4 from __future__
import absolute_import
, unicode_literals
28 from string
import ascii_letters
29 from zipimport
import zipimporter
34 compat_get_terminal_size
,
39 compat_tokenize_tokenize
,
41 compat_urllib_request
,
42 compat_urllib_request_DataHandler
,
53 DOT_DESKTOP_LINK_TEMPLATE
,
54 DOT_URL_LINK_TEMPLATE
,
55 DOT_WEBLOC_LINK_TEMPLATE
,
82 PerRequestProxyHandler
,
87 process_communicate_or_kill
,
89 register_socks_protocols
,
104 UnavailableVideoError
,
110 YoutubeDLCookieProcessor
,
112 YoutubeDLRedirectHandler
,
114 from .cache
import Cache
115 from .extractor
import (
116 gen_extractor_classes
,
121 from .extractor
.openload
import PhantomJSwrapper
122 from .downloader
import (
123 get_suitable_downloader
,
124 shorten_protocol_name
126 from .downloader
.rtmp
import rtmpdump_version
127 from .postprocessor
import (
130 FFmpegFixupStretchedPP
,
133 # FFmpegSubtitlesConvertorPP,
135 MoveFilesAfterDownloadPP
,
137 from .version
import __version__
139 if compat_os_name
== 'nt':
143 class YoutubeDL(object):
146 YoutubeDL objects are the ones responsible of downloading the
147 actual video file and writing it to disk if the user has requested
148 it, among some other tasks. In most cases there should be one per
149 program. As, given a video URL, the downloader doesn't know how to
150 extract all the needed information, task that InfoExtractors do, it
151 has to pass the URL to one of them.
153 For this, YoutubeDL objects have a method that allows
154 InfoExtractors to be registered in a given order. When it is passed
155 a URL, the YoutubeDL object handles it to the first InfoExtractor it
156 finds that reports being able to handle it. The InfoExtractor extracts
157 all the information about the video or videos the URL refers to, and
158 YoutubeDL process the extracted information, possibly using a File
159 Downloader to download the video.
161 YoutubeDL objects accept a lot of parameters. In order not to saturate
162 the object constructor with arguments, it receives a dictionary of
163 options instead. These options are available through the params
164 attribute for the InfoExtractors to use. The YoutubeDL also
165 registers itself as the downloader in charge for the InfoExtractors
166 that are added to it, so this is a "mutual registration".
170 username: Username for authentication purposes.
171 password: Password for authentication purposes.
172 videopassword: Password for accessing a video.
173 ap_mso: Adobe Pass multiple-system operator identifier.
174 ap_username: Multiple-system operator account username.
175 ap_password: Multiple-system operator account password.
176 usenetrc: Use netrc for authentication instead.
177 verbose: Print additional info to stdout.
178 quiet: Do not print messages to stdout.
179 no_warnings: Do not print out anything for warnings.
180 forceprint: A list of templates to force print
181 forceurl: Force printing final URL. (Deprecated)
182 forcetitle: Force printing title. (Deprecated)
183 forceid: Force printing ID. (Deprecated)
184 forcethumbnail: Force printing thumbnail URL. (Deprecated)
185 forcedescription: Force printing description. (Deprecated)
186 forcefilename: Force printing final filename. (Deprecated)
187 forceduration: Force printing duration. (Deprecated)
188 forcejson: Force printing info_dict as JSON.
189 dump_single_json: Force printing the info_dict of the whole playlist
190 (or video) as a single JSON line.
191 force_write_download_archive: Force writing download archive regardless
192 of 'skip_download' or 'simulate'.
193 simulate: Do not download the video files.
194 format: Video format code. see "FORMAT SELECTION" for more details.
195 allow_unplayable_formats: Allow unplayable formats to be extracted and downloaded.
196 ignore_no_formats_error: Ignore "No video formats" error. Usefull for
197 extracting metadata even if the video is not actually
198 available for download (experimental)
199 format_sort: How to sort the video formats. see "Sorting Formats"
201 format_sort_force: Force the given format_sort. see "Sorting Formats"
203 allow_multiple_video_streams: Allow multiple video streams to be merged
205 allow_multiple_audio_streams: Allow multiple audio streams to be merged
207 paths: Dictionary of output paths. The allowed keys are 'home'
208 'temp' and the keys of OUTTMPL_TYPES (in utils.py)
209 outtmpl: Dictionary of templates for output names. Allowed keys
210 are 'default' and the keys of OUTTMPL_TYPES (in utils.py).
211 A string a also accepted for backward compatibility
212 outtmpl_na_placeholder: Placeholder for unavailable meta fields.
213 restrictfilenames: Do not allow "&" and spaces in file names
214 trim_file_name: Limit length of filename (extension excluded)
215 windowsfilenames: Force the filenames to be windows compatible
216 ignoreerrors: Do not stop on download errors
217 (Default True when running yt-dlp,
218 but False when directly accessing YoutubeDL class)
219 skip_playlist_after_errors: Number of allowed failures until the rest of
220 the playlist is skipped
221 force_generic_extractor: Force downloader to use the generic extractor
222 overwrites: Overwrite all video and metadata files if True,
223 overwrite only non-video files if None
224 and don't overwrite any file if False
225 playliststart: Playlist item to start at.
226 playlistend: Playlist item to end at.
227 playlist_items: Specific indices of playlist to download.
228 playlistreverse: Download playlist items in reverse order.
229 playlistrandom: Download playlist items in random order.
230 matchtitle: Download only matching titles.
231 rejecttitle: Reject downloads for matching titles.
232 logger: Log messages to a logging.Logger instance.
233 logtostderr: Log messages to stderr instead of stdout.
234 writedescription: Write the video description to a .description file
235 writeinfojson: Write the video description to a .info.json file
236 clean_infojson: Remove private fields from the infojson
237 writecomments: Extract video comments. This will not be written to disk
238 unless writeinfojson is also given
239 writeannotations: Write the video annotations to a .annotations.xml file
240 writethumbnail: Write the thumbnail image to a file
241 allow_playlist_files: Whether to write playlists' description, infojson etc
242 also to disk when using the 'write*' options
243 write_all_thumbnails: Write all thumbnail formats to files
244 writelink: Write an internet shortcut file, depending on the
245 current platform (.url/.webloc/.desktop)
246 writeurllink: Write a Windows internet shortcut file (.url)
247 writewebloclink: Write a macOS internet shortcut file (.webloc)
248 writedesktoplink: Write a Linux internet shortcut file (.desktop)
249 writesubtitles: Write the video subtitles to a file
250 writeautomaticsub: Write the automatically generated subtitles to a file
251 allsubtitles: Deprecated - Use subtitlelangs = ['all']
252 Downloads all the subtitles of the video
253 (requires writesubtitles or writeautomaticsub)
254 listsubtitles: Lists all available subtitles for the video
255 subtitlesformat: The format code for subtitles
256 subtitleslangs: List of languages of the subtitles to download (can be regex).
257 The list may contain "all" to refer to all the available
258 subtitles. The language can be prefixed with a "-" to
259 exclude it from the requested languages. Eg: ['all', '-live_chat']
260 keepvideo: Keep the video file after post-processing
261 daterange: A DateRange object, download only if the upload_date is in the range.
262 skip_download: Skip the actual download of the video file
263 cachedir: Location of the cache files in the filesystem.
264 False to disable filesystem cache.
265 noplaylist: Download single video instead of a playlist if in doubt.
266 age_limit: An integer representing the user's age in years.
267 Unsuitable videos for the given age are skipped.
268 min_views: An integer representing the minimum view count the video
269 must have in order to not be skipped.
270 Videos without view count information are always
271 downloaded. None for no limit.
272 max_views: An integer representing the maximum view count.
273 Videos that are more popular than that are not
275 Videos without view count information are always
276 downloaded. None for no limit.
277 download_archive: File name of a file where all downloads are recorded.
278 Videos already present in the file are not downloaded
280 break_on_existing: Stop the download process after attempting to download a
281 file that is in the archive.
282 break_on_reject: Stop the download process when encountering a video that
283 has been filtered out.
284 cookiefile: File name where cookies should be read from and dumped to
285 nocheckcertificate:Do not verify SSL certificates
286 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
287 At the moment, this is only supported by YouTube.
288 proxy: URL of the proxy server to use
289 geo_verification_proxy: URL of the proxy to use for IP address verification
290 on geo-restricted sites.
291 socket_timeout: Time to wait for unresponsive hosts, in seconds
292 bidi_workaround: Work around buggy terminals without bidirectional text
293 support, using fridibi
294 debug_printtraffic:Print out sent and received HTTP traffic
295 include_ads: Download ads as well
296 default_search: Prepend this string if an input url is not valid.
297 'auto' for elaborate guessing
298 encoding: Use this encoding instead of the system-specified.
299 extract_flat: Do not resolve URLs, return the immediate result.
300 Pass in 'in_playlist' to only show this behavior for
302 postprocessors: A list of dictionaries, each with an entry
303 * key: The name of the postprocessor. See
304 yt_dlp/postprocessor/__init__.py for a list.
305 * when: When to run the postprocessor. Can be one of
306 pre_process|before_dl|post_process|after_move.
307 Assumed to be 'post_process' if not given
308 post_hooks: A list of functions that get called as the final step
309 for each video file, after all postprocessors have been
310 called. The filename will be passed as the only argument.
311 progress_hooks: A list of functions that get called on download
312 progress, with a dictionary with the entries
313 * status: One of "downloading", "error", or "finished".
314 Check this first and ignore unknown values.
316 If status is one of "downloading", or "finished", the
317 following properties may also be present:
318 * filename: The final filename (always present)
319 * tmpfilename: The filename we're currently writing to
320 * downloaded_bytes: Bytes on disk
321 * total_bytes: Size of the whole file, None if unknown
322 * total_bytes_estimate: Guess of the eventual file size,
324 * elapsed: The number of seconds since download started.
325 * eta: The estimated time in seconds, None if unknown
326 * speed: The download speed in bytes/second, None if
328 * fragment_index: The counter of the currently
329 downloaded video fragment.
330 * fragment_count: The number of fragments (= individual
331 files that will be merged)
333 Progress hooks are guaranteed to be called at least once
334 (with status "finished") if the download is successful.
335 merge_output_format: Extension to use when merging formats.
336 final_ext: Expected final extension; used to detect when the file was
337 already downloaded and converted. "merge_output_format" is
338 replaced by this extension when given
339 fixup: Automatically correct known faults of the file.
341 - "never": do nothing
342 - "warn": only emit a warning
343 - "detect_or_warn": check whether we can do anything
344 about it, warn otherwise (default)
345 source_address: Client-side IP address to bind to.
346 call_home: Boolean, true iff we are allowed to contact the
347 yt-dlp servers for debugging. (BROKEN)
348 sleep_interval_requests: Number of seconds to sleep between requests
350 sleep_interval: Number of seconds to sleep before each download when
351 used alone or a lower bound of a range for randomized
352 sleep before each download (minimum possible number
353 of seconds to sleep) when used along with
355 max_sleep_interval:Upper bound of a range for randomized sleep before each
356 download (maximum possible number of seconds to sleep).
357 Must only be used along with sleep_interval.
358 Actual sleep time will be a random float from range
359 [sleep_interval; max_sleep_interval].
360 sleep_interval_subtitles: Number of seconds to sleep before each subtitle download
361 listformats: Print an overview of available video formats and exit.
362 list_thumbnails: Print a table of all thumbnails and exit.
363 match_filter: A function that gets called with the info_dict of
365 If it returns a message, the video is ignored.
366 If it returns None, the video is downloaded.
367 match_filter_func in utils.py is one example for this.
368 no_color: Do not emit color codes in output.
369 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
372 Two-letter ISO 3166-2 country code that will be used for
373 explicit geographic restriction bypassing via faking
374 X-Forwarded-For HTTP header
376 IP range in CIDR notation that will be used similarly to
379 The following options determine which downloader is picked:
380 external_downloader: A dictionary of protocol keys and the executable of the
381 external downloader to use for it. The allowed protocols
382 are default|http|ftp|m3u8|dash|rtsp|rtmp|mms.
383 Set the value to 'native' to use the native downloader
384 hls_prefer_native: Deprecated - Use external_downloader = {'m3u8': 'native'}
385 or {'m3u8': 'ffmpeg'} instead.
386 Use the native HLS downloader instead of ffmpeg/avconv
387 if True, otherwise use ffmpeg/avconv if False, otherwise
388 use downloader suggested by extractor if None.
389 compat_opts: Compatibility options. See "Differences in default behavior".
390 Note that only format-sort, format-spec, no-live-chat, no-attach-info-json
391 playlist-index, list-formats, no-youtube-channel-redirect
392 and no-youtube-unavailable-videos works when used via the API
394 The following parameters are not used by YoutubeDL itself, they are used by
395 the downloader (see yt_dlp/downloader/common.py):
396 nopart, updatetime, buffersize, ratelimit, min_filesize, max_filesize, test,
397 noresizebuffer, retries, continuedl, noprogress, consoletitle,
398 xattr_set_filesize, external_downloader_args, hls_use_mpegts,
401 The following options are used by the post processors:
402 prefer_ffmpeg: If False, use avconv instead of ffmpeg if both are available,
403 otherwise prefer ffmpeg. (avconv support is deprecated)
404 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
405 to the binary or its containing directory.
406 postprocessor_args: A dictionary of postprocessor/executable keys (in lower case)
407 and a list of additional command-line arguments for the
408 postprocessor/executable. The dict can also have "PP+EXE" keys
409 which are used when the given exe is used by the given PP.
410 Use 'default' as the name for arguments to passed to all PP
412 The following options are used by the extractors:
413 extractor_retries: Number of times to retry for known errors
414 dynamic_mpd: Whether to process dynamic DASH manifests (default: True)
415 hls_split_discontinuity: Split HLS playlists to different formats at
416 discontinuities such as ad breaks (default: False)
417 youtube_include_dash_manifest: If True (default), DASH manifests and related
418 data will be downloaded and processed by extractor.
419 You can reduce network I/O by disabling it if you don't
420 care about DASH. (only for youtube)
421 youtube_include_hls_manifest: If True (default), HLS manifests and related
422 data will be downloaded and processed by extractor.
423 You can reduce network I/O by disabling it if you don't
424 care about HLS. (only for youtube)
427 _NUMERIC_FIELDS
= set((
428 'width', 'height', 'tbr', 'abr', 'asr', 'vbr', 'fps', 'filesize', 'filesize_approx',
429 'timestamp', 'upload_year', 'upload_month', 'upload_day',
430 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
431 'average_rating', 'comment_count', 'age_limit',
432 'start_time', 'end_time',
433 'chapter_number', 'season_number', 'episode_number',
434 'track_number', 'disc_number', 'release_year',
440 _pps
= {'pre_process': [], 'before_dl': [], 'after_move': [], 'post_process': []}
441 __prepare_filename_warned
= False
442 _first_webpage_request
= True
443 _download_retcode
= None
444 _num_downloads
= None
446 _playlist_urls
= set()
449 def __init__(self
, params
=None, auto_init
=True):
450 """Create a FileDownloader object with the given options."""
