported to json
1699 info_dict
['formats'] = formats
1700 if self
.params
.get('listformats'):
1701 self
.list_formats(info_dict
)
1704 req_format
= self
.params
.get('format')
1705 if req_format
is None:
1706 req_format
= self
._default
_format
_spec
(info_dict
, download
=download
)
1707 if self
.params
.get('verbose'):
1708 self
._write
_string
('[debug] Default format spec: %s\n' % req_format
)
1710 format_selector
= self
.build_format_selector(req_format
)
1712 # While in format selection we may need to have an access to the original
1713 # format set in order to calculate some metrics or do some processing.
1714 # For now we need to be able to guess whether original formats provided
1715 # by extractor are incomplete or not (i.e. whether extractor provides only
1716 # video-only or audio-only formats) for proper formats selection for
1717 # extractors with such incomplete formats (see
1718 # https://github.com/ytdl-org/youtube-dl/pull/5556).
1719 # Since formats may be filtered during format selection and may not match
1720 # the original formats the results may be incorrect. Thus original formats
1721 # or pre-calculated metrics should be passed to format selection routines
1723 # We will pass a context object containing all necessary additional data
1724 # instead of just formats.
1725 # This fixes incorrect format selection issue (see
1726 # https://github.com/ytdl-org/youtube-dl/issues/10083).
1727 incomplete_formats
= (
1728 # All formats are video-only or
1729 all(f
.get('vcodec') != 'none' and f
.get('acodec') == 'none' for f
in formats
)
1730 # all formats are audio-only
1731 or all(f
.get('vcodec') == 'none' and f
.get('acodec') != 'none' for f
in formats
))
1735 'incomplete_formats': incomplete_formats
,
1738 formats_to_download
= list(format_selector(ctx
))
1739 if not formats_to_download
:
1740 raise ExtractorError('requested format not available',
1744 self
.to_screen('[info] Downloading format(s) %s' % ", ".join([f
['format_id'] for f
in formats_to_download
]))
1745 if len(formats_to_download
) > 1:
1746 self
.to_screen('[info] %s: downloading video in %s formats' % (info_dict
['id'], len(formats_to_download
)))
1747 for format
in formats_to_download
:
1748 new_info
= dict(info_dict
)
1749 new_info
.update(format
)
1750 self
.process_info(new_info
)
1751 # We update the info dict with the best quality format (backwards compatibility)
1752 info_dict
.update(formats_to_download
[-1])
1755 def process_subtitles(self
, video_id
, normal_subtitles
, automatic_captions
):
1756 """Select the requested subtitles and their format"""
1758 if normal_subtitles
and self
.params
.get('writesubtitles'):
1759 available_subs
.update(normal_subtitles
)
1760 if automatic_captions
and self
.params
.get('writeautomaticsub'):
1761 for lang
, cap_info
in automatic_captions
.items():
1762 if lang
not in available_subs
:
1763 available_subs
[lang
] = cap_info
1765 if (not self
.params
.get('writesubtitles') and not
1766 self
.params
.get('writeautomaticsub') or not
1770 if self
.params
.get('allsubtitles', False):
1771 requested_langs
= available_subs
.keys()
1773 if self
.params
.get('subtitleslangs', False):
1774 requested_langs
= self
.params
.get('subtitleslangs')
1775 elif 'en' in available_subs
:
1776 requested_langs
= ['en']
1778 requested_langs
= [list(available_subs
.keys())[0]]
1780 formats_query
= self
.params
.get('subtitlesformat', 'best')
1781 formats_preference
= formats_query
.split('/') if formats_query
else []
1783 for lang
in requested_langs
:
1784 formats
= available_subs
.get(lang
)
1786 self
.report_warning('%s subtitles not available for %s' % (lang
, video_id
))
1788 for ext
in formats_preference
:
1792 matches
= list(filter(lambda f
: f
['ext'] == ext
, formats
))
1798 self
.report_warning(
1799 'No subtitle format found matching "%s" for language %s, '
1800 'using %s' % (formats_query
, lang
, f
['ext']))
1804 def __forced_printings(self
, info_dict
, filename
, incomplete
):
1805 def print_mandatory(field
):
1806 if (self
.params
.get('force%s' % field
, False)
1807 and (not incomplete
or info_dict
.get(field
) is not None)):
1808 self
.to_stdout(info_dict
[field
])
1810 def print_optional(field
):
1811 if (self
.params
.get('force%s' % field
, False)
1812 and info_dict
.get(field
) is not None):
1813 self
.to_stdout(info_dict
[field
])
1815 print_mandatory('title')
1816 print_mandatory('id')
1817 if self
.params
.get('forceurl', False) and not incomplete
:
1818 if info_dict
.get('requested_formats') is not None:
1819 for f
in info_dict
['requested_formats']:
1820 self
.to_stdout(f
['url'] + f
.get('play_path', ''))
1822 # For RTMP URLs, also include the playpath
1823 self
.to_stdout(info_dict
['url'] + info_dict
.get('play_path', ''))
1824 print_optional('thumbnail')
1825 print_optional('description')
1826 if self
.params
.get('forcefilename', False) and filename
is not None:
1827 self
.to_stdout(filename
)
1828 if self
.params
.get('forceduration', False) and info_dict
.get('duration') is not None:
1829 self
.to_stdout(formatSeconds(info_dict
['duration']))
1830 print_mandatory('format')
1831 if self
.params
.get('forcejson', False):
1832 self
.to_stdout(json
.dumps(info_dict
))
1834 def process_info(self
, info_dict
):
1835 """Process a single resolved IE result."""
