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. (Default False when running youtube-dlc, but True when directly accessing YoutubeDL class)
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 dictionary of postprocessor names (in lower case) and a list
337 of additional command-line arguments for the postprocessor.
338 Use 'default' as the name for arguments to passed to all PP.
340 The following options are used by the Youtube extractor:
341 youtube_include_dash_manifest: If True (default), DASH manifests and related
342 data will be downloaded and processed by extractor.
343 You can reduce network I/O by disabling it if you don't
347 _NUMERIC_FIELDS
= set((
348 'width', 'height', 'tbr', 'abr', 'asr', 'vbr', 'fps', 'filesize', 'filesize_approx',
349 'timestamp', 'upload_year', 'upload_month', 'upload_day',
350 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
351 'average_rating', 'comment_count', 'age_limit',
352 'start_time', 'end_time',
353 'chapter_number', 'season_number', 'episode_number',
354 'track_number', 'disc_number', 'release_year',
361 _download_retcode
= None
362 _num_downloads
= None
365 def __init__(self
, params
=None, auto_init
=True):
366 """Create a FileDownloader object with the given options."""
370 self
._ies
_instances
= {}
372 self
._progress
_hooks
= []
373 self
._download
_retcode
= 0
374 self
._num
_downloads
= 0
375 self
._screen
_file
= [sys
.stdout
, sys
.stderr
][params
.get('logtostderr', False)]
376 self
._err
_file
= sys
.stderr
379 'nocheckcertificate': False,
381 self
.params
.update(params
)
382 self
.cache
= Cache(self
)
385 """Preload the archive, if any is specified"""
386 def preload_download_archive(self
):
387 fn
= self
.params
.get('download_archive')
391 with locked_file(fn
, 'r', encoding
='utf-8') as archive_file
:
392 for line
in archive_file
:
393 self
.archive
.add(line
.strip())
394 except IOError as ioe
:
395 if ioe
.errno
!= errno
.ENOENT
:
400 def check_deprecated(param
, option
, suggestion
):
401 if self
.params
.get(param
) is not None:
403 '%s is deprecated. Use %s instead.' % (option
, suggestion
))
407 if self
.params
.get('verbose'):
408 self
.to_stdout('[debug] Loading archive file %r' % self
.params
.get('download_archive'))
410 preload_download_archive(self
)
412 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
413 if self
.params
.get('geo_verification_proxy') is None:
414 self
.params
['geo_verification_proxy'] = self
.params
['cn_verification_proxy']
416 check_deprecated('autonumber_size', '--autonumber-size', 'output template with %(autonumber)0Nd, where N in the number of digits')
417 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
418 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
420 if params
.get('bidi_workaround', False):
423 master
, slave
= pty
.openpty()
424 width
= compat_get_terminal_size().columns
428 width_args
= ['-w', str(width
)]
430 stdin
=subprocess
.PIPE
,
432 stderr
=self
._err
_file
)
434 self
._output
_process
= subprocess
.Popen(
435 ['bidiv'] + width_args
, **sp_kwargs
438 self
._output
_process
= subprocess
.Popen(
439 ['fribidi', '-c', 'UTF-8'] + width_args
, **sp_kwargs
)
440 self
._output
_channel
= os
.fdopen(master
, 'rb')
441 except OSError as ose
:
442 if ose
.errno
== errno
.ENOENT
:
443 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 $PATH.')
447 if (sys
.platform
!= 'win32'
448 and sys
.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
449 and not params
.get('restrictfilenames', False)):
450 # Unicode filesystem API will throw errors (#1474, #13027)
452 'Assuming --restrict-filenames since file system encoding '
453 'cannot encode all characters. '
454 'Set the LC_ALL environment variable to fix this.')
455 self
.params
['restrictfilenames'] = True
457 if isinstance(params
.get('outtmpl'), bytes):
459 'Parameter outtmpl is bytes, but should be a unicode string. '
460 'Put from __future__ import unicode_literals at the top of your code file or consider switching to Python 3.x.')
465 self
.print_debug_header()
466 self
.add_default_info_extractors()
468 for pp_def_raw
in self
.params
.get('postprocessors', []):
469 pp_class
= get_postprocessor(pp_def_raw
['key'])
470 pp_def
= dict(pp_def_raw
)
472 pp
= pp_class(self
, **compat_kwargs(pp_def
))
473 self
.add_post_processor(pp
)
475 for ph
in self
.params
.get('progress_hooks', []):
476 self
.add_progress_hook(ph
)
478 register_socks_protocols()
480 def warn_if_short_id(self
, argv
):
481 # short YouTube ID starting with dash?
483 i
for i
, a
in enumerate(argv
)
484 if re
.match(r
'^-[0-9A-Za-z_-]{10}$', a
)]
488 + [a
for i
, a
in enumerate(argv
) if i
not in idxs
]
489 + ['--'] + [argv
[i
] for i
in idxs
]
492 'Long argument string detected. '
493 'Use -- to separate parameters and URLs, like this:\n%s\n' %
494 args_to_str(correct_argv
))
496 def add_info_extractor(self
, ie
):
497 """Add an InfoExtractor object to the end of the list."""
499 if not isinstance(ie
, type):
500 self
._ies
_instances
[ie
.ie_key()] = ie
501 ie
.set_downloader(self
)
503 def get_info_extractor(self
, ie_key
):
505 Get an instance of an IE with name ie_key, it will try to get one from
506 the _ies list, if there's no instance it will create a new one and add
507 it to the extractor list.
509 ie
= self
._ies
_instances
.get(ie_key
)
511 ie
= get_info_extractor(ie_key
)()
512 self
.add_info_extractor(ie
)
515 def add_default_info_extractors(self
):
517 Add the InfoExtractors returned by gen_extractors to the end of the list
519 for ie
in gen_extractor_classes():
520 self
.add_info_extractor(ie
)
522 def add_post_processor(self
, pp
):
523 """Add a PostProcessor object to the end of the chain."""
525 pp
.set_downloader(self
)
527 def add_progress_hook(self
, ph
):
528 """Add the progress hook (currently only for the file downloader)"""
529 self
._progress
_hooks
.append(ph
)
531 def _bidi_workaround(self
, message
):
532 if not hasattr(self
, '_output_channel'):
535 assert hasattr(self
, '_output_process')
536 assert isinstance(message
, compat_str
)
537 line_count
= message
.count('\n') + 1
538 self
._output
_process
.stdin
.write((message
+ '\n').encode('utf-8'))
539 self
._output
_process
.stdin
.flush()
540 res
= ''.join(self
._output
_channel
.readline().decode('utf-8')
541 for _
in range(line_count
))
542 return res
[:-len('\n')]
544 def to_screen(self
, message
, skip_eol
=False):
545 """Print message to stdout if not in quiet mode."""
546 return self
.to_stdout(message
, skip_eol
, check_quiet
=True)
548 def _write_string(self
, s
, out
=None):
549 write_string(s
, out
=out
, encoding
=self
.params
.get('encoding'))
551 def to_stdout(self
, message
, skip_eol
=False, check_quiet
=False):
552 """Print message to stdout if not in quiet mode."""
553 if self
.params
.get('logger'):
554 self
.params
['logger'].debug(message
)
555 elif not check_quiet
or not self
.params
.get('quiet', False):
556 message
= self
._bidi
_workaround
(message
)
557 terminator
= ['\n', ''][skip_eol
]
558 output
= message
+ terminator
560 self
._write
_string
(output
, self
._screen
_file
)
562 def to_stderr(self
, message
):
563 """Print message to stderr."""
564 assert isinstance(message
, compat_str
)
565 if self
.params
.get('logger'):
566 self
.params
['logger'].error(message
)
568 message
= self
._bidi
_workaround
(message
)
569 output
= message
+ '\n'
570 self
._write
_string
(output
, self
._err
_file
)
572 def to_console_title(self
, message
):
573 if not self
.params
.get('consoletitle', False):
575 if compat_os_name
== 'nt':
576 if ctypes
.windll
.kernel32
.GetConsoleWindow():
577 # c_wchar_p() might not be necessary if `message` is
578 # already of type unicode()
579 ctypes
.windll
.kernel32
.SetConsoleTitleW(ctypes
.c_wchar_p(message
))
580 elif 'TERM' in os
.environ
:
581 self
._write
_string
('\033]0;%s\007' % message
, self
._screen
_file
)
583 def save_console_title(self
):
584 if not self
.params
.get('consoletitle', False):
586 if self
.params
.get('simulate', False):
588 if compat_os_name
!= 'nt' and 'TERM' in os
.environ
:
589 # Save the title on stack
590 self
._write
_string
('\033[22;0t', self
._screen
_file
)
592 def restore_console_title(self
):
593 if not self
.params
.get('consoletitle', False):
595 if self
.params
.get('simulate', False):
597 if compat_os_name
!= 'nt' and 'TERM' in os
.environ
:
598 # Restore the title from stack
599 self
._write
_string
('\033[23;0t', self
._screen
_file
)
602 self
.save_console_title()
605 def __exit__(self
, *args
):
606 self
.restore_console_title()
608 if self
.params
.get('cookiefile') is not None:
609 self
.cookiejar
.save(ignore_discard
=True, ignore_expires
=True)
611 def trouble(self
, message
=None, tb
=None):
612 """Determine action to take when a download problem appears.
614 Depending on if the downloader has been configured to ignore
615 download errors or not, this method may throw an exception or
616 not when errors are found, after printing the message.
618 tb, if given, is additional traceback information.
620 if message
is not None:
621 self
.to_stderr(message
)
622 if self
.params
.get('verbose'):
624 if sys
.exc_info()[0]: # if .trouble has been called from an except block
626 if hasattr(sys
.exc_info()[1], 'exc_info') and sys
.exc_info()[1].exc_info
[0]:
627 tb
+= ''.join(traceback
.format_exception(*sys
.exc_info()[1].exc_info
))
628 tb
+= encode_compat_str(traceback
.format_exc())
630 tb_data
= traceback
.format_list(traceback
.extract_stack())
631 tb
= ''.join(tb_data
)
633 if not self
.params
.get('ignoreerrors', False):
634 if sys
.exc_info()[0] and hasattr(sys
.exc_info()[1], 'exc_info') and sys
.exc_info()[1].exc_info
[0]:
635 exc_info
= sys
.exc_info()[1].exc_info
637 exc_info
= sys
.exc_info()
638 raise DownloadError(message
, exc_info
)
639 self
._download
_retcode
= 1
641 def report_warning(self
, message
):
643 Print the message to stderr, it will be prefixed with 'WARNING:'
644 If stderr is a tty file the 'WARNING:' will be colored
646 if self
.params
.get('logger') is not None:
647 self
.params
['logger'].warning(message
)
649 if self
.params
.get('no_warnings'):
651 if not self
.params
.get('no_color') and self
._err
_file
.isatty() and compat_os_name
!= 'nt':
652 _msg_header
= '\033[0;33mWARNING:\033[0m'
654 _msg_header
= 'WARNING:'
655 warning_message
= '%s %s' % (_msg_header
, message
)
656 self
.to_stderr(warning_message
)
658 def report_error(self
, message
, tb
=None):
660 Do the same as trouble, but prefixes the message with 'ERROR:', colored
661 in red if stderr is a tty file.