454 self
._ies
_instances
= {}
455 self
._pps
= {'pre_process': [], 'before_dl': [], 'after_move': [], 'post_process': []}
456 self
.__prepare
_filename
_warned
= False
457 self
._first
_webpage
_request
= True
458 self
._post
_hooks
= []
459 self
._progress
_hooks
= []
460 self
._download
_retcode
= 0
461 self
._num
_downloads
= 0
462 self
._screen
_file
= [sys
.stdout
, sys
.stderr
][params
.get('logtostderr', False)]
463 self
._err
_file
= sys
.stderr
466 'nocheckcertificate': False,
468 self
.params
.update(params
)
469 self
.cache
= Cache(self
)
471 if sys
.version_info
< (3, 6):
473 'Support for Python version %d.%d have been deprecated and will break in future versions of yt-dlp! '
474 'Update to Python 3.6 or above' % sys
.version_info
[:2])
476 def check_deprecated(param
, option
, suggestion
):
477 if self
.params
.get(param
) is not None:
478 self
.report_warning('%s is deprecated. Use %s instead' % (option
, suggestion
))
482 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
483 if self
.params
.get('geo_verification_proxy') is None:
484 self
.params
['geo_verification_proxy'] = self
.params
['cn_verification_proxy']
486 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
487 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
488 check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"')
490 for msg
in self
.params
.get('warnings', []):
491 self
.report_warning(msg
)
493 if self
.params
.get('final_ext'):
494 if self
.params
.get('merge_output_format'):
495 self
.report_warning('--merge-output-format will be ignored since --remux-video or --recode-video is given')
496 self
.params
['merge_output_format'] = self
.params
['final_ext']
498 if 'overwrites' in self
.params
and self
.params
['overwrites'] is None:
499 del self
.params
['overwrites']
501 if params
.get('bidi_workaround', False):
504 master
, slave
= pty
.openpty()
505 width
= compat_get_terminal_size().columns
509 width_args
= ['-w', str(width
)]
511 stdin
=subprocess
.PIPE
,
513 stderr
=self
._err
_file
)
515 self
._output
_process
= subprocess
.Popen(
516 ['bidiv'] + width_args
, **sp_kwargs
519 self
._output
_process
= subprocess
.Popen(
520 ['fribidi', '-c', 'UTF-8'] + width_args
, **sp_kwargs
)
521 self
._output
_channel
= os
.fdopen(master
, 'rb')
522 except OSError as ose
:
523 if ose
.errno
== errno
.ENOENT
:
524 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 $PATH.')
528 if (sys
.platform
!= 'win32'
529 and sys
.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
530 and not params
.get('restrictfilenames', False)):
531 # Unicode filesystem API will throw errors (#1474, #13027)
533 'Assuming --restrict-filenames since file system encoding '
534 'cannot encode all characters. '
535 'Set the LC_ALL environment variable to fix this.')
536 self
.params
['restrictfilenames'] = True
538 self
.outtmpl_dict
= self
.parse_outtmpl()
542 """Preload the archive, if any is specified"""
543 def preload_download_archive(fn
):
546 self
.write_debug('Loading archive file %r\n' % fn
)
548 with locked_file(fn
, 'r', encoding
='utf-8') as archive_file
:
549 for line
in archive_file
:
550 self
.archive
.add(line
.strip())
551 except IOError as ioe
:
552 if ioe
.errno
!= errno
.ENOENT
:
558 preload_download_archive(self
.params
.get('download_archive'))
561 self
.print_debug_header()
562 self
.add_default_info_extractors()
564 for pp_def_raw
in self
.params
.get('postprocessors', []):
565 pp_class
= get_postprocessor(pp_def_raw
['key'])
566 pp_def
= dict(pp_def_raw
)
569 when
= pp_def
['when']
572 when
= 'post_process'
573 pp
= pp_class(self
, **compat_kwargs(pp_def
))
574 self
.add_post_processor(pp
, when
=when
)
576 for ph
in self
.params
.get('post_hooks', []):
577 self
.add_post_hook(ph
)
579 for ph
in self
.params
.get('progress_hooks', []):
580 self
.add_progress_hook(ph
)
582 register_socks_protocols()
584 def warn_if_short_id(self
, argv
):
585 # short YouTube ID starting with dash?
587 i
for i
, a
in enumerate(argv
)
588 if re
.match(r
'^-[0-9A-Za-z_-]{10}$', a
)]
592 + [a
for i
, a
in enumerate(argv
) if i
not in idxs
]
593 + ['--'] + [argv
[i
] for i
in idxs
]
596 'Long argument string detected. '
597 'Use -- to separate parameters and URLs, like this:\n%s\n' %
598 args_to_str(correct_argv
))
600 def add_info_extractor(self
, ie
):
601 """Add an InfoExtractor object to the end of the list."""
603 if not isinstance(ie
, type):
604 self
._ies
_instances
[ie
.ie_key()] = ie
605 ie
.set_downloader(self
)
607 def get_info_extractor(self
, ie_key
):
609 Get an instance of an IE with name ie_key, it will try to get one from
610 the _ies list, if there's no instance it will create a new one and add
611 it to the extractor list.
613 ie
= self
._ies
_instances
.get(ie_key
)
615 ie
= get_info_extractor(ie_key
)()
616 self
.add_info_extractor(ie
)
619 def add_default_info_extractors(self
):
621 Add the InfoExtractors returned by gen_extractors to the end of the list
623 for ie
in gen_extractor_classes():
624 self
.add_info_extractor(ie
)
626 def add_post_processor(self
, pp
, when
='post_process'):
627 """Add a PostProcessor object to the end of the chain."""
628 self
._pps
[when
].append(pp
)
629 pp
.set_downloader(self
)
631 def add_post_hook(self
, ph
):
632 """Add the post hook"""
633 self
._post
_hooks
.append(ph
)
635 def add_progress_hook(self
, ph
):
636 """Add the progress hook (currently only for the file downloader)"""
637 self
._progress
_hooks
.append(ph
)
639 def _bidi_workaround(self
, message
):
640 if not hasattr(self
, '_output_channel'):
643 assert hasattr(self
, '_output_process')
644 assert isinstance(message
, compat_str
)
645 line_count
= message
.count('\n') + 1
646 self
._output
_process
.stdin
.write((message
+ '\n').encode('utf-8'))
647 self
._output
_process
.stdin
.flush()
648 res
= ''.join(self
._output
_channel
.readline().decode('utf-8')
649 for _
in range(line_count
))
650 return res
[:-len('\n')]
652 def _write_string(self
, s
, out
=None):
653 write_string(s
, out
=out
, encoding
=self
.params
.get('encoding'))
655 def to_stdout(self
, message
, skip_eol
=False, quiet
=False):
656 """Print message to stdout"""
657 if self
.params
.get('logger'):
658 self
.params
['logger'].debug(message
)
660 message
= self
._bidi
_workaround
(message
)
661 terminator
= ['\n', ''][skip_eol
]
662 output
= message
+ terminator
664 self
._write
_string
(output
, self
._screen
_file
)
666 def to_stderr(self
, message
):
667 """Print message to stderr"""
668 assert isinstance(message
, compat_str
)
669 if self
.params
.get('logger'):
670 self
.params
['logger'].error(message
)
672 message
= self
._bidi
_workaround
(message
)
673 output
= message
+ '\n'
674 self
._write
_string
(output
, self
._err
_file
)
676 def to_console_title(self
, message
):
677 if not self
.params
.get('consoletitle', False):
679 if compat_os_name
== 'nt':
680 if ctypes
.windll
.kernel32
.GetConsoleWindow():
681 # c_wchar_p() might not be necessary if `message` is
682 # already of type unicode()
683 ctypes
.windll
.kernel32
.SetConsoleTitleW(ctypes
.c_wchar_p(message
))
684 elif 'TERM' in os
.environ
:
685 self
._write
_string
('\033]0;%s\007' % message
, self
._screen
_file
)
687 def save_console_title(self
):
688 if not self
.params
.get('consoletitle', False):
690 if self
.params
.get('simulate', False):
692 if compat_os_name
!= 'nt' and 'TERM' in os
.environ
:
693 # Save the title on stack
694 self
._write
_string
('\033[22;0t', self
._screen
_file
)
696 def restore_console_title(self
):
697 if not self
.params
.get('consoletitle', False):
699 if self
.params
.get('simulate', False):
701 if compat_os_name
!= 'nt' and 'TERM' in os
.environ
:
702 # Restore the title from stack
703 self
._write
_string
('\033[23;0t', self
._screen
_file
)
706 self
.save_console_title()
709 def __exit__(self
, *args
):
710 self
.restore_console_title()
712 if self
.params
.get('cookiefile') is not None:
713 self
.cookiejar
.save(ignore_discard
=True, ignore_expires
=True)
715 def trouble(self
, message
=None, tb
=None):
716 """Determine action to take when a download problem appears.
718 Depending on if the downloader has been configured to ignore
719 download errors or not, this method may throw an exception or
720 not when errors are found, after printing the message.
722 tb, if given, is additional traceback information.
724 if message
is not None:
725 self
.to_stderr(message
)
726 if self
.params
.get('verbose'):
728 if sys
.exc_info()[0]: # if .trouble has been called from an except block
730 if hasattr(sys
.exc_info()[1], 'exc_info') and sys
.exc_info()[1].exc_info
[0]:
731 tb
+= ''.join(traceback
.format_exception(*sys
.exc_info()[1].exc_info
))
732 tb
+= encode_compat_str(traceback
.format_exc())
734 tb_data
= traceback
.format_list(traceback
.extract_stack())
735 tb
= ''.join(tb_data
)
737 if not self
.params
.get('ignoreerrors', False):
738 if sys
.exc_info()[0] and hasattr(sys
.exc_info()[1], 'exc_info') and sys
.exc_info()[1].exc_info
[0]:
739 exc_info
= sys
.exc_info()[1].exc_info
741 exc_info
= sys
.exc_info()
742 raise DownloadError(message
, exc_info
)
743 self
._download
_retcode
= 1
745 def to_screen(self
, message
, skip_eol
=False):
746 """Print message to stdout if not in quiet mode"""
748 message
, skip_eol
, quiet
=self
.params
.get('quiet', False))
750 def report_warning(self
, message
):
752 Print the message to stderr, it will be prefixed with 'WARNING:'
753 If stderr is a tty file the 'WARNING:' will be colored
755 if self
.params
.get('logger') is not None:
756 self
.params
['logger'].warning(message
)
758 if self
.params
.get('no_warnings'):
760 if not self
.params
.get('no_color') and self
._err
_file
.isatty() and compat_os_name
!= 'nt':
761 _msg_header
= '\033[0;33mWARNING:\033[0m'
763 _msg_header
= 'WARNING:'
764 warning_message
= '%s %s' % (_msg_header
, message
)
765 self
.to_stderr(warning_message
)
767 def report_error(self
, message
, tb
=None):
769 Do the same as trouble, but prefixes the message with 'ERROR:', colored
770 in red if stderr is a tty file.
772 if not self
.params
.get('no_color') and self
._err
_file
.isatty() and compat_os_name
!= 'nt':
773 _msg_header
= '\033[0;31mERROR:\033[0m'
775 _msg_header
= 'ERROR:'
776 error_message
= '%s %s' % (_msg_header
, message
)
777 self
.trouble(error_message
, tb
)
779 def write_debug(self
, message
):
780 '''Log debug message or Print message to stderr'''
781 if not self
.params
.get('verbose', False):
783 message
= '[debug] %s' % message
784 if self
.params
.get('logger'):
785 self
.params
['logger'].debug(message
)
787 self
._write
_string
('%s\n' % message
)
789 def report_file_already_downloaded(self
, file_name
):
790 """Report file has already been fully downloaded."""
792 self
.to_screen('[download] %s has already been downloaded' % file_name
)
793 except UnicodeEncodeError:
794 self
.to_screen('[download] The file has already been downloaded')
796 def report_file_delete(self
, file_name
):
797 """Report that existing file will be deleted."""
799 self
.to_screen('Deleting existing file %s' % file_name
)
800 except UnicodeEncodeError:
801 self
.to_screen('Deleting existing file')
803 def parse_outtmpl(self
):
804 outtmpl_dict
= self
.params
.get('outtmpl', {})
805 if not isinstance(outtmpl_dict
, dict):
806 outtmpl_dict
= {'default': outtmpl_dict}
807 outtmpl_dict
.update({
808 k
: v
for k
, v
in DEFAULT_OUTTMPL
.items()
809 if not outtmpl_dict
.get(k
)})
810 for key
, val
in outtmpl_dict
.items():
811 if isinstance(val
, bytes):
813 'Parameter outtmpl is bytes, but should be a unicode string. '
814 'Put from __future__ import unicode_literals at the top of your code file or consider switching to Python 3.x.')
817 def prepare_outtmpl(self
, outtmpl
, info_dict
, sanitize
=None):
818 """ Make the template and info_dict suitable for substitution (outtmpl % info_dict)"""
819 template_dict
= dict(info_dict
)
820 na
= self
.params
.get('outtmpl_na_placeholder', 'NA')
823 template_dict
['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
824 formatSeconds(info_dict
['duration'], '-' if sanitize
else ':')
825 if info_dict
.get('duration', None) is not None
829 template_dict
['epoch'] = int(time
.time())
832 autonumber_size
= self
.params
.get('autonumber_size')
833 if autonumber_size
is None:
835 template_dict
['autonumber'] = self
.params
.get('autonumber_start', 1) - 1 + self
._num
_downloads
837 # resolution if not defined
838 if template_dict
.get('resolution') is None:
839 if template_dict
.get('width') and template_dict
.get('height'):
840 template_dict
['resolution'] = '%dx%d' % (template_dict
['width'], template_dict
['height'])
841 elif template_dict
.get('height'):
842 template_dict
['resolution'] = '%sp' % template_dict
['height']
843 elif template_dict
.get('width'):
844 template_dict
['resolution'] = '%dx?' % template_dict
['width']
846 # For fields playlist_index and autonumber convert all occurrences
847 # of %(field)s to %(field)0Nd for backward compatibility
848 field_size_compat_map
= {
849 'playlist_index': len(str(template_dict
.get('_last_playlist_index') or '')),
850 'autonumber': autonumber_size
,
852 FIELD_SIZE_COMPAT_RE
= r
'(?<!%)%\((?P<field>autonumber|playlist_index)\)s'
853 mobj
= re
.search(FIELD_SIZE_COMPAT_RE
, outtmpl
)
856 FIELD_SIZE_COMPAT_RE
,
857 r
'%%(\1)0%dd' % field_size_compat_map
[mobj
.group('field')],
860 numeric_fields
= list(self
._NUMERIC
_FIELDS
)
862 sanitize
= lambda k
, v
: v
864 EXTERNAL_FORMAT_RE
= FORMAT_RE
.format('(?P<key>[^)]*)')
865 # Field is of the form key1.key2...
866 # where keys (except first) can be string, int or slice
867 FIELD_RE
= r
'\w+(?:\.(?:\w+|[-\d]*(?::[-\d]*){0,2}))*'
868 INTERNAL_FORMAT_RE
= re
.compile(r
'''(?x)
871 (?P<maths>(?:[-+]-?(?:\d+(?:\.\d+)?|{0}))*)
872 (?:>(?P<strf_format>.+?))?
873 (?:\|(?P<default>.*?))?
874 $'''.format(FIELD_RE
))
875 MATH_OPERATORS_RE
= re
.compile(r
'(?<![-+])([-+])')
880 for outer_mobj
in re
.finditer(EXTERNAL_FORMAT_RE
, outtmpl
):
881 final_key
= outer_mobj
.group('key')
882 str_type
= outer_mobj
.group('type')
884 mobj
= re
.match(INTERNAL_FORMAT_RE
, final_key
)
886 mobj
= mobj
.groupdict()
888 fields
= mobj
['fields'].split('.')