1837 assert info_dict
.get('_type', 'video') == 'video'
1839 max_downloads
= self
.params
.get('max_downloads')
1840 if max_downloads
is not None:
1841 if self
._num
_downloads
>= int(max_downloads
):
1842 raise MaxDownloadsReached()
1844 # TODO: backward compatibility, to be removed
1845 info_dict
['fulltitle'] = info_dict
['title']
1847 if 'format' not in info_dict
:
1848 info_dict
['format'] = info_dict
['ext']
1850 reason
= self
._match
_entry
(info_dict
, incomplete
=False)
1851 if reason
is not None:
1852 self
.to_screen('[download] ' + reason
)
1855 self
._num
_downloads
+= 1
1857 info_dict
['_filename'] = filename
= self
.prepare_filename(info_dict
)
1860 self
.__forced
_printings
(info_dict
, filename
, incomplete
=False)
1862 if self
.params
.get('simulate', False):
1863 if self
.params
.get('force_write_download_archive', False):
1864 self
.record_download_archive(info_dict
)
1866 # Do nothing else if in simulate mode
1869 if filename
is None:
1872 def ensure_dir_exists(path
):
1874 dn
= os
.path
.dirname(path
)
1875 if dn
and not os
.path
.exists(dn
):
1878 except (OSError, IOError) as err
:
1879 self
.report_error('unable to create directory ' + error_to_compat_str(err
))
1882 if not ensure_dir_exists(sanitize_path(encodeFilename(filename
))):
1885 if self
.params
.get('writedescription', False):
1886 descfn
= replace_extension(filename
, 'description', info_dict
.get('ext'))
1887 if self
.params
.get('nooverwrites', False) and os
.path
.exists(encodeFilename(descfn
)):
1888 self
.to_screen('[info] Video description is already present')
1889 elif info_dict
.get('description') is None:
1890 self
.report_warning('There\'s no description to write.')
1893 self
.to_screen('[info] Writing video description to: ' + descfn
)
1894 with io
.open(encodeFilename(descfn
), 'w', encoding
='utf-8') as descfile
:
1895 descfile
.write(info_dict
['description'])
1896 except (OSError, IOError):
1897 self
.report_error('Cannot write description file ' + descfn
)
1900 if self
.params
.get('writeannotations', False):
1901 annofn
= replace_extension(filename
, 'annotations.xml', info_dict
.get('ext'))
1902 if self
.params
.get('nooverwrites', False) and os
.path
.exists(encodeFilename(annofn
)):
1903 self
.to_screen('[info] Video annotations are already present')
1904 elif not info_dict
.get('annotations'):
1905 self
.report_warning('There are no annotations to write.')
1908 self
.to_screen('[info] Writing video annotations to: ' + annofn
)
1909 with io
.open(encodeFilename(annofn
), 'w', encoding
='utf-8') as annofile
:
1910 annofile
.write(info_dict
['annotations'])
1911 except (KeyError, TypeError):
1912 self
.report_warning('There are no annotations to write.')