663 if not self
.params
.get('no_color') and self
._err
_file
.isatty() and compat_os_name
!= 'nt':
664 _msg_header
= '\033[0;31mERROR:\033[0m'
666 _msg_header
= 'ERROR:'
667 error_message
= '%s %s' % (_msg_header
, message
)
668 self
.trouble(error_message
, tb
)
670 def report_file_already_downloaded(self
, file_name
):
671 """Report file has already been fully downloaded."""
673 self
.to_screen('[download] %s has already been downloaded' % file_name
)
674 except UnicodeEncodeError:
675 self
.to_screen('[download] The file has already been downloaded')
677 def prepare_filename(self
, info_dict
):
678 """Generate the output filename."""
680 template_dict
= dict(info_dict
)
682 template_dict
['epoch'] = int(time
.time())
683 autonumber_size
= self
.params
.get('autonumber_size')
684 if autonumber_size
is None:
686 template_dict
['autonumber'] = self
.params
.get('autonumber_start', 1) - 1 + self
._num
_downloads
687 if template_dict
.get('resolution') is None:
688 if template_dict
.get('width') and template_dict
.get('height'):
689 template_dict
['resolution'] = '%dx%d' % (template_dict
['width'], template_dict
['height'])
690 elif template_dict
.get('height'):
691 template_dict
['resolution'] = '%sp' % template_dict
['height']
692 elif template_dict
.get('width'):
693 template_dict
['resolution'] = '%dx?' % template_dict
['width']
695 sanitize
= lambda k
, v
: sanitize_filename(
697 restricted
=self
.params
.get('restrictfilenames'),
698 is_id
=(k
== 'id' or k
.endswith('_id')))
699 template_dict
= dict((k
, v
if isinstance(v
, compat_numeric_types
) else sanitize(k
, v
))
700 for k
, v
in template_dict
.items()
701 if v
is not None and not isinstance(v
, (list, tuple, dict)))
702 template_dict
= collections
.defaultdict(lambda: 'NA', template_dict
)
704 outtmpl
= self
.params
.get('outtmpl', DEFAULT_OUTTMPL
)
706 # For fields playlist_index and autonumber convert all occurrences
707 # of %(field)s to %(field)0Nd for backward compatibility
708 field_size_compat_map
= {
709 'playlist_index': len(str(template_dict
['n_entries'])),
710 'autonumber': autonumber_size
,
712 FIELD_SIZE_COMPAT_RE
= r
'(?<!%)%\((?P<field>autonumber|playlist_index)\)s'
713 mobj
= re
.search(FIELD_SIZE_COMPAT_RE
, outtmpl
)
716 FIELD_SIZE_COMPAT_RE
,
717 r
'%%(\1)0%dd' % field_size_compat_map
[mobj
.group('field')],
720 # Missing numeric fields used together with integer presentation types
721 # in format specification will break the argument substitution since
722 # string 'NA' is returned for missing fields. We will patch output
723 # template for missing fields to meet string presentation type.
724 for numeric_field
in self
._NUMERIC
_FIELDS
:
725 if numeric_field
not in template_dict
:
726 # As of [1] format syntax is:
727 # %[mapping_key][conversion_flags][minimum_width][.precision][length_modifier]type
728 # 1. https://docs.python.org/2/library/stdtypes.html#string-formatting
732 \({0}\) # mapping key
733 (?:[#0\-+ ]+)? # conversion flags (optional)
734 (?:\d+)? # minimum field width (optional)
735 (?:\.\d+)? # precision (optional)
736 [hlL]? # length modifier (optional)
737 [diouxXeEfFgGcrs%] # conversion type
740 FORMAT_RE
.format(numeric_field
),
741 r
'%({0})s'.format(numeric_field
), outtmpl
)
743 # expand_path translates '%%' into '%' and '$$' into '$'
744 # correspondingly that is not what we want since we need to keep
745 # '%%' intact for template dict substitution step. Working around
746 # with boundary-alike separator hack.
747 sep
= ''.join([random
.choice(ascii_letters
) for _
in range(32)])
748 outtmpl
= outtmpl
.replace('%%', '%{0}%'.format(sep
)).replace('$$', '${0}$'.format(sep
))
750 # outtmpl should be expand_path'ed before template dict substitution
751 # because meta fields may contain env variables we don't want to
752 # be expanded. For example, for outtmpl "%(title)s.%(ext)s" and
753 # title "Hello $PATH", we don't want `$PATH` to be expanded.
754 filename
= expand_path(outtmpl
).replace(sep
, '') % template_dict
756 # https://github.com/blackjack4494/youtube-dlc/issues/85
757 trim_file_name
= self
.params
.get('trim_file_name', False)
759 fn_groups
= filename
.rsplit('.')
762 if len(fn_groups
) > 2:
763 sub_ext
= fn_groups
[-2]
764 filename
= '.'.join(filter(None, [fn_groups
[0][:trim_file_name
], sub_ext
, ext
]))
766 # Temporary fix for #4787
767 # 'Treat' all problem characters by passing filename through preferredencoding
768 # to workaround encoding issues with subprocess on python2 @ Windows
769 if sys
.version_info
< (3, 0) and sys
.platform
== 'win32':
770 filename
= encodeFilename(filename
, True).decode(preferredencoding())
771 return sanitize_path(filename
)
772 except ValueError as err
:
773 self
.report_error('Error in output template: ' + str(err
) + ' (encoding: ' + repr(preferredencoding()) + ')')
776 def _match_entry(self
, info_dict
, incomplete
):
777 """ Returns None if the file should be downloaded """
779 video_title
= info_dict
.get('title', info_dict
.get('id', 'video'))
780 if 'title' in info_dict
:
781 # This can happen when we're just evaluating the playlist
782 title
= info_dict
['title']
783 matchtitle
= self
.params
.get('matchtitle', False)
785 if not re
.search(matchtitle
, title
, re
.IGNORECASE
):
786 return '"' + title
+ '" title did not match pattern "' + matchtitle
+ '"'
787 rejecttitle
= self
.params
.get('rejecttitle', False)
789 if re
.search(rejecttitle
, title
, re
.IGNORECASE
):
790 return '"' + title
+ '" title matched reject pattern "' + rejecttitle
+ '"'
791 date
= info_dict
.get('upload_date')
793 dateRange
= self
.params
.get('daterange', DateRange())
794 if date
not in dateRange
:
795 return '%s upload date is not in range %s' % (date_from_str(date
).isoformat(), dateRange
)
796 view_count
= info_dict
.get('view_count')
797 if view_count
is not None:
798 min_views
= self
.params
.get('min_views')
799 if min_views
is not None and view_count
< min_views
:
800 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title
, view_count
, min_views
)
801 max_views
= self
.params
.get('max_views')
802 if max_views
is not None and view_count
> max_views
:
803 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title
, view_count
, max_views
)
804 if age_restricted(info_dict
.get('age_limit'), self
.params
.get('age_limit')):
805 return 'Skipping "%s" because it is age restricted' % video_title
806 if self
.in_download_archive(info_dict
):
807 return '%s has already been recorded in archive' % video_title
810 match_filter
= self
.params
.get('match_filter')
811 if match_filter
is not None:
812 ret
= match_filter(info_dict
)
819 def add_extra_info(info_dict
, extra_info
):
820 '''Set the keys from extra_info in info dict if they are missing'''
821 for key
, value
in extra_info
.items():
822 info_dict
.setdefault(key
, value
)
824 def extract_info(self
, url
, download
=True, ie_key
=None, info_dict
=None, extra_info
={},
825 process
=True, force_generic_extractor
=False):
827 Returns a list with a dictionary for each video we find.
828 If 'download', also downloads the videos.
829 extra_info is a dict containing the extra values to add to each result
832 if not ie_key
and force_generic_extractor
:
836 ies
= [self
.get_info_extractor(ie_key
)]
841 if not ie
.suitable(url
):
845 ie
= self
.get_info_extractor(ie_key
)
847 self
.report_warning('The program functionality for this site has been marked as broken, '
848 'and will probably not work.')
851 temp_id
= ie
.extract_id(url
) if callable(getattr(ie
, 'extract_id', None)) else ie
._match
_id
(url
)
852 except (AssertionError, IndexError, AttributeError):
854 if temp_id
is not None and self
.in_download_archive({'id': temp_id, 'ie_key': ie_key}
):
855 self
.to_screen("[%s] %s: has already been recorded in archive" % (
859 return self
.__extract
_info
(url
, ie
, download
, extra_info
, process
, info_dict
)
862 self
.report_error('no suitable InfoExtractor for URL %s' % url
)
864 def __handle_extraction_exceptions(func
):
865 def wrapper(self
, *args
, **kwargs
):
867 return func(self
, *args
, **kwargs
)
868 except GeoRestrictedError
as e
:
871 msg
+= '\nThis video is available in %s.' % ', '.join(
872 map(ISO3166Utils
.short2full
, e
.countries
))
873 msg
+= '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
874 self
.report_error(msg
)
875 except ExtractorError
as e
: # An error we somewhat expected
876 self
.report_error(compat_str(e
), e
.format_traceback())
877 except MaxDownloadsReached
:
879 except Exception as e
:
880 if self
.params
.get('ignoreerrors', False):
881 self
.report_error(error_to_compat_str(e
), tb
=encode_compat_str(traceback
.format_exc()))
886 @__handle_extraction_exceptions
887 def __extract_info(self
, url
, ie
, download
, extra_info
, process
, info_dict
):
888 ie_result
= ie
.extract(url
)
889 if ie_result
is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
891 if isinstance(ie_result
, list):
892 # Backwards compatibility: old IE result format
894 '_type': 'compat_list',
895 'entries': ie_result
,
898 if info_dict
.get('id'):
899 ie_result
['id'] = info_dict
['id']
900 if info_dict
.get('title'):
901 ie_result
['title'] = info_dict
['title']
902 self
.add_default_extra_info(ie_result
, ie
, url
)
904 return self
.process_ie_result(ie_result
, download
, extra_info
)
908 def add_default_extra_info(self
, ie_result
, ie
, url
):
909 self
.add_extra_info(ie_result
, {
910 'extractor': ie
.IE_NAME
,
913 formatSeconds(ie_result
['duration'], '-')
914 if ie_result
.get('duration', None) is not None
916 'webpage_url_basename': url_basename(url
),
917 'extractor_key': ie
.ie_key(),
920 def process_ie_result(self
, ie_result
, download
=True, extra_info
={}):
922 Take the result of the ie(may be modified) and resolve all unresolved
923 references (URLs, playlist items).