889 value
= traverse_dict(template_dict
, fields
)
892 value
= float_or_none(value
)
893 if value
is not None:
897 value
= float_or_none(value
)
899 for item
in MATH_OPERATORS_RE
.split(mobj
['maths'])[1:]:
905 item
, multiplier
= (item
[1:], -1) if item
[0] == '-' else (item
, 1)
906 offset
= float_or_none(item
)
908 offset
= float_or_none(traverse_dict(template_dict
, item
.split('.')))
910 value
= operator(value
, multiplier
* offset
)
911 except (TypeError, ZeroDivisionError):
915 operator
= MATH_FUNCTIONS
[item
]
916 # Datetime formatting
917 if mobj
['strf_format']:
918 value
= strftime_or_none(value
, mobj
['strf_format'])
920 if value
is None and mobj
['default'] is not None:
921 value
= mobj
['default']
923 if str_type
in 'crs' and value
is not None: # string
924 value
= sanitize('%{}'.format(str_type
) % fields
[-1], value
)
926 numeric_fields
.append(final_key
)
927 value
= float_or_none(value
)
928 if value
is not None:
929 template_dict
[final_key
] = value
931 # Missing numeric fields used together with integer presentation types
932 # in format specification will break the argument substitution since
933 # string NA placeholder is returned for missing fields. We will patch
934 # output template for missing fields to meet string presentation type.
935 for numeric_field
in numeric_fields
:
936 if template_dict
.get(numeric_field
) is None:
938 FORMAT_RE
.format(re
.escape(numeric_field
)),
939 r
'%({0})s'.format(numeric_field
), outtmpl
)
941 template_dict
= collections
.defaultdict(lambda: na
, (
942 (k
, v
if isinstance(v
, compat_numeric_types
) else sanitize(k
, v
))
943 for k
, v
in template_dict
.items() if v
is not None))
944 return outtmpl
, template_dict
946 def _prepare_filename(self
, info_dict
, tmpl_type
='default'):
948 sanitize
= lambda k
, v
: sanitize_filename(
950 restricted
=self
.params
.get('restrictfilenames'),
951 is_id
=(k
== 'id' or k
.endswith('_id')))
952 outtmpl
= self
.outtmpl_dict
.get(tmpl_type
, self
.outtmpl_dict
['default'])
953 outtmpl
, template_dict
= self
.prepare_outtmpl(outtmpl
, info_dict
, sanitize
)
955 # expand_path translates '%%' into '%' and '$$' into '$'
956 # correspondingly that is not what we want since we need to keep
957 # '%%' intact for template dict substitution step. Working around
958 # with boundary-alike separator hack.
959 sep
= ''.join([random
.choice(ascii_letters
) for _
in range(32)])
960 outtmpl
= outtmpl
.replace('%%', '%{0}%'.format(sep
)).replace('$$', '${0}$'.format(sep
))
962 # outtmpl should be expand_path'ed before template dict substitution
963 # because meta fields may contain env variables we don't want to
964 # be expanded. For example, for outtmpl "%(title)s.%(ext)s" and
965 # title "Hello $PATH", we don't want `$PATH` to be expanded.
966 filename
= expand_path(outtmpl
).replace(sep
, '') % template_dict
968 force_ext
= OUTTMPL_TYPES
.get(tmpl_type
)
969 if force_ext
is not None:
970 filename
= replace_extension(filename
, force_ext
, template_dict
.get('ext'))
972 # https://github.com/blackjack4494/youtube-dlc/issues/85
973 trim_file_name
= self
.params
.get('trim_file_name', False)
975 fn_groups
= filename
.rsplit('.')
978 if len(fn_groups
) > 2:
979 sub_ext
= fn_groups
[-2]
980 filename
= '.'.join(filter(None, [fn_groups
[0][:trim_file_name
], sub_ext
, ext
]))
983 except ValueError as err
:
984 self
.report_error('Error in output template: ' + str(err
) + ' (encoding: ' + repr(preferredencoding()) + ')')
987 def prepare_filename(self
, info_dict
, dir_type
='', warn
=False):
988 """Generate the output filename."""
989 paths
= self
.params
.get('paths', {})
990 assert isinstance(paths
, dict)
991 filename
= self
._prepare
_filename
(info_dict
, dir_type
or 'default')
993 if warn
and not self
.__prepare
_filename
_warned
:
996 elif filename
== '-':
997 self
.report_warning('--paths is ignored when an outputting to stdout')
998 elif os
.path
.isabs(filename
):
999 self
.report_warning('--paths is ignored since an absolute path is given in output template')
1000 self
.__prepare
_filename
_warned
= True
1001 if filename
== '-' or not filename
:
1004 homepath
= expand_path(paths
.get('home', '').strip())
1005 assert isinstance(homepath
, compat_str
)
1006 subdir
= expand_path(paths
.get(dir_type
, '').strip()) if dir_type
else ''
1007 assert isinstance(subdir
, compat_str
)
1008 path
= os
.path
.join(homepath
, subdir
, filename
)
1010 # Temporary fix for #4787
1011 # 'Treat' all problem characters by passing filename through preferredencoding
1012 # to workaround encoding issues with subprocess on python2 @ Windows
1013 if sys
.version_info
< (3, 0) and sys
.platform
== 'win32':
1014 path
= encodeFilename(path
, True).decode(preferredencoding())
1015 return sanitize_path(path
, force
=self
.params
.get('windowsfilenames'))
1017 def _match_entry(self
, info_dict
, incomplete
):
1018 """ Returns None if the file should be downloaded """
1021 video_title
= info_dict
.get('title', info_dict
.get('id', 'video'))
1022 if 'title' in info_dict
:
1023 # This can happen when we're just evaluating the playlist
1024 title
= info_dict
['title']
1025 matchtitle
= self
.params
.get('matchtitle', False)
1027 if not re
.search(matchtitle
, title
, re
.IGNORECASE
):
1028 return '"' + title
+ '" title did not match pattern "' + matchtitle
+ '"'
1029 rejecttitle
= self
.params
.get('rejecttitle', False)
1031 if re
.search(rejecttitle
, title
, re
.IGNORECASE
):
1032 return '"' + title
+ '" title matched reject pattern "' + rejecttitle
+ '"'
1033 date
= info_dict
.get('upload_date')
1034 if date
is not None:
1035 dateRange
= self
.params
.get('daterange', DateRange())
1036 if date
not in dateRange
:
1037 return '%s upload date is not in range %s' % (date_from_str(date
).isoformat(), dateRange
)
1038 view_count
= info_dict
.get('view_count')
1039 if view_count
is not None:
1040 min_views
= self
.params
.get('min_views')
1041 if min_views
is not None and view_count
< min_views
:
1042 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title
, view_count
, min_views
)
1043 max_views
= self
.params
.get('max_views')
1044 if max_views
is not None and view_count
> max_views
:
1045 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title
, view_count
, max_views
)
1046 if age_restricted(info_dict
.get('age_limit'), self
.params
.get('age_limit')):
1047 return 'Skipping "%s" because it is age restricted' % video_title
1048 if self
.in_download_archive(info_dict
):
1049 return '%s has already been recorded in archive' % video_title
1052 match_filter
= self
.params
.get('match_filter')
1053 if match_filter
is not None:
1054 ret
= match_filter(info_dict
)
1059 reason
= check_filter()
1060 if reason
is not None:
1061 self
.to_screen('[download] ' + reason
)
1062 if reason
.endswith('has already been recorded in the archive') and self
.params
.get('break_on_existing', False):
1063 raise ExistingVideoReached()
1064 elif self
.params
.get('break_on_reject', False):
1065 raise RejectedVideoReached()
1069 def add_extra_info(info_dict
, extra_info
):
1070 '''Set the keys from extra_info in info dict if they are missing'''
1071 for key
, value
in extra_info
.items():
1072 info_dict
.setdefault(key
, value
)
1074 def extract_info(self
, url
, download
=True, ie_key
=None, extra_info
={},
1075 process
=True, force_generic_extractor
=False):
1077 Return a list with a dictionary for each video extracted.
1080 url -- URL to extract
1083 download -- whether to download videos during extraction
1084 ie_key -- extractor key hint
1085 extra_info -- dictionary containing the extra values to add to each result
1086 process -- whether to resolve all unresolved references (URLs, playlist items),
1087 must be True for download to work.
1088 force_generic_extractor -- force using the generic extractor
1091 if not ie_key
and force_generic_extractor
:
1095 ies
= [self
.get_info_extractor(ie_key
)]
1100 if not ie
.suitable(url
):
1103 ie_key
= ie
.ie_key()
1104 ie
= self
.get_info_extractor(ie_key
)
1105 if not ie
.working():
1106 self
.report_warning('The program functionality for this site has been marked as broken, '
1107 'and will probably not work.')
1110 temp_id
= str_or_none(
1111 ie
.extract_id(url
) if callable(getattr(ie
, 'extract_id', None))
1112 else ie
._match
_id
(url
))
1113 except (AssertionError, IndexError, AttributeError):
1115 if temp_id
is not None and self
.in_download_archive({'id': temp_id, 'ie_key': ie_key}
):
1116 self
.to_screen("[%s] %s: has already been recorded in archive" % (
1119 return self
.__extract
_info
(url
, ie
, download
, extra_info
, process
)
1121 self
.report_error('no suitable InfoExtractor for URL %s' % url
)
1123 def __handle_extraction_exceptions(func
):
1124 def wrapper(self
, *args
, **kwargs
):
1126 return func(self
, *args
, **kwargs
)
1127 except GeoRestrictedError
as e
:
1130 msg
+= '\nThis video is available in %s.' % ', '.join(
1131 map(ISO3166Utils
.short2full
, e
.countries
))
1132 msg
+= '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1133 self
.report_error(msg
)
1134 except ExtractorError
as e
: # An error we somewhat expected
1135 self
.report_error(compat_str(e
), e
.format_traceback())
1136 except (MaxDownloadsReached
, ExistingVideoReached
, RejectedVideoReached
):
1138 except Exception as e
:
1139 if self
.params
.get('ignoreerrors', False):
1140 self
.report_error(error_to_compat_str(e
), tb
=encode_compat_str(traceback
.format_exc()))
1145 @__handle_extraction_exceptions
1146 def __extract_info(self
, url
, ie
, download
, extra_info
, process
):
1147 ie_result
= ie
.extract(url
)
1148 if ie_result
is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
1150 if isinstance(ie_result
, list):
1151 # Backwards compatibility: old IE result format
1153 '_type': 'compat_list',
1154 'entries': ie_result
,
1156 self
.add_default_extra_info(ie_result
, ie
, url
)
1158 return self
.process_ie_result(ie_result
, download
, extra_info
)
1162 def add_default_extra_info(self
, ie_result
, ie
, url
):
1163 self
.add_extra_info(ie_result
, {
1164 'extractor': ie
.IE_NAME
,
1166 'webpage_url_basename': url_basename(url
),
1167 'extractor_key': ie
.ie_key(),
1170 def process_ie_result(self
, ie_result
, download
=True, extra_info
={}):
1172 Take the result of the ie(may be modified) and resolve all unresolved
1173 references (URLs, playlist items).
1175 It will also download the videos if 'download'.
1176 Returns the resolved ie_result.
1178 result_type
= ie_result
.get('_type', 'video')
1180 if result_type
in ('url', 'url_transparent'):
1181 ie_result
['url'] = sanitize_url(ie_result
['url'])
1182 extract_flat
= self
.params
.get('extract_flat', False)
1183 if ((extract_flat
== 'in_playlist' and 'playlist' in extra_info
)
1184 or extract_flat
is True):
1185 self
.__forced
_printings
(ie_result
, self
.prepare_filename(ie_result
), incomplete
=True)
1188 if result_type
== 'video':
1189 self
.add_extra_info(ie_result
, extra_info
)
1190 ie_result
= self
.process_video_result(ie_result
, download
=download
)
1191 additional_urls
= ie_result
.get('additional_urls')
1193 # TODO: Improve MetadataFromFieldPP to allow setting a list
1194 if isinstance(additional_urls
, compat_str
):
1195 additional_urls
= [additional_urls
]
1197 '[info] %s: %d additional URL(s) requested' % (ie_result
['id'], len(additional_urls
)))
1198 self
.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls
))
1199 ie_result
['additional_entries'] = [
1201 url
, download
, extra_info
,
1202 force_generic_extractor
=self
.params
.get('force_generic_extractor'))
1203 for url
in additional_urls
1206 elif result_type
== 'url':
1207 # We have to add extra_info to the results because it may be
1208 # contained in a playlist
1209 return self
.extract_info(
1210 ie_result
['url'], download
,
1211 ie_key
=ie_result
.get('ie_key'),
1212 extra_info
=extra_info
)
1213 elif result_type
== 'url_transparent':
1214 # Use the information from the embedding page
1215 info
= self
.extract_info(
1216 ie_result
['url'], ie_key
=ie_result
.get('ie_key'),
1217 extra_info
=extra_info
, download
=False, process
=False)
1219 # extract_info may return None when ignoreerrors is enabled and
1220 # extraction failed with an error, don't crash and return early
1225 force_properties
= dict(
1226 (k
, v
) for k
, v
in ie_result
.items() if v
is not None)
1227 for f
in ('_type', 'url', 'id', 'extractor', 'extractor_key', 'ie_key'):
1228 if f
in force_properties
:
1229 del force_properties
[f
]
1230 new_result
= info
.copy()
1231 new_result
.update(force_properties
)
1233 # Extracted info may not be a video result (i.e.