1913 except (OSError, IOError):
1914 self
.report_error('Cannot write annotations file: ' + annofn
)
1917 def dl(name
, info
, subtitle
=False):
1918 fd
= get_suitable_downloader(info
, self
.params
)(self
, self
.params
)
1919 for ph
in self
._progress
_hooks
:
1920 fd
.add_progress_hook(ph
)
1921 if self
.params
.get('verbose'):
1922 self
.to_screen('[debug] Invoking downloader on %r' % info
.get('url'))
1923 return fd
.download(name
, info
, subtitle
)
1925 subtitles_are_requested
= any([self
.params
.get('writesubtitles', False),
1926 self
.params
.get('writeautomaticsub')])
1928 if subtitles_are_requested
and info_dict
.get('requested_subtitles'):
1929 # subtitles download errors are already managed as troubles in relevant IE
1930 # that way it will silently go on when used with unsupporting IE
1931 subtitles
= info_dict
['requested_subtitles']
1932 # ie = self.get_info_extractor(info_dict['extractor_key'])
1933 for sub_lang
, sub_info
in subtitles
.items():
1934 sub_format
= sub_info
['ext']
1935 sub_filename
= subtitles_filename(filename
, sub_lang
, sub_format
, info_dict
.get('ext'))
1936 if self
.params
.get('nooverwrites', False) and os
.path
.exists(encodeFilename(sub_filename
)):
1937 self
.to_screen('[info] Video subtitle %s.%s is already present' % (sub_lang
, sub_format
))
1939 self
.to_screen('[info] Writing video subtitles to: ' + sub_filename
)
1940 if sub_info
.get('data') is not None:
1942 # Use newline='' to prevent conversion of newline characters
1943 # See https://github.com/ytdl-org/youtube-dl/issues/10268
1944 with io
.open(encodeFilename(sub_filename
), 'w', encoding
='utf-8', newline
='') as subfile
:
1945 subfile
.write(sub_info
['data'])
1946 except (OSError, IOError):
1947 self
.report_error('Cannot write subtitles file ' + sub_filename
)
1951 dl(sub_filename
, sub_info
, subtitle
=True)
1953 if self.params.get('sleep_interval_subtitles', False):
1954 dl(sub_filename, sub_info)
1956 sub_data = ie._request_webpage(
1957 sub_info['url'], info_dict['id'], note=False).read()
1958 with io.open(encodeFilename(sub_filename), 'wb') as subfile:
1959 subfile.write(sub_data)
1961 except (ExtractorError
, IOError, OSError, ValueError, compat_urllib_error
.URLError
, compat_http_client
.HTTPException
, socket
.error
) as err
:
1962 self
.report_warning('Unable to download subtitle for "%s": %s' %
1963 (sub_lang
, error_to_compat_str(err
)))
1966 if self
.params
.get('skip_download', False):
1967 if self
.params
.get('convertsubtitles', False):
1968 subconv
= FFmpegSubtitlesConvertorPP(self
, format
=self
.params
.get('convertsubtitles'))
1969 filename_real_ext
= os
.path
.splitext(filename
)[1][1:]
1971 os
.path
.splitext(filename
)[0]
1972 if filename_real_ext
== info_dict
['ext']
1974 afilename
= '%s.%s' % (filename_wo_ext
, self
.params
.get('convertsubtitles'))
1975 if subconv
.available
:
1976 info_dict
.setdefault('__postprocessors', [])
1977 # info_dict['__postprocessors'].append(subconv)
1978 if os
.path
.exists(encodeFilename(afilename
)):
1980 '[download] %s has already been downloaded and '
1981 'converted' % afilename
)
1984 self
.post_process(filename
, info_dict
)
1985 except (PostProcessingError
) as err
:
1986 self
.report_error('postprocessing: %s' % str(err
))
1989 if self
.params
.get('writeinfojson', False):
1990 infofn
= replace_extension(filename
, 'info.json', info_dict
.get('ext'))
1991 if self
.params
.get('nooverwrites', False) and os
.path
.exists(encodeFilename(infofn
)):
1992 self
.to_screen('[info] Video description metadata is already present')
1994 self
.to_screen('[info] Writing video description metadata as JSON to: ' + infofn
)
1996 write_json_file(self
.filter_requested_info(info_dict
), infofn
)
1997 except (OSError, IOError):
1998 self
.report_error('Cannot write metadata to JSON file ' + infofn
)
2001 self
._write
_thumbnails
(info_dict
, filename
)
2003 # Write internet shortcut files
2004 url_link
= webloc_link
= desktop_link
= False
2005 if self
.params
.get('writelink', False):
2006 if sys
.platform
== "darwin": # macOS.
2008 elif sys
.platform
.startswith("linux"):
2010 else: # if sys.platform in ['win32', 'cygwin']:
2012 if self
.params
.get('writeurllink', False):
2014 if self
.params
.get('writewebloclink', False):
2016 if self
.params
.get('writedesktoplink', False):
2019 if url_link
or webloc_link
or desktop_link
:
2020 if 'webpage_url' not in info_dict
:
2021 self
.report_error('Cannot write internet shortcut file because the "webpage_url" field is missing in the media information')
2023 ascii_url
= iri_to_uri(info_dict
['webpage_url'])
2025 def _write_link_file(extension
, template
, newline
, embed_filename
):
2026 linkfn
= replace_extension(filename
, extension
, info_dict
.get('ext'))
2027 if self
.params
.get('nooverwrites', False) and os
.path
.exists(encodeFilename(linkfn
)):
2028 self
.to_screen('[info] Internet shortcut is already present')
2031 self
.to_screen('[info] Writing internet shortcut to: ' + linkfn
)
2032 with io
.open(encodeFilename(to_high_limit_path(linkfn
)), 'w', encoding
='utf-8', newline
=newline
) as linkfile
:
2033 template_vars
= {'url': ascii_url}
2035 template_vars
['filename'] = linkfn
[:-(len(extension
) + 1)]
2036 linkfile
.write(template
% template_vars
)
2037 except (OSError, IOError):
2038 self
.report_error('Cannot write internet shortcut ' + linkfn
)
2043 if not _write_link_file('url', DOT_URL_LINK_TEMPLATE
, '\r\n', embed_filename
=False):
2046 if not _write_link_file('webloc', DOT_WEBLOC_LINK_TEMPLATE
, '\n', embed_filename
=False):
2049 if not _write_link_file('desktop', DOT_DESKTOP_LINK_TEMPLATE
, '\n', embed_filename
=True):
2053 must_record_download_archive
= False
2054 if not self
.params
.get('skip_download', False):
2056 if info_dict
.get('requested_formats') is not None:
2059 merger
= FFmpegMergerPP(self
)
2060 if not merger
.available
:
2062 self
.report_warning('You have requested multiple '
2063 'formats but ffmpeg or avconv are not installed.'