925 It will also download the videos if 'download'.
926 Returns the resolved ie_result.
928 result_type
= ie_result
.get('_type', 'video')
930 if result_type
in ('url', 'url_transparent'):
931 ie_result
['url'] = sanitize_url(ie_result
['url'])
932 extract_flat
= self
.params
.get('extract_flat', False)
933 if ((extract_flat
== 'in_playlist' and 'playlist' in extra_info
)
934 or extract_flat
is True):
935 self
.__forced
_printings
(
936 ie_result
, self
.prepare_filename(ie_result
),
940 if result_type
== 'video':
941 self
.add_extra_info(ie_result
, extra_info
)
942 return self
.process_video_result(ie_result
, download
=download
)
943 elif result_type
== 'url':
944 # We have to add extra_info to the results because it may be
945 # contained in a playlist
946 return self
.extract_info(ie_result
['url'],
947 download
, info_dict
=ie_result
,
948 ie_key
=ie_result
.get('ie_key'),
949 extra_info
=extra_info
)
950 elif result_type
== 'url_transparent':
951 # Use the information from the embedding page
952 info
= self
.extract_info(
953 ie_result
['url'], ie_key
=ie_result
.get('ie_key'),
954 extra_info
=extra_info
, download
=False, process
=False)
956 # extract_info may return None when ignoreerrors is enabled and
957 # extraction failed with an error, don't crash and return early
962 force_properties
= dict(
963 (k
, v
) for k
, v
in ie_result
.items() if v
is not None)
964 for f
in ('_type', 'url', 'id', 'extractor', 'extractor_key', 'ie_key'):
965 if f
in force_properties
:
966 del force_properties
[f
]
967 new_result
= info
.copy()
968 new_result
.update(force_properties
)
970 # Extracted info may not be a video result (i.e.
971 # info.get('_type', 'video') != video) but rather an url or
972 # url_transparent. In such cases outer metadata (from ie_result)
973 # should be propagated to inner one (info). For this to happen
974 # _type of info should be overridden with url_transparent. This
975 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
976 if new_result
.get('_type') == 'url':
977 new_result
['_type'] = 'url_transparent'
979 return self
.process_ie_result(
980 new_result
, download
=download
, extra_info
=extra_info
)
981 elif result_type
in ('playlist', 'multi_video'):
982 # We process each entry in the playlist
983 playlist
= ie_result
.get('title') or ie_result
.get('id')
984 self
.to_screen('[download] Downloading playlist: %s' % playlist
)
986 playlist_results
= []
988 playliststart
= self
.params
.get('playliststart', 1) - 1
989 playlistend
= self
.params
.get('playlistend')
990 # For backwards compatibility, interpret -1 as whole list
991 if playlistend
== -1:
994 playlistitems_str
= self
.params
.get('playlist_items')
996 if playlistitems_str
is not None:
997 def iter_playlistitems(format
):
998 for string_segment
in format
.split(','):
999 if '-' in string_segment
:
1000 start
, end
= string_segment
.split('-')
1001 for item
in range(int(start
), int(end
) + 1):
1004 yield int(string_segment
)
1005 playlistitems
= orderedSet(iter_playlistitems(playlistitems_str
))
1007 ie_entries
= ie_result
['entries']
1009 def make_playlistitems_entries(list_ie_entries
):
1010 num_entries
= len(list_ie_entries
)
1012 list_ie_entries
[i
- 1] for i
in playlistitems
1013 if -num_entries
<= i
- 1 < num_entries
]
1015 def report_download(num_entries
):
1017 '[%s] playlist %s: Downloading %d videos' %
1018 (ie_result
['extractor'], playlist
, num_entries
))
1020 if isinstance(ie_entries
, list):
1021 n_all_entries
= len(ie_entries
)
1023 entries
= make_playlistitems_entries(ie_entries
)
1025 entries
= ie_entries
[playliststart
:playlistend
]
1026 n_entries
= len(entries
)
1028 '[%s] playlist %s: Collected %d video ids (downloading %d of them)' %
1029 (ie_result
['extractor'], playlist
, n_all_entries
, n_entries
))
1030 elif isinstance(ie_entries
, PagedList
):
1033 for item
in playlistitems
:
1034 entries
.extend(ie_entries
.getslice(
1038 entries
= ie_entries
.getslice(
1039 playliststart
, playlistend
)
1040 n_entries
= len(entries
)
1041 report_download(n_entries
)
1044 entries
= make_playlistitems_entries(list(itertools
.islice(
1045 ie_entries
, 0, max(playlistitems
))))
1047 entries
= list(itertools
.islice(
1048 ie_entries
, playliststart
, playlistend
))
1049 n_entries
= len(entries
)
1050 report_download(n_entries
)
1052 if self
.params
.get('playlistreverse', False):
1053 entries
= entries
[::-1]
1055 if self
.params
.get('playlistrandom', False):
1056 random
.shuffle(entries
)
1058 x_forwarded_for
= ie_result
.get('__x_forwarded_for_ip')
1060 for i
, entry
in enumerate(entries
, 1):
1061 self
.to_screen('[download] Downloading video %s of %s' % (i
, n_entries
))
1062 # This __x_forwarded_for_ip thing is a bit ugly but requires
1065 entry
['__x_forwarded_for_ip'] = x_forwarded_for
1067 'n_entries': n_entries
,
1068 'playlist': playlist
,
1069 'playlist_id': ie_result
.get('id'),
1070 'playlist_title': ie_result
.get('title'),
1071 'playlist_uploader': ie_result
.get('uploader'),
1072 'playlist_uploader_id': ie_result
.get('uploader_id'),
1073 'playlist_index': playlistitems
[i
- 1] if playlistitems
else i
+ playliststart
,
1074 'extractor': ie_result
['extractor'],
1075 'webpage_url': ie_result
['webpage_url'],
1076 'webpage_url_basename': url_basename(ie_result
['webpage_url']),
1077 'extractor_key': ie_result
['extractor_key'],
1080 reason
= self
._match
_entry
(entry
, incomplete
=True)
1081 if reason
is not None:
1082 if reason
.endswith('has already been recorded in the archive') and self
.params
.get('break_on_existing'):
1083 print('[download] tried downloading a file that\'s already in the archive, stopping since --break-on-existing is set.')
1086 self
.to_screen('[download] ' + reason
)
1089 entry_result
= self
.__process
_iterable
_entry
(entry
, download
, extra
)
1090 # TODO: skip failed (empty) entries?
1091 playlist_results
.append(entry_result
)
1092 ie_result
['entries'] = playlist_results
1093 self
.to_screen('[download] Finished downloading playlist: %s' % playlist
)
1095 elif result_type
== 'compat_list':
1096 self
.report_warning(
1097 'Extractor %s returned a compat_list result. '
1098 'It needs to be updated.' % ie_result
.get('extractor'))
1101 self
.add_extra_info(
1104 'extractor': ie_result
['extractor'],
1105 'webpage_url': ie_result
['webpage_url'],
1106 'webpage_url_basename': url_basename(ie_result
['webpage_url']),
1107 'extractor_key': ie_result
['extractor_key'],
1111 ie_result
['entries'] = [
1112 self
.process_ie_result(_fixup(r
), download
, extra_info
)
1113 for r
in ie_result
['entries']
1117 raise Exception('Invalid result type: %s' % result_type
)
1119 @__handle_extraction_exceptions
1120 def __process_iterable_entry(self
, entry
, download
, extra_info
):
1121 return self
.process_ie_result(
1122 entry
, download
=download
, extra_info
=extra_info
)
1124 def _build_format_filter(self
, filter_spec
):
1125 " Returns a function to filter the formats according to the filter_spec "
1135 operator_rex
= re
.compile(r
'''(?x)\s*
1136 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)
1137 \s*(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1138 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)
1140 ''' % '|'.join(map(re
.escape
, OPERATORS
.keys())))
1141 m
= operator_rex
.search(filter_spec
)
1144 comparison_value
= int(m
.group('value'))
1146 comparison_value
= parse_filesize(m
.group('value'))
1147 if comparison_value
is None:
1148 comparison_value
= parse_filesize(m
.group('value') + 'B')
1149 if comparison_value
is None:
1151 'Invalid value %r in format specification %r' % (
1152 m
.group('value'), filter_spec
))
1153 op
= OPERATORS
[m
.group('op')]
1158 '^=': lambda attr
, value
: attr
.startswith(value
),
1159 '$=': lambda attr
, value
: attr
.endswith(value
),
1160 '*=': lambda attr
, value
: value
in attr
,
1162 str_operator_rex
= re
.compile(r
'''(?x)
1163 \s*(?P<key>[a-zA-Z0-9._-]+)
1164 \s*(?P<negation>!\s*)?(?P<op>%s)(?P<none_inclusive>\s*\?)?