1234 # info.get('_type', 'video') != video) but rather an url or
1235 # url_transparent. In such cases outer metadata (from ie_result)
1236 # should be propagated to inner one (info). For this to happen
1237 # _type of info should be overridden with url_transparent. This
1238 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
1239 if new_result
.get('_type') == 'url':
1240 new_result
['_type'] = 'url_transparent'
1242 return self
.process_ie_result(
1243 new_result
, download
=download
, extra_info
=extra_info
)
1244 elif result_type
in ('playlist', 'multi_video'):
1245 # Protect from infinite recursion due to recursively nested playlists
1246 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
1247 webpage_url
= ie_result
['webpage_url']
1248 if webpage_url
in self
._playlist
_urls
:
1250 '[download] Skipping already downloaded playlist: %s'
1251 % ie_result
.get('title') or ie_result
.get('id'))
1254 self
._playlist
_level
+= 1
1255 self
._playlist
_urls
.add(webpage_url
)
1257 return self
.__process
_playlist
(ie_result
, download
)
1259 self
._playlist
_level
-= 1
1260 if not self
._playlist
_level
:
1261 self
._playlist
_urls
.clear()
1262 elif result_type
== 'compat_list':
1263 self
.report_warning(
1264 'Extractor %s returned a compat_list result. '
1265 'It needs to be updated.' % ie_result
.get('extractor'))
1268 self
.add_extra_info(
1271 'extractor': ie_result
['extractor'],
1272 'webpage_url': ie_result
['webpage_url'],
1273 'webpage_url_basename': url_basename(ie_result
['webpage_url']),
1274 'extractor_key': ie_result
['extractor_key'],
1278 ie_result
['entries'] = [
1279 self
.process_ie_result(_fixup(r
), download
, extra_info
)
1280 for r
in ie_result
['entries']
1284 raise Exception('Invalid result type: %s' % result_type
)
1286 def _ensure_dir_exists(self
, path
):
1287 return make_dir(path
, self
.report_error
)
1289 def __process_playlist(self
, ie_result
, download
):
1290 # We process each entry in the playlist
1291 playlist
= ie_result
.get('title') or ie_result
.get('id')
1292 self
.to_screen('[download] Downloading playlist: %s' % playlist
)
1294 if 'entries' not in ie_result
:
1295 raise EntryNotInPlaylist()
1296 incomplete_entries
= bool(ie_result
.get('requested_entries'))
1297 if incomplete_entries
:
1298 def fill_missing_entries(entries
, indexes
):
1299 ret
= [None] * max(*indexes
)
1300 for i
, entry
in zip(indexes
, entries
):
1303 ie_result
['entries'] = fill_missing_entries(ie_result
['entries'], ie_result
['requested_entries'])
1305 playlist_results
= []
1307 playliststart
= self
.params
.get('playliststart', 1) - 1
1308 playlistend
= self
.params
.get('playlistend')
1309 # For backwards compatibility, interpret -1 as whole list
1310 if playlistend
== -1:
1313 playlistitems_str
= self
.params
.get('playlist_items')
1314 playlistitems
= None
1315 if playlistitems_str
is not None:
1316 def iter_playlistitems(format
):
1317 for string_segment
in format
.split(','):
1318 if '-' in string_segment
:
1319 start
, end
= string_segment
.split('-')
1320 for item
in range(int(start
), int(end
) + 1):
1323 yield int(string_segment
)
1324 playlistitems
= orderedSet(iter_playlistitems(playlistitems_str
))
1326 ie_entries
= ie_result
['entries']
1328 def make_playlistitems_entries(list_ie_entries
):
1329 num_entries
= len(list_ie_entries
)
1330 for i
in playlistitems
:
1331 if -num_entries
< i
<= num_entries
:
1332 yield list_ie_entries
[i
- 1]
1333 elif incomplete_entries
:
1334 raise EntryNotInPlaylist()
1336 if isinstance(ie_entries
, list):
1337 n_all_entries
= len(ie_entries
)
1339 entries
= list(make_playlistitems_entries(ie_entries
))
1341 entries
= ie_entries
[playliststart
:playlistend
]
1342 n_entries
= len(entries
)
1343 msg
= 'Collected %d videos; downloading %d of them' % (n_all_entries
, n_entries
)
1344 elif isinstance(ie_entries
, PagedList
):
1347 for item
in playlistitems
:
1348 entries
.extend(ie_entries
.getslice(
1352 entries
= ie_entries
.getslice(
1353 playliststart
, playlistend
)
1354 n_entries
= len(entries
)
1355 msg
= 'Downloading %d videos' % n_entries
1358 entries
= list(make_playlistitems_entries(list(itertools
.islice(
1359 ie_entries
, 0, max(playlistitems
)))))
1361 entries
= list(itertools
.islice(
1362 ie_entries
, playliststart
, playlistend
))
1363 n_entries
= len(entries
)
1364 msg
= 'Downloading %d videos' % n_entries
1366 if any((entry
is None for entry
in entries
)):
1367 raise EntryNotInPlaylist()
1368 if not playlistitems
and (playliststart
or playlistend
):
1369 playlistitems
= list(range(1 + playliststart
, 1 + playliststart
+ len(entries
)))
1370 ie_result
['entries'] = entries
1371 ie_result
['requested_entries'] = playlistitems
1373 if self
.params
.get('allow_playlist_files', True):
1375 'playlist': playlist
,
1376 'playlist_id': ie_result
.get('id'),
1377 'playlist_title': ie_result
.get('title'),
1378 'playlist_uploader': ie_result
.get('uploader'),
1379 'playlist_uploader_id': ie_result
.get('uploader_id'),
1380 'playlist_index': 0,
1382 ie_copy
.update(dict(ie_result
))
1384 if self
.params
.get('writeinfojson', False):
1385 infofn
= self
.prepare_filename(ie_copy
, 'pl_infojson')
1386 if not self
._ensure
_dir
_exists
(encodeFilename(infofn
)):
1388 if not self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(infofn
)):
1389 self
.to_screen('[info] Playlist metadata is already present')
1391 self
.to_screen('[info] Writing playlist metadata as JSON to: ' + infofn
)
1393 write_json_file(self
.filter_requested_info(ie_result
, self
.params
.get('clean_infojson', True)), infofn
)
1394 except (OSError, IOError):
1395 self
.report_error('Cannot write playlist metadata to JSON file ' + infofn
)
1397 # TODO: This should be passed to ThumbnailsConvertor if necessary
1398 self
._write
_thumbnails
(ie_copy
, self
.prepare_filename(ie_copy
, 'pl_thumbnail'))
1400 if self
.params
.get('writedescription', False):
1401 descfn
= self
.prepare_filename(ie_copy
, 'pl_description')
1402 if not self
._ensure
_dir
_exists
(encodeFilename(descfn
)):
1404 if not self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(descfn
)):
1405 self
.to_screen('[info] Playlist description is already present')
1406 elif ie_result
.get('description') is None:
1407 self
.report_warning('There\'s no playlist description to write.')
1410 self
.to_screen('[info] Writing playlist description to: ' + descfn
)
1411 with io
.open(encodeFilename(descfn
), 'w', encoding
='utf-8') as descfile
:
1412 descfile
.write(ie_result
['description'])
1413 except (OSError, IOError):
1414 self
.report_error('Cannot write playlist description file ' + descfn
)
1417 # Save playlist_index before re-ordering
1419 ((playlistitems
[i
- 1] if playlistitems
else i
), entry
)
1420 for i
, entry
in enumerate(entries
, 1)]
1422 if self
.params
.get('playlistreverse', False):
1423 entries
= entries
[::-1]
1424 if self
.params
.get('playlistrandom', False):
1425 random
.shuffle(entries
)
1427 x_forwarded_for
= ie_result
.get('__x_forwarded_for_ip')
1429 self
.to_screen('[%s] playlist %s: %s' % (ie_result
['extractor'], playlist
, msg
))
1431 max_failures
= self
.params
.get('skip_playlist_after_errors') or float('inf')
1432 for i
, entry_tuple
in enumerate(entries
, 1):
1433 playlist_index
, entry
= entry_tuple
1434 if 'playlist_index' in self
.params
.get('compat_options', []):
1435 playlist_index
= playlistitems
[i
- 1] if playlistitems
else i
1436 self
.to_screen('[download] Downloading video %s of %s' % (i
, n_entries
))
1437 # This __x_forwarded_for_ip thing is a bit ugly but requires
1440 entry
['__x_forwarded_for_ip'] = x_forwarded_for
1442 'n_entries': n_entries
,
1443 '_last_playlist_index': max(playlistitems
) if playlistitems
else (playlistend
or n_entries
),
1444 'playlist_index': playlist_index
,
1445 'playlist_autonumber': i
,
1446 'playlist': playlist
,
1447 'playlist_id': ie_result
.get('id'),
1448 'playlist_title': ie_result
.get('title'),
1449 'playlist_uploader': ie_result
.get('uploader'),
1450 'playlist_uploader_id': ie_result
.get('uploader_id'),
1451 'extractor': ie_result
['extractor'],
1452 'webpage_url': ie_result
['webpage_url'],
1453 'webpage_url_basename': url_basename(ie_result
['webpage_url']),
1454 'extractor_key': ie_result
['extractor_key'],
1457 if self
._match
_entry
(entry
, incomplete
=True) is not None:
1460 entry_result
= self
.__process
_iterable
_entry
(entry
, download
, extra
)
1461 if not entry_result
:
1463 if failures
>= max_failures
:
1465 'Skipping the remaining entries in playlist "%s" since %d items failed extraction' % (playlist
, failures
))
1467 # TODO: skip failed (empty) entries?
1468 playlist_results
.append(entry_result
)
1469 ie_result
['entries'] = playlist_results
1470 self
.to_screen('[download] Finished downloading playlist: %s' % playlist
)
1473 @__handle_extraction_exceptions
1474 def __process_iterable_entry(self
, entry
, download
, extra_info
):
1475 return self
.process_ie_result(
1476 entry
, download
=download
, extra_info
=extra_info
)
1478 def _build_format_filter(self
, filter_spec
):
1479 " Returns a function to filter the formats according to the filter_spec "
1489 operator_rex
= re
.compile(r
'''(?x)\s*
1490 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)
1491 \s*(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1492 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)
1494 ''' % '|'.join(map(re
.escape
, OPERATORS
.keys())))
1495 m
= operator_rex
.search(filter_spec
)
1498 comparison_value
= int(m
.group('value'))
1500 comparison_value
= parse_filesize(m
.group('value'))
1501 if comparison_value
is None:
1502 comparison_value
= parse_filesize(m
.group('value') + 'B')
1503 if comparison_value
is None:
1505 'Invalid value %r in format specification %r' % (
1506 m
.group('value'), filter_spec
))
1507 op
= OPERATORS
[m
.group('op')]
1512 '^=': lambda attr
, value
: attr
.startswith(value
),
1513 '$=': lambda attr
, value
: attr
.endswith(value
),
1514 '*=': lambda attr
, value
: value
in attr
,
1516 str_operator_rex
= re
.compile(r
'''(?x)
1517 \s*(?P<key>[a-zA-Z0-9._-]+)
1518 \s*(?P<negation>!\s*)?(?P<op>%s)(?P<none_inclusive>\s*\?)?
1519 \s*(?P<value>[a-zA-Z0-9._-]+)
1521 ''' % '|'.join(map(re
.escape
, STR_OPERATORS
.keys())))
1522 m
= str_operator_rex
.search(filter_spec
)
1524 comparison_value
= m
.group('value')
1525 str_op
= STR_OPERATORS
[m
.group('op')]
1526 if m
.group('negation'):
1527 op
= lambda attr
, value
: not str_op(attr
, value
)
1532 raise ValueError('Invalid filter specification %r' % filter_spec
)
1535 actual_value
= f
.get(m
.group('key'))
1536 if actual_value
is None:
1537 return m
.group('none_inclusive')
1538 return op(actual_value
, comparison_value
)
1541 def _default_format_spec(self
, info_dict
, download
=True):
1544 merger
= FFmpegMergerPP(self
)
1545 return merger
.available
and merger
.can_merge()
1548 not self
.params
.get('simulate', False)
1552 or info_dict
.get('is_live', False)
1553 or self
.outtmpl_dict
['default'] == '-'))
1556 or self
.params
.get('allow_multiple_audio_streams', False)
1557 or 'format-spec' in self
.params
.get('compat_opts', []))
1560 'best/bestvideo+bestaudio' if prefer_best
1561 else 'bestvideo*+bestaudio/best' if not compat
1562 else 'bestvideo+bestaudio/best')
1564 def build_format_selector(self
, format_spec
):
1565 def syntax_error(note
, start
):
1567 'Invalid format specification: '
1568 '{0}\n\t{1}\n\t{2}^'.format(note
, format_spec
, ' ' * start
[1]))
1569 return SyntaxError(message
)
1571 PICKFIRST
= 'PICKFIRST'
1575 FormatSelector
= collections
.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
1577 allow_multiple_streams
= {'audio': self
.params
.get('allow_multiple_audio_streams', False),
1578 'video': self
.params
.get('allow_multiple_video_streams', False)}
1580 check_formats
= self
.params
.get('check_formats')
1582 def _parse_filter(tokens
):
1584 for type, string
, start
, _
, _
in tokens
:
1585 if type == tokenize
.OP
and string
== ']':
1586 return ''.join(filter_parts
)
1588 filter_parts
.append(string
)
1590 def _remove_unused_ops(tokens
):
1591 # Remove operators that we don't use and join them with the surrounding strings
1592 # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
1593 ALLOWED_OPS
= ('/', '+', ',', '(', ')')
1594 last_string
, last_start
, last_end
, last_line
= None, None, None, None
1595 for type, string
, start
, end
, line
in tokens
:
1596 if type == tokenize
.OP
and string
== '[':
1598 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
1600 yield type, string
, start
, end
, line
1601 # everything inside brackets will be handled by _parse_filter
1602 for type, string
, start
, end
, line
in tokens
:
1603 yield type, string
, start
, end
, line
1604 if type == tokenize
.OP
and string
== ']':
1606 elif type == tokenize
.OP
and string
in ALLOWED_OPS
:
1608 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
1610 yield type, string
, start
, end
, line
1611 elif type in [tokenize
.NAME
, tokenize
.NUMBER
, tokenize
.OP
]:
1613 last_string
= string
1617 last_string
+= string
1619 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
1621 def _parse_format_selection(tokens
, inside_merge
=False, inside_choice
=False, inside_group
=False):
1623 current_selector
= None
1624 for type, string
, start
, _
, _
in tokens
:
1625 # ENCODING is only defined in python 3.x
1626 if type == getattr(tokenize
, 'ENCODING', None):
1628 elif type in [tokenize
.NAME
, tokenize
.NUMBER
]:
1629 current_selector
= FormatSelector(SINGLE
, string
, [])
1630 elif type == tokenize
.OP
:
1632 if not inside_group
:
1633 # ')' will be handled by the parentheses group
1634 tokens
.restore_last_token()
1636 elif inside_merge
and string
in ['/', ',']:
1637 tokens
.restore_last_token()
1639 elif inside_choice
and string
== ',':
1640 tokens
.restore_last_token()
1643 if not current_selector
:
1644 raise syntax_error('"," must follow a format selector', start
)
1645 selectors
.append(current_selector
)
1646 current_selector
= None
1648 if not current_selector
:
1649 raise syntax_error('"/" must follow a format selector', start
)
1650 first_choice
= current_selector
1651 second_choice
= _parse_format_selection(tokens
, inside_choice
=True)
1652 current_selector
= FormatSelector(PICKFIRST
, (first_choice
, second_choice
), [])
1654 if not current_selector
:
1655 current_selector
= FormatSelector(SINGLE
, 'best', [])
1656 format_filter
= _parse_filter(tokens
)
1657 current_selector
.filters
.append(format_filter
)
1659 if current_selector
:
1660 raise syntax_error('Unexpected "("', start
)
1661 group
= _parse_format_selection(tokens
, inside_group
=True)
1662 current_selector
= FormatSelector(GROUP
, group
, [])
1664 if not current_selector
:
1665 raise syntax_error('Unexpected "+"', start
)
1666 selector_1
= current_selector
1667 selector_2
= _parse_format_selection(tokens
, inside_merge
=True)
1669 raise syntax_error('Expected a selector', start
)
1670 current_selector
= FormatSelector(MERGE
, (selector_1
, selector_2
), [])
1672 raise syntax_error('Operator not recognized: "{0}"'.format(string