2064 ' The formats won\'t be merged.')
2066 postprocessors
= [merger
]
2068 def compatible_formats(formats
):
2069 # TODO: some formats actually allow this (mkv, webm, ogg, mp4), but not all of them.
2070 video_formats
= [format
for format
in formats
if format
.get('vcodec') != 'none']
2071 audio_formats
= [format
for format
in formats
if format
.get('acodec') != 'none']
2072 if len(video_formats
) > 2 or len(audio_formats
) > 2:
2076 exts
= set(format
.get('ext') for format
in formats
)
2078 set(('mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma')),
2081 for ext_sets
in COMPATIBLE_EXTS
:
2082 if ext_sets
.issuperset(exts
):
2084 # TODO: Check acodec/vcodec
2087 filename_real_ext
= os
.path
.splitext(filename
)[1][1:]
2089 os
.path
.splitext(filename
)[0]
2090 if filename_real_ext
== info_dict
['ext']
2092 requested_formats
= info_dict
['requested_formats']
2093 if self
.params
.get('merge_output_format') is None and not compatible_formats(requested_formats
):
2094 info_dict
['ext'] = 'mkv'
2095 self
.report_warning(
2096 'Requested formats are incompatible for merge and will be merged into mkv.')
2097 # Ensure filename always has a correct extension for successful merge
2098 filename
= '%s.%s' % (filename_wo_ext
, info_dict
['ext'])
2099 if os
.path
.exists(encodeFilename(filename
)):
2101 '[download] %s has already been downloaded and '
2102 'merged' % filename
)
2104 for f
in requested_formats
:
2105 new_info
= dict(info_dict
)
2107 fname
= prepend_extension(
2108 self
.prepare_filename(new_info
),
2109 'f%s' % f
['format_id'], new_info
['ext'])
2110 if not ensure_dir_exists(fname
):
2112 downloaded
.append(fname
)
2113 partial_success
, real_download
= dl(fname
, new_info
)
2114 success
= success
and partial_success
2115 info_dict
['__postprocessors'] = postprocessors
2116 info_dict
['__files_to_merge'] = downloaded
2117 # Even if there were no downloads, it is being merged only now
2118 info_dict
['__real_download'] = True
2120 # Just a single file
2121 success
, real_download
= dl(filename
, info_dict
)
2122 info_dict
['__real_download'] = real_download
2123 except (compat_urllib_error
.URLError
, compat_http_client
.HTTPException
, socket
.error
) as err
:
2124 self
.report_error('unable to download video data: %s' % error_to_compat_str(err
))
2126 except (OSError, IOError) as err
:
2127 raise UnavailableVideoError(err
)
2128 except (ContentTooShortError
, ) as err
:
2129 self
.report_error('content too short (expected %s bytes and served %s)' % (err
.expected
, err
.downloaded
))
2132 if success
and filename
!= '-':
2134 fixup_policy
= self
.params
.get('fixup')
2135 if fixup_policy
is None:
2136 fixup_policy
= 'detect_or_warn'
2138 INSTALL_FFMPEG_MESSAGE
= 'Install ffmpeg or avconv to fix this automatically.'
2140 stretched_ratio
= info_dict
.get('stretched_ratio')
2141 if stretched_ratio
is not None and stretched_ratio
!= 1:
2142 if fixup_policy
== 'warn':
2143 self
.report_warning('%s: Non-uniform pixel ratio (%s)' % (
2144 info_dict
['id'], stretched_ratio
))
2145 elif fixup_policy
== 'detect_or_warn':
2146 stretched_pp
= FFmpegFixupStretchedPP(self
)
2147 if stretched_pp
.available
:
2148 info_dict
.setdefault('__postprocessors', [])
2149 info_dict
['__postprocessors'].append(stretched_pp
)
2151 self
.report_warning(
2152 '%s: Non-uniform pixel ratio (%s). %s'
2153 % (info_dict
['id'], stretched_ratio
, INSTALL_FFMPEG_MESSAGE
))
2155 assert fixup_policy
in ('ignore', 'never')
2157 if (info_dict
.get('requested_formats') is None
2158 and info_dict
.get('container') == 'm4a_dash'):
2159 if fixup_policy
== 'warn':
2160 self
.report_warning(
2161 '%s: writing DASH m4a. '
2162 'Only some players support this container.'