1165 \s*(?P<value>[a-zA-Z0-9._-]+)
1167 ''' % '|'.join(map(re
.escape
, STR_OPERATORS
.keys())))
1168 m
= str_operator_rex
.search(filter_spec
)
1170 comparison_value
= m
.group('value')
1171 str_op
= STR_OPERATORS
[m
.group('op')]
1172 if m
.group('negation'):
1173 op
= lambda attr
, value
: not str_op(attr
, value
)
1178 raise ValueError('Invalid filter specification %r' % filter_spec
)
1181 actual_value
= f
.get(m
.group('key'))
1182 if actual_value
is None:
1183 return m
.group('none_inclusive')
1184 return op(actual_value
, comparison_value
)
1187 def _default_format_spec(self
, info_dict
, download
=True):
1190 merger
= FFmpegMergerPP(self
)
1191 return merger
.available
and merger
.can_merge()
1194 not self
.params
.get('simulate', False)
1198 or info_dict
.get('is_live', False)
1199 or self
.params
.get('outtmpl', DEFAULT_OUTTMPL
) == '-'))
1202 'best/bestvideo+bestaudio'
1204 else 'bestvideo*+bestaudio/best'
1205 if not self
.params
.get('allow_multiple_audio_streams', False)
1206 else 'bestvideo+bestaudio/best')
1208 def build_format_selector(self
, format_spec
):
1209 def syntax_error(note
, start
):
1211 'Invalid format specification: '
1212 '{0}\n\t{1}\n\t{2}^'.format(note
, format_spec
, ' ' * start
[1]))
1213 return SyntaxError(message
)
1215 PICKFIRST
= 'PICKFIRST'
1219 FormatSelector
= collections
.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
1221 allow_multiple_streams
= {'audio': self
.params
.get('allow_multiple_audio_streams', False),
1222 'video': self
.params
.get('allow_multiple_video_streams', False)}
1224 def _parse_filter(tokens
):
1226 for type, string
, start
, _
, _
in tokens
:
1227 if type == tokenize
.OP
and string
== ']':
1228 return ''.join(filter_parts
)
1230 filter_parts
.append(string
)
1232 def _remove_unused_ops(tokens
):
1233 # Remove operators that we don't use and join them with the surrounding strings
1234 # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
1235 ALLOWED_OPS
= ('/', '+', ',', '(', ')')
1236 last_string
, last_start
, last_end
, last_line
= None, None, None, None
1237 for type, string
, start
, end
, line
in tokens
:
1238 if type == tokenize
.OP
and string
== '[':
1240 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
1242 yield type, string
, start
, end
, line
1243 # everything inside brackets will be handled by _parse_filter
1244 for type, string
, start
, end
, line
in tokens
:
1245 yield type, string
, start
, end
, line
1246 if type == tokenize
.OP
and string
== ']':
1248 elif type == tokenize
.OP
and string
in ALLOWED_OPS
:
1250 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
1252 yield type, string
, start
, end
, line
1253 elif type in [tokenize
.NAME
, tokenize
.NUMBER
, tokenize
.OP
]:
1255 last_string
= string
1259 last_string
+= string
1261 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
1263 def _parse_format_selection(tokens
, inside_merge
=False, inside_choice
=False, inside_group
=False):
1265 current_selector
= None
1266 for type, string
, start
, _
, _
in tokens
:
1267 # ENCODING is only defined in python 3.x
1268 if type == getattr(tokenize
, 'ENCODING', None):
1270 elif type in [tokenize
.NAME
, tokenize
.NUMBER
]:
1271 current_selector
= FormatSelector(SINGLE
, string
, [])
1272 elif type == tokenize
.OP
:
1274 if not inside_group
:
1275 # ')' will be handled by the parentheses group
1276 tokens
.restore_last_token()
1278 elif inside_merge
and string
in ['/', ',']:
1279 tokens
.restore_last_token()
1281 elif inside_choice
and string
== ',':
1282 tokens
.restore_last_token()
1285 if not current_selector
:
1286 raise syntax_error('"," must follow a format selector', start
)
1287 selectors
.append(current_selector
)
1288 current_selector
= None
1290 if not current_selector
:
1291 raise syntax_error('"/" must follow a format selector', start
)
1292 first_choice
= current_selector
1293 second_choice
= _parse_format_selection(tokens
, inside_choice
=True)
1294 current_selector
= FormatSelector(PICKFIRST
, (first_choice
, second_choice
), [])
1296 if not current_selector
:
1297 current_selector
= FormatSelector(SINGLE
, 'best', [])
1298 format_filter
= _parse_filter(tokens
)
1299 current_selector
.filters
.append(format_filter
)
1301 if current_selector
:
1302 raise syntax_error('Unexpected "("', start
)
1303 group
= _parse_format_selection(tokens
, inside_group
=True)
1304 current_selector
= FormatSelector(GROUP
, group
, [])
1306 if not current_selector
:
1307 raise syntax_error('Unexpected "+"', start
)
1308 selector_1
= current_selector
1309 selector_2
= _parse_format_selection(tokens
, inside_merge
=True)
1311 raise syntax_error('Expected a selector', start
)
1312 current_selector
= FormatSelector(MERGE
, (selector_1
, selector_2
), [])
1314 raise syntax_error('Operator not recognized: "{0}"'.format(string
), start
)
1315 elif type == tokenize
.ENDMARKER
:
1317 if current_selector
:
1318 selectors
.append(current_selector
)
1321 def _build_selector_function(selector
):
1322 if isinstance(selector
, list): # ,
1323 fs
= [_build_selector_function(s
) for s
in selector
]
1325 def selector_function(ctx
):
1327 for format
in f(ctx
):
1329 return selector_function
1331 elif selector
.type == GROUP
: # ()
1332 selector_function
= _build_selector_function(selector
.selector
)
1334 elif selector
.type == PICKFIRST
: # /
1335 fs
= [_build_selector_function(s
) for s
in selector
.selector
]
1337 def selector_function(ctx
):
1339 picked_formats
= list(f(ctx
))
1341 return picked_formats
1344 elif selector
.type == SINGLE
: # atom
1345 format_spec
= selector
.selector
if selector
.selector
is not None else 'best'
1347 if format_spec
== 'all':
1348 def selector_function(ctx
):
1349 formats
= list(ctx
['formats'])
1355 format_fallback
= False
1356 format_spec_obj
= re
.match(r
'(best|worst|b|w)(video|audio|v|a)?(\*)?$', format_spec
)
1357 if format_spec_obj
is not None:
1358 format_idx
= 0 if format_spec_obj
.group(1)[0] == 'w' else -1
1359 format_type
= format_spec_obj
.group(2)[0] if format_spec_obj
.group(2) else False
1360 not_format_type
= 'v' if format_type
== 'a' else 'a'
1361 format_modified
= format_spec_obj
.group(3) is not None
1363 format_fallback
= not format_type
and not format_modified
# for b, w
1364 filter_f
= ((lambda f
: f
.get(format_type
+ 'codec') != 'none')
1365 if format_type
and format_modified
# bv*, ba*, wv*, wa*
1366 else (lambda f
: f
.get(not_format_type
+ 'codec') == 'none')
1367 if format_type
# bv, ba, wv, wa
1368 else (lambda f
: f
.get('vcodec') != 'none' and f
.get('acodec') != 'none')
1369 if not format_modified
# b, w
1373 filter_f
= ((lambda f
: f
.get('ext') == format_spec
)
1374 if format_spec
in ['mp4', 'flv', 'webm', '3gp', 'm4a', 'mp3', 'ogg', 'aac', 'wav'] # extension
1375 else (lambda f
: f
.get('format_id') == format_spec
)) # id
1377 def selector_function(ctx
):
1378 formats
= list(ctx
['formats'])
1381 matches
= list(filter(filter_f
, formats
)) if filter_f
is not None else formats
1383 yield matches
[format_idx
]
1384 elif format_fallback
== 'force' or (format_fallback
and ctx
['incomplete_formats']):
1385 # for extractors with incomplete formats (audio only (soundcloud)
1386 # or video only (imgur)) best/worst will fallback to
1387 # best/worst {video,audio}-only format
1388 yield formats
[format_idx
]
1390 elif selector
.type == MERGE
: # +
1391 def _merge(formats_pair
):
1392 format_1
, format_2
= formats_pair
1395 formats_info
.extend(format_1
.get('requested_formats', (format_1
,)))
1396 formats_info
.extend(format_2
.get('requested_formats', (format_2
,)))
1398 if not allow_multiple_streams
['video'] or not allow_multiple_streams
['audio']:
1399 get_no_more
= {"video": False, "audio": False}
1400 for (i
, fmt_info
) in enumerate(formats_info
):
1401 for aud_vid
in ["audio", "video"]:
1402 if not allow_multiple_streams
[aud_vid
] and fmt_info
.get(aud_vid
[0] + 'codec') != 'none':
1403 if get_no_more
[aud_vid
]:
1405 get_no_more
[aud_vid
] = True
1407 if len(formats_info
) == 1:
1408 return formats_info
[0]
1410 video_fmts
= [fmt_info
for fmt_info
in formats_info
if fmt_info
.get('vcodec') != 'none']
1411 audio_fmts
= [fmt_info
for fmt_info
in formats_info
if fmt_info
.get('acodec') != 'none']
1413 the_only_video
= video_fmts
[0] if len(video_fmts
) == 1 else None
1414 the_only_audio
= audio_fmts
[0] if len(audio_fmts
) == 1 else None
1416 output_ext
= self
.params
.get('merge_output_format')
1419 output_ext
= the_only_video
['ext']
1420 elif the_only_audio
and not video_fmts
:
1421 output_ext
= the_only_audio
['ext']
1426 'requested_formats': formats_info
,
1427 'format': '+'.join(fmt_info
.get('format') for fmt_info
in formats_info
),
1428 'format_id': '+'.join(fmt_info
.get('format_id') for fmt_info
in formats_info
),
1434 'width': the_only_video
.get('width'),
1435 'height': the_only_video
.get('height'),
1436 'resolution': the_only_video
.get('resolution'),
1437 'fps': the_only_video
.get('fps'),
1438 'vcodec': the_only_video
.get('vcodec'),
1439 'vbr': the_only_video
.get('vbr'),
1440 'stretched_ratio': the_only_video
.get('stretched_ratio'),
1445 'acodec': the_only_audio
.get('acodec'),
1446 'abr': the_only_audio
.get('abr'),
1451 selector_1
, selector_2
= map(_build_selector_function
, selector
.selector
)
1453 def selector_function(ctx
):
1454 for pair
in itertools
.product(
1455 selector_1(copy
.deepcopy(ctx
)), selector_2(copy
.deepcopy(ctx
))):
1458 filters
= [self
._build
_format
_filter
(f
) for f
in selector
.filters
]
1460 def final_selector(ctx
):
1461 ctx_copy
= copy
.deepcopy(ctx
)
1462 for _filter
in filters
:
1463 ctx_copy
['formats'] = list(filter(_filter
, ctx_copy
['formats']))
1464 return selector_function(ctx_copy
)
1465 return final_selector
1467 stream
= io
.BytesIO(format_spec
.encode('utf-8'))
1469 tokens
= list(_remove_unused_ops(compat_tokenize_tokenize(stream
.readline
)))
1470 except tokenize
.TokenError
:
1471 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec
)))
1473 class TokenIterator(object):
1474 def __init__(self
, tokens
):
1475 self
.tokens
= tokens
1482 if self
.counter
>= len(self
.tokens
):
1483 raise StopIteration()
1484 value
= self
.tokens
[self
.counter
]
1490 def restore_last_token(self
):
1493 parsed_selector
= _parse_format_selection(iter(TokenIterator(tokens
)))
1494 return _build_selector_function(parsed_selector
)
1496 def _calc_headers(self
, info_dict
):
1497 res
= std_headers
.copy()
1499 add_headers
= info_dict
.get('http_headers')
1501 res
.update(add_headers
)
1503 cookies
= self
._calc
_cookies
(info_dict
)
1505 res
['Cookie'] = cookies
1507 if 'X-Forwarded-For' not in res
:
1508 x_forwarded_for_ip
= info_dict
.get('__x_forwarded_for_ip')
1509 if x_forwarded_for_ip
:
1510 res
['X-Forwarded-For'] = x_forwarded_for_ip
1514 def _calc_cookies(self
, info_dict
):
1515 pr
= sanitized_Request(info_dict
['url'])
1516 self
.cookiejar
.add_cookie_header(pr
)
1517 return pr
.get_header('Cookie')
1519 def process_video_result(self
, info_dict
, download
=True):
1520 assert info_dict
.get('_type', 'video') == 'video'
1522 if 'id' not in info_dict
:
1523 raise ExtractorError('Missing "id" field in extractor result')
1524 if 'title' not in info_dict
:
1525 raise ExtractorError('Missing "title" field in extractor result')
1527 def report_force_conversion(field
, field_not
, conversion
):
1528 self
.report_warning(
1529 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
1530 % (field
, field_not
, conversion
))
1532 def sanitize_string_field(info
, string_field
):
1533 field
= info
.get(string_field
)
1534 if field
is None or isinstance(field
, compat_str
):
1536 report_force_conversion(string_field
, 'a string', 'string')
1537 info
[string_field
] = compat_str(field
)
1539 def sanitize_numeric_fields(info
):
1540 for numeric_field
in self
._NUMERIC
_FIELDS
:
1541 field
= info
.get(numeric_field
)
1542 if field
is None or isinstance(field
, compat_numeric_types
):
1544 report_force_conversion(numeric_field
, 'numeric', 'int')
1545 info
[numeric_field
] = int_or_none(field
)
1547 sanitize_string_field(info_dict
, 'id')
1548 sanitize_numeric_fields(info_dict
)
1550 if 'playlist' not in info_dict
:
1551 # It isn't part of a playlist
1552 info_dict
['playlist'] = None
1553 info_dict
['playlist_index'] = None
1555 thumbnails
= info_dict
.get('thumbnails')
1556 if thumbnails
is None:
1557 thumbnail
= info_dict
.get('thumbnail')
1559 info_dict
['thumbnails'] = thumbnails
= [{'url': thumbnail}
]
1561 thumbnails
.sort(key
=lambda t
: (
1562 t
.get('preference') if t
.get('preference') is not None else -1,
1563 t
.get('width') if t
.get('width') is not None else -1,
1564 t
.get('height') if t
.get('height') is not None else -1,
1565 t
.get('id') if t
.get('id') is not None else '', t
.get('url')))
1566 for i
, t
in enumerate(thumbnails
):
1567 t
['url'] = sanitize_url(t
['url'])
1568 if t
.get('width') and t
.get('height'):
1569 t
['resolution'] = '%dx%d' % (t
['width'], t
['height'])
1570 if t
.get('id') is None:
1573 if self
.params
.get('list_thumbnails'):
1574 self
.list_thumbnails(info_dict
)
1577 thumbnail
= info_dict
.get('thumbnail')
1579 info_dict
['thumbnail'] = sanitize_url(thumbnail
)
1581 info_dict
['thumbnail'] = thumbnails
[-1]['url']
1583 if 'display_id' not in info_dict
and 'id' in info_dict
:
1584 info_dict
['display_id'] = info_dict
['id']
1586 if info_dict
.get('upload_date') is None and info_dict
.get('timestamp') is not None:
1587 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
1588 # see http://bugs.python.org/issue1646728)
1590 upload_date
= datetime
.datetime
.utcfromtimestamp(info_dict
['timestamp'])
1591 info_dict
['upload_date'] = upload_date
.strftime('%Y%m%d')
1592 except (ValueError, OverflowError, OSError):
1595 # Auto generate title fields corresponding to the *_number fields when missing
1596 # in order to always have clean titles. This is very common for TV series.