), start
)
1673 elif type == tokenize
.ENDMARKER
:
1675 if current_selector
:
1676 selectors
.append(current_selector
)
1679 def _merge(formats_pair
):
1680 format_1
, format_2
= formats_pair
1683 formats_info
.extend(format_1
.get('requested_formats', (format_1
,)))
1684 formats_info
.extend(format_2
.get('requested_formats', (format_2
,)))
1686 if not allow_multiple_streams
['video'] or not allow_multiple_streams
['audio']:
1687 get_no_more
= {"video": False, "audio": False}
1688 for (i
, fmt_info
) in enumerate(formats_info
):
1689 for aud_vid
in ["audio", "video"]:
1690 if not allow_multiple_streams
[aud_vid
] and fmt_info
.get(aud_vid
[0] + 'codec') != 'none':
1691 if get_no_more
[aud_vid
]:
1693 get_no_more
[aud_vid
] = True
1695 if len(formats_info
) == 1:
1696 return formats_info
[0]
1698 video_fmts
= [fmt_info
for fmt_info
in formats_info
if fmt_info
.get('vcodec') != 'none']
1699 audio_fmts
= [fmt_info
for fmt_info
in formats_info
if fmt_info
.get('acodec') != 'none']
1701 the_only_video
= video_fmts
[0] if len(video_fmts
) == 1 else None
1702 the_only_audio
= audio_fmts
[0] if len(audio_fmts
) == 1 else None
1704 output_ext
= self
.params
.get('merge_output_format')
1707 output_ext
= the_only_video
['ext']
1708 elif the_only_audio
and not video_fmts
:
1709 output_ext
= the_only_audio
['ext']
1714 'requested_formats': formats_info
,
1715 'format': '+'.join(fmt_info
.get('format') for fmt_info
in formats_info
),
1716 'format_id': '+'.join(fmt_info
.get('format_id') for fmt_info
in formats_info
),
1722 'width': the_only_video
.get('width'),
1723 'height': the_only_video
.get('height'),
1724 'resolution': the_only_video
.get('resolution') or self
.format_resolution(the_only_video
),
1725 'fps': the_only_video
.get('fps'),
1726 'vcodec': the_only_video
.get('vcodec'),
1727 'vbr': the_only_video
.get('vbr'),
1728 'stretched_ratio': the_only_video
.get('stretched_ratio'),
1733 'acodec': the_only_audio
.get('acodec'),
1734 'abr': the_only_audio
.get('abr'),
1739 def _check_formats(formats
):
1741 self
.to_screen('[info] Testing format %s' % f
['format_id'])
1742 paths
= self
.params
.get('paths', {})
1743 temp_file
= os
.path
.join(
1744 expand_path(paths
.get('home', '').strip()),
1745 expand_path(paths
.get('temp', '').strip()),
1746 'ytdl.%s.f%s.check-format' % (random_uuidv4(), f
['format_id']))
1748 dl
, _
= self
.dl(temp_file
, f
, test
=True)
1749 except (ExtractorError
, IOError, OSError, ValueError) + network_exceptions
:
1752 if os
.path
.exists(temp_file
):
1753 os
.remove(temp_file
)
1757 self
.to_screen('[info] Unable to download format %s. Skipping...' % f
['format_id'])
1759 def _build_selector_function(selector
):
1760 if isinstance(selector
, list): # ,
1761 fs
= [_build_selector_function(s
) for s
in selector
]
1763 def selector_function(ctx
):
1765 for format
in f(ctx
):
1767 return selector_function
1769 elif selector
.type == GROUP
: # ()
1770 selector_function
= _build_selector_function(selector
.selector
)
1772 elif selector
.type == PICKFIRST
: # /
1773 fs
= [_build_selector_function(s
) for s
in selector
.selector
]
1775 def selector_function(ctx
):
1777 picked_formats
= list(f(ctx
))
1779 return picked_formats
1782 elif selector
.type == SINGLE
: # atom
1783 format_spec
= selector
.selector
or 'best'
1785 # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
1786 if format_spec
== 'all':
1787 def selector_function(ctx
):
1788 formats
= list(ctx
['formats'])
1790 formats
= _check_formats(formats
)
1793 elif format_spec
== 'mergeall':
1794 def selector_function(ctx
):
1795 formats
= list(_check_formats(ctx
['formats']))
1798 merged_format
= formats
[-1]
1799 for f
in formats
[-2::-1]:
1800 merged_format
= _merge((merged_format
, f
))
1804 format_fallback
, format_reverse
, format_idx
= False, True, 1
1806 r
'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
1808 if mobj
is not None:
1809 format_idx
= int_or_none(mobj
.group('n'), default
=1)
1810 format_reverse
= mobj
.group('bw')[0] == 'b'
1811 format_type
= (mobj
.group('type') or [None])[0]
1812 not_format_type
= {'v': 'a', 'a': 'v'}
.get(format_type
)
1813 format_modified
= mobj
.group('mod') is not None
1815 format_fallback
= not format_type
and not format_modified
# for b, w
1817 (lambda f
: f
.get('%scodec' % format_type
) != 'none')
1818 if format_type
and format_modified
# bv*, ba*, wv*, wa*
1819 else (lambda f
: f
.get('%scodec' % not_format_type
) == 'none')
1820 if format_type
# bv, ba, wv, wa
1821 else (lambda f
: f
.get('vcodec') != 'none' and f
.get('acodec') != 'none')
1822 if not format_modified
# b, w
1825 filter_f
= ((lambda f
: f
.get('ext') == format_spec
)
1826 if format_spec
in ['mp4', 'flv', 'webm', '3gp', 'm4a', 'mp3', 'ogg', 'aac', 'wav'] # extension
1827 else (lambda f
: f
.get('format_id') == format_spec
)) # id
1829 def selector_function(ctx
):
1830 formats
= list(ctx
['formats'])
1833 matches
= list(filter(filter_f
, formats
)) if filter_f
is not None else formats
1834 if format_fallback
and ctx
['incomplete_formats'] and not matches
:
1835 # for extractors with incomplete formats (audio only (soundcloud)
1836 # or video only (imgur)) best/worst will fallback to
1837 # best/worst {video,audio}-only format
1840 matches
= matches
[::-1]
1842 matches
= list(itertools
.islice(_check_formats(matches
), format_idx
))
1844 if -n
<= format_idx
- 1 < n
:
1845 yield matches
[format_idx
- 1]
1847 elif selector
.type == MERGE
: # +
1848 selector_1
, selector_2
= map(_build_selector_function
, selector
.selector
)
1850 def selector_function(ctx
):
1851 for pair
in itertools
.product(
1852 selector_1(copy
.deepcopy(ctx
)), selector_2(copy
.deepcopy(ctx
))):
1855 filters
= [self
._build
_format
_filter
(f
) for f
in selector
.filters
]
1857 def final_selector(ctx
):
1858 ctx_copy
= copy
.deepcopy(ctx
)
1859 for _filter
in filters
:
1860 ctx_copy
['formats'] = list(filter(_filter
, ctx_copy
['formats']))
1861 return selector_function(ctx_copy
)
1862 return final_selector
1864 stream
= io
.BytesIO(format_spec
.encode('utf-8'))
1866 tokens
= list(_remove_unused_ops(compat_tokenize_tokenize(stream
.readline
)))
1867 except tokenize
.TokenError
:
1868 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec
)))
1870 class TokenIterator(object):
1871 def __init__(self
, tokens
):
1872 self
.tokens
= tokens
1879 if self
.counter
>= len(self
.tokens
):
1880 raise StopIteration()
1881 value
= self
.tokens
[self
.counter
]
1887 def restore_last_token(self
):
1890 parsed_selector
= _parse_format_selection(iter(TokenIterator(tokens
)))
1891 return _build_selector_function(parsed_selector
)
1893 def _calc_headers(self
, info_dict
):
1894 res
= std_headers
.copy()
1896 add_headers
= info_dict
.get('http_headers')
1898 res
.update(add_headers
)
1900 cookies
= self
._calc
_cookies
(info_dict
)
1902 res
['Cookie'] = cookies
1904 if 'X-Forwarded-For' not in res
:
1905 x_forwarded_for_ip
= info_dict
.get('__x_forwarded_for_ip')
1906 if x_forwarded_for_ip
:
1907 res
['X-Forwarded-For'] = x_forwarded_for_ip
1911 def _calc_cookies(self
, info_dict
):
1912 pr
= sanitized_Request(info_dict
['url'])
1913 self
.cookiejar
.add_cookie_header(pr
)
1914 return pr
.get_header('Cookie')
1916 def process_video_result(self
, info_dict
, download
=True):
1917 assert info_dict
.get('_type', 'video') == 'video'
1919 if 'id' not in info_dict
:
1920 raise ExtractorError('Missing "id" field in extractor result')
1921 if 'title' not in info_dict
:
1922 raise ExtractorError('Missing "title" field in extractor result')
1924 def report_force_conversion(field
, field_not
, conversion
):
1925 self
.report_warning(
1926 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
1927 % (field
, field_not
, conversion
))
1929 def sanitize_string_field(info
, string_field
):
1930 field
= info
.get(string_field
)
1931 if field
is None or isinstance(field
, compat_str
):
1933 report_force_conversion(string_field
, 'a string', 'string')
1934 info
[string_field
] = compat_str(field
)
1936 def sanitize_numeric_fields(info
):
1937 for numeric_field
in self
._NUMERIC
_FIELDS
:
1938 field
= info
.get(numeric_field
)
1939 if field
is None or isinstance(field
, compat_numeric_types
):
1941 report_force_conversion(numeric_field
, 'numeric', 'int')
1942 info
[numeric_field
] = int_or_none(field
)
1944 sanitize_string_field(info_dict
, 'id')
1945 sanitize_numeric_fields(info_dict
)
1947 if 'playlist' not in info_dict
:
1948 # It isn't part of a playlist
1949 info_dict
['playlist'] = None
1950 info_dict
['playlist_index'] = None
1952 thumbnails
= info_dict
.get('thumbnails')
1953 if thumbnails
is None:
1954 thumbnail
= info_dict
.get('thumbnail')
1956 info_dict
['thumbnails'] = thumbnails
= [{'url': thumbnail}
]
1958 thumbnails
.sort(key
=lambda t
: (
1959 t
.get('preference') if t
.get('preference') is not None else -1,
1960 t
.get('width') if t
.get('width') is not None else -1,
1961 t
.get('height') if t
.get('height') is not None else -1,
1962 t
.get('id') if t
.get('id') is not None else '',
1964 for i
, t
in enumerate(thumbnails
):
1965 t
['url'] = sanitize_url(t
['url'])
1966 if t
.get('width') and t
.get('height'):
1967 t
['resolution'] = '%dx%d' % (t
['width'], t
['height'])
1968 if t
.get('id') is None:
1971 if self
.params
.get('list_thumbnails'):
1972 self
.list_thumbnails(info_dict
)
1975 thumbnail
= info_dict
.get('thumbnail')
1977 info_dict
['thumbnail'] = sanitize_url(thumbnail
)
1979 info_dict
['thumbnail'] = thumbnails
[-1]['url']
1981 if 'display_id' not in info_dict
and 'id' in info_dict
:
1982 info_dict
['display_id'] = info_dict
['id']
1984 for ts_key
, date_key
in (
1985 ('timestamp', 'upload_date'),
1986 ('release_timestamp', 'release_date'),
1988 if info_dict
.get(date_key
) is None and info_dict
.get(ts_key
) is not None:
1989 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
1990 # see http://bugs.python.org/issue1646728)
1992 upload_date
= datetime
.datetime
.utcfromtimestamp(info_dict
[ts_key
])
1993 info_dict
[date_key
] = upload_date
.strftime('%Y%m%d')
1994 except (ValueError, OverflowError, OSError):
1997 # Auto generate title fields corresponding to the *_number fields when missing
1998 # in order to always have clean titles. This is very common for TV series.
1999 for field
in ('chapter', 'season', 'episode'):
2000 if info_dict
.get('%s_number' % field
) is not None and not info_dict
.get(field
):
2001 info_dict
[field
] = '%s %d' % (field
.capitalize(), info_dict
['%s_number' % field
])
2003 for cc_kind
in ('subtitles', 'automatic_captions'):
2004 cc
= info_dict
.get(cc_kind
)
2006 for _
, subtitle
in cc
.items():
2007 for subtitle_format
in subtitle
:
2008 if subtitle_format
.get('url'):
2009 subtitle_format
['url'] = sanitize_url(subtitle_format
['url'])
2010 if subtitle_format
.get('ext') is None:
2011 subtitle_format
['ext'] = determine_ext(subtitle_format
['url']).lower()
2013 automatic_captions
= info_dict
.get('automatic_captions')
2014 subtitles
= info_dict
.get('subtitles')
2016 if self
.params
.get('listsubtitles', False):
2017 if 'automatic_captions' in info_dict
:
2018 self
.list_subtitles(
2019 info_dict
['id'], automatic_captions
, 'automatic captions')
2020 self
.list_subtitles(info_dict
['id'], subtitles
, 'subtitles')
2023 info_dict
['requested_subtitles'] = self
.process_subtitles(
2024 info_dict
['id'], subtitles
, automatic_captions
)
2026 # We now pick which formats have to be downloaded
2027 if info_dict
.get('formats') is None:
2028 # There's only one format available
2029 formats
= [info_dict
]
2031 formats
= info_dict
['formats']
2034 if not self
.params
.get('ignore_no_formats_error'):
2035 raise ExtractorError('No video formats found!')
2037 self
.report_warning('No video formats found!')
2039 def is_wellformed(f
):
2042 self
.report_warning(
2043 '"url" field is missing or empty - skipping format, '
2044 'there is an error in extractor')
2046 if isinstance(url
, bytes):
2047 sanitize_string_field(f
, 'url')
2050 # Filter out malformed formats for better extraction robustness
2051 formats
= list(filter(is_wellformed
, formats
))
2055 # We check that all the formats have the format and format_id fields
2056 for i
, format
in enumerate(formats
):
2057 sanitize_string_field(format
, 'format_id')
2058 sanitize_numeric_fields(format
)
2059 format
['url'] = sanitize_url(format
['url'])
2060 if not format
.get('format_id'):
2061 format
['format_id'] = compat_str(i
)
2063 # Sanitize format_id from characters used in format selector expression
2064 format
['format_id'] = re
.sub(r
'[\s,/+\[\]()]', '_', format
['format_id'])
2065 format_id
= format
['format_id']
2066 if format_id
not in formats_dict
:
2067 formats_dict
[format_id
] = []
2068 formats_dict
[format_id
].append(format
)
2070 # Make sure all formats have unique format_id
2071 for format_id
, ambiguous_formats
in formats_dict
.items():
2072 if len(ambiguous_formats
) > 1:
2073 for i
, format
in enumerate(ambiguous_formats
):
2074 format
['format_id'] = '%s-%d' % (format_id
, i
)
2076 for i
, format
in enumerate(formats
):
2077 if format
.get('format') is None:
2078 format
['format'] = '{id} - {res}{note}'.format(
2079 id=format
['format_id'],
2080 res
=self
.format_resolution(format
),
2081 note
=' ({0})'.format(format
['format_note']) if format
.get('format_note') is not None else '',
2083 # Automatically determine file extension if missing
2084 if format
.get('ext') is None:
2085 format
['ext'] = determine_ext(format
['url']).lower()
2086 # Automatically determine protocol if missing (useful for format
2087 # selection purposes)
2088 if format
.get('protocol') is None:
2089 format
['protocol'] = determine_protocol(format
)
2090 # Add HTTP headers, so that external programs can use them from the
2092 full_format_info
= info_dict
.copy()
2093 full_format_info
.update(format
)
2094 format
['http_headers'] = self
._calc
_headers
(full_format_info
)
2095 # Remove private housekeeping stuff
2096 if '__x_forwarded_for_ip' in info_dict
:
2097 del info_dict
['__x_forwarded_for_ip']
2099 # TODO Central sorting goes here
2101 if formats
and formats
[0] is not info_dict
:
2102 # only set the 'formats' fields if the original info_dict list them
2103 # otherwise we end up with a circular reference, the first (and unique)
2104 # element in the 'formats' field in info_dict is info_dict itself,
2105 # which can't be exported to json
2106 info_dict
['formats'] = formats
2108 info_dict
, _
= self
.pre_process(info_dict
)
2110 if self
.params
.get('listformats'):
2111 if not info_dict
.get('formats'):
2112 raise ExtractorError('No video formats found', expected
=True)
2113 self
.list_formats(info_dict
)
2116 req_format
= self
.params
.get('format')
2117 if req_format
is None:
2118 req_format
= self
._default
_format
_spec
(info_dict
, download
=download
)
2119 self
.write_debug('Default format spec: %s' % req_format
)
2121 format_selector
= self
.build_format_selector(req_format
)
2123 # While in format selection we may need to have an access to the original
2124 # format set in order to calculate some metrics or do some processing.