2164 elif fixup_policy
== 'detect_or_warn':
2165 fixup_pp
= FFmpegFixupM4aPP(self
)
2166 if fixup_pp
.available
:
2167 info_dict
.setdefault('__postprocessors', [])
2168 info_dict
['__postprocessors'].append(fixup_pp
)
2170 self
.report_warning(
2171 '%s: writing DASH m4a. '
2172 'Only some players support this container. %s'
2173 % (info_dict
['id'], INSTALL_FFMPEG_MESSAGE
))
2175 assert fixup_policy
in ('ignore', 'never')
2177 if (info_dict
.get('protocol') == 'm3u8_native'
2178 or info_dict
.get('protocol') == 'm3u8'
2179 and self
.params
.get('hls_prefer_native')):
2180 if fixup_policy
== 'warn':
2181 self
.report_warning('%s: malformed AAC bitstream detected.' % (
2183 elif fixup_policy
== 'detect_or_warn':
2184 fixup_pp
= FFmpegFixupM3u8PP(self
)
2185 if fixup_pp
.available
:
2186 info_dict
.setdefault('__postprocessors', [])
2187 info_dict
['__postprocessors'].append(fixup_pp
)
2189 self
.report_warning(
2190 '%s: malformed AAC bitstream detected. %s'
2191 % (info_dict
['id'], INSTALL_FFMPEG_MESSAGE
))
2193 assert fixup_policy
in ('ignore', 'never')
2196 self
.post_process(filename
, info_dict
)
2197 except (PostProcessingError
) as err
:
2198 self
.report_error('postprocessing: %s' % str(err
))
2200 must_record_download_archive
= True
2202 if must_record_download_archive
or self
.params
.get('force_write_download_archive', False):
2203 self
.record_download_archive(info_dict
)
2205 def download(self
, url_list
):
2206 """Download a given list of URLs."""
2207 outtmpl
= self
.params
.get('outtmpl', DEFAULT_OUTTMPL
)
2208 if (len(url_list
) > 1
2210 and '%' not in outtmpl
2211 and self
.params
.get('max_downloads') != 1):
2212 raise SameFileError(outtmpl
)
2214 for url
in url_list
:
2216 # It also downloads the videos
2217 res
= self
.extract_info(
2218 url
, force_generic_extractor
=self
.params
.get('force_generic_extractor', False))
2219 except UnavailableVideoError
:
2220 self
.report_error('unable to download video')
2221 except MaxDownloadsReached
:
2222 self
.to_screen('[info] Maximum number of downloaded files reached.')
2225 if self
.params
.get('dump_single_json', False):
2226 self
.to_stdout(json
.dumps(res
))
2228 return self
._download
_retcode
2230 def download_with_info_file(self
, info_filename
):
2231 with contextlib
.closing(fileinput
.FileInput(
2232 [info_filename
], mode
='r',
2233 openhook
=fileinput
.hook_encoded('utf-8'))) as f
:
2234 # FileInput doesn't have a read method, we can't call json.load
2235 info
= self
.filter_requested_info(json
.loads('\n'.join(f
)))
2237 self
.process_ie_result(info
, download
=True)
2238 except DownloadError
:
2239 webpage_url
= info
.get('webpage_url')
2240 if webpage_url
is not None:
2241 self
.report_warning('The info failed to download, trying with "%s"' % webpage_url
)
2242 return self
.download([webpage_url
])
2245 return self
._download
_retcode
2248 def filter_requested_info(info_dict
):
2250 (k
, v
) for k
, v
in info_dict
.items()
2251 if k
not in ['requested_formats', 'requested_subtitles'])
2253 def post_process(self
, filename
, ie_info
):
2254 """Run all the postprocessors on the given file."""