1597 for field
in ('chapter', 'season', 'episode'):
1598 if info_dict
.get('%s_number' % field
) is not None and not info_dict
.get(field
):
1599 info_dict
[field
] = '%s %d' % (field
.capitalize(), info_dict
['%s_number' % field
])
1601 for cc_kind
in ('subtitles', 'automatic_captions'):
1602 cc
= info_dict
.get(cc_kind
)
1604 for _
, subtitle
in cc
.items():
1605 for subtitle_format
in subtitle
:
1606 if subtitle_format
.get('url'):
1607 subtitle_format
['url'] = sanitize_url(subtitle_format
['url'])
1608 if subtitle_format
.get('ext') is None:
1609 subtitle_format
['ext'] = determine_ext(subtitle_format
['url']).lower()
1611 automatic_captions
= info_dict
.get('automatic_captions')
1612 subtitles
= info_dict
.get('subtitles')
1614 if self
.params
.get('listsubtitles', False):
1615 if 'automatic_captions' in info_dict
:
1616 self
.list_subtitles(
1617 info_dict
['id'], automatic_captions
, 'automatic captions')
1618 self
.list_subtitles(info_dict
['id'], subtitles
, 'subtitles')
1621 info_dict
['requested_subtitles'] = self
.process_subtitles(
1622 info_dict
['id'], subtitles
, automatic_captions
)
1624 # We now pick which formats have to be downloaded
1625 if info_dict
.get('formats') is None:
1626 # There's only one format available
1627 formats
= [info_dict
]
1629 formats
= info_dict
['formats']
1632 raise ExtractorError('No video formats found!')
1634 def is_wellformed(f
):
1637 self
.report_warning(
1638 '"url" field is missing or empty - skipping format, '
1639 'there is an error in extractor')
1641 if isinstance(url
, bytes):
1642 sanitize_string_field(f
, 'url')
1645 # Filter out malformed formats for better extraction robustness
1646 formats
= list(filter(is_wellformed
, formats
))
1650 # We check that all the formats have the format and format_id fields
1651 for i
, format
in enumerate(formats
):
1652 sanitize_string_field(format
, 'format_id')
1653 sanitize_numeric_fields(format
)
1654 format
['url'] = sanitize_url(format
['url'])
1655 if not format
.get('format_id'):
1656 format
['format_id'] = compat_str(i
)
1658 # Sanitize format_id from characters used in format selector expression
1659 format
['format_id'] = re
.sub(r
'[\s,/+\[\]()]', '_', format
['format_id'])
1660 format_id
= format
['format_id']
1661 if format_id
not in formats_dict
:
1662 formats_dict
[format_id
] = []
1663 formats_dict
[format_id
].append(format
)
1665 # Make sure all formats have unique format_id
1666 for format_id
, ambiguous_formats
in formats_dict
.items():
1667 if len(ambiguous_formats
) > 1:
1668 for i
, format
in enumerate(ambiguous_formats
):
1669 format
['format_id'] = '%s-%d' % (format_id
, i
)
1671 for i
, format
in enumerate(formats
):
1672 if format
.get('format') is None:
1673 format
['format'] = '{id} - {res}{note}'.format(
1674 id=format
['format_id'],
1675 res
=self
.format_resolution(format
),
1676 note
=' ({0})'.format(format
['format_note']) if format
.get('format_note') is not None else '',
1678 # Automatically determine file extension if missing
1679 if format
.get('ext') is None:
1680 format
['ext'] = determine_ext(format
['url']).lower()
1681 # Automatically determine protocol if missing (useful for format
1682 # selection purposes)
1683 if format
.get('protocol') is None:
1684 format
['protocol'] = determine_protocol(format
)
1685 # Add HTTP headers, so that external programs can use them from the
1687 full_format_info
= info_dict
.copy()
1688 full_format_info
.update(format
)
1689 format
['http_headers'] = self
._calc
_headers
(full_format_info
)
1690 # Remove private housekeeping stuff
1691 if '__x_forwarded_for_ip' in info_dict
:
1692 del info_dict
['__x_forwarded_for_ip']
1694 # TODO Central sorting goes here
1696 if formats
[0] is not info_dict
:
1697 # only set the 'formats' fields if the original info_dict list them
1698 # otherwise we end up with a circular reference, the first (and unique)
1699 # element in the 'formats' field in info_dict is info_dict itself,
1700 # which can't be exported to json
1701 info_dict
['formats'] = formats
1702 if self
.params
.get('listformats'):
1703 self
.list_formats(info_dict
)
1706 req_format
= self
.params
.get('format')
1707 if req_format
is None:
1708 req_format
= self
._default
_format
_spec
(info_dict
, download
=download
)
1709 if self
.params
.get('verbose'):
1710 self
._write
_string
('[debug] Default format spec: %s\n' % req_format
)
1712 format_selector
= self
.build_format_selector(req_format
)
1714 # While in format selection we may need to have an access to the original
1715 # format set in order to calculate some metrics or do some processing.
1716 # For now we need to be able to guess whether original formats provided
1717 # by extractor are incomplete or not (i.e. whether extractor provides only
1718 # video-only or audio-only formats) for proper formats selection for
1719 # extractors with such incomplete formats (see
1720 # https://github.com/ytdl-org/youtube-dl/pull/5556).
1721 # Since formats may be filtered during format selection and may not match
1722 # the original formats the results may be incorrect. Thus original formats
1723 # or pre-calculated metrics should be passed to format selection routines
1725 # We will pass a context object containing all necessary additional data
1726 # instead of just formats.
1727 # This fixes incorrect format selection issue (see
1728 # https://github.com/ytdl-org/youtube-dl/issues/10083).
1729 incomplete_formats
= (
1730 # All formats are video-only or
1731 all(f
.get('vcodec') != 'none' and f
.get('acodec') == 'none' for f
in formats
)
1732 # all formats are audio-only
1733 or all(f
.get('vcodec') == 'none' and f
.get('acodec') != 'none' for f
in formats
))
1737 'incomplete_formats': incomplete_formats
,
1740 formats_to_download
= list(format_selector(ctx
))
1741 if not formats_to_download
:
1742 raise ExtractorError('requested format not available',
1746 self
.to_screen('[info] Downloading format(s) %s' % ", ".join([f
['format_id'] for f
in formats_to_download
]))
1747 if len(formats_to_download
) > 1:
1748 self
.to_screen('[info] %s: downloading video in %s formats' % (info_dict
['id'], len(formats_to_download
)))
1749 for format
in formats_to_download
:
1750 new_info
= dict(info_dict
)
1751 new_info
.update(format
)
1752 self
.process_info(new_info
)
1753 # We update the info dict with the best quality format (backwards compatibility)
1754 info_dict
.update(formats_to_download
[-1])
1757 def process_subtitles(self
, video_id
, normal_subtitles
, automatic_captions
):
1758 """Select the requested subtitles and their format"""
1760 if normal_subtitles
and self
.params
.get('writesubtitles'):
1761 available_subs
.update(normal_subtitles
)
1762 if automatic_captions
and self
.params
.get('writeautomaticsub'):
1763 for lang
, cap_info
in automatic_captions
.items():
1764 if lang
not in available_subs
:
1765 available_subs
[lang
] = cap_info
1767 if (not self
.params
.get('writesubtitles') and not
1768 self
.params
.get('writeautomaticsub') or not
1772 if self
.params
.get('allsubtitles', False):
1773 requested_langs
= available_subs
.keys()
1775 if self
.params
.get('subtitleslangs', False):
1776 requested_langs
= self
.params
.get('subtitleslangs')
1777 elif 'en' in available_subs
:
1778 requested_langs
= ['en']
1780 requested_langs
= [list(available_subs
.keys())[0]]
1782 formats_query
= self
.params
.get('subtitlesformat', 'best')
1783 formats_preference
= formats_query
.split('/') if formats_query
else []
1785 for lang
in requested_langs
:
1786 formats
= available_subs
.get(lang
)
1788 self
.report_warning('%s subtitles not available for %s' % (lang
, video_id
))
1790 for ext
in formats_preference
:
1794 matches
= list(filter(lambda f
: f
['ext'] == ext
, formats
))
1800 self
.report_warning(
1801 'No subtitle format found matching "%s" for language %s, '
1802 'using %s' % (formats_query
, lang
, f
['ext']))
1806 def __forced_printings(self
, info_dict
, filename
, incomplete
):
1807 def print_mandatory(field
):
1808 if (self
.params
.get('force%s' % field
, False)
1809 and (not incomplete
or info_dict
.get(field
) is not None)):
1810 self
.to_stdout(info_dict
[field
])
1812 def print_optional(field
):
1813 if (self
.params
.get('force%s' % field
, False)
1814 and info_dict
.get(field
) is not None):
1815 self
.to_stdout(info_dict
[field
])
1817 print_mandatory('title')
1818 print_mandatory('id')
1819 if self
.params
.get('forceurl', False) and not incomplete
:
1820 if info_dict
.get('requested_formats') is not None:
1821 for f
in info_dict
['requested_formats']:
1822 self
.to_stdout(f
['url'] + f
.get('play_path', ''))
1824 # For RTMP URLs, also include the playpath
1825 self
.to_stdout(info_dict
['url'] + info_dict
.get('play_path', ''))
1826 print_optional('thumbnail')
1827 print_optional('description')
1828 if self
.params
.get('forcefilename', False) and filename
is not None:
1829 self
.to_stdout(filename
)
1830 if self
.params
.get('forceduration', False) and info_dict
.get('duration') is not None:
1831 self
.to_stdout(formatSeconds(info_dict
['duration']))
1832 print_mandatory('format')
1833 if self
.params
.get('forcejson', False):
1834 self
.to_stdout(json
.dumps(info_dict
))
1836 def process_info(self
, info_dict
):
1837 """Process a single resolved IE result."""
1839 assert info_dict
.get('_type', 'video') == 'video'
1841 max_downloads
= self
.params
.get('max_downloads')
1842 if max_downloads
is not None:
1843 if self
._num
_downloads
>= int(max_downloads
):
1844 raise MaxDownloadsReached()
1846 # TODO: backward compatibility, to be removed
1847 info_dict
['fulltitle'] = info_dict
['title']
1849 if 'format' not in info_dict
:
1850 info_dict
['format'] = info_dict
['ext']
1852 reason
= self
._match
_entry
(info_dict
, incomplete
=False)
1853 if reason
is not None:
1854 self
.to_screen('[download] ' + reason
)
1857 self
._num
_downloads
+= 1
1859 info_dict
['_filename'] = filename
= self
.prepare_filename(info_dict
)
1862 self
.__forced
_printings
(info_dict
, filename
, incomplete
=False)
1864 if self
.params
.get('simulate', False):
1865 if self
.params
.get('force_write_download_archive', False):
1866 self
.record_download_archive(info_dict
)
1868 # Do nothing else if in simulate mode
1871 if filename
is None:
1874 def ensure_dir_exists(path
):
1876 dn
= os
.path
.dirname(path
)
1877 if dn
and not os
.path
.exists(dn
):
1880 except (OSError, IOError) as err
:
1881 self
.report_error('unable to create directory ' + error_to_compat_str(err
))
1884 if not ensure_dir_exists(sanitize_path(encodeFilename(filename
))):
1887 if self
.params
.get('writedescription', False):
1888 descfn
= replace_extension(filename
, 'description', info_dict
.get('ext'))
1889 if self
.params
.get('nooverwrites', False) and os
.path
.exists(encodeFilename(descfn
)):
1890 self
.to_screen('[info] Video description is already present')
1891 elif info_dict
.get('description') is None:
1892 self
.report_warning('There\'s no description to write.')
1895 self
.to_screen('[info] Writing video description to: ' + descfn
)
1896 with io
.open(encodeFilename(descfn
), 'w', encoding
='utf-8') as descfile
:
1897 descfile
.write(info_dict
['description'])
1898 except (OSError, IOError):
1899 self
.report_error('Cannot write description file ' + descfn
)
1902 if self
.params
.get('writeannotations', False):
1903 annofn
= replace_extension(filename
, 'annotations.xml', info_dict
.get('ext'))
1904 if self
.params
.get('nooverwrites', False) and os
.path
.exists(encodeFilename(annofn
)):
1905 self
.to_screen('[info] Video annotations are already present')
1906 elif not info_dict
.get('annotations'):
1907 self
.report_warning('There are no annotations to write.')
1910 self
.to_screen('[info] Writing video annotations to: ' + annofn
)
1911 with io
.open(encodeFilename(annofn
), 'w', encoding
='utf-8') as annofile
:
1912 annofile
.write(info_dict
['annotations'])
1913 except (KeyError, TypeError):
1914 self
.report_warning('There are no annotations to write.')
1915 except (OSError, IOError):
1916 self
.report_error('Cannot write annotations file: ' + annofn
)
1919 def dl(name
, info
, subtitle
=False):
1920 fd
= get_suitable_downloader(info
, self
.params
)(self
, self
.params
)
1921 for ph
in self
._progress
_hooks
:
1922 fd
.add_progress_hook(ph
)
1923 if self
.params
.get('verbose'):
1924 self
.to_screen('[debug] Invoking downloader on %r' % info
.get('url'))
1925 return fd
.download(name
, info
, subtitle
)
1927 subtitles_are_requested
= any([self
.params
.get('writesubtitles', False),
1928 self
.params
.get('writeautomaticsub')])
1930 if subtitles_are_requested
and info_dict
.get('requested_subtitles'):
1931 # subtitles download errors are already managed as troubles in relevant IE
1932 # that way it will silently go on when used with unsupporting IE
1933 subtitles
= info_dict
['requested_subtitles']
1934 # ie = self.get_info_extractor(info_dict['extractor_key'])
1935 for sub_lang
, sub_info
in subtitles
.items():
1936 sub_format
= sub_info
['ext']
1937 sub_filename
= subtitles_filename(filename
, sub_lang
, sub_format
, info_dict
.get('ext'))
1938 if self
.params
.get('nooverwrites', False) and os
.path
.exists(encodeFilename(sub_filename
)):
1939 self
.to_screen('[info] Video subtitle %s.%s is already present' % (sub_lang
, sub_format
))
1941 self
.to_screen('[info] Writing video subtitles to: ' + sub_filename
)
1942 if sub_info
.get('data') is not None:
1944 # Use newline='' to prevent conversion of newline characters
1945 # See https://github.com/ytdl-org/youtube-dl/issues/10268
1946 with io
.open(encodeFilename(sub_filename
), 'w', encoding
='utf-8', newline
='') as subfile
:
1947 subfile
.write(sub_info
['data'])
1948 except (OSError, IOError):
1949 self
.report_error('Cannot write subtitles file ' + sub_filename
)
1953 dl(sub_filename
, sub_info
, subtitle
=True)
1955 if self.params.get('sleep_interval_subtitles', False):
1956 dl(sub_filename, sub_info)
1958 sub_data = ie._request_webpage(
1959 sub_info['url'], info_dict['id'], note=False).read()
1960 with io.open(encodeFilename(sub_filename), 'wb') as subfile:
1961 subfile.write(sub_data)
1963 except (ExtractorError
, IOError, OSError, ValueError, compat_urllib_error
.URLError
, compat_http_client
.HTTPException
, socket
.error
) as err
:
1964 self
.report_warning('Unable to download subtitle for "%s": %s' %
1965 (sub_lang
, error_to_compat_str(err
)))
1968 if self
.params
.get('skip_download', False):
1969 if self
.params
.get('convertsubtitles', False):
1970 subconv
= FFmpegSubtitlesConvertorPP(self
, format
=self
.params
.get('convertsubtitles'))
1971 filename_real_ext
= os
.path
.splitext(filename
)[1][1:]
1973 os
.path
.splitext(filename
)[0]
1974 if filename_real_ext
== info_dict
['ext']
1976 afilename
= '%s.%s' % (filename_wo_ext
, self
.params
.get('convertsubtitles'))
1977 if subconv
.available
:
1978 info_dict
.setdefault('__postprocessors', [])
1979 # info_dict['__postprocessors'].append(subconv)
1980 if os
.path
.exists(encodeFilename(afilename
)):
1982 '[download] %s has already been downloaded and '
1983 'converted' % afilename
)
1986 self
.post_process(filename
, info_dict
)
1987 except (PostProcessingError
) as err
:
1988 self
.report_error('postprocessing: %s' % str(err
))
1991 if self
.params
.get('writeinfojson', False):
1992 infofn
= replace_extension(filename
, 'info.json', info_dict
.get('ext'))
1993 if self
.params
.get('nooverwrites', False) and os
.path
.exists(encodeFilename(infofn
)):
1994 self
.to_screen('[info] Video description metadata is already present')
1996 self
.to_screen('[info] Writing video description metadata as JSON to: ' + infofn
)
1998 write_json_file(self
.filter_requested_info(info_dict
), infofn
)
1999 except (OSError, IOError):
2000 self
.report_error('Cannot write metadata to JSON file ' + infofn
)
2003 self
._write
_thumbnails
(info_dict
, filename
)
2005 # Write internet shortcut files
2006 url_link
= webloc_link
= desktop_link
= False
2007 if self
.params
.get('writelink', False):
2008 if sys
.platform
== "darwin": # macOS.
2010 elif sys
.platform
.startswith("linux"):
2012 else: # if sys.platform in ['win32', 'cygwin']:
2014 if self
.params
.get('writeurllink', False):
2016 if self
.params
.get('writewebloclink', False):
2018 if self
.params
.get('writedesktoplink', False):
2021 if url_link
or webloc_link
or desktop_link
:
2022 if 'webpage_url' not in info_dict
:
2023 self
.report_error('Cannot write internet shortcut file because the "webpage_url" field is missing in the media information')
2025 ascii_url
= iri_to_uri(info_dict
['webpage_url'])
2027 def _write_link_file(extension
, template
, newline
, embed_filename
):
2028 linkfn
= replace_extension(filename
, extension
, info_dict
.get('ext'))
2029 if self
.params
.get('nooverwrites', False) and os
.path
.exists(encodeFilename(linkfn
)):
2030 self
.to_screen('[info] Internet shortcut is already present')
2033 self
.to_screen('[info] Writing internet shortcut to: ' + linkfn
)
2034 with io
.open(encodeFilename(to_high_limit_path(linkfn
)), 'w', encoding
='utf-8', newline
=newline
) as linkfile
:
2035 template_vars
= {'url': ascii_url}
2037 template_vars
['filename'] = linkfn
[:-(len(extension
) + 1)]
2038 linkfile
.write(template
% template_vars
)
2039 except (OSError, IOError):
2040 self
.report_error('Cannot write internet shortcut ' + linkfn
)
2045 if not _write_link_file('url', DOT_URL_LINK_TEMPLATE
, '\r\n', embed_filename
=False):
2048 if not _write_link_file('webloc', DOT_WEBLOC_LINK_TEMPLATE
, '\n', embed_filename
=False):
2051 if not _write_link_file('desktop', DOT_DESKTOP_LINK_TEMPLATE
, '\n', embed_filename
=True):
2055 must_record_download_archive
= False
2056 if not self
.params
.get('skip_download', False):
2058 if info_dict
.get('requested_formats') is not None:
2061 merger
= FFmpegMergerPP(self
)
2062 if not merger
.available
:
2064 self
.report_warning('You have requested multiple '
2065 'formats but ffmpeg or avconv are not installed.'
2066 ' The formats won\'t be merged.')
2068 postprocessors
= [merger
]
2070 def compatible_formats(formats
):
2071 # TODO: some formats actually allow this (mkv, webm, ogg, mp4), but not all of them.
2072 video_formats
= [format
for format
in formats
if format
.get('vcodec') != 'none']
2073 audio_formats
= [format
for format
in formats
if format
.get('acodec') != 'none']
2074 if len(video_formats
) > 2 or len(audio_formats
) > 2:
2078 exts
= set(format
.get('ext') for format
in formats
)
2080 set(('mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma')),
2083 for ext_sets
in COMPATIBLE_EXTS
:
2084 if ext_sets
.issuperset(exts
):
2086 # TODO: Check acodec/vcodec
2089 filename_real_ext
= os
.path
.splitext(filename
)[1][1:]
2091 os
.path
.splitext(filename
)[0]
2092 if filename_real_ext
== info_dict
['ext']
2094 requested_formats
= info_dict
['requested_formats']
2095 if self
.params
.get('merge_output_format') is None and not compatible_formats(requested_formats
):
2096 info_dict
['ext'] = 'mkv'
2097 self
.report_warning(
2098 'Requested formats are incompatible for merge and will be merged into mkv.')
2099 # Ensure filename always has a correct extension for successful merge
2100 filename
= '%s.%s' % (filename_wo_ext
, info_dict
['ext'])
2101 if os
.path
.exists(encodeFilename(filename
)):
2103 '[download] %s has already been downloaded and '
2104 'merged' % filename
)
2106 for f
in requested_formats
:
2107 new_info
= dict(info_dict
)
2109 fname
= prepend_extension(
2110 self
.prepare_filename(new_info
),
2111 'f%s' % f
['format_id'], new_info
['ext'])
2112 if not ensure_dir_exists(fname
):
2114 downloaded
.append(fname
)
2115 partial_success
, real_download
= dl(fname
, new_info
)
2116 success
= success
and partial_success
2117 info_dict
['__postprocessors'] = postprocessors
2118 info_dict
['__files_to_merge'] = downloaded
2119 # Even if there were no downloads, it is being merged only now
2120 info_dict
['__real_download'] = True
2122 # Just a single file
2123 success
, real_download
= dl(filename
, info_dict
)
2124 info_dict
['__real_download'] = real_download
2125 except (compat_urllib_error
.URLError
, compat_http_client
.HTTPException
, socket
.error
) as err
:
2126 self
.report_error('unable to download video data: %s' % error_to_compat_str(err
))
2128 except (OSError, IOError) as err
:
2129 raise UnavailableVideoError(err
)
2130 except (ContentTooShortError
, ) as err
:
2131 self
.report_error('content too short (expected %s bytes and served %s)' % (err
.expected
, err
.downloaded
))
2134 if success
and filename
!= '-':
2136 fixup_policy
= self
.params
.get('fixup')
2137 if fixup_policy
is None:
2138 fixup_policy
= 'detect_or_warn'
2140 INSTALL_FFMPEG_MESSAGE
= 'Install ffmpeg or avconv to fix this automatically.'
2142 stretched_ratio
= info_dict
.get('stretched_ratio')
2143 if stretched_ratio
is not None and stretched_ratio
!= 1:
2144 if fixup_policy
== 'warn':
2145 self
.report_warning('%s: Non-uniform pixel ratio (%s)' % (
2146 info_dict
['id'], stretched_ratio
))
2147 elif fixup_policy
== 'detect_or_warn':
2148 stretched_pp
= FFmpegFixupStretchedPP(self
)
2149 if stretched_pp
.available
:
2150 info_dict
.setdefault('__postprocessors', [])
2151 info_dict
['__postprocessors'].append(stretched_pp
)
2153 self
.report_warning(
2154 '%s: Non-uniform pixel ratio (%s). %s'
2155 % (info_dict
['id'], stretched_ratio
, INSTALL_FFMPEG_MESSAGE
))
2157 assert fixup_policy
in ('ignore', 'never')
2159 if (info_dict
.get('requested_formats') is None
2160 and info_dict
.get('container') == 'm4a_dash'):
2161 if fixup_policy
== 'warn':
2162 self
.report_warning(
2163 '%s: writing DASH m4a. '
2164 'Only some players support this container.'
2166 elif fixup_policy
== 'detect_or_warn':
2167 fixup_pp
= FFmpegFixupM4aPP(self
)
2168 if fixup_pp
.available
:
2169 info_dict
.setdefault('__postprocessors', [])
2170 info_dict
['__postprocessors'].append(fixup_pp
)
2172 self
.report_warning(
2173 '%s: writing DASH m4a. '
2174 'Only some players support this container. %s'
2175 % (info_dict
['id'], INSTALL_FFMPEG_MESSAGE
))
2177 assert fixup_policy
in ('ignore', 'never')
2179 if (info_dict
.get('protocol') == 'm3u8_native'
2180 or info_dict
.get('protocol') == 'm3u8'
2181 and self
.params
.get('hls_prefer_native')):
2182 if fixup_policy
== 'warn':
2183 self
.report_warning('%s: malformed AAC bitstream detected.' % (
2185 elif fixup_policy
== 'detect_or_warn':
2186 fixup_pp
= FFmpegFixupM3u8PP(self
)
2187 if fixup_pp
.available
:
2188 info_dict
.setdefault('__postprocessors', [])
2189 info_dict
['__postprocessors'].append(fixup_pp
)
2191 self
.report_warning(
2192 '%s: malformed AAC bitstream detected. %s'
2193 % (info_dict
['id'], INSTALL_FFMPEG_MESSAGE
))
2195 assert fixup_policy
in ('ignore', 'never')
2198 self
.post_process(filename
, info_dict
)
2199 except (PostProcessingError
) as err
:
2200 self
.report_error('postprocessing: %s' % str(err
))
2202 must_record_download_archive
= True
2204 if must_record_download_archive
or self
.params
.get('force_write_download_archive', False):
2205 self
.record_download_archive(info_dict
)
2207 def download(self
, url_list
):
2208 """Download a given list of URLs."""
2209 outtmpl
= self
.params
.get('outtmpl', DEFAULT_OUTTMPL
)
2210 if (len(url_list
) > 1
2212 and '%' not in outtmpl
2213 and self
.params
.get('max_downloads') != 1):
2214 raise SameFileError(outtmpl
)
2216 for url
in url_list
:
2218 # It also downloads the videos
2219 res
= self
.extract_info(
2220 url
, force_generic_extractor
=self
.params
.get('force_generic_extractor', False))
2221 except UnavailableVideoError
:
2222 self
.report_error('unable to download video')
2223 except MaxDownloadsReached
:
2224 self
.to_screen('[info] Maximum number of downloaded files reached.')
2227 if self
.params
.get('dump_single_json', False):
2228 self
.to_stdout(json
.dumps(res
))
2230 return self
._download
_retcode
2232 def download_with_info_file(self
, info_filename
):
2233 with contextlib
.closing(fileinput
.FileInput(
2234 [info_filename
], mode
='r',
2235 openhook
=fileinput
.hook_encoded('utf-8'))) as f
:
2236 # FileInput doesn't have a read method, we can't call json.load
2237 info
= self
.filter_requested_info(json
.loads('\n'.join(f
)))
2239 self
.process_ie_result(info
, download
=True)
2240 except DownloadError
:
2241 webpage_url
= info
.get('webpage_url')
2242 if webpage_url
is not None:
2243 self
.report_warning('The info failed to download, trying with "%s"' % webpage_url
)
2244 return self
.download([webpage_url
])
2247 return self
._download
_retcode
2250 def filter_requested_info(info_dict
):
2252 (k
, v
) for k
, v
in info_dict
.items()
2253 if k
not in ['requested_formats', 'requested_subtitles'])
2255 def post_process(self
, filename
, ie_info
):
2256 """Run all the postprocessors on the given file."""
2257 info
= dict(ie_info
)
2258 info
['filepath'] = filename
2260 if ie_info
.get('__postprocessors') is not None:
2261 pps_chain
.extend(ie_info
['__postprocessors'])
2262 pps_chain
.extend(self
._pps
)
2263 for pp
in pps_chain
:
2264 files_to_delete
= []
2266 files_to_delete
, info
= pp
.run(info
)
2267 except PostProcessingError
as e
:
2268 self
.report_error(e
.msg
)
2269 if files_to_delete
and not self
.params
.get('keepvideo', False):
2270 for old_filename
in set(files_to_delete
):
2271 self
.to_screen('Deleting original file %s (pass -k to keep)' % old_filename
)
2273 os
.remove(encodeFilename(old_filename
))
2274 except (IOError, OSError):
2275 self
.report_warning('Unable to remove downloaded original file')
2277 def _make_archive_id(self
, info_dict
):
2278 video_id
= info_dict
.get('id')
2281 # Future-proof against any change in case
2282 # and backwards compatibility with prior versions
2283 extractor
= info_dict
.get('extractor_key') or info_dict
.get('ie_key') # key in a playlist
2284 if extractor
is None:
2285 url
= str_or_none(info_dict
.get('url'))
2288 # Try to find matching extractor for the URL and take its ie_key
2289 for ie
in self
._ies
:
2290 if ie
.suitable(url
):
2291 extractor
= ie
.ie_key()
2295 return extractor
.lower() + ' ' + video_id
2297 def in_download_archive(self
, info_dict
):
2298 fn
= self
.params
.get('download_archive')
2302 vid_id
= self
._make
_archive
_id
(info_dict
)
2304 return False # Incomplete video information
2306 return vid_id
in self
.archive
2308 def record_download_archive(self
, info_dict
):
2309 fn
= self
.params
.get('download_archive')
2312 vid_id
= self
._make
_archive
_id
(info_dict
)
2314 with locked_file(fn
, 'a', encoding
='utf-8') as archive_file
:
2315 archive_file
.write(vid_id
+ '\n')
2316 self
.archive
.add(vid_id
)
2319 def format_resolution(format
, default
='unknown'):
2320 if format
.get('vcodec') == 'none':
2322 if format
.get('resolution') is not None:
2323 return format
['resolution']
2324 if format
.get('height') is not None:
2325 if format
.get('width') is not None:
2326 res
= '%sx%s' % (format
['width'], format
['height'])
2328 res
= '%sp' % format
['height']
2329 elif format
.get('width') is not None:
2330 res
= '%dx?' % format
['width']
2335 def _format_note(self
, fdict
):
2337 if fdict
.get('ext') in ['f4f', 'f4m']:
2338 res
+= '(unsupported) '
2339 if fdict
.get('language'):
2342 res
+= '[%s] ' % fdict
['language']
2343 if fdict
.get('format_note') is not None:
2344 res
+= fdict
['format_note'] + ' '
2345 if fdict
.get('tbr') is not None:
2346 res
+= '%4dk ' % fdict
['tbr']
2347 if fdict
.get('container') is not None:
2350 res
+= '%s container' % fdict
['container']
2351 if (fdict
.get('vcodec') is not None
2352 and fdict
.get('vcodec') != 'none'):
2355 res
+= fdict
['vcodec']
2356 if fdict
.get('vbr') is not None:
2358 elif fdict
.get('vbr') is not None and fdict
.get('abr') is not None:
2360 if fdict
.get('vbr') is not None:
2361 res
+= '%4dk' % fdict
['vbr']
2362 if fdict
.get('fps') is not None:
2365 res
+= '%sfps' % fdict
['fps']
2366 if fdict
.get('acodec') is not None:
2369 if fdict
['acodec'] == 'none':
2372 res
+= '%-5s' % fdict
['acodec']
2373 elif fdict
.get('abr') is not None:
2377 if fdict
.get('abr') is not None:
2378 res
+= '@%3dk' % fdict
['abr']
2379 if fdict
.get('asr') is not None:
2380 res
+= ' (%5dHz)' % fdict
['asr']
2381 if fdict
.get('filesize') is not None:
2384 res
+= format_bytes(fdict
['filesize'])
2385 elif fdict
.get('filesize_approx') is not None:
2388 res
+= '~' + format_bytes(fdict
['filesize_approx'])
2391 def _format_note_table(self
, f
):
2392 def join_fields(*vargs
):
2393 return ', '.join((val
for val
in vargs
if val
!= ''))
2396 'UNSUPPORTED' if f
.get('ext') in ('f4f', 'f4m') else '',
2397 format_field(f
, 'language', '[%s]'),
2398 format_field(f
, 'format_note'),
2399 format_field(f
, 'container', ignore
=(None, f
.get('ext'))),
2400 format_field(f
, 'asr', '%5dHz'))
2402 def list_formats(self
, info_dict
):
2403 formats
= info_dict
.get('formats', [info_dict
])
2404 new_format
= self
.params
.get('listformats_table', False)
2408 format_field(f
, 'format_id'),
2409 format_field(f
, 'ext'),
2410 self
.format_resolution(f
),
2411 format_field(f
, 'fps', '%d'),
2413 format_field(f
, 'filesize', ' %s', func
=format_bytes
) + format_field(f
, 'filesize_approx', '~%s', func
=format_bytes
),
2414 format_field(f
, 'tbr', '%4dk'),
2415 f
.get('protocol').replace('http_dash_segments', 'dash').replace("native", "n"),
2417 format_field(f
, 'vcodec', default
='unknown').replace('none', ''),
2418 format_field(f
, 'vbr', '%4dk'),
2419 format_field(f
, 'acodec', default
='unknown').replace('none', ''),
2420 format_field(f
, 'abr', '%3dk'),
2421 format_field(f
, 'asr', '%5dHz'),
2422 self
._format
_note
_table
(f
)]
2424 if f
.get('preference') is None or f
['preference'] >= -1000]
2425 header_line
= ['ID', 'EXT', 'RESOLUTION', 'FPS', '|', ' FILESIZE', ' TBR', 'PROTO',
2426 '|', 'VCODEC', ' VBR', 'ACODEC', ' ABR', ' ASR', 'NOTE']
2430 format_field(f
, 'format_id'),
2431 format_field(f
, 'ext'),
2432 self
.format_resolution(f
),
2433 self
._format
_note
(f
)]
2435 if f
.get('preference') is None or f
['preference'] >= -1000]
2436 header_line
= ['format code', 'extension', 'resolution', 'note']
2438 # if len(formats) > 1:
2439 # table[-1][-1] += (' ' if table[-1][-1] else '') + '(best)'
2441 '[info] Available formats for %s:\n%s' % (info_dict
['id'], render_table(
2445 extraGap
=(0 if new_format
else 1),
2446 hideEmpty
=new_format
)))
2448 def list_thumbnails(self
, info_dict
):
2449 thumbnails
= info_dict
.get('thumbnails')
2451 self
.to_screen('[info] No thumbnails present for %s' % info_dict
['id'])
2455 '[info] Thumbnails for %s:' % info_dict
['id'])
2456 self
.to_screen(render_table(
2457 ['ID', 'width', 'height', 'URL'],
2458 [[t
['id'], t
.get('width', 'unknown'), t
.get('height', 'unknown'), t
['url']] for t
in thumbnails
]))
2460 def list_subtitles(self
, video_id
, subtitles
, name
='subtitles'):
2462 self
.to_screen('%s has no %s' % (video_id
, name
))
2465 'Available %s for %s:' % (name
, video_id
))
2466 self
.to_screen(render_table(
2467 ['Language', 'formats'],
2468 [[lang
, ', '.join(f
['ext'] for f
in reversed(formats
))]
2469 for lang
, formats
in subtitles
.items()]))
2471 def urlopen(self
, req
):
2472 """ Start an HTTP download """
2473 if isinstance(req
, compat_basestring
):
2474 req
= sanitized_Request(req
)
2475 return self
._opener
.open(req
, timeout
=self
._socket
_timeout
)
2477 def print_debug_header(self
):
2478 if not self
.params
.get('verbose'):
2481 if type('') is not compat_str
:
2482 # Python 2.6 on SLES11 SP1 (https://github.com/ytdl-org/youtube-dl/issues/3326)
2483 self
.report_warning(
2484 'Your Python is broken! Update to a newer and supported version')
2486 stdout_encoding
= getattr(
2487 sys
.stdout
, 'encoding', 'missing (%s)' % type(sys
.stdout
).__name
__)
2489 '[debug] Encodings: locale %s, fs %s, out %s, pref %s\n' % (
2490 locale
.getpreferredencoding(),
2491 sys
.getfilesystemencoding(),
2493 self
.get_encoding()))
2494 write_string(encoding_str
, encoding
=None)
2496 self
._write
_string
('[debug] youtube-dlc version ' + __version__
+ '\n')
2498 self
._write
_string
('[debug] Lazy loading extractors enabled' + '\n')
2500 sp
= subprocess
.Popen(
2501 ['git', 'rev-parse', '--short', 'HEAD'],
2502 stdout
=subprocess
.PIPE
, stderr
=subprocess
.PIPE
,
2503 cwd
=os
.path
.dirname(os
.path
.abspath(__file__
)))
2504 out
, err
= sp
.communicate()
2505 out
= out
.decode().strip()
2506 if re
.match('[0-9a-f]+', out
):
2507 self
._write
_string
('[debug] Git HEAD: ' + out
+ '\n')
2514 def python_implementation():
2515 impl_name
= platform
.python_implementation()
2516 if impl_name
== 'PyPy' and hasattr(sys
, 'pypy_version_info'):
2517 return impl_name
+ ' version %d.%d.%d' % sys
.pypy_version_info
[:3]
2520 self
._write
_string
('[debug] Python version %s (%s) - %s\n' % (
2521 platform
.python_version(), python_implementation(),
2524 exe_versions
= FFmpegPostProcessor
.get_versions(self
)
2525 exe_versions
['rtmpdump'] = rtmpdump_version()
2526 exe_versions
['phantomjs'] = PhantomJSwrapper
._version
()
2527 exe_str
= ', '.join(
2529 for exe
, v
in sorted(exe_versions
.items())
2534 self
._write
_string
('[debug] exe versions: %s\n' % exe_str
)
2537 for handler
in self
._opener
.handlers
:
2538 if hasattr(handler
, 'proxies'):
2539 proxy_map
.update(handler
.proxies
)
2540 self
._write
_string
('[debug] Proxy map: ' + compat_str(proxy_map
) + '\n')
2542 if self
.params
.get('call_home', False):
2543 ipaddr
= self
.urlopen('https://yt-dl.org/ip').read().decode('utf-8')
2544 self
._write
_string
('[debug] Public IP address: %s\n' % ipaddr
)
2545 latest_version
= self
.urlopen(
2546 'https://yt-dl.org/latest/version').read().decode('utf-8')
2547 if version_tuple(latest_version
) > version_tuple(__version__
):
2548 self
.report_warning(
2549 'You are using an outdated version (newest version: %s)! '
2550 'See https://yt-dl.org/update if you need help updating.' %
2553 def _setup_opener(self
):
2554 timeout_val
= self
.params
.get('socket_timeout')
2555 self
._socket
_timeout
= 600 if timeout_val
is None else float(timeout_val
)
2557 opts_cookiefile
= self
.params
.get('cookiefile')
2558 opts_proxy
= self
.params
.get('proxy')
2560 if opts_cookiefile
is None:
2561 self
.cookiejar
= compat_cookiejar
.CookieJar()
2563 opts_cookiefile
= expand_path(opts_cookiefile
)
2564 self
.cookiejar
= YoutubeDLCookieJar(opts_cookiefile
)
2565 if os
.access(opts_cookiefile
, os
.R_OK
):
2566 self
.cookiejar
.load(ignore_discard
=True, ignore_expires
=True)
2568 cookie_processor
= YoutubeDLCookieProcessor(self
.cookiejar
)
2569 if opts_proxy
is not None:
2570 if opts_proxy
== '':
2573 proxies
= {'http': opts_proxy, 'https': opts_proxy}
2575 proxies
= compat_urllib_request
.getproxies()
2576 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
2577 if 'http' in proxies
and 'https' not in proxies
:
2578 proxies
['https'] = proxies
['http']
2579 proxy_handler
= PerRequestProxyHandler(proxies
)
2581 debuglevel
= 1 if self
.params
.get('debug_printtraffic') else 0
2582 https_handler
= make_HTTPS_handler(self
.params
, debuglevel
=debuglevel
)
2583 ydlh
= YoutubeDLHandler(self
.params
, debuglevel
=debuglevel
)
2584 redirect_handler
= YoutubeDLRedirectHandler()
2585 data_handler
= compat_urllib_request_DataHandler()
2587 # When passing our own FileHandler instance, build_opener won't add the
2588 # default FileHandler and allows us to disable the file protocol, which
2589 # can be used for malicious purposes (see
2590 # https://github.com/ytdl-org/youtube-dl/issues/8227)
2591 file_handler
= compat_urllib_request
.FileHandler()
2593 def file_open(*args
, **kwargs
):
2594 raise compat_urllib_error
.URLError('file:// scheme is explicitly disabled in youtube-dlc for security reasons')
2595 file_handler
.file_open
= file_open
2597 opener
= compat_urllib_request
.build_opener(
2598 proxy_handler
, https_handler
, cookie_processor
, ydlh
, redirect_handler
, data_handler
, file_handler
)
2600 # Delete the default user-agent header, which would otherwise apply in
2601 # cases where our custom HTTP handler doesn't come into play
2602 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
2603 opener
.addheaders
= []
2604 self
._opener
= opener
2606 def encode(self
, s
):
2607 if isinstance(s
, bytes):
2608 return s
# Already encoded
2611 return s
.encode(self
.get_encoding())
2612 except UnicodeEncodeError as err
:
2613 err
.reason
= err
.reason
+ '. Check your system encoding configuration or use the --encoding option.'
2616 def get_encoding(self
):
2617 encoding
= self
.params
.get('encoding')
2618 if encoding
is None:
2619 encoding
= preferredencoding()
2622 def _write_thumbnails(self
, info_dict
, filename
):
2623 if self
.params
.get('writethumbnail', False):
2624 thumbnails
= info_dict
.get('thumbnails')
2626 thumbnails
= [thumbnails
[-1]]
2627 elif self
.params
.get('write_all_thumbnails', False):
2628 thumbnails
= info_dict
.get('thumbnails')
2633 # No thumbnails present, so return immediately
2636 for t
in thumbnails
:
2637 thumb_ext
= determine_ext(t
['url'], 'jpg')
2638 suffix
= '_%s' % t
['id'] if len(thumbnails
) > 1 else ''
2639 thumb_display_id
= '%s ' % t
['id'] if len(thumbnails
) > 1 else ''
2640 t
['filename'] = thumb_filename
= replace_extension(filename
+ suffix
, thumb_ext
, info_dict
.get('ext'))
2642 if self
.params
.get('nooverwrites', False) and os
.path
.exists(encodeFilename(thumb_filename
)):
2643 self
.to_screen('[%s] %s: Thumbnail %sis already present' %
2644 (info_dict
['extractor'], info_dict
['id'], thumb_display_id
))
2646 self
.to_screen('[%s] %s: Downloading thumbnail %s...' %
2647 (info_dict
['extractor'], info_dict
['id'], thumb_display_id
))
2649 uf
= self
.urlopen(t
['url'])
2650 with open(encodeFilename(thumb_filename
), 'wb') as thumbf
:
2651 shutil
.copyfileobj(uf
, thumbf
)
2652 self
.to_screen('[%s] %s: Writing thumbnail %sto: %s' %
2653 (info_dict
['extractor'], info_dict
['id'], thumb_display_id
, thumb_filename
))
2654 except (compat_urllib_error
.URLError
, compat_http_client
.HTTPException
, socket
.error
) as err
:
2655 self
.report_warning('Unable to download thumbnail "%s": %s' %
2656 (t
['url'], error_to_compat_str(err
)))