2125 # For now we need to be able to guess whether original formats provided
2126 # by extractor are incomplete or not (i.e. whether extractor provides only
2127 # video-only or audio-only formats) for proper formats selection for
2128 # extractors with such incomplete formats (see
2129 # https://github.com/ytdl-org/youtube-dl/pull/5556).
2130 # Since formats may be filtered during format selection and may not match
2131 # the original formats the results may be incorrect. Thus original formats
2132 # or pre-calculated metrics should be passed to format selection routines
2134 # We will pass a context object containing all necessary additional data
2135 # instead of just formats.
2136 # This fixes incorrect format selection issue (see
2137 # https://github.com/ytdl-org/youtube-dl/issues/10083).
2138 incomplete_formats
= (
2139 # All formats are video-only or
2140 all(f
.get('vcodec') != 'none' and f
.get('acodec') == 'none' for f
in formats
)
2141 # all formats are audio-only
2142 or all(f
.get('vcodec') == 'none' and f
.get('acodec') != 'none' for f
in formats
))
2146 'incomplete_formats': incomplete_formats
,
2149 formats_to_download
= list(format_selector(ctx
))
2150 if not formats_to_download
:
2151 if not self
.params
.get('ignore_no_formats_error'):
2152 raise ExtractorError('Requested format is not available', expected
=True)
2154 self
.report_warning('Requested format is not available')
2157 '[info] %s: Downloading %d format(s): %s' % (
2158 info_dict
['id'], len(formats_to_download
),
2159 ", ".join([f
['format_id'] for f
in formats_to_download
])))
2160 for fmt
in formats_to_download
:
2161 new_info
= dict(info_dict
)
2162 # Save a reference to the original info_dict so that it can be modified in process_info if needed
2163 new_info
['__original_infodict'] = info_dict
2164 new_info
.update(fmt
)
2165 self
.process_info(new_info
)
2166 # We update the info dict with the best quality format (backwards compatibility)
2167 if formats_to_download
:
2168 info_dict
.update(formats_to_download
[-1])
2171 def process_subtitles(self
, video_id
, normal_subtitles
, automatic_captions
):
2172 """Select the requested subtitles and their format"""
2174 if normal_subtitles
and self
.params
.get('writesubtitles'):
2175 available_subs
.update(normal_subtitles
)
2176 if automatic_captions
and self
.params
.get('writeautomaticsub'):
2177 for lang
, cap_info
in automatic_captions
.items():
2178 if lang
not in available_subs
:
2179 available_subs
[lang
] = cap_info
2181 if (not self
.params
.get('writesubtitles') and not
2182 self
.params
.get('writeautomaticsub') or not
2186 all_sub_langs
= available_subs
.keys()
2187 if self
.params
.get('allsubtitles', False):
2188 requested_langs
= all_sub_langs
2189 elif self
.params
.get('subtitleslangs', False):
2190 requested_langs
= set()
2191 for lang
in self
.params
.get('subtitleslangs'):
2193 requested_langs
.update(all_sub_langs
)
2195 discard
= lang
[0] == '-'
2198 current_langs
= filter(re
.compile(lang
+ '$').match
, all_sub_langs
)
2200 for lang
in current_langs
:
2201 requested_langs
.discard(lang
)
2203 requested_langs
.update(current_langs
)
2204 elif 'en' in available_subs
:
2205 requested_langs
= ['en']
2207 requested_langs
= [list(all_sub_langs
)[0]]
2208 self
.write_debug('Downloading subtitles: %s' % ', '.join(requested_langs
))
2210 formats_query
= self
.params
.get('subtitlesformat', 'best')
2211 formats_preference
= formats_query
.split('/') if formats_query
else []
2213 for lang
in requested_langs
:
2214 formats
= available_subs
.get(lang
)
2216 self
.report_warning('%s subtitles not available for %s' % (lang
, video_id
))
2218 for ext
in formats_preference
:
2222 matches
= list(filter(lambda f
: f
['ext'] == ext
, formats
))
2228 self
.report_warning(
2229 'No subtitle format found matching "%s" for language %s, '
2230 'using %s' % (formats_query
, lang
, f
['ext']))
2234 def __forced_printings(self
, info_dict
, filename
, incomplete
):
2235 def print_mandatory(field
, actual_field
=None):
2236 if actual_field
is None:
2237 actual_field
= field
2238 if (self
.params
.get('force%s' % field
, False)
2239 and (not incomplete
or info_dict
.get(actual_field
) is not None)):
2240 self
.to_stdout(info_dict
[actual_field
])
2242 def print_optional(field
):
2243 if (self
.params
.get('force%s' % field
, False)
2244 and info_dict
.get(field
) is not None):
2245 self
.to_stdout(info_dict
[field
])
2247 info_dict
= info_dict
.copy()
2248 if filename
is not None:
2249 info_dict
['filename'] = filename
2250 if info_dict
.get('requested_formats') is not None:
2251 # For RTMP URLs, also include the playpath
2252 info_dict
['urls'] = '\n'.join(f
['url'] + f
.get('play_path', '') for f
in info_dict
['requested_formats'])
2253 elif 'url' in info_dict
:
2254 info_dict
['urls'] = info_dict
['url'] + info_dict
.get('play_path', '')
2256 for tmpl
in self
.params
.get('forceprint', []):
2257 if re
.match(r
'\w+$', tmpl
):
2258 tmpl
= '%({})s'.format(tmpl
)
2259 tmpl
, info_copy
= self
.prepare_outtmpl(tmpl
, info_dict
)
2260 self
.to_stdout(tmpl
% info_copy
)
2262 print_mandatory('title')
2263 print_mandatory('id')
2264 print_mandatory('url', 'urls')
2265 print_optional('thumbnail')
2266 print_optional('description')
2267 print_optional('filename')
2268 if self
.params
.get('forceduration', False) and info_dict
.get('duration') is not None:
2269 self
.to_stdout(formatSeconds(info_dict
['duration']))
2270 print_mandatory('format')
2272 if self
.params
.get('forcejson', False):
2273 self
.post_extract(info_dict
)
2274 self
.to_stdout(json
.dumps(info_dict
, default
=repr))
2276 def dl(self
, name
, info
, subtitle
=False, test
=False):
2279 verbose
= self
.params
.get('verbose')
2282 'quiet': not verbose
,
2284 'noprogress': not verbose
,
2286 'skip_unavailable_fragments': False,
2287 'keep_fragments': False,
2289 '_no_ytdl_file': True,
2292 params
= self
.params
2293 fd
= get_suitable_downloader(info
, params
)(self
, params
)
2295 for ph
in self
._progress
_hooks
:
2296 fd
.add_progress_hook(ph
)
2297 self
.write_debug('Invoking downloader on %r' % info
.get('url'))
2298 new_info
= dict(info
)
2299 if new_info
.get('http_headers') is None:
2300 new_info
['http_headers'] = self
._calc
_headers
(new_info
)
2301 return fd
.download(name
, new_info
, subtitle
)
2303 def process_info(self
, info_dict
):
2304 """Process a single resolved IE result."""
2306 assert info_dict
.get('_type', 'video') == 'video'
2308 info_dict
.setdefault('__postprocessors', [])
2310 max_downloads
= self
.params
.get('max_downloads')
2311 if max_downloads
is not None:
2312 if self
._num
_downloads
>= int(max_downloads
):
2313 raise MaxDownloadsReached()
2315 # TODO: backward compatibility, to be removed
2316 info_dict
['fulltitle'] = info_dict
['title']
2318 if 'format' not in info_dict
:
2319 info_dict
['format'] = info_dict
['ext']
2321 if self
._match
_entry
(info_dict
, incomplete
=False) is not None:
2324 self
.post_extract(info_dict
)
2325 self
._num
_downloads
+= 1
2327 # info_dict['_filename'] needs to be set for backward compatibility
2328 info_dict
['_filename'] = full_filename
= self
.prepare_filename(info_dict
, warn
=True)
2329 temp_filename
= self
.prepare_filename(info_dict
, 'temp')
2333 self
.__forced
_printings
(info_dict
, full_filename
, incomplete
=False)
2335 if self
.params
.get('simulate', False):
2336 if self
.params
.get('force_write_download_archive', False):
2337 self
.record_download_archive(info_dict
)
2339 # Do nothing else if in simulate mode
2342 if full_filename
is None:
2345 if not self
._ensure
_dir
_exists
(encodeFilename(full_filename
)):
2347 if not self
._ensure
_dir
_exists
(encodeFilename(temp_filename
)):
2350 if self
.params
.get('writedescription', False):
2351 descfn
= self
.prepare_filename(info_dict
, 'description')
2352 if not self
._ensure
_dir
_exists
(encodeFilename(descfn
)):
2354 if not self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(descfn
)):
2355 self
.to_screen('[info] Video description is already present')
2356 elif info_dict
.get('description') is None:
2357 self
.report_warning('There\'s no description to write.')
2360 self
.to_screen('[info] Writing video description to: ' + descfn
)
2361 with io
.open(encodeFilename(descfn
), 'w', encoding
='utf-8') as descfile
:
2362 descfile
.write(info_dict
['description'])
2363 except (OSError, IOError):
2364 self
.report_error('Cannot write description file ' + descfn
)
2367 if self
.params
.get('writeannotations', False):
2368 annofn
= self
.prepare_filename(info_dict
, 'annotation')
2369 if not self
._ensure
_dir
_exists
(encodeFilename(annofn
)):
2371 if not self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(annofn
)):
2372 self
.to_screen('[info] Video annotations are already present')
2373 elif not info_dict
.get('annotations'):
2374 self
.report_warning('There are no annotations to write.')
2377 self
.to_screen('[info] Writing video annotations to: ' + annofn
)
2378 with io
.open(encodeFilename(annofn
), 'w', encoding
='utf-8') as annofile
:
2379 annofile
.write(info_dict
['annotations'])
2380 except (KeyError, TypeError):
2381 self
.report_warning('There are no annotations to write.')
2382 except (OSError, IOError):
2383 self
.report_error('Cannot write annotations file: ' + annofn
)
2386 subtitles_are_requested
= any([self
.params
.get('writesubtitles', False),
2387 self
.params
.get('writeautomaticsub')])
2389 if subtitles_are_requested
and info_dict
.get('requested_subtitles'):
2390 # subtitles download errors are already managed as troubles in relevant IE
2391 # that way it will silently go on when used with unsupporting IE
2392 subtitles
= info_dict
['requested_subtitles']
2393 # ie = self.get_info_extractor(info_dict['extractor_key'])
2394 for sub_lang
, sub_info
in subtitles
.items():
2395 sub_format
= sub_info
['ext']
2396 sub_filename
= subtitles_filename(temp_filename
, sub_lang
, sub_format
, info_dict
.get('ext'))
2397 sub_filename_final
= subtitles_filename(
2398 self
.prepare_filename(info_dict
, 'subtitle'), sub_lang
, sub_format
, info_dict
.get('ext'))
2399 if not self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(sub_filename
)):
2400 self
.to_screen('[info] Video subtitle %s.%s is already present' % (sub_lang
, sub_format
))
2401 sub_info
['filepath'] = sub_filename
2402 files_to_move
[sub_filename
] = sub_filename_final
2404 self
.to_screen('[info] Writing video subtitles to: ' + sub_filename
)
2405 if sub_info
.get('data') is not None:
2407 # Use newline='' to prevent conversion of newline characters
2408 # See https://github.com/ytdl-org/youtube-dl/issues/10268
2409 with io
.open(encodeFilename(sub_filename
), 'w', encoding
='utf-8', newline
='') as subfile
:
2410 subfile
.write(sub_info
['data'])
2411 sub_info
['filepath'] = sub_filename
2412 files_to_move
[sub_filename
] = sub_filename_final
2413 except (OSError, IOError):
2414 self
.report_error('Cannot write subtitles file ' + sub_filename
)
2418 self
.dl(sub_filename
, sub_info
.copy(), subtitle
=True)
2419 sub_info
['filepath'] = sub_filename
2420 files_to_move
[sub_filename
] = sub_filename_final
2421 except (ExtractorError
, IOError, OSError, ValueError) + network_exceptions
as err
:
2422 self
.report_warning('Unable to download subtitle for "%s": %s' %
2423 (sub_lang
, error_to_compat_str(err
)))
2426 if self
.params
.get('writeinfojson', False):
2427 infofn
= self
.prepare_filename(info_dict
, 'infojson')
2428 if not self
._ensure
_dir
_exists
(encodeFilename(infofn
)):
2430 if not self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(infofn
)):
2431 self
.to_screen('[info] Video metadata is already present')
2433 self
.to_screen('[info] Writing video metadata as JSON to: ' + infofn
)
2435 write_json_file(self
.filter_requested_info(info_dict
, self
.params
.get('clean_infojson', True)), infofn
)
2436 except (OSError, IOError):
2437 self
.report_error('Cannot write video metadata to JSON file ' + infofn
)
2439 info_dict
['__infojson_filename'] = infofn
2441 for thumb_ext
in self
._write
_thumbnails
(info_dict
, temp_filename
):
2442 thumb_filename_temp
= replace_extension(temp_filename
, thumb_ext
, info_dict
.get('ext'))
2443 thumb_filename
= replace_extension(
2444 self
.prepare_filename(info_dict
, 'thumbnail'), thumb_ext
, info_dict
.get('ext'))
2445 files_to_move
[thumb_filename_temp
] = thumb_filename
2447 # Write internet shortcut files
2448 url_link
= webloc_link
= desktop_link
= False
2449 if self
.params
.get('writelink', False):
2450 if sys
.platform
== "darwin": # macOS.