2255 info
= dict(ie_info
)
2256 info
['filepath'] = filename
2258 if ie_info
.get('__postprocessors') is not None:
2259 pps_chain
.extend(ie_info
['__postprocessors'])
2260 pps_chain
.extend(self
._pps
)
2261 for pp
in pps_chain
:
2262 files_to_delete
= []
2264 files_to_delete
, info
= pp
.run(info
)
2265 except PostProcessingError
as e
:
2266 self
.report_error(e
.msg
)
2267 if files_to_delete
and not self
.params
.get('keepvideo', False):
2268 for old_filename
in set(files_to_delete
):
2269 self
.to_screen('Deleting original file %s (pass -k to keep)' % old_filename
)
2271 os
.remove(encodeFilename(old_filename
))
2272 except (IOError, OSError):
2273 self
.report_warning('Unable to remove downloaded original file')
2275 def _make_archive_id(self
, info_dict
):
2276 video_id
= info_dict
.get('id')
2279 # Future-proof against any change in case
2280 # and backwards compatibility with prior versions
2281 extractor
= info_dict
.get('extractor_key') or info_dict
.get('ie_key') # key in a playlist
2282 if extractor
is None:
2283 url
= str_or_none(info_dict
.get('url'))
2286 # Try to find matching extractor for the URL and take its ie_key
2287 for ie
in self
._ies
:
2288 if ie
.suitable(url
):
2289 extractor
= ie
.ie_key()
2293 return extractor
.lower() + ' ' + video_id
2295 def in_download_archive(self
, info_dict
):
2296 fn
= self
.params
.get('download_archive')
2300 vid_id
= self
._make
_archive
_id
(info_dict
)
2302 return False # Incomplete video information
2304 return vid_id
in self
.archive
2306 def record_download_archive(self
, info_dict
):
2307 fn
= self
.params
.get('download_archive')
2310 vid_id
= self
._make
_archive
_id
(info_dict
)
2312 with locked_file(fn
, 'a', encoding
='utf-8') as archive_file
:
2313 archive_file
.write(vid_id
+ '\n')
2314 self
.archive
.add(vid_id
)
2317 def format_resolution(format
, default
='unknown'):
2318 if format
.get('vcodec') == 'none':
2320 if format
.get('resolution') is not None:
2321 return format
['resolution']
2322 if format
.get('height') is not None:
2323 if format
.get('width') is not None:
2324 res
= '%sx%s' % (format
['width'], format
['height'])
2326 res
= '%sp' % format
['height']
2327 elif format
.get('width') is not None:
2328 res
= '%dx?' % format
['width']
2333 def _format_note(self
, fdict
):
2335 if fdict
.get('ext') in ['f4f', 'f4m']:
2336 res
+= '(unsupported) '
2337 if fdict
.get('language'):
2340 res
+= '[%s] ' % fdict
['language']
2341 if fdict
.get('format_note') is not None:
2342 res
+= fdict
['format_note'] + ' '
2343 if fdict
.get('tbr') is not None:
2344 res
+= '%4dk ' % fdict
['tbr']
2345 if fdict
.get('container') is not None:
2348 res
+= '%s container' % fdict
['container']
2349 if (fdict
.get('vcodec') is not None
2350 and fdict
.get('vcodec') != 'none'):
2353 res
+= fdict
['vcodec']
2354 if fdict
.get('vbr') is not None:
2356 elif fdict
.get('vbr') is not None and fdict
.get('abr') is not None:
2358 if fdict
.get('vbr') is not None:
2359 res
+= '%4dk' % fdict
['vbr']
2360 if fdict
.get('fps') is not None:
2363 res
+= '%sfps' % fdict
['fps']
2364 if fdict
.get('acodec') is not None:
2367 if fdict
['acodec'] == 'none':
2370 res
+= '%-5s' % fdict
['acodec']
2371 elif fdict
.get('abr') is not None:
2375 if fdict
.get('abr') is not None:
2376 res
+= '@%3dk' % fdict
['abr']
2377 if fdict
.get('asr') is not None:
2378 res
+= ' (%5dHz)' % fdict
['asr']
2379 if fdict
.get('filesize') is not None:
2382 res
+= format_bytes(fdict
['filesize'])
2383 elif fdict
.get('filesize_approx') is not None:
2386 res
+= '~' + format_bytes(fdict
['filesize_approx'])
2389 def _format_note_table(self
, f
):
2390 def join_fields(*vargs
):
2391 return ', '.join((val
for val
in vargs
if val
!= ''))
2394 'UNSUPPORTED' if f
.get('ext') in ('f4f', 'f4m') else '',
2395 format_field(f
, 'language', '[%s]'),
2396 format_field(f
, 'format_note'),
2397 format_field(f
, 'container', ignore
=(None, f
.get('ext'))),
2398 format_field(f
, 'asr', '%5dHz'))
2400 def list_formats(self
, info_dict
):
2401 formats
= info_dict
.get('formats', [info_dict
])
2402 new_format
= self
.params
.get('listformats_table', False)
2406 format_field(f
, 'format_id'),
2407 format_field(f
, 'ext'),
2408 self
.format_resolution(f
),
2409 format_field(f
, 'fps', '%d'),
2411 format_field(f
, 'filesize', ' %s', func
=format_bytes
) + format_field(f
, 'filesize_approx', '~%s', func
=format_bytes
),
2412 format_field(f
, 'tbr', '%4dk'),
2413 f
.get('protocol').replace('http_dash_segments', 'dash').replace("native", "n"),
2415 format_field(f
, 'vcodec', default
='unknown').replace('none', ''),
2416 format_field(f
, 'vbr', '%4dk'),
2417 format_field(f
, 'acodec', default
='unknown').replace('none', ''),
2418 format_field(f
, 'abr', '%3dk'),
2419 format_field(f
, 'asr', '%5dHz'),
2420 self
._format
_note
_table
(f
)]
2422 if f
.get('preference') is None or f
['preference'] >= -1000]
2423 header_line
= ['ID', 'EXT', 'RESOLUTION', 'FPS', '|', ' FILESIZE', ' TBR', 'PROTO',
2424 '|', 'VCODEC', ' VBR', 'ACODEC', ' ABR', ' ASR', 'NOTE']
2428 format_field(f
, 'format_id'),
2429 format_field(f
, 'ext'),
2430 self
.format_resolution(f
),
2431 self
._format
_note
(f
)]
2433 if f
.get('preference') is None or f
['preference'] >= -1000]
2434 header_line
= ['format code', 'extension', 'resolution', 'note']
2436 # if len(formats) > 1:
2437 # table[-1][-1] += (' ' if table[-1][-1] else '') + '(best)'
2439 '[info] Available formats for %s:\n%s' % (info_dict
['id'], render_table(
2443 extraGap
=(0 if new_format
else 1),
2444 hideEmpty
=new_format
)))
2446 def list_thumbnails(self
, info_dict
):
2447 thumbnails
= info_dict
.get('thumbnails')
2449 self
.to_screen('[info] No thumbnails present for %s' % info_dict
['id'])
2453 '[info] Thumbnails for %s:' % info_dict
['id'])
2454 self
.to_screen(render_table(
2455 ['ID', 'width', 'height', 'URL'],
2456 [[t
['id'], t
.get('width', 'unknown'), t
.get('height', 'unknown'), t
['url']] for t
in thumbnails
]))
2458 def list_subtitles(self
, video_id
, subtitles
, name
='subtitles'):
2460 self
.to_screen('%s has no %s' % (video_id
, name
))
2463 'Available %s for %s:' % (name
, video_id
))
2464 self
.to_screen(render_table(
2465 ['Language', 'formats'],
2466 [[lang
, ', '.join(f
['ext'] for f
in reversed(formats
))]
2467 for lang
, formats
in subtitles
.items()]))
2469 def urlopen(self
, req
):
2470 """ Start an HTTP download """
2471 if isinstance(req
, compat_basestring
):
2472 req
= sanitized_Request(req
)
2473 return self
._opener
.open(req
, timeout
=self
._socket
_timeout
)
2475 def print_debug_header(self
):
2476 if not self
.params
.get('verbose'):
2479 if type('') is not compat_str
:
2480 # Python 2.6 on SLES11 SP1 (https://github.com/ytdl-org/youtube-dl/issues/3326)
2481 self
.report_warning(
2482 'Your Python is broken! Update to a newer and supported version')
2484 stdout_encoding
= getattr(
2485 sys
.stdout
, 'encoding', 'missing (%s)' % type(sys
.stdout
).__name
__)
2487 '[debug] Encodings: locale %s, fs %s, out %s, pref %s\n' % (
2488 locale
.getpreferredencoding(),
2489 sys
.getfilesystemencoding(),
2491 self
.get_encoding()))
2492 write_string(encoding_str
, encoding
=None)
2494 self
._write
_string
('[debug] youtube-dlc version ' + __version__
+ '\n')
2496 self
._write
_string
('[debug] Lazy loading extractors enabled' + '\n')
2498 sp
= subprocess
.Popen(
2499 ['git', 'rev-parse', '--short', 'HEAD'],
2500 stdout
=subprocess
.PIPE
, stderr
=subprocess
.PIPE
,
2501 cwd
=os
.path
.dirname(os
.path
.abspath(__file__
)))
2502 out
, err
= sp
.communicate()
2503 out
= out
.decode().strip()
2504 if re
.match('[0-9a-f]+', out
):
2505 self
._write
_string
('[debug] Git HEAD: ' + out
+ '\n')
2512 def python_implementation():
2513 impl_name
= platform
.python_implementation()
2514 if impl_name
== 'PyPy' and hasattr(sys
, 'pypy_version_info'):
2515 return impl_name
+ ' version %d.%d.%d' % sys
.pypy_version_info
[:3]
2518 self
._write
_string
('[debug] Python version %s (%s) - %s\n' % (
2519 platform
.python_version(), python_implementation(),
2522 exe_versions
= FFmpegPostProcessor
.get_versions(self
)
2523 exe_versions
['rtmpdump'] = rtmpdump_version()
2524 exe_versions
['phantomjs'] = PhantomJSwrapper
._version
()
2525 exe_str
= ', '.join(
2527 for exe
, v
in sorted(exe_versions
.items())
2532 self
._write
_string
('[debug] exe versions: %s\n' % exe_str
)
2535 for handler
in self
._opener
.handlers
:
2536 if hasattr(handler
, 'proxies'):
2537 proxy_map
.update(handler
.proxies
)
2538 self
._write
_string
('[debug] Proxy map: ' + compat_str(proxy_map
) + '\n')
2540 if self
.params
.get('call_home', False):
2541 ipaddr
= self
.urlopen('https://yt-dl.org/ip').read().decode('utf-8')
2542 self
._write
_string
('[debug] Public IP address: %s\n' % ipaddr
)
2543 latest_version
= self
.urlopen(
2544 'https://yt-dl.org/latest/version').read().decode('utf-8')
2545 if version_tuple(latest_version
) > version_tuple(__version__
):
2546 self
.report_warning(
2547 'You are using an outdated version (newest version: %s)! '
2548 'See https://yt-dl.org/update if you need help updating.' %
2551 def _setup_opener(self
):
2552 timeout_val
= self
.params
.get('socket_timeout')
2553 self
._socket
_timeout
= 600 if timeout_val
is None else float(timeout_val
)
2555 opts_cookiefile
= self
.params
.get('cookiefile')
2556 opts_proxy
= self
.params
.get('proxy')
2558 if opts_cookiefile
is None:
2559 self
.cookiejar
= compat_cookiejar
.CookieJar()
2561 opts_cookiefile
= expand_path(opts_cookiefile
)
2562 self
.cookiejar
= YoutubeDLCookieJar(opts_cookiefile
)
2563 if os
.access(opts_cookiefile
, os
.R_OK
):
2564 self
.cookiejar
.load(ignore_discard
=True, ignore_expires
=True)
2566 cookie_processor
= YoutubeDLCookieProcessor(self
.cookiejar
)
2567 if opts_proxy
is not None:
2568 if opts_proxy
== '':
2571 proxies
= {'http': opts_proxy, 'https': opts_proxy}
2573 proxies
= compat_urllib_request
.getproxies()
2574 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
2575 if 'http' in proxies
and 'https' not in proxies
:
2576 proxies
['https'] = proxies
['http']
2577 proxy_handler
= PerRequestProxyHandler(proxies
)
2579 debuglevel
= 1 if self
.params
.get('debug_printtraffic') else 0
2580 https_handler
= make_HTTPS_handler(self
.params
, debuglevel
=debuglevel
)
2581 ydlh
= YoutubeDLHandler(self
.params
, debuglevel
=debuglevel
)
2582 redirect_handler
= YoutubeDLRedirectHandler()
2583 data_handler
= compat_urllib_request_DataHandler()
2585 # When passing our own FileHandler instance, build_opener won't add the
2586 # default FileHandler and allows us to disable the file protocol, which
2587 # can be used for malicious purposes (see
2588 # https://github.com/ytdl-org/youtube-dl/issues/8227)
2589 file_handler
= compat_urllib_request
.FileHandler()
2591 def file_open(*args
, **kwargs
):
2592 raise compat_urllib_error
.URLError('file:// scheme is explicitly disabled in youtube-dlc for security reasons')
2593 file_handler
.file_open
= file_open
2595 opener
= compat_urllib_request
.build_opener(
2596 proxy_handler
, https_handler
, cookie_processor
, ydlh
, redirect_handler
, data_handler
, file_handler
)
2598 # Delete the default user-agent header, which would otherwise apply in
2599 # cases where our custom HTTP handler doesn't come into play
2600 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
2601 opener
.addheaders
= []
2602 self
._opener
= opener
2604 def encode(self
, s
):
2605 if isinstance(s
, bytes):
2606 return s
# Already encoded
2609 return s
.encode(self
.get_encoding())
2610 except UnicodeEncodeError as err
:
2611 err
.reason
= err
.reason
+ '. Check your system encoding configuration or use the --encoding option.'
2614 def get_encoding(self
):
2615 encoding
= self
.params
.get('encoding')
2616 if encoding
is None:
2617 encoding
= preferredencoding()
2620 def _write_thumbnails(self
, info_dict
, filename
):
2621 if self
.params
.get('writethumbnail', False):
2622 thumbnails
= info_dict
.get('thumbnails')
2624 thumbnails
= [thumbnails
[-1]]
2625 elif self
.params
.get('write_all_thumbnails', False):
2626 thumbnails
= info_dict
.get('thumbnails')
2631 # No thumbnails present, so return immediately
2634 for t
in thumbnails
:
2635 thumb_ext
= determine_ext(t
['url'], 'jpg')
2636 suffix
= '_%s' % t
['id'] if len(thumbnails
) > 1 else ''
2637 thumb_display_id
= '%s ' % t
['id'] if len(thumbnails
) > 1 else ''
2638 t
['filename'] = thumb_filename
= replace_extension(filename
+ suffix
, thumb_ext
, info_dict
.get('ext'))
2640 if self
.params
.get('nooverwrites', False) and os
.path
.exists(encodeFilename(thumb_filename
)):
2641 self
.to_screen('[%s] %s: Thumbnail %sis already present' %
2642 (info_dict
['extractor'], info_dict
['id'], thumb_display_id
))
2644 self
.to_screen('[%s] %s: Downloading thumbnail %s...' %
2645 (info_dict
['extractor'], info_dict
['id'], thumb_display_id
))
2647 uf
= self
.urlopen(t
['url'])
2648 with open(encodeFilename(thumb_filename
), 'wb') as thumbf
:
2649 shutil
.copyfileobj(uf
, thumbf
)
2650 self
.to_screen('[%s] %s: Writing thumbnail %sto: %s' %
2651 (info_dict
['extractor'], info_dict
['id'], thumb_display_id
, thumb_filename
))
2652 except (compat_urllib_error
.URLError
, compat_http_client
.HTTPException
, socket
.error
) as err
:
2653 self
.report_warning('Unable to download thumbnail "%s": %s' %
2654 (t
['url'], error_to_compat_str(err
)))