2452 elif sys
.platform
.startswith("linux"):
2454 else: # if sys.platform in ['win32', 'cygwin']:
2456 if self
.params
.get('writeurllink', False):
2458 if self
.params
.get('writewebloclink', False):
2460 if self
.params
.get('writedesktoplink', False):
2463 if url_link
or webloc_link
or desktop_link
:
2464 if 'webpage_url' not in info_dict
:
2465 self
.report_error('Cannot write internet shortcut file because the "webpage_url" field is missing in the media information')
2467 ascii_url
= iri_to_uri(info_dict
['webpage_url'])
2469 def _write_link_file(extension
, template
, newline
, embed_filename
):
2470 linkfn
= replace_extension(full_filename
, extension
, info_dict
.get('ext'))
2471 if self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(linkfn
)):
2472 self
.to_screen('[info] Internet shortcut is already present')
2475 self
.to_screen('[info] Writing internet shortcut to: ' + linkfn
)
2476 with io
.open(encodeFilename(to_high_limit_path(linkfn
)), 'w', encoding
='utf-8', newline
=newline
) as linkfile
:
2477 template_vars
= {'url': ascii_url}
2479 template_vars
['filename'] = linkfn
[:-(len(extension
) + 1)]
2480 linkfile
.write(template
% template_vars
)
2481 except (OSError, IOError):
2482 self
.report_error('Cannot write internet shortcut ' + linkfn
)
2487 if not _write_link_file('url', DOT_URL_LINK_TEMPLATE
, '\r\n', embed_filename
=False):
2490 if not _write_link_file('webloc', DOT_WEBLOC_LINK_TEMPLATE
, '\n', embed_filename
=False):
2493 if not _write_link_file('desktop', DOT_DESKTOP_LINK_TEMPLATE
, '\n', embed_filename
=True):
2497 info_dict
, files_to_move
= self
.pre_process(info_dict
, 'before_dl', files_to_move
)
2498 except PostProcessingError
as err
:
2499 self
.report_error('Preprocessing: %s' % str(err
))
2502 must_record_download_archive
= False
2503 if self
.params
.get('skip_download', False):
2504 info_dict
['filepath'] = temp_filename
2505 info_dict
['__finaldir'] = os
.path
.dirname(os
.path
.abspath(encodeFilename(full_filename
)))
2506 info_dict
['__files_to_move'] = files_to_move
2507 info_dict
= self
.run_pp(MoveFilesAfterDownloadPP(self
, False), info_dict
)
2512 def existing_file(*filepaths
):
2513 ext
= info_dict
.get('ext')
2514 final_ext
= self
.params
.get('final_ext', ext
)
2516 for file in orderedSet(filepaths
):
2517 if final_ext
!= ext
:
2518 converted
= replace_extension(file, final_ext
, ext
)
2519 if os
.path
.exists(encodeFilename(converted
)):
2520 existing_files
.append(converted
)
2521 if os
.path
.exists(encodeFilename(file)):
2522 existing_files
.append(file)
2524 if not existing_files
or self
.params
.get('overwrites', False):
2525 for file in orderedSet(existing_files
):
2526 self
.report_file_delete(file)
2527 os
.remove(encodeFilename(file))
2530 self
.report_file_already_downloaded(existing_files
[0])
2531 info_dict
['ext'] = os
.path
.splitext(existing_files
[0])[1][1:]
2532 return existing_files
[0]
2535 if info_dict
.get('requested_formats') is not None:
2537 merger
= FFmpegMergerPP(self
)
2538 if self
.params
.get('allow_unplayable_formats'):
2539 self
.report_warning(
2540 'You have requested merging of multiple formats '
2541 'while also allowing unplayable formats to be downloaded. '
2542 'The formats won\'t be merged to prevent data corruption.')
2543 elif not merger
.available
:
2544 self
.report_warning(
2545 'You have requested merging of multiple formats but ffmpeg is not installed. '
2546 'The formats won\'t be merged.')
2548 def compatible_formats(formats
):
2549 # TODO: some formats actually allow this (mkv, webm, ogg, mp4), but not all of them.
2550 video_formats
= [format
for format
in formats
if format
.get('vcodec') != 'none']
2551 audio_formats
= [format
for format
in formats
if format
.get('acodec') != 'none']
2552 if len(video_formats
) > 2 or len(audio_formats
) > 2:
2556 exts
= set(format
.get('ext') for format
in formats
)
2558 set(('mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma')),
2561 for ext_sets
in COMPATIBLE_EXTS
:
2562 if ext_sets
.issuperset(exts
):
2564 # TODO: Check acodec/vcodec
2567 requested_formats
= info_dict
['requested_formats']
2568 old_ext
= info_dict
['ext']
2569 if self
.params
.get('merge_output_format') is None:
2570 if not compatible_formats(requested_formats
):
2571 info_dict
['ext'] = 'mkv'
2572 self
.report_warning(
2573 'Requested formats are incompatible for merge and will be merged into mkv.')
2574 if (info_dict
['ext'] == 'webm'
2575 and self
.params
.get('writethumbnail', False)
2576 and info_dict
.get('thumbnails')):
2577 info_dict
['ext'] = 'mkv'
2578 self
.report_warning(
2579 'webm doesn\'t support embedding a thumbnail, mkv will be used.')
2581 def correct_ext(filename
):
2582 filename_real_ext
= os
.path
.splitext(filename
)[1][1:]
2584 os
.path
.splitext(filename
)[0]
2585 if filename_real_ext
== old_ext
2587 return '%s.%s' % (filename_wo_ext
, info_dict
['ext'])
2589 # Ensure filename always has a correct extension for successful merge
2590 full_filename
= correct_ext(full_filename
)
2591 temp_filename
= correct_ext(temp_filename
)
2592 dl_filename
= existing_file(full_filename
, temp_filename
)
2593 info_dict
['__real_download'] = False
2594 if dl_filename
is None:
2595 for f
in requested_formats
:
2596 new_info
= dict(info_dict
)
2598 fname
= prepend_extension(
2599 self
.prepare_filename(new_info
, 'temp'),
2600 'f%s' % f
['format_id'], new_info
['ext'])
2601 if not self
._ensure
_dir
_exists
(fname
):
2603 downloaded
.append(fname
)
2604 partial_success
, real_download
= self
.dl(fname
, new_info
)
2605 info_dict
['__real_download'] = info_dict
['__real_download'] or real_download
2606 success
= success
and partial_success
2607 if merger
.available
and not self
.params
.get('allow_unplayable_formats'):
2608 info_dict
['__postprocessors'].append(merger
)
2609 info_dict
['__files_to_merge'] = downloaded
2610 # Even if there were no downloads, it is being merged only now
2611 info_dict
['__real_download'] = True
2613 for file in downloaded
:
2614 files_to_move
[file] = None
2616 # Just a single file
2617 dl_filename
= existing_file(full_filename
, temp_filename
)
2618 if dl_filename
is None:
2619 success
, real_download
= self
.dl(temp_filename
, info_dict
)
2620 info_dict
['__real_download'] = real_download
2622 dl_filename
= dl_filename
or temp_filename
2623 info_dict
['__finaldir'] = os
.path
.dirname(os
.path
.abspath(encodeFilename(full_filename
)))
2625 except network_exceptions
as err
:
2626 self
.report_error('unable to download video data: %s' % error_to_compat_str(err
))
2628 except (OSError, IOError) as err
:
2629 raise UnavailableVideoError(err
)
2630 except (ContentTooShortError
, ) as err
:
2631 self
.report_error('content too short (expected %s bytes and served %s)' % (err
.expected
, err
.downloaded
))
2634 if success
and full_filename
!= '-':
2636 fixup_policy
= self
.params
.get('fixup')
2637 if fixup_policy
is None:
2638 fixup_policy
= 'detect_or_warn'
2640 INSTALL_FFMPEG_MESSAGE
= 'Install ffmpeg to fix this automatically.'
2642 stretched_ratio
= info_dict
.get('stretched_ratio')
2643 if stretched_ratio
is not None and stretched_ratio
!= 1:
2644 if fixup_policy
== 'warn':
2645 self
.report_warning('%s: Non-uniform pixel ratio (%s)' % (
2646 info_dict
['id'], stretched_ratio
))
2647 elif fixup_policy
== 'detect_or_warn':
2648 stretched_pp
= FFmpegFixupStretchedPP(self
)
2649 if stretched_pp
.available
:
2650 info_dict
['__postprocessors'].append(stretched_pp
)
2652 self
.report_warning(
2653 '%s: Non-uniform pixel ratio (%s). %s'
2654 % (info_dict
['id'], stretched_ratio
, INSTALL_FFMPEG_MESSAGE
))
2656 assert fixup_policy
in ('ignore', 'never')
2658 if (info_dict
.get('requested_formats') is None
2659 and info_dict
.get('container') == 'm4a_dash'
2660 and info_dict
.get('ext') == 'm4a'):
2661 if fixup_policy
== 'warn':
2662 self
.report_warning(
2663 '%s: writing DASH m4a. '
2664 'Only some players support this container.'
2666 elif fixup_policy
== 'detect_or_warn':
2667 fixup_pp
= FFmpegFixupM4aPP(self
)
2668 if fixup_pp
.available
:
2669 info_dict
['__postprocessors'].append(fixup_pp
)
2671 self
.report_warning(
2672 '%s: writing DASH m4a. '
2673 'Only some players support this container. %s'
2674 % (info_dict
['id'], INSTALL_FFMPEG_MESSAGE
))
2676 assert fixup_policy
in ('ignore', 'never')
2678 if ('protocol' in info_dict
2679 and get_suitable_downloader(info_dict
, self
.params
).__name
__ == 'HlsFD'):
2680 if fixup_policy
== 'warn':
2681 self
.report_warning('%s: malformed AAC bitstream detected.' % (
2683 elif fixup_policy
== 'detect_or_warn':
2684 fixup_pp
= FFmpegFixupM3u8PP(self
)
2685 if fixup_pp
.available
:
2686 info_dict
['__postprocessors'].append(fixup_pp
)
2688 self
.report_warning(
2689 '%s: malformed AAC bitstream detected. %s'
2690 % (info_dict
['id'], INSTALL_FFMPEG_MESSAGE
))
2692 assert fixup_policy
in ('ignore', 'never')
2695 info_dict
= self
.post_process(dl_filename
, info_dict
, files_to_move
)
2696 except PostProcessingError
as err
:
2697 self
.report_error('Postprocessing: %s' % str(err
))
2700 for ph
in self
._post
_hooks
:
2701 ph(info_dict
['filepath'])
2702 except Exception as err
:
2703 self
.report_error('post hooks: %s' % str(err
))
2705 must_record_download_archive
= True
2707 if must_record_download_archive
or self
.params
.get('force_write_download_archive', False):
2708 self
.record_download_archive(info_dict
)
2709 max_downloads
= self
.params
.get('max_downloads')
2710 if max_downloads
is not None and self
._num
_downloads
>= int(max_downloads
):
2711 raise MaxDownloadsReached()
2713 def download(self
, url_list
):
2714 """Download a given list of URLs."""
2715 outtmpl
= self
.outtmpl_dict
['default']
2716 if (len(url_list
) > 1
2718 and '%' not in outtmpl
2719 and self
.params
.get('max_downloads') != 1):
2720 raise SameFileError(outtmpl
)
2722 for url
in url_list
:
2724 # It also downloads the videos
2725 res
= self
.extract_info(
2726 url
, force_generic_extractor
=self
.params
.get('force_generic_extractor', False))
2727 except UnavailableVideoError
:
2728 self
.report_error('unable to download video')
2729 except MaxDownloadsReached
:
2730 self
.to_screen('[info] Maximum number of downloaded files reached')
2732 except ExistingVideoReached
:
2733 self
.to_screen('[info] Encountered a file that is already in the archive, stopping due to --break-on-existing')
2735 except RejectedVideoReached
:
2736 self
.to_screen('[info] Encountered a file that did not match filter, stopping due to --break-on-reject')
2739 if self
.params
.get('dump_single_json', False):
2740 self
.post_extract(res
)
2741 self
.to_stdout(json
.dumps(res
, default
=repr))
2743 return self
._download
_retcode
2745 def download_with_info_file(self
, info_filename
):
2746 with contextlib
.closing(fileinput
.FileInput(
2747 [info_filename
], mode
='r',
2748 openhook
=fileinput
.hook_encoded('utf-8'))) as f
:
2749 # FileInput doesn't have a read method, we can't call json.load
2750 info
= self
.filter_requested_info(json
.loads('\n'.join(f
)), self
.params
.get('clean_infojson', True))
2752 self
.process_ie_result(info
, download
=True)
2753 except (DownloadError
, EntryNotInPlaylist
):
2754 webpage_url
= info
.get('webpage_url')
2755 if webpage_url
is not None:
2756 self
.report_warning('The info failed to download, trying with "%s"' % webpage_url
)
2757 return self
.download([webpage_url
])
2760 return self
._download
_retcode
2763 def filter_requested_info(info_dict
, actually_filter
=True):
2764 info_dict
.pop('__original_infodict', None) # Always remove this
2765 if not actually_filter
:
2766 info_dict
['epoch'] = int(time
.time())
2769 'remove': ['requested_formats', 'requested_subtitles', 'requested_entries', 'filepath', 'entries'],
2772 keep_key
= lambda k
: k
in exceptions
['keep'] or not (k
.startswith('_') or k
in exceptions
['remove'])
2773 filter_fn
= lambda obj
: (
2774 list(map(filter_fn
, obj
)) if isinstance(obj
, (list, tuple))
2775 else obj
if not isinstance(obj
, dict)
2776 else dict((k
, filter_fn(v
)) for k
, v
in obj
.items() if keep_key(k
)))
2777 return filter_fn(info_dict
)
2779 def run_pp(self
, pp
, infodict
):
2780 files_to_delete
= []
2781 if '__files_to_move' not in infodict
:
2782 infodict
['__files_to_move'] = {}
2783 files_to_delete
, infodict
= pp
.run(infodict
)
2784 if not files_to_delete
:
2787 if self
.params
.get('keepvideo', False):
2788 for f
in files_to_delete
:
2789 infodict
['__files_to_move'].setdefault(f
, '')
2791 for old_filename
in set(files_to_delete
):
2792 self
.to_screen('Deleting original file %s (pass -k to keep)' % old_filename
)
2794 os
.remove(encodeFilename(old_filename
))
2795 except (IOError, OSError):
2796 self
.report_warning('Unable to remove downloaded original file')
2797 if old_filename
in infodict
['__files_to_move']:
2798 del infodict
['__files_to_move'][old_filename
]
2802 def post_extract(info_dict
):
2803 def actual_post_extract(info_dict
):
2804 if info_dict
.get('_type') in ('playlist', 'multi_video'):
2805 for video_dict
in info_dict
.get('entries', {}):
2806 actual_post_extract(video_dict
or {})
2809 post_extractor
= info_dict
.get('__post_extractor') or (lambda: {})
2810 extra
= post_extractor().items()
2811 info_dict
.update(extra
)
2812 info_dict
.pop('__post_extractor', None)
2814 original_infodict
= info_dict
.get('__original_infodict') or {}
2815 original_infodict
.update(extra
)
2816 original_infodict
.pop('__post_extractor', None)
2818 actual_post_extract(info_dict
or {})
2820 def pre_process(self
, ie_info
, key
='pre_process', files_to_move
=None):
2821 info
= dict(ie_info
)
2822 info
['__files_to_move'] = files_to_move
or {}
2823 for pp
in self
._pps
[key
]:
2824 info
= self
.run_pp(pp
, info
)
2825 return info
, info
.pop('__files_to_move', None)
2827 def post_process(self
, filename
, ie_info
, files_to_move
=None):
2828 """Run all the postprocessors on the given file."""
2829 info
= dict(ie_info
)
2830 info
['filepath'] = filename
2831 info
['__files_to_move'] = files_to_move
or {}
2833 for pp
in ie_info
.get('__postprocessors', []) + self
._pps
['post_process']:
2834 info
= self
.run_pp(pp
, info
)
2835 info
= self
.run_pp(MoveFilesAfterDownloadPP(self
), info
)
2836 del info
['__files_to_move']
2837 for pp
in self
._pps
['after_move']:
2838 info
= self
.run_pp(pp
, info
)
2841 def _make_archive_id(self
, info_dict
):
2842 video_id
= info_dict
.get('id')
2845 # Future-proof against any change in case
2846 # and backwards compatibility with prior versions
2847 extractor
= info_dict
.get('extractor_key') or info_dict
.get('ie_key') # key in a playlist
2848 if extractor
is None:
2849 url
= str_or_none(info_dict
.get('url'))
2852 # Try to find matching extractor for the URL and take its ie_key
2853 for ie
in self
._ies
:
2854 if ie
.suitable(url
):
2855 extractor
= ie
.ie_key()
2859 return '%s %s' % (extractor
.lower(), video_id
)
2861 def in_download_archive(self
, info_dict
):
2862 fn
= self
.params
.get('download_archive')
2866 vid_id
= self
._make
_archive
_id
(info_dict
)
2868 return False # Incomplete video information
2870 return vid_id
in self
.archive
2872 def record_download_archive(self
, info_dict
):
2873 fn
= self
.params
.get('download_archive')
2876 vid_id
= self
._make
_archive
_id
(info_dict
)
2878 with locked_file(fn
, 'a', encoding
='utf-8') as archive_file
:
2879 archive_file
.write(vid_id
+ '\n')
2880 self
.archive
.add(vid_id
)
2883 def format_resolution(format
, default
='unknown'):
2884 if format
.get('vcodec') == 'none':
2886 if format
.get('resolution') is not None:
2887 return format
['resolution']
2888 if format
.get('width') and format
.get('height'):
2889 res
= '%dx%d' % (format
['width'], format
['height'])
2890 elif format
.get('height'):
2891 res
= '%sp' % format
['height']
2892 elif format
.get('width'):
2893 res
= '%dx?' % format
['width']
2898 def _format_note(self
, fdict
):
2900 if fdict
.get('ext') in ['f4f', 'f4m']:
2901 res
+= '(unsupported) '
2902 if fdict
.get('language'):
2905 res
+= '[%s] ' % fdict
['language']
2906 if fdict
.get('format_note') is not None:
2907 res
+= fdict
['format_note'] + ' '
2908 if fdict
.get('tbr') is not None:
2909 res
+= '%4dk ' % fdict
['tbr']
2910 if fdict
.get('container') is not None:
2913 res
+= '%s container' % fdict
['container']
2914 if (fdict
.get('vcodec') is not None
2915 and fdict
.get('vcodec') != 'none'):
2918 res
+= fdict
['vcodec']
2919 if fdict
.get('vbr') is not None:
2921 elif fdict
.get('vbr') is not None and fdict
.get('abr') is not None:
2923 if fdict
.get('vbr') is not None:
2924 res
+= '%4dk' % fdict
['vbr']
2925 if fdict
.get('fps') is not None:
2928 res
+= '%sfps' % fdict
['fps']
2929 if fdict
.get('acodec') is not None:
2932 if fdict
['acodec'] == 'none':
2935 res
+= '%-5s' % fdict
['acodec']
2936 elif fdict
.get('abr') is not None:
2940 if fdict
.get('abr') is not None:
2941 res
+= '@%3dk' % fdict
['abr']
2942 if fdict
.get('asr') is not None:
2943 res
+= ' (%5dHz)' % fdict
['asr']
2944 if fdict
.get('filesize') is not None:
2947 res
+= format_bytes(fdict
['filesize'])
2948 elif fdict
.get('filesize_approx') is not None:
2951 res
+= '~' + format_bytes(fdict
['filesize_approx'])
2954 def _format_note_table(self
, f
):
2955 def join_fields(*vargs
):
2956 return ', '.join((val
for val
in vargs
if val
!= ''))
2959 'UNSUPPORTED' if f
.get('ext') in ('f4f', 'f4m') else '',
2960 format_field(f
, 'language', '[%s]'),
2961 format_field(f
, 'format_note'),
2962 format_field(f
, 'container', ignore
=(None, f
.get('ext'))),
2963 format_field(f
, 'asr', '%5dHz'))
2965 def list_formats(self
, info_dict
):
2966 formats
= info_dict
.get('formats', [info_dict
])
2968 'list-formats' not in self
.params
.get('compat_opts', [])
2969 and self
.params
.get('list_formats_as_table', True) is not False)
2973 format_field(f
, 'format_id'),
2974 format_field(f
, 'ext'),
2975 self
.format_resolution(f
),
2976 format_field(f
, 'fps', '%d'),
2978 format_field(f
, 'filesize', ' %s', func
=format_bytes
) + format_field(f
, 'filesize_approx', '~%s', func
=format_bytes
),
2979 format_field(f
, 'tbr', '%4dk'),
2980 shorten_protocol_name(f
.get('protocol', '').replace("native", "n")),
2982 format_field(f
, 'vcodec', default
='unknown').replace('none', ''),
2983 format_field(f
, 'vbr', '%4dk'),
2984 format_field(f
, 'acodec', default
='unknown').replace('none', ''),
2985 format_field(f
, 'abr', '%3dk'),
2986 format_field(f
, 'asr', '%5dHz'),
2987 self
._format
_note
_table
(f
)]
2989 if f
.get('preference') is None or f
['preference'] >= -1000]
2990 header_line
= ['ID', 'EXT', 'RESOLUTION', 'FPS', '|', ' FILESIZE', ' TBR', 'PROTO',
2991 '|', 'VCODEC', ' VBR', 'ACODEC', ' ABR', ' ASR', 'NOTE']
2995 format_field(f
, 'format_id'),
2996 format_field(f
, 'ext'),
2997 self
.format_resolution(f
),
2998 self
._format
_note
(f
)]
3000 if f
.get('preference') is None or f
['preference'] >= -1000]
3001 header_line
= ['format code', 'extension', 'resolution', 'note']
3004 '[info] Available formats for %s:\n%s' % (info_dict
['id'], render_table(
3008 extraGap
=(0 if new_format
else 1),
3009 hideEmpty
=new_format
)))
3011 def list_thumbnails(self
, info_dict
):
3012 thumbnails
= info_dict
.get('thumbnails')
3014 self
.to_screen('[info] No thumbnails present for %s' % info_dict
['id'])
3018 '[info] Thumbnails for %s:' % info_dict
['id'])
3019 self
.to_screen(render_table(
3020 ['ID', 'width', 'height', 'URL'],
3021 [[t
['id'], t
.get('width', 'unknown'), t
.get('height', 'unknown'), t
['url']] for t
in thumbnails
]))
3023 def list_subtitles(self
, video_id
, subtitles
, name
='subtitles'):
3025 self
.to_screen('%s has no %s' % (video_id
, name
))
3028 'Available %s for %s:' % (name
, video_id
))
3030 def _row(lang
, formats
):
3031 exts
, names
= zip(*((f
['ext'], f
.get('name', 'unknown')) for f
in reversed(formats
)))
3032 if len(set(names
)) == 1:
3033 names
= [] if names
[0] == 'unknown' else names
[:1]
3034 return [lang
, ', '.join(names
), ', '.join(exts
)]
3036 self
.to_screen(render_table(
3037 ['Language', 'Name', 'Formats'],
3038 [_row(lang
, formats
) for lang
, formats
in subtitles
.items()],
3041 def urlopen(self
, req
):
3042 """ Start an HTTP download """
3043 if isinstance(req
, compat_basestring
):
3044 req
= sanitized_Request(req
)
3045 return self
._opener
.open(req
, timeout
=self
._socket
_timeout
)
3047 def print_debug_header(self
):
3048 if not self
.params
.get('verbose'):
3051 if type('') is not compat_str
:
3052 # Python 2.6 on SLES11 SP1 (https://github.com/ytdl-org/youtube-dl/issues/3326)
3053 self
.report_warning(
3054 'Your Python is broken! Update to a newer and supported version')
3056 stdout_encoding
= getattr(
3057 sys
.stdout
, 'encoding', 'missing (%s)' % type(sys
.stdout
).__name
__)
3059 '[debug] Encodings: locale %s, fs %s, out %s, pref %s\n' % (
3060 locale
.getpreferredencoding(),
3061 sys
.getfilesystemencoding(),
3063 self
.get_encoding()))
3064 write_string(encoding_str
, encoding
=None)
3067 '(exe)' if hasattr(sys
, 'frozen')
3068 else '(zip)' if isinstance(globals().get('__loader__'), zipimporter
)
3069 else '(source)' if os
.path
.basename(sys
.argv
[0]) == '__main__.py'
3071 self
._write
_string
('[debug] yt-dlp version %s %s\n' % (__version__
, source
))
3073 self
._write
_string
('[debug] Lazy loading extractors enabled\n')
3076 '[debug] Plugin Extractors: %s\n' % [ie
.ie_key() for ie
in _PLUGIN_CLASSES
])
3077 if self
.params
.get('compat_opts'):
3079 '[debug] Compatibility options: %s\n' % ', '.join(self
.params
.get('compat_opts')))
3081 sp
= subprocess
.Popen(
3082 ['git', 'rev-parse', '--short', 'HEAD'],
3083 stdout
=subprocess
.PIPE
, stderr
=subprocess
.PIPE
,
3084 cwd
=os
.path
.dirname(os
.path
.abspath(__file__
)))
3085 out
, err
= process_communicate_or_kill(sp
)
3086 out
= out
.decode().strip()
3087 if re
.match('[0-9a-f]+', out
):
3088 self
._write
_string
('[debug] Git HEAD: %s\n' % out
)
3095 def python_implementation():
3096 impl_name
= platform
.python_implementation()
3097 if impl_name
== 'PyPy' and hasattr(sys
, 'pypy_version_info'):
3098 return impl_name
+ ' version %d.%d.%d' % sys
.pypy_version_info
[:3]
3101 self
._write
_string
('[debug] Python version %s (%s %s) - %s\n' % (
3102 platform
.python_version(),
3103 python_implementation(),
3104 platform
.architecture()[0],
3107 exe_versions
= FFmpegPostProcessor
.get_versions(self
)
3108 exe_versions
['rtmpdump'] = rtmpdump_version()
3109 exe_versions
['phantomjs'] = PhantomJSwrapper
._version
()
3110 exe_str
= ', '.join(
3112 for exe
, v
in sorted(exe_versions
.items())
3117 self
._write
_string
('[debug] exe versions: %s\n' % exe_str
)
3120 for handler
in self
._opener
.handlers
:
3121 if hasattr(handler
, 'proxies'):
3122 proxy_map
.update(handler
.proxies
)
3123 self
._write
_string
('[debug] Proxy map: ' + compat_str(proxy_map
) + '\n')
3125 if self
.params
.get('call_home', False):
3126 ipaddr
= self
.urlopen('https://yt-dl.org/ip').read().decode('utf-8')
3127 self
._write
_string
('[debug] Public IP address: %s\n' % ipaddr
)
3129 latest_version
= self
.urlopen(
3130 'https://yt-dl.org/latest/version').read().decode('utf-8')
3131 if version_tuple(latest_version
) > version_tuple(__version__
):
3132 self
.report_warning(
3133 'You are using an outdated version (newest version: %s)! '
3134 'See https://yt-dl.org/update if you need help updating.' %
3137 def _setup_opener(self
):
3138 timeout_val
= self
.params
.get('socket_timeout')
3139 self
._socket
_timeout
= 600 if timeout_val
is None else float(timeout_val
)
3141 opts_cookiefile
= self
.params
.get('cookiefile')
3142 opts_proxy
= self
.params
.get('proxy')
3144 if opts_cookiefile
is None:
3145 self
.cookiejar
= compat_cookiejar
.CookieJar()
3147 opts_cookiefile
= expand_path(opts_cookiefile
)
3148 self
.cookiejar
= YoutubeDLCookieJar(opts_cookiefile
)
3149 if os
.access(opts_cookiefile
, os
.R_OK
):
3150 self
.cookiejar
.load(ignore_discard
=True, ignore_expires
=True)
3152 cookie_processor
= YoutubeDLCookieProcessor(self
.cookiejar
)
3153 if opts_proxy
is not None:
3154 if opts_proxy
== '':
3157 proxies
= {'http': opts_proxy, 'https': opts_proxy}
3159 proxies
= compat_urllib_request
.getproxies()
3160 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
3161 if 'http' in proxies
and 'https' not in proxies
:
3162 proxies
['https'] = proxies
['http']
3163 proxy_handler
= PerRequestProxyHandler(proxies
)
3165 debuglevel
= 1 if self
.params
.get('debug_printtraffic') else 0
3166 https_handler
= make_HTTPS_handler(self
.params
, debuglevel
=debuglevel
)
3167 ydlh
= YoutubeDLHandler(self
.params
, debuglevel
=debuglevel
)
3168 redirect_handler
= YoutubeDLRedirectHandler()
3169 data_handler
= compat_urllib_request_DataHandler()
3171 # When passing our own FileHandler instance, build_opener won't add the
3172 # default FileHandler and allows us to disable the file protocol, which
3173 # can be used for malicious purposes (see
3174 # https://github.com/ytdl-org/youtube-dl/issues/8227)
3175 file_handler
= compat_urllib_request
.FileHandler()
3177 def file_open(*args
, **kwargs
):
3178 raise compat_urllib_error
.URLError('file:// scheme is explicitly disabled in yt-dlp for security reasons')
3179 file_handler
.file_open
= file_open
3181 opener
= compat_urllib_request
.build_opener(
3182 proxy_handler
, https_handler
, cookie_processor
, ydlh
, redirect_handler
, data_handler
, file_handler
)
3184 # Delete the default user-agent header, which would otherwise apply in
3185 # cases where our custom HTTP handler doesn't come into play
3186 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
3187 opener
.addheaders
= []
3188 self
._opener
= opener
3190 def encode(self
, s
):
3191 if isinstance(s
, bytes):
3192 return s
# Already encoded
3195 return s
.encode(self
.get_encoding())
3196 except UnicodeEncodeError as err
:
3197 err
.reason
= err
.reason
+ '. Check your system encoding configuration or use the --encoding option.'
3200 def get_encoding(self
):
3201 encoding
= self
.params
.get('encoding')
3202 if encoding
is None:
3203 encoding
= preferredencoding()
3206 def _write_thumbnails(self
, info_dict
, filename
): # return the extensions
3207 write_all
= self
.params
.get('write_all_thumbnails', False)
3209 if write_all
or self
.params
.get('writethumbnail', False):
3210 thumbnails
= info_dict
.get('thumbnails') or []
3211 multiple
= write_all
and len(thumbnails
) > 1
3214 for t
in thumbnails
[::1 if write_all
else -1]:
3215 thumb_ext
= determine_ext(t
['url'], 'jpg')
3216 suffix
= '%s.' % t
['id'] if multiple
else ''
3217 thumb_display_id
= '%s ' % t
['id'] if multiple
else ''
3218 t
['filepath'] = thumb_filename
= replace_extension(filename
, suffix
+ thumb_ext
, info_dict
.get('ext'))
3220 if not self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(thumb_filename
)):
3221 ret
.append(suffix
+ thumb_ext
)
3222 self
.to_screen('[%s] %s: Thumbnail %sis already present' %
3223 (info_dict
['extractor'], info_dict
['id'], thumb_display_id
))
3225 self
.to_screen('[%s] %s: Downloading thumbnail %s ...' %
3226 (info_dict
['extractor'], info_dict
['id'], thumb_display_id
))
3228 uf
= self
.urlopen(t
['url'])
3229 with open(encodeFilename(thumb_filename
), 'wb') as thumbf
:
3230 shutil
.copyfileobj(uf
, thumbf
)
3231 ret
.append(suffix
+ thumb_ext
)
3232 self
.to_screen('[%s] %s: Writing thumbnail %sto: %s' %
3233 (info_dict
['extractor'], info_dict
['id'], thumb_display_id
, thumb_filename
))
3234 except network_exceptions
as err
:
3235 self
.report_warning('Unable to download thumbnail "%s": %s' %
3236 (t
['url'], error_to_compat_str(err
)))
3237 if ret
and not write_all
: