]> jfr.im git - yt-dlp.git/blob - yt_dlp/YoutubeDL.py
[cleanup] Misc
[yt-dlp.git] / yt_dlp / YoutubeDL.py
1 #!/usr/bin/env python3
2 import collections
3 import contextlib
4 import datetime
5 import errno
6 import fileinput
7 import functools
8 import io
9 import itertools
10 import json
11 import locale
12 import operator
13 import os
14 import platform
15 import random
16 import re
17 import shutil
18 import subprocess
19 import sys
20 import tempfile
21 import time
22 import tokenize
23 import traceback
24 import unicodedata
25 import urllib.request
26 from string import ascii_letters
27
28 from .cache import Cache
29 from .compat import (
30 compat_get_terminal_size,
31 compat_os_name,
32 compat_shlex_quote,
33 compat_str,
34 compat_urllib_error,
35 compat_urllib_request,
36 )
37 from .cookies import load_cookies
38 from .downloader import FFmpegFD, get_suitable_downloader, shorten_protocol_name
39 from .downloader.rtmp import rtmpdump_version
40 from .extractor import _LAZY_LOADER
41 from .extractor import _PLUGIN_CLASSES as plugin_extractors
42 from .extractor import gen_extractor_classes, get_info_extractor
43 from .extractor.openload import PhantomJSwrapper
44 from .minicurses import format_text
45 from .postprocessor import _PLUGIN_CLASSES as plugin_postprocessors
46 from .postprocessor import (
47 EmbedThumbnailPP,
48 FFmpegFixupDuplicateMoovPP,
49 FFmpegFixupDurationPP,
50 FFmpegFixupM3u8PP,
51 FFmpegFixupM4aPP,
52 FFmpegFixupStretchedPP,
53 FFmpegFixupTimestampPP,
54 FFmpegMergerPP,
55 FFmpegPostProcessor,
56 MoveFilesAfterDownloadPP,
57 get_postprocessor,
58 )
59 from .update import detect_variant
60 from .utils import (
61 DEFAULT_OUTTMPL,
62 LINK_TEMPLATES,
63 NO_DEFAULT,
64 NUMBER_RE,
65 OUTTMPL_TYPES,
66 POSTPROCESS_WHEN,
67 STR_FORMAT_RE_TMPL,
68 STR_FORMAT_TYPES,
69 ContentTooShortError,
70 DateRange,
71 DownloadCancelled,
72 DownloadError,
73 EntryNotInPlaylist,
74 ExistingVideoReached,
75 ExtractorError,
76 GeoRestrictedError,
77 HEADRequest,
78 InAdvancePagedList,
79 ISO3166Utils,
80 LazyList,
81 MaxDownloadsReached,
82 Namespace,
83 PagedList,
84 PerRequestProxyHandler,
85 Popen,
86 PostProcessingError,
87 ReExtractInfo,
88 RejectedVideoReached,
89 SameFileError,
90 UnavailableVideoError,
91 YoutubeDLCookieProcessor,
92 YoutubeDLHandler,
93 YoutubeDLRedirectHandler,
94 age_restricted,
95 args_to_str,
96 date_from_str,
97 determine_ext,
98 determine_protocol,
99 encode_compat_str,
100 encodeFilename,
101 error_to_compat_str,
102 expand_path,
103 filter_dict,
104 float_or_none,
105 format_bytes,
106 format_decimal_suffix,
107 format_field,
108 formatSeconds,
109 get_domain,
110 int_or_none,
111 iri_to_uri,
112 join_nonempty,
113 locked_file,
114 make_dir,
115 make_HTTPS_handler,
116 merge_headers,
117 network_exceptions,
118 number_of_digits,
119 orderedSet,
120 parse_filesize,
121 platform_name,
122 preferredencoding,
123 prepend_extension,
124 register_socks_protocols,
125 remove_terminal_sequences,
126 render_table,
127 replace_extension,
128 sanitize_filename,
129 sanitize_path,
130 sanitize_url,
131 sanitized_Request,
132 std_headers,
133 str_or_none,
134 strftime_or_none,
135 subtitles_filename,
136 supports_terminal_sequences,
137 timetuple_from_msec,
138 to_high_limit_path,
139 traverse_obj,
140 try_get,
141 url_basename,
142 variadic,
143 version_tuple,
144 windows_enable_vt_mode,
145 write_json_file,
146 write_string,
147 )
148 from .version import RELEASE_GIT_HEAD, __version__
149
150 if compat_os_name == 'nt':
151 import ctypes
152
153
154 class YoutubeDL:
155 """YoutubeDL class.
156
157 YoutubeDL objects are the ones responsible of downloading the
158 actual video file and writing it to disk if the user has requested
159 it, among some other tasks. In most cases there should be one per
160 program. As, given a video URL, the downloader doesn't know how to
161 extract all the needed information, task that InfoExtractors do, it
162 has to pass the URL to one of them.
163
164 For this, YoutubeDL objects have a method that allows
165 InfoExtractors to be registered in a given order. When it is passed
166 a URL, the YoutubeDL object handles it to the first InfoExtractor it
167 finds that reports being able to handle it. The InfoExtractor extracts
168 all the information about the video or videos the URL refers to, and
169 YoutubeDL process the extracted information, possibly using a File
170 Downloader to download the video.
171
172 YoutubeDL objects accept a lot of parameters. In order not to saturate
173 the object constructor with arguments, it receives a dictionary of
174 options instead. These options are available through the params
175 attribute for the InfoExtractors to use. The YoutubeDL also
176 registers itself as the downloader in charge for the InfoExtractors
177 that are added to it, so this is a "mutual registration".
178
179 Available options:
180
181 username: Username for authentication purposes.
182 password: Password for authentication purposes.
183 videopassword: Password for accessing a video.
184 ap_mso: Adobe Pass multiple-system operator identifier.
185 ap_username: Multiple-system operator account username.
186 ap_password: Multiple-system operator account password.
187 usenetrc: Use netrc for authentication instead.
188 verbose: Print additional info to stdout.
189 quiet: Do not print messages to stdout.
190 no_warnings: Do not print out anything for warnings.
191 forceprint: A dict with keys WHEN mapped to a list of templates to
192 print to stdout. The allowed keys are video or any of the
193 items in utils.POSTPROCESS_WHEN.
194 For compatibility, a single list is also accepted
195 print_to_file: A dict with keys WHEN (same as forceprint) mapped to
196 a list of tuples with (template, filename)
197 forceurl: Force printing final URL. (Deprecated)
198 forcetitle: Force printing title. (Deprecated)
199 forceid: Force printing ID. (Deprecated)
200 forcethumbnail: Force printing thumbnail URL. (Deprecated)
201 forcedescription: Force printing description. (Deprecated)
202 forcefilename: Force printing final filename. (Deprecated)
203 forceduration: Force printing duration. (Deprecated)
204 forcejson: Force printing info_dict as JSON.
205 dump_single_json: Force printing the info_dict of the whole playlist
206 (or video) as a single JSON line.
207 force_write_download_archive: Force writing download archive regardless
208 of 'skip_download' or 'simulate'.
209 simulate: Do not download the video files. If unset (or None),
210 simulate only if listsubtitles, listformats or list_thumbnails is used
211 format: Video format code. see "FORMAT SELECTION" for more details.
212 You can also pass a function. The function takes 'ctx' as
213 argument and returns the formats to download.
214 See "build_format_selector" for an implementation
215 allow_unplayable_formats: Allow unplayable formats to be extracted and downloaded.
216 ignore_no_formats_error: Ignore "No video formats" error. Usefull for
217 extracting metadata even if the video is not actually
218 available for download (experimental)
219 format_sort: A list of fields by which to sort the video formats.
220 See "Sorting Formats" for more details.
221 format_sort_force: Force the given format_sort. see "Sorting Formats"
222 for more details.
223 prefer_free_formats: Whether to prefer video formats with free containers
224 over non-free ones of same quality.
225 allow_multiple_video_streams: Allow multiple video streams to be merged
226 into a single file
227 allow_multiple_audio_streams: Allow multiple audio streams to be merged
228 into a single file
229 check_formats Whether to test if the formats are downloadable.
230 Can be True (check all), False (check none),
231 'selected' (check selected formats),
232 or None (check only if requested by extractor)
233 paths: Dictionary of output paths. The allowed keys are 'home'
234 'temp' and the keys of OUTTMPL_TYPES (in utils.py)
235 outtmpl: Dictionary of templates for output names. Allowed keys
236 are 'default' and the keys of OUTTMPL_TYPES (in utils.py).
237 For compatibility with youtube-dl, a single string can also be used
238 outtmpl_na_placeholder: Placeholder for unavailable meta fields.
239 restrictfilenames: Do not allow "&" and spaces in file names
240 trim_file_name: Limit length of filename (extension excluded)
241 windowsfilenames: Force the filenames to be windows compatible
242 ignoreerrors: Do not stop on download/postprocessing errors.
243 Can be 'only_download' to ignore only download errors.
244 Default is 'only_download' for CLI, but False for API
245 skip_playlist_after_errors: Number of allowed failures until the rest of
246 the playlist is skipped
247 force_generic_extractor: Force downloader to use the generic extractor
248 overwrites: Overwrite all video and metadata files if True,
249 overwrite only non-video files if None
250 and don't overwrite any file if False
251 For compatibility with youtube-dl,
252 "nooverwrites" may also be used instead
253 playliststart: Playlist item to start at.
254 playlistend: Playlist item to end at.
255 playlist_items: Specific indices of playlist to download.
256 playlistreverse: Download playlist items in reverse order.
257 playlistrandom: Download playlist items in random order.
258 matchtitle: Download only matching titles.
259 rejecttitle: Reject downloads for matching titles.
260 logger: Log messages to a logging.Logger instance.
261 logtostderr: Log messages to stderr instead of stdout.
262 consoletitle: Display progress in console window's titlebar.
263 writedescription: Write the video description to a .description file
264 writeinfojson: Write the video description to a .info.json file
265 clean_infojson: Remove private fields from the infojson
266 getcomments: Extract video comments. This will not be written to disk
267 unless writeinfojson is also given
268 writeannotations: Write the video annotations to a .annotations.xml file
269 writethumbnail: Write the thumbnail image to a file
270 allow_playlist_files: Whether to write playlists' description, infojson etc
271 also to disk when using the 'write*' options
272 write_all_thumbnails: Write all thumbnail formats to files
273 writelink: Write an internet shortcut file, depending on the
274 current platform (.url/.webloc/.desktop)
275 writeurllink: Write a Windows internet shortcut file (.url)
276 writewebloclink: Write a macOS internet shortcut file (.webloc)
277 writedesktoplink: Write a Linux internet shortcut file (.desktop)
278 writesubtitles: Write the video subtitles to a file
279 writeautomaticsub: Write the automatically generated subtitles to a file
280 allsubtitles: Deprecated - Use subtitleslangs = ['all']
281 Downloads all the subtitles of the video
282 (requires writesubtitles or writeautomaticsub)
283 listsubtitles: Lists all available subtitles for the video
284 subtitlesformat: The format code for subtitles
285 subtitleslangs: List of languages of the subtitles to download (can be regex).
286 The list may contain "all" to refer to all the available
287 subtitles. The language can be prefixed with a "-" to
288 exclude it from the requested languages. Eg: ['all', '-live_chat']
289 keepvideo: Keep the video file after post-processing
290 daterange: A DateRange object, download only if the upload_date is in the range.
291 skip_download: Skip the actual download of the video file
292 cachedir: Location of the cache files in the filesystem.
293 False to disable filesystem cache.
294 noplaylist: Download single video instead of a playlist if in doubt.
295 age_limit: An integer representing the user's age in years.
296 Unsuitable videos for the given age are skipped.
297 min_views: An integer representing the minimum view count the video
298 must have in order to not be skipped.
299 Videos without view count information are always
300 downloaded. None for no limit.
301 max_views: An integer representing the maximum view count.
302 Videos that are more popular than that are not
303 downloaded.
304 Videos without view count information are always
305 downloaded. None for no limit.
306 download_archive: File name of a file where all downloads are recorded.
307 Videos already present in the file are not downloaded
308 again.
309 break_on_existing: Stop the download process after attempting to download a
310 file that is in the archive.
311 break_on_reject: Stop the download process when encountering a video that
312 has been filtered out.
313 break_per_url: Whether break_on_reject and break_on_existing
314 should act on each input URL as opposed to for the entire queue
315 cookiefile: File name or text stream from where cookies should be read and dumped to
316 cookiesfrombrowser: A tuple containing the name of the browser, the profile
317 name/pathfrom where cookies are loaded, and the name of the
318 keyring. Eg: ('chrome', ) or ('vivaldi', 'default', 'BASICTEXT')
319 legacyserverconnect: Explicitly allow HTTPS connection to servers that do not
320 support RFC 5746 secure renegotiation
321 nocheckcertificate: Do not verify SSL certificates
322 client_certificate: Path to client certificate file in PEM format. May include the private key
323 client_certificate_key: Path to private key file for client certificate
324 client_certificate_password: Password for client certificate private key, if encrypted.
325 If not provided and the key is encrypted, yt-dlp will ask interactively
326 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
327 At the moment, this is only supported by YouTube.
328 http_headers: A dictionary of custom headers to be used for all requests
329 proxy: URL of the proxy server to use
330 geo_verification_proxy: URL of the proxy to use for IP address verification
331 on geo-restricted sites.
332 socket_timeout: Time to wait for unresponsive hosts, in seconds
333 bidi_workaround: Work around buggy terminals without bidirectional text
334 support, using fridibi
335 debug_printtraffic:Print out sent and received HTTP traffic
336 include_ads: Download ads as well (deprecated)
337 default_search: Prepend this string if an input url is not valid.
338 'auto' for elaborate guessing
339 encoding: Use this encoding instead of the system-specified.
340 extract_flat: Do not resolve URLs, return the immediate result.
341 Pass in 'in_playlist' to only show this behavior for
342 playlist items.
343 wait_for_video: If given, wait for scheduled streams to become available.
344 The value should be a tuple containing the range
345 (min_secs, max_secs) to wait between retries
346 postprocessors: A list of dictionaries, each with an entry
347 * key: The name of the postprocessor. See
348 yt_dlp/postprocessor/__init__.py for a list.
349 * when: When to run the postprocessor. Allowed values are
350 the entries of utils.POSTPROCESS_WHEN
351 Assumed to be 'post_process' if not given
352 post_hooks: Deprecated - Register a custom postprocessor instead
353 A list of functions that get called as the final step
354 for each video file, after all postprocessors have been
355 called. The filename will be passed as the only argument.
356 progress_hooks: A list of functions that get called on download
357 progress, with a dictionary with the entries
358 * status: One of "downloading", "error", or "finished".
359 Check this first and ignore unknown values.
360 * info_dict: The extracted info_dict
361
362 If status is one of "downloading", or "finished", the
363 following properties may also be present:
364 * filename: The final filename (always present)
365 * tmpfilename: The filename we're currently writing to
366 * downloaded_bytes: Bytes on disk
367 * total_bytes: Size of the whole file, None if unknown
368 * total_bytes_estimate: Guess of the eventual file size,
369 None if unavailable.
370 * elapsed: The number of seconds since download started.
371 * eta: The estimated time in seconds, None if unknown
372 * speed: The download speed in bytes/second, None if
373 unknown
374 * fragment_index: The counter of the currently
375 downloaded video fragment.
376 * fragment_count: The number of fragments (= individual
377 files that will be merged)
378
379 Progress hooks are guaranteed to be called at least once
380 (with status "finished") if the download is successful.
381 postprocessor_hooks: A list of functions that get called on postprocessing
382 progress, with a dictionary with the entries
383 * status: One of "started", "processing", or "finished".
384 Check this first and ignore unknown values.
385 * postprocessor: Name of the postprocessor
386 * info_dict: The extracted info_dict
387
388 Progress hooks are guaranteed to be called at least twice
389 (with status "started" and "finished") if the processing is successful.
390 merge_output_format: Extension to use when merging formats.
391 final_ext: Expected final extension; used to detect when the file was
392 already downloaded and converted
393 fixup: Automatically correct known faults of the file.
394 One of:
395 - "never": do nothing
396 - "warn": only emit a warning
397 - "detect_or_warn": check whether we can do anything
398 about it, warn otherwise (default)
399 source_address: Client-side IP address to bind to.
400 call_home: Boolean, true iff we are allowed to contact the
401 yt-dlp servers for debugging. (BROKEN)
402 sleep_interval_requests: Number of seconds to sleep between requests
403 during extraction
404 sleep_interval: Number of seconds to sleep before each download when
405 used alone or a lower bound of a range for randomized
406 sleep before each download (minimum possible number
407 of seconds to sleep) when used along with
408 max_sleep_interval.
409 max_sleep_interval:Upper bound of a range for randomized sleep before each
410 download (maximum possible number of seconds to sleep).
411 Must only be used along with sleep_interval.
412 Actual sleep time will be a random float from range
413 [sleep_interval; max_sleep_interval].
414 sleep_interval_subtitles: Number of seconds to sleep before each subtitle download
415 listformats: Print an overview of available video formats and exit.
416 list_thumbnails: Print a table of all thumbnails and exit.
417 match_filter: A function that gets called for every video with the signature
418 (info_dict, *, incomplete: bool) -> Optional[str]
419 For backward compatibility with youtube-dl, the signature
420 (info_dict) -> Optional[str] is also allowed.
421 - If it returns a message, the video is ignored.
422 - If it returns None, the video is downloaded.
423 - If it returns utils.NO_DEFAULT, the user is interactively
424 asked whether to download the video.
425 match_filter_func in utils.py is one example for this.
426 no_color: Do not emit color codes in output.
427 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
428 HTTP header
429 geo_bypass_country:
430 Two-letter ISO 3166-2 country code that will be used for
431 explicit geographic restriction bypassing via faking
432 X-Forwarded-For HTTP header
433 geo_bypass_ip_block:
434 IP range in CIDR notation that will be used similarly to
435 geo_bypass_country
436
437 The following options determine which downloader is picked:
438 external_downloader: A dictionary of protocol keys and the executable of the
439 external downloader to use for it. The allowed protocols
440 are default|http|ftp|m3u8|dash|rtsp|rtmp|mms.
441 Set the value to 'native' to use the native downloader
442 hls_prefer_native: Deprecated - Use external_downloader = {'m3u8': 'native'}
443 or {'m3u8': 'ffmpeg'} instead.
444 Use the native HLS downloader instead of ffmpeg/avconv
445 if True, otherwise use ffmpeg/avconv if False, otherwise
446 use downloader suggested by extractor if None.
447 compat_opts: Compatibility options. See "Differences in default behavior".
448 The following options do not work when used through the API:
449 filename, abort-on-error, multistreams, no-live-chat, format-sort
450 no-clean-infojson, no-playlist-metafiles, no-keep-subs, no-attach-info-json.
451 Refer __init__.py for their implementation
452 progress_template: Dictionary of templates for progress outputs.
453 Allowed keys are 'download', 'postprocess',
454 'download-title' (console title) and 'postprocess-title'.
455 The template is mapped on a dictionary with keys 'progress' and 'info'
456 retry_sleep_functions: Dictionary of functions that takes the number of attempts
457 as argument and returns the time to sleep in seconds.
458 Allowed keys are 'http', 'fragment', 'file_access'
459
460 The following parameters are not used by YoutubeDL itself, they are used by
461 the downloader (see yt_dlp/downloader/common.py):
462 nopart, updatetime, buffersize, ratelimit, throttledratelimit, min_filesize,
463 max_filesize, test, noresizebuffer, retries, file_access_retries, fragment_retries,
464 continuedl, noprogress, xattr_set_filesize, hls_use_mpegts, http_chunk_size,
465 external_downloader_args, concurrent_fragment_downloads.
466
467 The following options are used by the post processors:
468 prefer_ffmpeg: If False, use avconv instead of ffmpeg if both are available,
469 otherwise prefer ffmpeg. (avconv support is deprecated)
470 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
471 to the binary or its containing directory.
472 postprocessor_args: A dictionary of postprocessor/executable keys (in lower case)
473 and a list of additional command-line arguments for the
474 postprocessor/executable. The dict can also have "PP+EXE" keys
475 which are used when the given exe is used by the given PP.
476 Use 'default' as the name for arguments to passed to all PP
477 For compatibility with youtube-dl, a single list of args
478 can also be used
479
480 The following options are used by the extractors:
481 extractor_retries: Number of times to retry for known errors
482 dynamic_mpd: Whether to process dynamic DASH manifests (default: True)
483 hls_split_discontinuity: Split HLS playlists to different formats at
484 discontinuities such as ad breaks (default: False)
485 extractor_args: A dictionary of arguments to be passed to the extractors.
486 See "EXTRACTOR ARGUMENTS" for details.
487 Eg: {'youtube': {'skip': ['dash', 'hls']}}
488 mark_watched: Mark videos watched (even with --simulate). Only for YouTube
489 youtube_include_dash_manifest: Deprecated - Use extractor_args instead.
490 If True (default), DASH manifests and related
491 data will be downloaded and processed by extractor.
492 You can reduce network I/O by disabling it if you don't
493 care about DASH. (only for youtube)
494 youtube_include_hls_manifest: Deprecated - Use extractor_args instead.
495 If True (default), HLS manifests and related
496 data will be downloaded and processed by extractor.
497 You can reduce network I/O by disabling it if you don't
498 care about HLS. (only for youtube)
499 """
500
501 _NUMERIC_FIELDS = {
502 'width', 'height', 'tbr', 'abr', 'asr', 'vbr', 'fps', 'filesize', 'filesize_approx',
503 'timestamp', 'release_timestamp',
504 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
505 'average_rating', 'comment_count', 'age_limit',
506 'start_time', 'end_time',
507 'chapter_number', 'season_number', 'episode_number',
508 'track_number', 'disc_number', 'release_year',
509 }
510
511 _format_fields = {
512 # NB: Keep in sync with the docstring of extractor/common.py
513 'url', 'manifest_url', 'manifest_stream_number', 'ext', 'format', 'format_id', 'format_note',
514 'width', 'height', 'resolution', 'dynamic_range', 'tbr', 'abr', 'acodec', 'asr',
515 'vbr', 'fps', 'vcodec', 'container', 'filesize', 'filesize_approx',
516 'player_url', 'protocol', 'fragment_base_url', 'fragments', 'is_from_start',
517 'preference', 'language', 'language_preference', 'quality', 'source_preference',
518 'http_headers', 'stretched_ratio', 'no_resume', 'has_drm', 'downloader_options',
519 'page_url', 'app', 'play_path', 'tc_url', 'flash_version', 'rtmp_live', 'rtmp_conn', 'rtmp_protocol', 'rtmp_real_time'
520 }
521 _format_selection_exts = {
522 'audio': {'m4a', 'mp3', 'ogg', 'aac'},
523 'video': {'mp4', 'flv', 'webm', '3gp'},
524 'storyboards': {'mhtml'},
525 }
526
527 def __init__(self, params=None, auto_init=True):
528 """Create a FileDownloader object with the given options.
529 @param auto_init Whether to load the default extractors and print header (if verbose).
530 Set to 'no_verbose_header' to not print the header
531 """
532 if params is None:
533 params = {}
534 self.params = params
535 self._ies = {}
536 self._ies_instances = {}
537 self._pps = {k: [] for k in POSTPROCESS_WHEN}
538 self._printed_messages = set()
539 self._first_webpage_request = True
540 self._post_hooks = []
541 self._progress_hooks = []
542 self._postprocessor_hooks = []
543 self._download_retcode = 0
544 self._num_downloads = 0
545 self._num_videos = 0
546 self._playlist_level = 0
547 self._playlist_urls = set()
548 self.cache = Cache(self)
549
550 windows_enable_vt_mode()
551 stdout = sys.stderr if self.params.get('logtostderr') else sys.stdout
552 self._out_files = Namespace(
553 out=stdout,
554 error=sys.stderr,
555 screen=sys.stderr if self.params.get('quiet') else stdout,
556 console=None if compat_os_name == 'nt' else next(
557 filter(supports_terminal_sequences, (sys.stderr, sys.stdout)), None)
558 )
559 self._allow_colors = Namespace(**{
560 type_: not self.params.get('no_color') and supports_terminal_sequences(stream)
561 for type_, stream in self._out_files if type_ != 'console'
562 })
563
564 if sys.version_info < (3, 6):
565 self.report_warning(
566 'Python version %d.%d is not supported! Please update to Python 3.6 or above' % sys.version_info[:2])
567
568 if self.params.get('allow_unplayable_formats'):
569 self.report_warning(
570 f'You have asked for {self._format_err("UNPLAYABLE", self.Styles.EMPHASIS)} formats to be listed/downloaded. '
571 'This is a developer option intended for debugging. \n'
572 ' If you experience any issues while using this option, '
573 f'{self._format_err("DO NOT", self.Styles.ERROR)} open a bug report')
574
575 def check_deprecated(param, option, suggestion):
576 if self.params.get(param) is not None:
577 self.report_warning(f'{option} is deprecated. Use {suggestion} instead')
578 return True
579 return False
580
581 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
582 if self.params.get('geo_verification_proxy') is None:
583 self.params['geo_verification_proxy'] = self.params['cn_verification_proxy']
584
585 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
586 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
587 check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"')
588
589 for msg in self.params.get('_warnings', []):
590 self.report_warning(msg)
591 for msg in self.params.get('_deprecation_warnings', []):
592 self.deprecation_warning(msg)
593
594 if 'list-formats' in self.params.get('compat_opts', []):
595 self.params['listformats_table'] = False
596
597 if 'overwrites' not in self.params and self.params.get('nooverwrites') is not None:
598 # nooverwrites was unnecessarily changed to overwrites
599 # in 0c3d0f51778b153f65c21906031c2e091fcfb641
600 # This ensures compatibility with both keys
601 self.params['overwrites'] = not self.params['nooverwrites']
602 elif self.params.get('overwrites') is None:
603 self.params.pop('overwrites', None)
604 else:
605 self.params['nooverwrites'] = not self.params['overwrites']
606
607 self.params.setdefault('forceprint', {})
608 self.params.setdefault('print_to_file', {})
609
610 # Compatibility with older syntax
611 if not isinstance(params['forceprint'], dict):
612 self.params['forceprint'] = {'video': params['forceprint']}
613
614 if self.params.get('bidi_workaround', False):
615 try:
616 import pty
617 master, slave = pty.openpty()
618 width = compat_get_terminal_size().columns
619 width_args = [] if width is None else ['-w', str(width)]
620 sp_kwargs = {'stdin': subprocess.PIPE, 'stdout': slave, 'stderr': self._out_files.error}
621 try:
622 self._output_process = Popen(['bidiv'] + width_args, **sp_kwargs)
623 except OSError:
624 self._output_process = Popen(['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
625 self._output_channel = os.fdopen(master, 'rb')
626 except OSError as ose:
627 if ose.errno == errno.ENOENT:
628 self.report_warning(
629 'Could not find fribidi executable, ignoring --bidi-workaround. '
630 'Make sure that fribidi is an executable file in one of the directories in your $PATH.')
631 else:
632 raise
633
634 if auto_init:
635 if auto_init != 'no_verbose_header':
636 self.print_debug_header()
637 self.add_default_info_extractors()
638
639 if (sys.platform != 'win32'
640 and sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
641 and not self.params.get('restrictfilenames', False)):
642 # Unicode filesystem API will throw errors (#1474, #13027)
643 self.report_warning(
644 'Assuming --restrict-filenames since file system encoding '
645 'cannot encode all characters. '
646 'Set the LC_ALL environment variable to fix this.')
647 self.params['restrictfilenames'] = True
648
649 self.outtmpl_dict = self.parse_outtmpl()
650
651 # Creating format selector here allows us to catch syntax errors before the extraction
652 self.format_selector = (
653 self.params.get('format') if self.params.get('format') in (None, '-')
654 else self.params['format'] if callable(self.params['format'])
655 else self.build_format_selector(self.params['format']))
656
657 # Set http_headers defaults according to std_headers
658 self.params['http_headers'] = merge_headers(std_headers, self.params.get('http_headers', {}))
659
660 hooks = {
661 'post_hooks': self.add_post_hook,
662 'progress_hooks': self.add_progress_hook,
663 'postprocessor_hooks': self.add_postprocessor_hook,
664 }
665 for opt, fn in hooks.items():
666 for ph in self.params.get(opt, []):
667 fn(ph)
668
669 for pp_def_raw in self.params.get('postprocessors', []):
670 pp_def = dict(pp_def_raw)
671 when = pp_def.pop('when', 'post_process')
672 self.add_post_processor(
673 get_postprocessor(pp_def.pop('key'))(self, **pp_def),
674 when=when)
675
676 self._setup_opener()
677 register_socks_protocols()
678
679 def preload_download_archive(fn):
680 """Preload the archive, if any is specified"""
681 if fn is None:
682 return False
683 self.write_debug(f'Loading archive file {fn!r}')
684 try:
685 with locked_file(fn, 'r', encoding='utf-8') as archive_file:
686 for line in archive_file:
687 self.archive.add(line.strip())
688 except OSError as ioe:
689 if ioe.errno != errno.ENOENT:
690 raise
691 return False
692 return True
693
694 self.archive = set()
695 preload_download_archive(self.params.get('download_archive'))
696
697 def warn_if_short_id(self, argv):
698 # short YouTube ID starting with dash?
699 idxs = [
700 i for i, a in enumerate(argv)
701 if re.match(r'^-[0-9A-Za-z_-]{10}$', a)]
702 if idxs:
703 correct_argv = (
704 ['yt-dlp']
705 + [a for i, a in enumerate(argv) if i not in idxs]
706 + ['--'] + [argv[i] for i in idxs]
707 )
708 self.report_warning(
709 'Long argument string detected. '
710 'Use -- to separate parameters and URLs, like this:\n%s' %
711 args_to_str(correct_argv))
712
713 def add_info_extractor(self, ie):
714 """Add an InfoExtractor object to the end of the list."""
715 ie_key = ie.ie_key()
716 self._ies[ie_key] = ie
717 if not isinstance(ie, type):
718 self._ies_instances[ie_key] = ie
719 ie.set_downloader(self)
720
721 def _get_info_extractor_class(self, ie_key):
722 ie = self._ies.get(ie_key)
723 if ie is None:
724 ie = get_info_extractor(ie_key)
725 self.add_info_extractor(ie)
726 return ie
727
728 def get_info_extractor(self, ie_key):
729 """
730 Get an instance of an IE with name ie_key, it will try to get one from
731 the _ies list, if there's no instance it will create a new one and add
732 it to the extractor list.
733 """
734 ie = self._ies_instances.get(ie_key)
735 if ie is None:
736 ie = get_info_extractor(ie_key)()
737 self.add_info_extractor(ie)
738 return ie
739
740 def add_default_info_extractors(self):
741 """
742 Add the InfoExtractors returned by gen_extractors to the end of the list
743 """
744 for ie in gen_extractor_classes():
745 self.add_info_extractor(ie)
746
747 def add_post_processor(self, pp, when='post_process'):
748 """Add a PostProcessor object to the end of the chain."""
749 self._pps[when].append(pp)
750 pp.set_downloader(self)
751
752 def add_post_hook(self, ph):
753 """Add the post hook"""
754 self._post_hooks.append(ph)
755
756 def add_progress_hook(self, ph):
757 """Add the download progress hook"""
758 self._progress_hooks.append(ph)
759
760 def add_postprocessor_hook(self, ph):
761 """Add the postprocessing progress hook"""
762 self._postprocessor_hooks.append(ph)
763 for pps in self._pps.values():
764 for pp in pps:
765 pp.add_progress_hook(ph)
766
767 def _bidi_workaround(self, message):
768 if not hasattr(self, '_output_channel'):
769 return message
770
771 assert hasattr(self, '_output_process')
772 assert isinstance(message, compat_str)
773 line_count = message.count('\n') + 1
774 self._output_process.stdin.write((message + '\n').encode())
775 self._output_process.stdin.flush()
776 res = ''.join(self._output_channel.readline().decode()
777 for _ in range(line_count))
778 return res[:-len('\n')]
779
780 def _write_string(self, message, out=None, only_once=False):
781 if only_once:
782 if message in self._printed_messages:
783 return
784 self._printed_messages.add(message)
785 write_string(message, out=out, encoding=self.params.get('encoding'))
786
787 def to_stdout(self, message, skip_eol=False, quiet=None):
788 """Print message to stdout"""
789 if quiet is not None:
790 self.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument quiet. Use "YoutubeDL.to_screen" instead')
791 self._write_string(
792 '%s%s' % (self._bidi_workaround(message), ('' if skip_eol else '\n')),
793 self._out_files.out)
794
795 def to_screen(self, message, skip_eol=False, quiet=None):
796 """Print message to screen if not in quiet mode"""
797 if self.params.get('logger'):
798 self.params['logger'].debug(message)
799 return
800 if (self.params.get('quiet') if quiet is None else quiet) and not self.params.get('verbose'):
801 return
802 self._write_string(
803 '%s%s' % (self._bidi_workaround(message), ('' if skip_eol else '\n')),
804 self._out_files.screen)
805
806 def to_stderr(self, message, only_once=False):
807 """Print message to stderr"""
808 assert isinstance(message, compat_str)
809 if self.params.get('logger'):
810 self.params['logger'].error(message)
811 else:
812 self._write_string(f'{self._bidi_workaround(message)}\n', self._out_files.error, only_once=only_once)
813
814 def _send_console_code(self, code):
815 if compat_os_name == 'nt' or not self._out_files.console:
816 return
817 self._write_string(code, self._out_files.console)
818
819 def to_console_title(self, message):
820 if not self.params.get('consoletitle', False):
821 return
822 message = remove_terminal_sequences(message)
823 if compat_os_name == 'nt':
824 if ctypes.windll.kernel32.GetConsoleWindow():
825 # c_wchar_p() might not be necessary if `message` is
826 # already of type unicode()
827 ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
828 else:
829 self._send_console_code(f'\033]0;{message}\007')
830
831 def save_console_title(self):
832 if not self.params.get('consoletitle') or self.params.get('simulate'):
833 return
834 self._send_console_code('\033[22;0t') # Save the title on stack
835
836 def restore_console_title(self):
837 if not self.params.get('consoletitle') or self.params.get('simulate'):
838 return
839 self._send_console_code('\033[23;0t') # Restore the title from stack
840
841 def __enter__(self):
842 self.save_console_title()
843 return self
844
845 def __exit__(self, *args):
846 self.restore_console_title()
847
848 if self.params.get('cookiefile') is not None:
849 self.cookiejar.save(ignore_discard=True, ignore_expires=True)
850
851 def trouble(self, message=None, tb=None, is_error=True):
852 """Determine action to take when a download problem appears.
853
854 Depending on if the downloader has been configured to ignore
855 download errors or not, this method may throw an exception or
856 not when errors are found, after printing the message.
857
858 @param tb If given, is additional traceback information
859 @param is_error Whether to raise error according to ignorerrors
860 """
861 if message is not None:
862 self.to_stderr(message)
863 if self.params.get('verbose'):
864 if tb is None:
865 if sys.exc_info()[0]: # if .trouble has been called from an except block
866 tb = ''
867 if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
868 tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
869 tb += encode_compat_str(traceback.format_exc())
870 else:
871 tb_data = traceback.format_list(traceback.extract_stack())
872 tb = ''.join(tb_data)
873 if tb:
874 self.to_stderr(tb)
875 if not is_error:
876 return
877 if not self.params.get('ignoreerrors'):
878 if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
879 exc_info = sys.exc_info()[1].exc_info
880 else:
881 exc_info = sys.exc_info()
882 raise DownloadError(message, exc_info)
883 self._download_retcode = 1
884
885 Styles = Namespace(
886 HEADERS='yellow',
887 EMPHASIS='light blue',
888 FILENAME='green',
889 ID='green',
890 DELIM='blue',
891 ERROR='red',
892 WARNING='yellow',
893 SUPPRESS='light black',
894 )
895
896 def _format_text(self, handle, allow_colors, text, f, fallback=None, *, test_encoding=False):
897 text = str(text)
898 if test_encoding:
899 original_text = text
900 # handle.encoding can be None. See https://github.com/yt-dlp/yt-dlp/issues/2711
901 encoding = self.params.get('encoding') or getattr(handle, 'encoding', None) or 'ascii'
902 text = text.encode(encoding, 'ignore').decode(encoding)
903 if fallback is not None and text != original_text:
904 text = fallback
905 return format_text(text, f) if allow_colors else text if fallback is None else fallback
906
907 def _format_out(self, *args, **kwargs):
908 return self._format_text(self._out_files.out, self._allow_colors.out, *args, **kwargs)
909
910 def _format_screen(self, *args, **kwargs):
911 return self._format_text(self._out_files.screen, self._allow_colors.screen, *args, **kwargs)
912
913 def _format_err(self, *args, **kwargs):
914 return self._format_text(self._out_files.error, self._allow_colors.error, *args, **kwargs)
915
916 def report_warning(self, message, only_once=False):
917 '''
918 Print the message to stderr, it will be prefixed with 'WARNING:'
919 If stderr is a tty file the 'WARNING:' will be colored
920 '''
921 if self.params.get('logger') is not None:
922 self.params['logger'].warning(message)
923 else:
924 if self.params.get('no_warnings'):
925 return
926 self.to_stderr(f'{self._format_err("WARNING:", self.Styles.WARNING)} {message}', only_once)
927
928 def deprecation_warning(self, message):
929 if self.params.get('logger') is not None:
930 self.params['logger'].warning(f'DeprecationWarning: {message}')
931 else:
932 self.to_stderr(f'{self._format_err("DeprecationWarning:", self.Styles.ERROR)} {message}', True)
933
934 def report_error(self, message, *args, **kwargs):
935 '''
936 Do the same as trouble, but prefixes the message with 'ERROR:', colored
937 in red if stderr is a tty file.
938 '''
939 self.trouble(f'{self._format_err("ERROR:", self.Styles.ERROR)} {message}', *args, **kwargs)
940
941 def write_debug(self, message, only_once=False):
942 '''Log debug message or Print message to stderr'''
943 if not self.params.get('verbose', False):
944 return
945 message = '[debug] %s' % message
946 if self.params.get('logger'):
947 self.params['logger'].debug(message)
948 else:
949 self.to_stderr(message, only_once)
950
951 def report_file_already_downloaded(self, file_name):
952 """Report file has already been fully downloaded."""
953 try:
954 self.to_screen('[download] %s has already been downloaded' % file_name)
955 except UnicodeEncodeError:
956 self.to_screen('[download] The file has already been downloaded')
957
958 def report_file_delete(self, file_name):
959 """Report that existing file will be deleted."""
960 try:
961 self.to_screen('Deleting existing file %s' % file_name)
962 except UnicodeEncodeError:
963 self.to_screen('Deleting existing file')
964
965 def raise_no_formats(self, info, forced=False, *, msg=None):
966 has_drm = info.get('_has_drm')
967 ignored, expected = self.params.get('ignore_no_formats_error'), bool(msg)
968 msg = msg or has_drm and 'This video is DRM protected' or 'No video formats found!'
969 if forced or not ignored:
970 raise ExtractorError(msg, video_id=info['id'], ie=info['extractor'],
971 expected=has_drm or ignored or expected)
972 else:
973 self.report_warning(msg)
974
975 def parse_outtmpl(self):
976 outtmpl_dict = self.params.get('outtmpl', {})
977 if not isinstance(outtmpl_dict, dict):
978 outtmpl_dict = {'default': outtmpl_dict}
979 # Remove spaces in the default template
980 if self.params.get('restrictfilenames'):
981 sanitize = lambda x: x.replace(' - ', ' ').replace(' ', '-')
982 else:
983 sanitize = lambda x: x
984 outtmpl_dict.update({
985 k: sanitize(v) for k, v in DEFAULT_OUTTMPL.items()
986 if outtmpl_dict.get(k) is None})
987 for _, val in outtmpl_dict.items():
988 if isinstance(val, bytes):
989 self.report_warning('Parameter outtmpl is bytes, but should be a unicode string')
990 return outtmpl_dict
991
992 def get_output_path(self, dir_type='', filename=None):
993 paths = self.params.get('paths', {})
994 assert isinstance(paths, dict)
995 path = os.path.join(
996 expand_path(paths.get('home', '').strip()),
997 expand_path(paths.get(dir_type, '').strip()) if dir_type else '',
998 filename or '')
999 return sanitize_path(path, force=self.params.get('windowsfilenames'))
1000
1001 @staticmethod
1002 def _outtmpl_expandpath(outtmpl):
1003 # expand_path translates '%%' into '%' and '$$' into '$'
1004 # correspondingly that is not what we want since we need to keep
1005 # '%%' intact for template dict substitution step. Working around
1006 # with boundary-alike separator hack.
1007 sep = ''.join([random.choice(ascii_letters) for _ in range(32)])
1008 outtmpl = outtmpl.replace('%%', f'%{sep}%').replace('$$', f'${sep}$')
1009
1010 # outtmpl should be expand_path'ed before template dict substitution
1011 # because meta fields may contain env variables we don't want to
1012 # be expanded. For example, for outtmpl "%(title)s.%(ext)s" and
1013 # title "Hello $PATH", we don't want `$PATH` to be expanded.
1014 return expand_path(outtmpl).replace(sep, '')
1015
1016 @staticmethod
1017 def escape_outtmpl(outtmpl):
1018 ''' Escape any remaining strings like %s, %abc% etc. '''
1019 return re.sub(
1020 STR_FORMAT_RE_TMPL.format('', '(?![%(\0])'),
1021 lambda mobj: ('' if mobj.group('has_key') else '%') + mobj.group(0),
1022 outtmpl)
1023
1024 @classmethod
1025 def validate_outtmpl(cls, outtmpl):
1026 ''' @return None or Exception object '''
1027 outtmpl = re.sub(
1028 STR_FORMAT_RE_TMPL.format('[^)]*', '[ljqBUDS]'),
1029 lambda mobj: f'{mobj.group(0)[:-1]}s',
1030 cls._outtmpl_expandpath(outtmpl))
1031 try:
1032 cls.escape_outtmpl(outtmpl) % collections.defaultdict(int)
1033 return None
1034 except ValueError as err:
1035 return err
1036
1037 @staticmethod
1038 def _copy_infodict(info_dict):
1039 info_dict = dict(info_dict)
1040 info_dict.pop('__postprocessors', None)
1041 info_dict.pop('__pending_error', None)
1042 return info_dict
1043
1044 def prepare_outtmpl(self, outtmpl, info_dict, sanitize=False):
1045 """ Make the outtmpl and info_dict suitable for substitution: ydl.escape_outtmpl(outtmpl) % info_dict
1046 @param sanitize Whether to sanitize the output as a filename.
1047 For backward compatibility, a function can also be passed
1048 """
1049
1050 info_dict.setdefault('epoch', int(time.time())) # keep epoch consistent once set
1051
1052 info_dict = self._copy_infodict(info_dict)
1053 info_dict['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
1054 formatSeconds(info_dict['duration'], '-' if sanitize else ':')
1055 if info_dict.get('duration', None) is not None
1056 else None)
1057 info_dict['autonumber'] = int(self.params.get('autonumber_start', 1) - 1 + self._num_downloads)
1058 info_dict['video_autonumber'] = self._num_videos
1059 if info_dict.get('resolution') is None:
1060 info_dict['resolution'] = self.format_resolution(info_dict, default=None)
1061
1062 # For fields playlist_index, playlist_autonumber and autonumber convert all occurrences
1063 # of %(field)s to %(field)0Nd for backward compatibility
1064 field_size_compat_map = {
1065 'playlist_index': number_of_digits(info_dict.get('__last_playlist_index') or 0),
1066 'playlist_autonumber': number_of_digits(info_dict.get('n_entries') or 0),
1067 'autonumber': self.params.get('autonumber_size') or 5,
1068 }
1069
1070 TMPL_DICT = {}
1071 EXTERNAL_FORMAT_RE = re.compile(STR_FORMAT_RE_TMPL.format('[^)]*', f'[{STR_FORMAT_TYPES}ljqBUDS]'))
1072 MATH_FUNCTIONS = {
1073 '+': float.__add__,
1074 '-': float.__sub__,
1075 }
1076 # Field is of the form key1.key2...
1077 # where keys (except first) can be string, int or slice
1078 FIELD_RE = r'\w*(?:\.(?:\w+|{num}|{num}?(?::{num}?){{1,2}}))*'.format(num=r'(?:-?\d+)')
1079 MATH_FIELD_RE = rf'(?:{FIELD_RE}|-?{NUMBER_RE})'
1080 MATH_OPERATORS_RE = r'(?:%s)' % '|'.join(map(re.escape, MATH_FUNCTIONS.keys()))
1081 INTERNAL_FORMAT_RE = re.compile(rf'''(?x)
1082 (?P<negate>-)?
1083 (?P<fields>{FIELD_RE})
1084 (?P<maths>(?:{MATH_OPERATORS_RE}{MATH_FIELD_RE})*)
1085 (?:>(?P<strf_format>.+?))?
1086 (?P<remaining>
1087 (?P<alternate>(?<!\\),[^|&)]+)?
1088 (?:&(?P<replacement>.*?))?
1089 (?:\|(?P<default>.*?))?
1090 )$''')
1091
1092 def _traverse_infodict(k):
1093 k = k.split('.')
1094 if k[0] == '':
1095 k.pop(0)
1096 return traverse_obj(info_dict, k, is_user_input=True, traverse_string=True)
1097
1098 def get_value(mdict):
1099 # Object traversal
1100 value = _traverse_infodict(mdict['fields'])
1101 # Negative
1102 if mdict['negate']:
1103 value = float_or_none(value)
1104 if value is not None:
1105 value *= -1
1106 # Do maths
1107 offset_key = mdict['maths']
1108 if offset_key:
1109 value = float_or_none(value)
1110 operator = None
1111 while offset_key:
1112 item = re.match(
1113 MATH_FIELD_RE if operator else MATH_OPERATORS_RE,
1114 offset_key).group(0)
1115 offset_key = offset_key[len(item):]
1116 if operator is None:
1117 operator = MATH_FUNCTIONS[item]
1118 continue
1119 item, multiplier = (item[1:], -1) if item[0] == '-' else (item, 1)
1120 offset = float_or_none(item)
1121 if offset is None:
1122 offset = float_or_none(_traverse_infodict(item))
1123 try:
1124 value = operator(value, multiplier * offset)
1125 except (TypeError, ZeroDivisionError):
1126 return None
1127 operator = None
1128 # Datetime formatting
1129 if mdict['strf_format']:
1130 value = strftime_or_none(value, mdict['strf_format'].replace('\\,', ','))
1131
1132 return value
1133
1134 na = self.params.get('outtmpl_na_placeholder', 'NA')
1135
1136 def filename_sanitizer(key, value, restricted=self.params.get('restrictfilenames')):
1137 return sanitize_filename(str(value), restricted=restricted, is_id=(
1138 bool(re.search(r'(^|[_.])id(\.|$)', key))
1139 if 'filename-sanitization' in self.params.get('compat_opts', [])
1140 else NO_DEFAULT))
1141
1142 sanitizer = sanitize if callable(sanitize) else filename_sanitizer
1143 sanitize = bool(sanitize)
1144
1145 def _dumpjson_default(obj):
1146 if isinstance(obj, (set, LazyList)):
1147 return list(obj)
1148 return repr(obj)
1149
1150 def create_key(outer_mobj):
1151 if not outer_mobj.group('has_key'):
1152 return outer_mobj.group(0)
1153 key = outer_mobj.group('key')
1154 mobj = re.match(INTERNAL_FORMAT_RE, key)
1155 initial_field = mobj.group('fields') if mobj else ''
1156 value, replacement, default = None, None, na
1157 while mobj:
1158 mobj = mobj.groupdict()
1159 default = mobj['default'] if mobj['default'] is not None else default
1160 value = get_value(mobj)
1161 replacement = mobj['replacement']
1162 if value is None and mobj['alternate']:
1163 mobj = re.match(INTERNAL_FORMAT_RE, mobj['remaining'][1:])
1164 else:
1165 break
1166
1167 fmt = outer_mobj.group('format')
1168 if fmt == 's' and value is not None and key in field_size_compat_map.keys():
1169 fmt = f'0{field_size_compat_map[key]:d}d'
1170
1171 value = default if value is None else value if replacement is None else replacement
1172
1173 flags = outer_mobj.group('conversion') or ''
1174 str_fmt = f'{fmt[:-1]}s'
1175 if fmt[-1] == 'l': # list
1176 delim = '\n' if '#' in flags else ', '
1177 value, fmt = delim.join(map(str, variadic(value, allowed_types=(str, bytes)))), str_fmt
1178 elif fmt[-1] == 'j': # json
1179 value, fmt = json.dumps(value, default=_dumpjson_default, indent=4 if '#' in flags else None), str_fmt
1180 elif fmt[-1] == 'q': # quoted
1181 value = map(str, variadic(value) if '#' in flags else [value])
1182 value, fmt = ' '.join(map(compat_shlex_quote, value)), str_fmt
1183 elif fmt[-1] == 'B': # bytes
1184 value = f'%{str_fmt}'.encode() % str(value).encode()
1185 value, fmt = value.decode('utf-8', 'ignore'), 's'
1186 elif fmt[-1] == 'U': # unicode normalized
1187 value, fmt = unicodedata.normalize(
1188 # "+" = compatibility equivalence, "#" = NFD
1189 'NF%s%s' % ('K' if '+' in flags else '', 'D' if '#' in flags else 'C'),
1190 value), str_fmt
1191 elif fmt[-1] == 'D': # decimal suffix
1192 num_fmt, fmt = fmt[:-1].replace('#', ''), 's'
1193 value = format_decimal_suffix(value, f'%{num_fmt}f%s' if num_fmt else '%d%s',
1194 factor=1024 if '#' in flags else 1000)
1195 elif fmt[-1] == 'S': # filename sanitization
1196 value, fmt = filename_sanitizer(initial_field, value, restricted='#' in flags), str_fmt
1197 elif fmt[-1] == 'c':
1198 if value:
1199 value = str(value)[0]
1200 else:
1201 fmt = str_fmt
1202 elif fmt[-1] not in 'rs': # numeric
1203 value = float_or_none(value)
1204 if value is None:
1205 value, fmt = default, 's'
1206
1207 if sanitize:
1208 if fmt[-1] == 'r':
1209 # If value is an object, sanitize might convert it to a string
1210 # So we convert it to repr first
1211 value, fmt = repr(value), str_fmt
1212 if fmt[-1] in 'csr':
1213 value = sanitizer(initial_field, value)
1214
1215 key = '%s\0%s' % (key.replace('%', '%\0'), outer_mobj.group('format'))
1216 TMPL_DICT[key] = value
1217 return '{prefix}%({key}){fmt}'.format(key=key, fmt=fmt, prefix=outer_mobj.group('prefix'))
1218
1219 return EXTERNAL_FORMAT_RE.sub(create_key, outtmpl), TMPL_DICT
1220
1221 def evaluate_outtmpl(self, outtmpl, info_dict, *args, **kwargs):
1222 outtmpl, info_dict = self.prepare_outtmpl(outtmpl, info_dict, *args, **kwargs)
1223 return self.escape_outtmpl(outtmpl) % info_dict
1224
1225 def _prepare_filename(self, info_dict, *, outtmpl=None, tmpl_type=None):
1226 assert None in (outtmpl, tmpl_type), 'outtmpl and tmpl_type are mutually exclusive'
1227 if outtmpl is None:
1228 outtmpl = self.outtmpl_dict.get(tmpl_type or 'default', self.outtmpl_dict['default'])
1229 try:
1230 outtmpl = self._outtmpl_expandpath(outtmpl)
1231 filename = self.evaluate_outtmpl(outtmpl, info_dict, True)
1232 if not filename:
1233 return None
1234
1235 if tmpl_type in ('', 'temp'):
1236 final_ext, ext = self.params.get('final_ext'), info_dict.get('ext')
1237 if final_ext and ext and final_ext != ext and filename.endswith(f'.{final_ext}'):
1238 filename = replace_extension(filename, ext, final_ext)
1239 elif tmpl_type:
1240 force_ext = OUTTMPL_TYPES[tmpl_type]
1241 if force_ext:
1242 filename = replace_extension(filename, force_ext, info_dict.get('ext'))
1243
1244 # https://github.com/blackjack4494/youtube-dlc/issues/85
1245 trim_file_name = self.params.get('trim_file_name', False)
1246 if trim_file_name:
1247 no_ext, *ext = filename.rsplit('.', 2)
1248 filename = join_nonempty(no_ext[:trim_file_name], *ext, delim='.')
1249
1250 return filename
1251 except ValueError as err:
1252 self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')')
1253 return None
1254
1255 def prepare_filename(self, info_dict, dir_type='', *, outtmpl=None, warn=False):
1256 """Generate the output filename"""
1257 if outtmpl:
1258 assert not dir_type, 'outtmpl and dir_type are mutually exclusive'
1259 dir_type = None
1260 filename = self._prepare_filename(info_dict, tmpl_type=dir_type, outtmpl=outtmpl)
1261 if not filename and dir_type not in ('', 'temp'):
1262 return ''
1263
1264 if warn:
1265 if not self.params.get('paths'):
1266 pass
1267 elif filename == '-':
1268 self.report_warning('--paths is ignored when an outputting to stdout', only_once=True)
1269 elif os.path.isabs(filename):
1270 self.report_warning('--paths is ignored since an absolute path is given in output template', only_once=True)
1271 if filename == '-' or not filename:
1272 return filename
1273
1274 return self.get_output_path(dir_type, filename)
1275
1276 def _match_entry(self, info_dict, incomplete=False, silent=False):
1277 """ Returns None if the file should be downloaded """
1278
1279 video_title = info_dict.get('title', info_dict.get('id', 'video'))
1280
1281 def check_filter():
1282 if 'title' in info_dict:
1283 # This can happen when we're just evaluating the playlist
1284 title = info_dict['title']
1285 matchtitle = self.params.get('matchtitle', False)
1286 if matchtitle:
1287 if not re.search(matchtitle, title, re.IGNORECASE):
1288 return '"' + title + '" title did not match pattern "' + matchtitle + '"'
1289 rejecttitle = self.params.get('rejecttitle', False)
1290 if rejecttitle:
1291 if re.search(rejecttitle, title, re.IGNORECASE):
1292 return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
1293 date = info_dict.get('upload_date')
1294 if date is not None:
1295 dateRange = self.params.get('daterange', DateRange())
1296 if date not in dateRange:
1297 return f'{date_from_str(date).isoformat()} upload date is not in range {dateRange}'
1298 view_count = info_dict.get('view_count')
1299 if view_count is not None:
1300 min_views = self.params.get('min_views')
1301 if min_views is not None and view_count < min_views:
1302 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
1303 max_views = self.params.get('max_views')
1304 if max_views is not None and view_count > max_views:
1305 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
1306 if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')):
1307 return 'Skipping "%s" because it is age restricted' % video_title
1308
1309 match_filter = self.params.get('match_filter')
1310 if match_filter is not None:
1311 try:
1312 ret = match_filter(info_dict, incomplete=incomplete)
1313 except TypeError:
1314 # For backward compatibility
1315 ret = None if incomplete else match_filter(info_dict)
1316 if ret is NO_DEFAULT:
1317 while True:
1318 filename = self._format_screen(self.prepare_filename(info_dict), self.Styles.FILENAME)
1319 reply = input(self._format_screen(
1320 f'Download "{filename}"? (Y/n): ', self.Styles.EMPHASIS)).lower().strip()
1321 if reply in {'y', ''}:
1322 return None
1323 elif reply == 'n':
1324 return f'Skipping {video_title}'
1325 elif ret is not None:
1326 return ret
1327 return None
1328
1329 if self.in_download_archive(info_dict):
1330 reason = '%s has already been recorded in the archive' % video_title
1331 break_opt, break_err = 'break_on_existing', ExistingVideoReached
1332 else:
1333 reason = check_filter()
1334 break_opt, break_err = 'break_on_reject', RejectedVideoReached
1335 if reason is not None:
1336 if not silent:
1337 self.to_screen('[download] ' + reason)
1338 if self.params.get(break_opt, False):
1339 raise break_err()
1340 return reason
1341
1342 @staticmethod
1343 def add_extra_info(info_dict, extra_info):
1344 '''Set the keys from extra_info in info dict if they are missing'''
1345 for key, value in extra_info.items():
1346 info_dict.setdefault(key, value)
1347
1348 def extract_info(self, url, download=True, ie_key=None, extra_info=None,
1349 process=True, force_generic_extractor=False):
1350 """
1351 Return a list with a dictionary for each video extracted.
1352
1353 Arguments:
1354 url -- URL to extract
1355
1356 Keyword arguments:
1357 download -- whether to download videos during extraction
1358 ie_key -- extractor key hint
1359 extra_info -- dictionary containing the extra values to add to each result
1360 process -- whether to resolve all unresolved references (URLs, playlist items),
1361 must be True for download to work.
1362 force_generic_extractor -- force using the generic extractor
1363 """
1364
1365 if extra_info is None:
1366 extra_info = {}
1367
1368 if not ie_key and force_generic_extractor:
1369 ie_key = 'Generic'
1370
1371 if ie_key:
1372 ies = {ie_key: self._get_info_extractor_class(ie_key)}
1373 else:
1374 ies = self._ies
1375
1376 for ie_key, ie in ies.items():
1377 if not ie.suitable(url):
1378 continue
1379
1380 if not ie.working():
1381 self.report_warning('The program functionality for this site has been marked as broken, '
1382 'and will probably not work.')
1383
1384 temp_id = ie.get_temp_id(url)
1385 if temp_id is not None and self.in_download_archive({'id': temp_id, 'ie_key': ie_key}):
1386 self.to_screen(f'[{ie_key}] {temp_id}: has already been recorded in the archive')
1387 if self.params.get('break_on_existing', False):
1388 raise ExistingVideoReached()
1389 break
1390 return self.__extract_info(url, self.get_info_extractor(ie_key), download, extra_info, process)
1391 else:
1392 self.report_error('no suitable InfoExtractor for URL %s' % url)
1393
1394 def __handle_extraction_exceptions(func):
1395 @functools.wraps(func)
1396 def wrapper(self, *args, **kwargs):
1397 while True:
1398 try:
1399 return func(self, *args, **kwargs)
1400 except (DownloadCancelled, LazyList.IndexError, PagedList.IndexError):
1401 raise
1402 except ReExtractInfo as e:
1403 if e.expected:
1404 self.to_screen(f'{e}; Re-extracting data')
1405 else:
1406 self.to_stderr('\r')
1407 self.report_warning(f'{e}; Re-extracting data')
1408 continue
1409 except GeoRestrictedError as e:
1410 msg = e.msg
1411 if e.countries:
1412 msg += '\nThis video is available in %s.' % ', '.join(
1413 map(ISO3166Utils.short2full, e.countries))
1414 msg += '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1415 self.report_error(msg)
1416 except ExtractorError as e: # An error we somewhat expected
1417 self.report_error(str(e), e.format_traceback())
1418 except Exception as e:
1419 if self.params.get('ignoreerrors'):
1420 self.report_error(str(e), tb=encode_compat_str(traceback.format_exc()))
1421 else:
1422 raise
1423 break
1424 return wrapper
1425
1426 def _wait_for_video(self, ie_result):
1427 if (not self.params.get('wait_for_video')
1428 or ie_result.get('_type', 'video') != 'video'
1429 or ie_result.get('formats') or ie_result.get('url')):
1430 return
1431
1432 format_dur = lambda dur: '%02d:%02d:%02d' % timetuple_from_msec(dur * 1000)[:-1]
1433 last_msg = ''
1434
1435 def progress(msg):
1436 nonlocal last_msg
1437 self.to_screen(msg + ' ' * (len(last_msg) - len(msg)) + '\r', skip_eol=True)
1438 last_msg = msg
1439
1440 min_wait, max_wait = self.params.get('wait_for_video')
1441 diff = try_get(ie_result, lambda x: x['release_timestamp'] - time.time())
1442 if diff is None and ie_result.get('live_status') == 'is_upcoming':
1443 diff = round(random.uniform(min_wait, max_wait) if (max_wait and min_wait) else (max_wait or min_wait), 0)
1444 self.report_warning('Release time of video is not known')
1445 elif (diff or 0) <= 0:
1446 self.report_warning('Video should already be available according to extracted info')
1447 diff = min(max(diff or 0, min_wait or 0), max_wait or float('inf'))
1448 self.to_screen(f'[wait] Waiting for {format_dur(diff)} - Press Ctrl+C to try now')
1449
1450 wait_till = time.time() + diff
1451 try:
1452 while True:
1453 diff = wait_till - time.time()
1454 if diff <= 0:
1455 progress('')
1456 raise ReExtractInfo('[wait] Wait period ended', expected=True)
1457 progress(f'[wait] Remaining time until next attempt: {self._format_screen(format_dur(diff), self.Styles.EMPHASIS)}')
1458 time.sleep(1)
1459 except KeyboardInterrupt:
1460 progress('')
1461 raise ReExtractInfo('[wait] Interrupted by user', expected=True)
1462 except BaseException as e:
1463 if not isinstance(e, ReExtractInfo):
1464 self.to_screen('')
1465 raise
1466
1467 @__handle_extraction_exceptions
1468 def __extract_info(self, url, ie, download, extra_info, process):
1469 ie_result = ie.extract(url)
1470 if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
1471 return
1472 if isinstance(ie_result, list):
1473 # Backwards compatibility: old IE result format
1474 ie_result = {
1475 '_type': 'compat_list',
1476 'entries': ie_result,
1477 }
1478 if extra_info.get('original_url'):
1479 ie_result.setdefault('original_url', extra_info['original_url'])
1480 self.add_default_extra_info(ie_result, ie, url)
1481 if process:
1482 self._wait_for_video(ie_result)
1483 return self.process_ie_result(ie_result, download, extra_info)
1484 else:
1485 return ie_result
1486
1487 def add_default_extra_info(self, ie_result, ie, url):
1488 if url is not None:
1489 self.add_extra_info(ie_result, {
1490 'webpage_url': url,
1491 'original_url': url,
1492 })
1493 webpage_url = ie_result.get('webpage_url')
1494 if webpage_url:
1495 self.add_extra_info(ie_result, {
1496 'webpage_url_basename': url_basename(webpage_url),
1497 'webpage_url_domain': get_domain(webpage_url),
1498 })
1499 if ie is not None:
1500 self.add_extra_info(ie_result, {
1501 'extractor': ie.IE_NAME,
1502 'extractor_key': ie.ie_key(),
1503 })
1504
1505 def process_ie_result(self, ie_result, download=True, extra_info=None):
1506 """
1507 Take the result of the ie(may be modified) and resolve all unresolved
1508 references (URLs, playlist items).
1509
1510 It will also download the videos if 'download'.
1511 Returns the resolved ie_result.
1512 """
1513 if extra_info is None:
1514 extra_info = {}
1515 result_type = ie_result.get('_type', 'video')
1516
1517 if result_type in ('url', 'url_transparent'):
1518 ie_result['url'] = sanitize_url(ie_result['url'])
1519 if ie_result.get('original_url'):
1520 extra_info.setdefault('original_url', ie_result['original_url'])
1521
1522 extract_flat = self.params.get('extract_flat', False)
1523 if ((extract_flat == 'in_playlist' and 'playlist' in extra_info)
1524 or extract_flat is True):
1525 info_copy = ie_result.copy()
1526 ie = try_get(ie_result.get('ie_key'), self.get_info_extractor)
1527 if ie and not ie_result.get('id'):
1528 info_copy['id'] = ie.get_temp_id(ie_result['url'])
1529 self.add_default_extra_info(info_copy, ie, ie_result['url'])
1530 self.add_extra_info(info_copy, extra_info)
1531 info_copy, _ = self.pre_process(info_copy)
1532 self.__forced_printings(info_copy, self.prepare_filename(info_copy), incomplete=True)
1533 self._raise_pending_errors(info_copy)
1534 if self.params.get('force_write_download_archive', False):
1535 self.record_download_archive(info_copy)
1536 return ie_result
1537
1538 if result_type == 'video':
1539 self.add_extra_info(ie_result, extra_info)
1540 ie_result = self.process_video_result(ie_result, download=download)
1541 self._raise_pending_errors(ie_result)
1542 additional_urls = (ie_result or {}).get('additional_urls')
1543 if additional_urls:
1544 # TODO: Improve MetadataParserPP to allow setting a list
1545 if isinstance(additional_urls, compat_str):
1546 additional_urls = [additional_urls]
1547 self.to_screen(
1548 '[info] %s: %d additional URL(s) requested' % (ie_result['id'], len(additional_urls)))
1549 self.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls))
1550 ie_result['additional_entries'] = [
1551 self.extract_info(
1552 url, download, extra_info=extra_info,
1553 force_generic_extractor=self.params.get('force_generic_extractor'))
1554 for url in additional_urls
1555 ]
1556 return ie_result
1557 elif result_type == 'url':
1558 # We have to add extra_info to the results because it may be
1559 # contained in a playlist
1560 return self.extract_info(
1561 ie_result['url'], download,
1562 ie_key=ie_result.get('ie_key'),
1563 extra_info=extra_info)
1564 elif result_type == 'url_transparent':
1565 # Use the information from the embedding page
1566 info = self.extract_info(
1567 ie_result['url'], ie_key=ie_result.get('ie_key'),
1568 extra_info=extra_info, download=False, process=False)
1569
1570 # extract_info may return None when ignoreerrors is enabled and
1571 # extraction failed with an error, don't crash and return early
1572 # in this case
1573 if not info:
1574 return info
1575
1576 new_result = info.copy()
1577 new_result.update(filter_dict(ie_result, lambda k, v: (
1578 v is not None and k not in {'_type', 'url', 'id', 'extractor', 'extractor_key', 'ie_key'})))
1579
1580 # Extracted info may not be a video result (i.e.
1581 # info.get('_type', 'video') != video) but rather an url or
1582 # url_transparent. In such cases outer metadata (from ie_result)
1583 # should be propagated to inner one (info). For this to happen
1584 # _type of info should be overridden with url_transparent. This
1585 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
1586 if new_result.get('_type') == 'url':
1587 new_result['_type'] = 'url_transparent'
1588
1589 return self.process_ie_result(
1590 new_result, download=download, extra_info=extra_info)
1591 elif result_type in ('playlist', 'multi_video'):
1592 # Protect from infinite recursion due to recursively nested playlists
1593 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
1594 webpage_url = ie_result['webpage_url']
1595 if webpage_url in self._playlist_urls:
1596 self.to_screen(
1597 '[download] Skipping already downloaded playlist: %s'
1598 % ie_result.get('title') or ie_result.get('id'))
1599 return
1600
1601 self._playlist_level += 1
1602 self._playlist_urls.add(webpage_url)
1603 self._fill_common_fields(ie_result, False)
1604 self._sanitize_thumbnails(ie_result)
1605 try:
1606 return self.__process_playlist(ie_result, download)
1607 finally:
1608 self._playlist_level -= 1
1609 if not self._playlist_level:
1610 self._playlist_urls.clear()
1611 elif result_type == 'compat_list':
1612 self.report_warning(
1613 'Extractor %s returned a compat_list result. '
1614 'It needs to be updated.' % ie_result.get('extractor'))
1615
1616 def _fixup(r):
1617 self.add_extra_info(r, {
1618 'extractor': ie_result['extractor'],
1619 'webpage_url': ie_result['webpage_url'],
1620 'webpage_url_basename': url_basename(ie_result['webpage_url']),
1621 'webpage_url_domain': get_domain(ie_result['webpage_url']),
1622 'extractor_key': ie_result['extractor_key'],
1623 })
1624 return r
1625 ie_result['entries'] = [
1626 self.process_ie_result(_fixup(r), download, extra_info)
1627 for r in ie_result['entries']
1628 ]
1629 return ie_result
1630 else:
1631 raise Exception('Invalid result type: %s' % result_type)
1632
1633 def _ensure_dir_exists(self, path):
1634 return make_dir(path, self.report_error)
1635
1636 @staticmethod
1637 def _playlist_infodict(ie_result, **kwargs):
1638 return {
1639 **ie_result,
1640 'playlist': ie_result.get('title') or ie_result.get('id'),
1641 'playlist_id': ie_result.get('id'),
1642 'playlist_title': ie_result.get('title'),
1643 'playlist_uploader': ie_result.get('uploader'),
1644 'playlist_uploader_id': ie_result.get('uploader_id'),
1645 'playlist_index': 0,
1646 **kwargs,
1647 }
1648
1649 def __process_playlist(self, ie_result, download):
1650 # We process each entry in the playlist
1651 playlist = ie_result.get('title') or ie_result.get('id')
1652 self.to_screen('[download] Downloading playlist: %s' % playlist)
1653
1654 if 'entries' not in ie_result:
1655 raise EntryNotInPlaylist('There are no entries')
1656
1657 MissingEntry = object()
1658 incomplete_entries = bool(ie_result.get('requested_entries'))
1659 if incomplete_entries:
1660 def fill_missing_entries(entries, indices):
1661 ret = [MissingEntry] * max(indices)
1662 for i, entry in zip(indices, entries):
1663 ret[i - 1] = entry
1664 return ret
1665 ie_result['entries'] = fill_missing_entries(ie_result['entries'], ie_result['requested_entries'])
1666
1667 playlist_results = []
1668
1669 playliststart = self.params.get('playliststart', 1)
1670 playlistend = self.params.get('playlistend')
1671 # For backwards compatibility, interpret -1 as whole list
1672 if playlistend == -1:
1673 playlistend = None
1674
1675 playlistitems_str = self.params.get('playlist_items')
1676 playlistitems = None
1677 if playlistitems_str is not None:
1678 def iter_playlistitems(format):
1679 for string_segment in format.split(','):
1680 if '-' in string_segment:
1681 start, end = string_segment.split('-')
1682 for item in range(int(start), int(end) + 1):
1683 yield int(item)
1684 else:
1685 yield int(string_segment)
1686 playlistitems = orderedSet(iter_playlistitems(playlistitems_str))
1687
1688 ie_entries = ie_result['entries']
1689 if isinstance(ie_entries, list):
1690 playlist_count = len(ie_entries)
1691 msg = f'Collected {playlist_count} videos; downloading %d of them'
1692 ie_result['playlist_count'] = ie_result.get('playlist_count') or playlist_count
1693
1694 def get_entry(i):
1695 return ie_entries[i - 1]
1696 else:
1697 msg = 'Downloading %d videos'
1698 if not isinstance(ie_entries, (PagedList, LazyList)):
1699 ie_entries = LazyList(ie_entries)
1700 elif isinstance(ie_entries, InAdvancePagedList):
1701 if ie_entries._pagesize == 1:
1702 playlist_count = ie_entries._pagecount
1703
1704 def get_entry(i):
1705 return YoutubeDL.__handle_extraction_exceptions(
1706 lambda self, i: ie_entries[i - 1]
1707 )(self, i)
1708
1709 entries, broken = [], False
1710 items = playlistitems if playlistitems is not None else itertools.count(playliststart)
1711 for i in items:
1712 if i == 0:
1713 continue
1714 if playlistitems is None and playlistend is not None and playlistend < i:
1715 break
1716 entry = None
1717 try:
1718 entry = get_entry(i)
1719 if entry is MissingEntry:
1720 raise EntryNotInPlaylist()
1721 except (IndexError, EntryNotInPlaylist):
1722 if incomplete_entries:
1723 raise EntryNotInPlaylist(f'Entry {i} cannot be found')
1724 elif not playlistitems:
1725 break
1726 entries.append(entry)
1727 try:
1728 if entry is not None:
1729 # TODO: Add auto-generated fields
1730 self._match_entry(entry, incomplete=True, silent=True)
1731 except (ExistingVideoReached, RejectedVideoReached):
1732 broken = True
1733 break
1734 ie_result['entries'] = entries
1735
1736 # Save playlist_index before re-ordering
1737 entries = [
1738 ((playlistitems[i - 1] if playlistitems else i + playliststart - 1), entry)
1739 for i, entry in enumerate(entries, 1)
1740 if entry is not None]
1741 n_entries = len(entries)
1742
1743 if not (ie_result.get('playlist_count') or broken or playlistitems or playlistend):
1744 ie_result['playlist_count'] = n_entries
1745
1746 if not playlistitems and (playliststart != 1 or playlistend):
1747 playlistitems = list(range(playliststart, playliststart + n_entries))
1748 ie_result['requested_entries'] = playlistitems
1749
1750 _infojson_written = False
1751 write_playlist_files = self.params.get('allow_playlist_files', True)
1752 if write_playlist_files and self.params.get('list_thumbnails'):
1753 self.list_thumbnails(ie_result)
1754 if write_playlist_files and not self.params.get('simulate'):
1755 ie_copy = self._playlist_infodict(ie_result, n_entries=n_entries)
1756 _infojson_written = self._write_info_json(
1757 'playlist', ie_result, self.prepare_filename(ie_copy, 'pl_infojson'))
1758 if _infojson_written is None:
1759 return
1760 if self._write_description('playlist', ie_result,
1761 self.prepare_filename(ie_copy, 'pl_description')) is None:
1762 return
1763 # TODO: This should be passed to ThumbnailsConvertor if necessary
1764 self._write_thumbnails('playlist', ie_copy, self.prepare_filename(ie_copy, 'pl_thumbnail'))
1765
1766 if self.params.get('playlistreverse', False):
1767 entries = entries[::-1]
1768 if self.params.get('playlistrandom', False):
1769 random.shuffle(entries)
1770
1771 x_forwarded_for = ie_result.get('__x_forwarded_for_ip')
1772
1773 self.to_screen(f'[{ie_result["extractor"]}] playlist {playlist}: {msg % n_entries}')
1774 failures = 0
1775 max_failures = self.params.get('skip_playlist_after_errors') or float('inf')
1776 for i, entry_tuple in enumerate(entries, 1):
1777 playlist_index, entry = entry_tuple
1778 if 'playlist-index' in self.params.get('compat_opts', []):
1779 playlist_index = playlistitems[i - 1] if playlistitems else i + playliststart - 1
1780 self.to_screen('[download] Downloading video %s of %s' % (
1781 self._format_screen(i, self.Styles.ID), self._format_screen(n_entries, self.Styles.EMPHASIS)))
1782 # This __x_forwarded_for_ip thing is a bit ugly but requires
1783 # minimal changes
1784 if x_forwarded_for:
1785 entry['__x_forwarded_for_ip'] = x_forwarded_for
1786 extra = {
1787 'n_entries': n_entries,
1788 '__last_playlist_index': max(playlistitems) if playlistitems else (playlistend or n_entries),
1789 'playlist_count': ie_result.get('playlist_count'),
1790 'playlist_index': playlist_index,
1791 'playlist_autonumber': i,
1792 'playlist': playlist,
1793 'playlist_id': ie_result.get('id'),
1794 'playlist_title': ie_result.get('title'),
1795 'playlist_uploader': ie_result.get('uploader'),
1796 'playlist_uploader_id': ie_result.get('uploader_id'),
1797 'extractor': ie_result['extractor'],
1798 'webpage_url': ie_result['webpage_url'],
1799 'webpage_url_basename': url_basename(ie_result['webpage_url']),
1800 'webpage_url_domain': get_domain(ie_result['webpage_url']),
1801 'extractor_key': ie_result['extractor_key'],
1802 }
1803
1804 if self._match_entry(entry, incomplete=True) is not None:
1805 continue
1806
1807 entry_result = self.__process_iterable_entry(entry, download, extra)
1808 if not entry_result:
1809 failures += 1
1810 if failures >= max_failures:
1811 self.report_error(
1812 'Skipping the remaining entries in playlist "%s" since %d items failed extraction' % (playlist, failures))
1813 break
1814 playlist_results.append(entry_result)
1815 ie_result['entries'] = playlist_results
1816
1817 # Write the updated info to json
1818 if _infojson_written is True and self._write_info_json(
1819 'updated playlist', ie_result,
1820 self.prepare_filename(ie_copy, 'pl_infojson'), overwrite=True) is None:
1821 return
1822
1823 ie_result = self.run_all_pps('playlist', ie_result)
1824 self.to_screen(f'[download] Finished downloading playlist: {playlist}')
1825 return ie_result
1826
1827 @__handle_extraction_exceptions
1828 def __process_iterable_entry(self, entry, download, extra_info):
1829 return self.process_ie_result(
1830 entry, download=download, extra_info=extra_info)
1831
1832 def _build_format_filter(self, filter_spec):
1833 " Returns a function to filter the formats according to the filter_spec "
1834
1835 OPERATORS = {
1836 '<': operator.lt,
1837 '<=': operator.le,
1838 '>': operator.gt,
1839 '>=': operator.ge,
1840 '=': operator.eq,
1841 '!=': operator.ne,
1842 }
1843 operator_rex = re.compile(r'''(?x)\s*
1844 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)\s*
1845 (?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1846 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)\s*
1847 ''' % '|'.join(map(re.escape, OPERATORS.keys())))
1848 m = operator_rex.fullmatch(filter_spec)
1849 if m:
1850 try:
1851 comparison_value = int(m.group('value'))
1852 except ValueError:
1853 comparison_value = parse_filesize(m.group('value'))
1854 if comparison_value is None:
1855 comparison_value = parse_filesize(m.group('value') + 'B')
1856 if comparison_value is None:
1857 raise ValueError(
1858 'Invalid value %r in format specification %r' % (
1859 m.group('value'), filter_spec))
1860 op = OPERATORS[m.group('op')]
1861
1862 if not m:
1863 STR_OPERATORS = {
1864 '=': operator.eq,
1865 '^=': lambda attr, value: attr.startswith(value),
1866 '$=': lambda attr, value: attr.endswith(value),
1867 '*=': lambda attr, value: value in attr,
1868 '~=': lambda attr, value: value.search(attr) is not None
1869 }
1870 str_operator_rex = re.compile(r'''(?x)\s*
1871 (?P<key>[a-zA-Z0-9._-]+)\s*
1872 (?P<negation>!\s*)?(?P<op>%s)\s*(?P<none_inclusive>\?\s*)?
1873 (?P<quote>["'])?
1874 (?P<value>(?(quote)(?:(?!(?P=quote))[^\\]|\\.)+|[\w.-]+))
1875 (?(quote)(?P=quote))\s*
1876 ''' % '|'.join(map(re.escape, STR_OPERATORS.keys())))
1877 m = str_operator_rex.fullmatch(filter_spec)
1878 if m:
1879 if m.group('op') == '~=':
1880 comparison_value = re.compile(m.group('value'))
1881 else:
1882 comparison_value = re.sub(r'''\\([\\"'])''', r'\1', m.group('value'))
1883 str_op = STR_OPERATORS[m.group('op')]
1884 if m.group('negation'):
1885 op = lambda attr, value: not str_op(attr, value)
1886 else:
1887 op = str_op
1888
1889 if not m:
1890 raise SyntaxError('Invalid filter specification %r' % filter_spec)
1891
1892 def _filter(f):
1893 actual_value = f.get(m.group('key'))
1894 if actual_value is None:
1895 return m.group('none_inclusive')
1896 return op(actual_value, comparison_value)
1897 return _filter
1898
1899 def _check_formats(self, formats):
1900 for f in formats:
1901 self.to_screen('[info] Testing format %s' % f['format_id'])
1902 path = self.get_output_path('temp')
1903 if not self._ensure_dir_exists(f'{path}/'):
1904 continue
1905 temp_file = tempfile.NamedTemporaryFile(suffix='.tmp', delete=False, dir=path or None)
1906 temp_file.close()
1907 try:
1908 success, _ = self.dl(temp_file.name, f, test=True)
1909 except (DownloadError, IOError, OSError, ValueError) + network_exceptions:
1910 success = False
1911 finally:
1912 if os.path.exists(temp_file.name):
1913 try:
1914 os.remove(temp_file.name)
1915 except OSError:
1916 self.report_warning('Unable to delete temporary file "%s"' % temp_file.name)
1917 if success:
1918 yield f
1919 else:
1920 self.to_screen('[info] Unable to download format %s. Skipping...' % f['format_id'])
1921
1922 def _default_format_spec(self, info_dict, download=True):
1923
1924 def can_merge():
1925 merger = FFmpegMergerPP(self)
1926 return merger.available and merger.can_merge()
1927
1928 prefer_best = (
1929 not self.params.get('simulate')
1930 and download
1931 and (
1932 not can_merge()
1933 or info_dict.get('is_live') and not self.params.get('live_from_start')
1934 or self.outtmpl_dict['default'] == '-'))
1935 compat = (
1936 prefer_best
1937 or self.params.get('allow_multiple_audio_streams', False)
1938 or 'format-spec' in self.params.get('compat_opts', []))
1939
1940 return (
1941 'best/bestvideo+bestaudio' if prefer_best
1942 else 'bestvideo*+bestaudio/best' if not compat
1943 else 'bestvideo+bestaudio/best')
1944
1945 def build_format_selector(self, format_spec):
1946 def syntax_error(note, start):
1947 message = (
1948 'Invalid format specification: '
1949 '{}\n\t{}\n\t{}^'.format(note, format_spec, ' ' * start[1]))
1950 return SyntaxError(message)
1951
1952 PICKFIRST = 'PICKFIRST'
1953 MERGE = 'MERGE'
1954 SINGLE = 'SINGLE'
1955 GROUP = 'GROUP'
1956 FormatSelector = collections.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
1957
1958 allow_multiple_streams = {'audio': self.params.get('allow_multiple_audio_streams', False),
1959 'video': self.params.get('allow_multiple_video_streams', False)}
1960
1961 check_formats = self.params.get('check_formats') == 'selected'
1962
1963 def _parse_filter(tokens):
1964 filter_parts = []
1965 for type, string, start, _, _ in tokens:
1966 if type == tokenize.OP and string == ']':
1967 return ''.join(filter_parts)
1968 else:
1969 filter_parts.append(string)
1970
1971 def _remove_unused_ops(tokens):
1972 # Remove operators that we don't use and join them with the surrounding strings
1973 # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
1974 ALLOWED_OPS = ('/', '+', ',', '(', ')')
1975 last_string, last_start, last_end, last_line = None, None, None, None
1976 for type, string, start, end, line in tokens:
1977 if type == tokenize.OP and string == '[':
1978 if last_string:
1979 yield tokenize.NAME, last_string, last_start, last_end, last_line
1980 last_string = None
1981 yield type, string, start, end, line
1982 # everything inside brackets will be handled by _parse_filter
1983 for type, string, start, end, line in tokens:
1984 yield type, string, start, end, line
1985 if type == tokenize.OP and string == ']':
1986 break
1987 elif type == tokenize.OP and string in ALLOWED_OPS:
1988 if last_string:
1989 yield tokenize.NAME, last_string, last_start, last_end, last_line
1990 last_string = None
1991 yield type, string, start, end, line
1992 elif type in [tokenize.NAME, tokenize.NUMBER, tokenize.OP]:
1993 if not last_string:
1994 last_string = string
1995 last_start = start
1996 last_end = end
1997 else:
1998 last_string += string
1999 if last_string:
2000 yield tokenize.NAME, last_string, last_start, last_end, last_line
2001
2002 def _parse_format_selection(tokens, inside_merge=False, inside_choice=False, inside_group=False):
2003 selectors = []
2004 current_selector = None
2005 for type, string, start, _, _ in tokens:
2006 # ENCODING is only defined in python 3.x
2007 if type == getattr(tokenize, 'ENCODING', None):
2008 continue
2009 elif type in [tokenize.NAME, tokenize.NUMBER]:
2010 current_selector = FormatSelector(SINGLE, string, [])
2011 elif type == tokenize.OP:
2012 if string == ')':
2013 if not inside_group:
2014 # ')' will be handled by the parentheses group
2015 tokens.restore_last_token()
2016 break
2017 elif inside_merge and string in ['/', ',']:
2018 tokens.restore_last_token()
2019 break
2020 elif inside_choice and string == ',':
2021 tokens.restore_last_token()
2022 break
2023 elif string == ',':
2024 if not current_selector:
2025 raise syntax_error('"," must follow a format selector', start)
2026 selectors.append(current_selector)
2027 current_selector = None
2028 elif string == '/':
2029 if not current_selector:
2030 raise syntax_error('"/" must follow a format selector', start)
2031 first_choice = current_selector
2032 second_choice = _parse_format_selection(tokens, inside_choice=True)
2033 current_selector = FormatSelector(PICKFIRST, (first_choice, second_choice), [])
2034 elif string == '[':
2035 if not current_selector:
2036 current_selector = FormatSelector(SINGLE, 'best', [])
2037 format_filter = _parse_filter(tokens)
2038 current_selector.filters.append(format_filter)
2039 elif string == '(':
2040 if current_selector:
2041 raise syntax_error('Unexpected "("', start)
2042 group = _parse_format_selection(tokens, inside_group=True)
2043 current_selector = FormatSelector(GROUP, group, [])
2044 elif string == '+':
2045 if not current_selector:
2046 raise syntax_error('Unexpected "+"', start)
2047 selector_1 = current_selector
2048 selector_2 = _parse_format_selection(tokens, inside_merge=True)
2049 if not selector_2:
2050 raise syntax_error('Expected a selector', start)
2051 current_selector = FormatSelector(MERGE, (selector_1, selector_2), [])
2052 else:
2053 raise syntax_error(f'Operator not recognized: "{string}"', start)
2054 elif type == tokenize.ENDMARKER:
2055 break
2056 if current_selector:
2057 selectors.append(current_selector)
2058 return selectors
2059
2060 def _merge(formats_pair):
2061 format_1, format_2 = formats_pair
2062
2063 formats_info = []
2064 formats_info.extend(format_1.get('requested_formats', (format_1,)))
2065 formats_info.extend(format_2.get('requested_formats', (format_2,)))
2066
2067 if not allow_multiple_streams['video'] or not allow_multiple_streams['audio']:
2068 get_no_more = {'video': False, 'audio': False}
2069 for (i, fmt_info) in enumerate(formats_info):
2070 if fmt_info.get('acodec') == fmt_info.get('vcodec') == 'none':
2071 formats_info.pop(i)
2072 continue
2073 for aud_vid in ['audio', 'video']:
2074 if not allow_multiple_streams[aud_vid] and fmt_info.get(aud_vid[0] + 'codec') != 'none':
2075 if get_no_more[aud_vid]:
2076 formats_info.pop(i)
2077 break
2078 get_no_more[aud_vid] = True
2079
2080 if len(formats_info) == 1:
2081 return formats_info[0]
2082
2083 video_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('vcodec') != 'none']
2084 audio_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('acodec') != 'none']
2085
2086 the_only_video = video_fmts[0] if len(video_fmts) == 1 else None
2087 the_only_audio = audio_fmts[0] if len(audio_fmts) == 1 else None
2088
2089 output_ext = self.params.get('merge_output_format')
2090 if not output_ext:
2091 if the_only_video:
2092 output_ext = the_only_video['ext']
2093 elif the_only_audio and not video_fmts:
2094 output_ext = the_only_audio['ext']
2095 else:
2096 output_ext = 'mkv'
2097
2098 filtered = lambda *keys: filter(None, (traverse_obj(fmt, *keys) for fmt in formats_info))
2099
2100 new_dict = {
2101 'requested_formats': formats_info,
2102 'format': '+'.join(filtered('format')),
2103 'format_id': '+'.join(filtered('format_id')),
2104 'ext': output_ext,
2105 'protocol': '+'.join(map(determine_protocol, formats_info)),
2106 'language': '+'.join(orderedSet(filtered('language'))) or None,
2107 'format_note': '+'.join(orderedSet(filtered('format_note'))) or None,
2108 'filesize_approx': sum(filtered('filesize', 'filesize_approx')) or None,
2109 'tbr': sum(filtered('tbr', 'vbr', 'abr')),
2110 }
2111
2112 if the_only_video:
2113 new_dict.update({
2114 'width': the_only_video.get('width'),
2115 'height': the_only_video.get('height'),
2116 'resolution': the_only_video.get('resolution') or self.format_resolution(the_only_video),
2117 'fps': the_only_video.get('fps'),
2118 'dynamic_range': the_only_video.get('dynamic_range'),
2119 'vcodec': the_only_video.get('vcodec'),
2120 'vbr': the_only_video.get('vbr'),
2121 'stretched_ratio': the_only_video.get('stretched_ratio'),
2122 })
2123
2124 if the_only_audio:
2125 new_dict.update({
2126 'acodec': the_only_audio.get('acodec'),
2127 'abr': the_only_audio.get('abr'),
2128 'asr': the_only_audio.get('asr'),
2129 })
2130
2131 return new_dict
2132
2133 def _check_formats(formats):
2134 if not check_formats:
2135 yield from formats
2136 return
2137 yield from self._check_formats(formats)
2138
2139 def _build_selector_function(selector):
2140 if isinstance(selector, list): # ,
2141 fs = [_build_selector_function(s) for s in selector]
2142
2143 def selector_function(ctx):
2144 for f in fs:
2145 yield from f(ctx)
2146 return selector_function
2147
2148 elif selector.type == GROUP: # ()
2149 selector_function = _build_selector_function(selector.selector)
2150
2151 elif selector.type == PICKFIRST: # /
2152 fs = [_build_selector_function(s) for s in selector.selector]
2153
2154 def selector_function(ctx):
2155 for f in fs:
2156 picked_formats = list(f(ctx))
2157 if picked_formats:
2158 return picked_formats
2159 return []
2160
2161 elif selector.type == MERGE: # +
2162 selector_1, selector_2 = map(_build_selector_function, selector.selector)
2163
2164 def selector_function(ctx):
2165 for pair in itertools.product(selector_1(ctx), selector_2(ctx)):
2166 yield _merge(pair)
2167
2168 elif selector.type == SINGLE: # atom
2169 format_spec = selector.selector or 'best'
2170
2171 # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
2172 if format_spec == 'all':
2173 def selector_function(ctx):
2174 yield from _check_formats(ctx['formats'][::-1])
2175 elif format_spec == 'mergeall':
2176 def selector_function(ctx):
2177 formats = list(_check_formats(
2178 f for f in ctx['formats'] if f.get('vcodec') != 'none' or f.get('acodec') != 'none'))
2179 if not formats:
2180 return
2181 merged_format = formats[-1]
2182 for f in formats[-2::-1]:
2183 merged_format = _merge((merged_format, f))
2184 yield merged_format
2185
2186 else:
2187 format_fallback, seperate_fallback, format_reverse, format_idx = False, None, True, 1
2188 mobj = re.match(
2189 r'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
2190 format_spec)
2191 if mobj is not None:
2192 format_idx = int_or_none(mobj.group('n'), default=1)
2193 format_reverse = mobj.group('bw')[0] == 'b'
2194 format_type = (mobj.group('type') or [None])[0]
2195 not_format_type = {'v': 'a', 'a': 'v'}.get(format_type)
2196 format_modified = mobj.group('mod') is not None
2197
2198 format_fallback = not format_type and not format_modified # for b, w
2199 _filter_f = (
2200 (lambda f: f.get('%scodec' % format_type) != 'none')
2201 if format_type and format_modified # bv*, ba*, wv*, wa*
2202 else (lambda f: f.get('%scodec' % not_format_type) == 'none')
2203 if format_type # bv, ba, wv, wa
2204 else (lambda f: f.get('vcodec') != 'none' and f.get('acodec') != 'none')
2205 if not format_modified # b, w
2206 else lambda f: True) # b*, w*
2207 filter_f = lambda f: _filter_f(f) and (
2208 f.get('vcodec') != 'none' or f.get('acodec') != 'none')
2209 else:
2210 if format_spec in self._format_selection_exts['audio']:
2211 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none'
2212 elif format_spec in self._format_selection_exts['video']:
2213 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none' and f.get('vcodec') != 'none'
2214 seperate_fallback = lambda f: f.get('ext') == format_spec and f.get('vcodec') != 'none'
2215 elif format_spec in self._format_selection_exts['storyboards']:
2216 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') == 'none' and f.get('vcodec') == 'none'
2217 else:
2218 filter_f = lambda f: f.get('format_id') == format_spec # id
2219
2220 def selector_function(ctx):
2221 formats = list(ctx['formats'])
2222 matches = list(filter(filter_f, formats)) if filter_f is not None else formats
2223 if not matches:
2224 if format_fallback and ctx['incomplete_formats']:
2225 # for extractors with incomplete formats (audio only (soundcloud)
2226 # or video only (imgur)) best/worst will fallback to
2227 # best/worst {video,audio}-only format
2228 matches = formats
2229 elif seperate_fallback and not ctx['has_merged_format']:
2230 # for compatibility with youtube-dl when there is no pre-merged format
2231 matches = list(filter(seperate_fallback, formats))
2232 matches = LazyList(_check_formats(matches[::-1 if format_reverse else 1]))
2233 try:
2234 yield matches[format_idx - 1]
2235 except LazyList.IndexError:
2236 return
2237
2238 filters = [self._build_format_filter(f) for f in selector.filters]
2239
2240 def final_selector(ctx):
2241 ctx_copy = dict(ctx)
2242 for _filter in filters:
2243 ctx_copy['formats'] = list(filter(_filter, ctx_copy['formats']))
2244 return selector_function(ctx_copy)
2245 return final_selector
2246
2247 stream = io.BytesIO(format_spec.encode())
2248 try:
2249 tokens = list(_remove_unused_ops(tokenize.tokenize(stream.readline)))
2250 except tokenize.TokenError:
2251 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec)))
2252
2253 class TokenIterator:
2254 def __init__(self, tokens):
2255 self.tokens = tokens
2256 self.counter = 0
2257
2258 def __iter__(self):
2259 return self
2260
2261 def __next__(self):
2262 if self.counter >= len(self.tokens):
2263 raise StopIteration()
2264 value = self.tokens[self.counter]
2265 self.counter += 1
2266 return value
2267
2268 next = __next__
2269
2270 def restore_last_token(self):
2271 self.counter -= 1
2272
2273 parsed_selector = _parse_format_selection(iter(TokenIterator(tokens)))
2274 return _build_selector_function(parsed_selector)
2275
2276 def _calc_headers(self, info_dict):
2277 res = merge_headers(self.params['http_headers'], info_dict.get('http_headers') or {})
2278
2279 cookies = self._calc_cookies(info_dict['url'])
2280 if cookies:
2281 res['Cookie'] = cookies
2282
2283 if 'X-Forwarded-For' not in res:
2284 x_forwarded_for_ip = info_dict.get('__x_forwarded_for_ip')
2285 if x_forwarded_for_ip:
2286 res['X-Forwarded-For'] = x_forwarded_for_ip
2287
2288 return res
2289
2290 def _calc_cookies(self, url):
2291 pr = sanitized_Request(url)
2292 self.cookiejar.add_cookie_header(pr)
2293 return pr.get_header('Cookie')
2294
2295 def _sort_thumbnails(self, thumbnails):
2296 thumbnails.sort(key=lambda t: (
2297 t.get('preference') if t.get('preference') is not None else -1,
2298 t.get('width') if t.get('width') is not None else -1,
2299 t.get('height') if t.get('height') is not None else -1,
2300 t.get('id') if t.get('id') is not None else '',
2301 t.get('url')))
2302
2303 def _sanitize_thumbnails(self, info_dict):
2304 thumbnails = info_dict.get('thumbnails')
2305 if thumbnails is None:
2306 thumbnail = info_dict.get('thumbnail')
2307 if thumbnail:
2308 info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}]
2309 if not thumbnails:
2310 return
2311
2312 def check_thumbnails(thumbnails):
2313 for t in thumbnails:
2314 self.to_screen(f'[info] Testing thumbnail {t["id"]}')
2315 try:
2316 self.urlopen(HEADRequest(t['url']))
2317 except network_exceptions as err:
2318 self.to_screen(f'[info] Unable to connect to thumbnail {t["id"]} URL {t["url"]!r} - {err}. Skipping...')
2319 continue
2320 yield t
2321
2322 self._sort_thumbnails(thumbnails)
2323 for i, t in enumerate(thumbnails):
2324 if t.get('id') is None:
2325 t['id'] = '%d' % i
2326 if t.get('width') and t.get('height'):
2327 t['resolution'] = '%dx%d' % (t['width'], t['height'])
2328 t['url'] = sanitize_url(t['url'])
2329
2330 if self.params.get('check_formats') is True:
2331 info_dict['thumbnails'] = LazyList(check_thumbnails(thumbnails[::-1]), reverse=True)
2332 else:
2333 info_dict['thumbnails'] = thumbnails
2334
2335 def _fill_common_fields(self, info_dict, is_video=True):
2336 # TODO: move sanitization here
2337 if is_video:
2338 # playlists are allowed to lack "title"
2339 title = info_dict.get('title', NO_DEFAULT)
2340 if title is NO_DEFAULT:
2341 raise ExtractorError('Missing "title" field in extractor result',
2342 video_id=info_dict['id'], ie=info_dict['extractor'])
2343 info_dict['fulltitle'] = title
2344 if not title:
2345 if title == '':
2346 self.write_debug('Extractor gave empty title. Creating a generic title')
2347 else:
2348 self.report_warning('Extractor failed to obtain "title". Creating a generic title instead')
2349 info_dict['title'] = f'{info_dict["extractor"].replace(":", "-")} video #{info_dict["id"]}'
2350
2351 if info_dict.get('duration') is not None:
2352 info_dict['duration_string'] = formatSeconds(info_dict['duration'])
2353
2354 for ts_key, date_key in (
2355 ('timestamp', 'upload_date'),
2356 ('release_timestamp', 'release_date'),
2357 ('modified_timestamp', 'modified_date'),
2358 ):
2359 if info_dict.get(date_key) is None and info_dict.get(ts_key) is not None:
2360 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
2361 # see http://bugs.python.org/issue1646728)
2362 with contextlib.suppress(ValueError, OverflowError, OSError):
2363 upload_date = datetime.datetime.utcfromtimestamp(info_dict[ts_key])
2364 info_dict[date_key] = upload_date.strftime('%Y%m%d')
2365
2366 live_keys = ('is_live', 'was_live')
2367 live_status = info_dict.get('live_status')
2368 if live_status is None:
2369 for key in live_keys:
2370 if info_dict.get(key) is False:
2371 continue
2372 if info_dict.get(key):
2373 live_status = key
2374 break
2375 if all(info_dict.get(key) is False for key in live_keys):
2376 live_status = 'not_live'
2377 if live_status:
2378 info_dict['live_status'] = live_status
2379 for key in live_keys:
2380 if info_dict.get(key) is None:
2381 info_dict[key] = (live_status == key)
2382
2383 # Auto generate title fields corresponding to the *_number fields when missing
2384 # in order to always have clean titles. This is very common for TV series.
2385 for field in ('chapter', 'season', 'episode'):
2386 if info_dict.get('%s_number' % field) is not None and not info_dict.get(field):
2387 info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field])
2388
2389 def _raise_pending_errors(self, info):
2390 err = info.pop('__pending_error', None)
2391 if err:
2392 self.report_error(err, tb=False)
2393
2394 def process_video_result(self, info_dict, download=True):
2395 assert info_dict.get('_type', 'video') == 'video'
2396 self._num_videos += 1
2397
2398 if 'id' not in info_dict:
2399 raise ExtractorError('Missing "id" field in extractor result', ie=info_dict['extractor'])
2400 elif not info_dict.get('id'):
2401 raise ExtractorError('Extractor failed to obtain "id"', ie=info_dict['extractor'])
2402
2403 def report_force_conversion(field, field_not, conversion):
2404 self.report_warning(
2405 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
2406 % (field, field_not, conversion))
2407
2408 def sanitize_string_field(info, string_field):
2409 field = info.get(string_field)
2410 if field is None or isinstance(field, compat_str):
2411 return
2412 report_force_conversion(string_field, 'a string', 'string')
2413 info[string_field] = compat_str(field)
2414
2415 def sanitize_numeric_fields(info):
2416 for numeric_field in self._NUMERIC_FIELDS:
2417 field = info.get(numeric_field)
2418 if field is None or isinstance(field, (int, float)):
2419 continue
2420 report_force_conversion(numeric_field, 'numeric', 'int')
2421 info[numeric_field] = int_or_none(field)
2422
2423 sanitize_string_field(info_dict, 'id')
2424 sanitize_numeric_fields(info_dict)
2425 if (info_dict.get('duration') or 0) <= 0 and info_dict.pop('duration', None):
2426 self.report_warning('"duration" field is negative, there is an error in extractor')
2427
2428 if 'playlist' not in info_dict:
2429 # It isn't part of a playlist
2430 info_dict['playlist'] = None
2431 info_dict['playlist_index'] = None
2432
2433 self._sanitize_thumbnails(info_dict)
2434
2435 thumbnail = info_dict.get('thumbnail')
2436 thumbnails = info_dict.get('thumbnails')
2437 if thumbnail:
2438 info_dict['thumbnail'] = sanitize_url(thumbnail)
2439 elif thumbnails:
2440 info_dict['thumbnail'] = thumbnails[-1]['url']
2441
2442 if info_dict.get('display_id') is None and 'id' in info_dict:
2443 info_dict['display_id'] = info_dict['id']
2444
2445 self._fill_common_fields(info_dict)
2446
2447 for cc_kind in ('subtitles', 'automatic_captions'):
2448 cc = info_dict.get(cc_kind)
2449 if cc:
2450 for _, subtitle in cc.items():
2451 for subtitle_format in subtitle:
2452 if subtitle_format.get('url'):
2453 subtitle_format['url'] = sanitize_url(subtitle_format['url'])
2454 if subtitle_format.get('ext') is None:
2455 subtitle_format['ext'] = determine_ext(subtitle_format['url']).lower()
2456
2457 automatic_captions = info_dict.get('automatic_captions')
2458 subtitles = info_dict.get('subtitles')
2459
2460 info_dict['requested_subtitles'] = self.process_subtitles(
2461 info_dict['id'], subtitles, automatic_captions)
2462
2463 if info_dict.get('formats') is None:
2464 # There's only one format available
2465 formats = [info_dict]
2466 else:
2467 formats = info_dict['formats']
2468
2469 # or None ensures --clean-infojson removes it
2470 info_dict['_has_drm'] = any(f.get('has_drm') for f in formats) or None
2471 if not self.params.get('allow_unplayable_formats'):
2472 formats = [f for f in formats if not f.get('has_drm')]
2473 if info_dict['_has_drm'] and all(
2474 f.get('acodec') == f.get('vcodec') == 'none' for f in formats):
2475 self.report_warning(
2476 'This video is DRM protected and only images are available for download. '
2477 'Use --list-formats to see them')
2478
2479 get_from_start = not info_dict.get('is_live') or bool(self.params.get('live_from_start'))
2480 if not get_from_start:
2481 info_dict['title'] += ' ' + datetime.datetime.now().strftime('%Y-%m-%d %H:%M')
2482 if info_dict.get('is_live') and formats:
2483 formats = [f for f in formats if bool(f.get('is_from_start')) == get_from_start]
2484 if get_from_start and not formats:
2485 self.raise_no_formats(info_dict, msg=(
2486 '--live-from-start is passed, but there are no formats that can be downloaded from the start. '
2487 'If you want to download from the current time, use --no-live-from-start'))
2488
2489 if not formats:
2490 self.raise_no_formats(info_dict)
2491
2492 def is_wellformed(f):
2493 url = f.get('url')
2494 if not url:
2495 self.report_warning(
2496 '"url" field is missing or empty - skipping format, '
2497 'there is an error in extractor')
2498 return False
2499 if isinstance(url, bytes):
2500 sanitize_string_field(f, 'url')
2501 return True
2502
2503 # Filter out malformed formats for better extraction robustness
2504 formats = list(filter(is_wellformed, formats))
2505
2506 formats_dict = {}
2507
2508 # We check that all the formats have the format and format_id fields
2509 for i, format in enumerate(formats):
2510 sanitize_string_field(format, 'format_id')
2511 sanitize_numeric_fields(format)
2512 format['url'] = sanitize_url(format['url'])
2513 if not format.get('format_id'):
2514 format['format_id'] = compat_str(i)
2515 else:
2516 # Sanitize format_id from characters used in format selector expression
2517 format['format_id'] = re.sub(r'[\s,/+\[\]()]', '_', format['format_id'])
2518 format_id = format['format_id']
2519 if format_id not in formats_dict:
2520 formats_dict[format_id] = []
2521 formats_dict[format_id].append(format)
2522
2523 # Make sure all formats have unique format_id
2524 common_exts = set(itertools.chain(*self._format_selection_exts.values()))
2525 for format_id, ambiguous_formats in formats_dict.items():
2526 ambigious_id = len(ambiguous_formats) > 1
2527 for i, format in enumerate(ambiguous_formats):
2528 if ambigious_id:
2529 format['format_id'] = '%s-%d' % (format_id, i)
2530 if format.get('ext') is None:
2531 format['ext'] = determine_ext(format['url']).lower()
2532 # Ensure there is no conflict between id and ext in format selection
2533 # See https://github.com/yt-dlp/yt-dlp/issues/1282
2534 if format['format_id'] != format['ext'] and format['format_id'] in common_exts:
2535 format['format_id'] = 'f%s' % format['format_id']
2536
2537 for i, format in enumerate(formats):
2538 if format.get('format') is None:
2539 format['format'] = '{id} - {res}{note}'.format(
2540 id=format['format_id'],
2541 res=self.format_resolution(format),
2542 note=format_field(format, 'format_note', ' (%s)'),
2543 )
2544 if format.get('protocol') is None:
2545 format['protocol'] = determine_protocol(format)
2546 if format.get('resolution') is None:
2547 format['resolution'] = self.format_resolution(format, default=None)
2548 if format.get('dynamic_range') is None and format.get('vcodec') != 'none':
2549 format['dynamic_range'] = 'SDR'
2550 if (info_dict.get('duration') and format.get('tbr')
2551 and not format.get('filesize') and not format.get('filesize_approx')):
2552 format['filesize_approx'] = info_dict['duration'] * format['tbr'] * (1024 / 8)
2553
2554 # Add HTTP headers, so that external programs can use them from the
2555 # json output
2556 full_format_info = info_dict.copy()
2557 full_format_info.update(format)
2558 format['http_headers'] = self._calc_headers(full_format_info)
2559 # Remove private housekeeping stuff
2560 if '__x_forwarded_for_ip' in info_dict:
2561 del info_dict['__x_forwarded_for_ip']
2562
2563 if self.params.get('check_formats') is True:
2564 formats = LazyList(self._check_formats(formats[::-1]), reverse=True)
2565
2566 if not formats or formats[0] is not info_dict:
2567 # only set the 'formats' fields if the original info_dict list them
2568 # otherwise we end up with a circular reference, the first (and unique)
2569 # element in the 'formats' field in info_dict is info_dict itself,
2570 # which can't be exported to json
2571 info_dict['formats'] = formats
2572
2573 info_dict, _ = self.pre_process(info_dict)
2574
2575 if self._match_entry(info_dict, incomplete=self._format_fields) is not None:
2576 return info_dict
2577
2578 self.post_extract(info_dict)
2579 info_dict, _ = self.pre_process(info_dict, 'after_filter')
2580
2581 # The pre-processors may have modified the formats
2582 formats = info_dict.get('formats', [info_dict])
2583
2584 list_only = self.params.get('simulate') is None and (
2585 self.params.get('list_thumbnails') or self.params.get('listformats') or self.params.get('listsubtitles'))
2586 interactive_format_selection = not list_only and self.format_selector == '-'
2587 if self.params.get('list_thumbnails'):
2588 self.list_thumbnails(info_dict)
2589 if self.params.get('listsubtitles'):
2590 if 'automatic_captions' in info_dict:
2591 self.list_subtitles(
2592 info_dict['id'], automatic_captions, 'automatic captions')
2593 self.list_subtitles(info_dict['id'], subtitles, 'subtitles')
2594 if self.params.get('listformats') or interactive_format_selection:
2595 self.list_formats(info_dict)
2596 if list_only:
2597 # Without this printing, -F --print-json will not work
2598 self.__forced_printings(info_dict, self.prepare_filename(info_dict), incomplete=True)
2599 return info_dict
2600
2601 format_selector = self.format_selector
2602 if format_selector is None:
2603 req_format = self._default_format_spec(info_dict, download=download)
2604 self.write_debug('Default format spec: %s' % req_format)
2605 format_selector = self.build_format_selector(req_format)
2606
2607 while True:
2608 if interactive_format_selection:
2609 req_format = input(
2610 self._format_screen('\nEnter format selector: ', self.Styles.EMPHASIS))
2611 try:
2612 format_selector = self.build_format_selector(req_format)
2613 except SyntaxError as err:
2614 self.report_error(err, tb=False, is_error=False)
2615 continue
2616
2617 formats_to_download = list(format_selector({
2618 'formats': formats,
2619 'has_merged_format': any('none' not in (f.get('acodec'), f.get('vcodec')) for f in formats),
2620 'incomplete_formats': (
2621 # All formats are video-only or
2622 all(f.get('vcodec') != 'none' and f.get('acodec') == 'none' for f in formats)
2623 # all formats are audio-only
2624 or all(f.get('vcodec') == 'none' and f.get('acodec') != 'none' for f in formats)),
2625 }))
2626 if interactive_format_selection and not formats_to_download:
2627 self.report_error('Requested format is not available', tb=False, is_error=False)
2628 continue
2629 break
2630
2631 if not formats_to_download:
2632 if not self.params.get('ignore_no_formats_error'):
2633 raise ExtractorError(
2634 'Requested format is not available. Use --list-formats for a list of available formats',
2635 expected=True, video_id=info_dict['id'], ie=info_dict['extractor'])
2636 self.report_warning('Requested format is not available')
2637 # Process what we can, even without any available formats.
2638 formats_to_download = [{}]
2639
2640 best_format = formats_to_download[-1]
2641 if download:
2642 if best_format:
2643 self.to_screen(
2644 f'[info] {info_dict["id"]}: Downloading {len(formats_to_download)} format(s): '
2645 + ', '.join([f['format_id'] for f in formats_to_download]))
2646 max_downloads_reached = False
2647 for i, fmt in enumerate(formats_to_download):
2648 formats_to_download[i] = new_info = self._copy_infodict(info_dict)
2649 new_info.update(fmt)
2650 try:
2651 self.process_info(new_info)
2652 except MaxDownloadsReached:
2653 max_downloads_reached = True
2654 self._raise_pending_errors(new_info)
2655 # Remove copied info
2656 for key, val in tuple(new_info.items()):
2657 if info_dict.get(key) == val:
2658 new_info.pop(key)
2659 if max_downloads_reached:
2660 break
2661
2662 write_archive = {f.get('__write_download_archive', False) for f in formats_to_download}
2663 assert write_archive.issubset({True, False, 'ignore'})
2664 if True in write_archive and False not in write_archive:
2665 self.record_download_archive(info_dict)
2666
2667 info_dict['requested_downloads'] = formats_to_download
2668 info_dict = self.run_all_pps('after_video', info_dict)
2669 if max_downloads_reached:
2670 raise MaxDownloadsReached()
2671
2672 # We update the info dict with the selected best quality format (backwards compatibility)
2673 info_dict.update(best_format)
2674 return info_dict
2675
2676 def process_subtitles(self, video_id, normal_subtitles, automatic_captions):
2677 """Select the requested subtitles and their format"""
2678 available_subs, normal_sub_langs = {}, []
2679 if normal_subtitles and self.params.get('writesubtitles'):
2680 available_subs.update(normal_subtitles)
2681 normal_sub_langs = tuple(normal_subtitles.keys())
2682 if automatic_captions and self.params.get('writeautomaticsub'):
2683 for lang, cap_info in automatic_captions.items():
2684 if lang not in available_subs:
2685 available_subs[lang] = cap_info
2686
2687 if (not self.params.get('writesubtitles') and not
2688 self.params.get('writeautomaticsub') or not
2689 available_subs):
2690 return None
2691
2692 all_sub_langs = tuple(available_subs.keys())
2693 if self.params.get('allsubtitles', False):
2694 requested_langs = all_sub_langs
2695 elif self.params.get('subtitleslangs', False):
2696 # A list is used so that the order of languages will be the same as
2697 # given in subtitleslangs. See https://github.com/yt-dlp/yt-dlp/issues/1041
2698 requested_langs = []
2699 for lang_re in self.params.get('subtitleslangs'):
2700 discard = lang_re[0] == '-'
2701 if discard:
2702 lang_re = lang_re[1:]
2703 if lang_re == 'all':
2704 if discard:
2705 requested_langs = []
2706 else:
2707 requested_langs.extend(all_sub_langs)
2708 continue
2709 current_langs = filter(re.compile(lang_re + '$').match, all_sub_langs)
2710 if discard:
2711 for lang in current_langs:
2712 while lang in requested_langs:
2713 requested_langs.remove(lang)
2714 else:
2715 requested_langs.extend(current_langs)
2716 requested_langs = orderedSet(requested_langs)
2717 elif normal_sub_langs:
2718 requested_langs = ['en'] if 'en' in normal_sub_langs else normal_sub_langs[:1]
2719 else:
2720 requested_langs = ['en'] if 'en' in all_sub_langs else all_sub_langs[:1]
2721 if requested_langs:
2722 self.write_debug('Downloading subtitles: %s' % ', '.join(requested_langs))
2723
2724 formats_query = self.params.get('subtitlesformat', 'best')
2725 formats_preference = formats_query.split('/') if formats_query else []
2726 subs = {}
2727 for lang in requested_langs:
2728 formats = available_subs.get(lang)
2729 if formats is None:
2730 self.report_warning(f'{lang} subtitles not available for {video_id}')
2731 continue
2732 for ext in formats_preference:
2733 if ext == 'best':
2734 f = formats[-1]
2735 break
2736 matches = list(filter(lambda f: f['ext'] == ext, formats))
2737 if matches:
2738 f = matches[-1]
2739 break
2740 else:
2741 f = formats[-1]
2742 self.report_warning(
2743 'No subtitle format found matching "%s" for language %s, '
2744 'using %s' % (formats_query, lang, f['ext']))
2745 subs[lang] = f
2746 return subs
2747
2748 def _forceprint(self, key, info_dict):
2749 if info_dict is None:
2750 return
2751 info_copy = info_dict.copy()
2752 info_copy['formats_table'] = self.render_formats_table(info_dict)
2753 info_copy['thumbnails_table'] = self.render_thumbnails_table(info_dict)
2754 info_copy['subtitles_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('subtitles'))
2755 info_copy['automatic_captions_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('automatic_captions'))
2756
2757 def format_tmpl(tmpl):
2758 mobj = re.match(r'\w+(=?)$', tmpl)
2759 if mobj and mobj.group(1):
2760 return f'{tmpl[:-1]} = %({tmpl[:-1]})r'
2761 elif mobj:
2762 return f'%({tmpl})s'
2763 return tmpl
2764
2765 for tmpl in self.params['forceprint'].get(key, []):
2766 self.to_stdout(self.evaluate_outtmpl(format_tmpl(tmpl), info_copy))
2767
2768 for tmpl, file_tmpl in self.params['print_to_file'].get(key, []):
2769 filename = self.prepare_filename(info_dict, outtmpl=file_tmpl)
2770 tmpl = format_tmpl(tmpl)
2771 self.to_screen(f'[info] Writing {tmpl!r} to: {filename}')
2772 if self._ensure_dir_exists(filename):
2773 with open(filename, 'a', encoding='utf-8') as f:
2774 f.write(self.evaluate_outtmpl(tmpl, info_copy) + '\n')
2775
2776 def __forced_printings(self, info_dict, filename, incomplete):
2777 def print_mandatory(field, actual_field=None):
2778 if actual_field is None:
2779 actual_field = field
2780 if (self.params.get('force%s' % field, False)
2781 and (not incomplete or info_dict.get(actual_field) is not None)):
2782 self.to_stdout(info_dict[actual_field])
2783
2784 def print_optional(field):
2785 if (self.params.get('force%s' % field, False)
2786 and info_dict.get(field) is not None):
2787 self.to_stdout(info_dict[field])
2788
2789 info_dict = info_dict.copy()
2790 if filename is not None:
2791 info_dict['filename'] = filename
2792 if info_dict.get('requested_formats') is not None:
2793 # For RTMP URLs, also include the playpath
2794 info_dict['urls'] = '\n'.join(f['url'] + f.get('play_path', '') for f in info_dict['requested_formats'])
2795 elif info_dict.get('url'):
2796 info_dict['urls'] = info_dict['url'] + info_dict.get('play_path', '')
2797
2798 if (self.params.get('forcejson')
2799 or self.params['forceprint'].get('video')
2800 or self.params['print_to_file'].get('video')):
2801 self.post_extract(info_dict)
2802 self._forceprint('video', info_dict)
2803
2804 print_mandatory('title')
2805 print_mandatory('id')
2806 print_mandatory('url', 'urls')
2807 print_optional('thumbnail')
2808 print_optional('description')
2809 print_optional('filename')
2810 if self.params.get('forceduration') and info_dict.get('duration') is not None:
2811 self.to_stdout(formatSeconds(info_dict['duration']))
2812 print_mandatory('format')
2813
2814 if self.params.get('forcejson'):
2815 self.to_stdout(json.dumps(self.sanitize_info(info_dict)))
2816
2817 def dl(self, name, info, subtitle=False, test=False):
2818 if not info.get('url'):
2819 self.raise_no_formats(info, True)
2820
2821 if test:
2822 verbose = self.params.get('verbose')
2823 params = {
2824 'test': True,
2825 'quiet': self.params.get('quiet') or not verbose,
2826 'verbose': verbose,
2827 'noprogress': not verbose,
2828 'nopart': True,
2829 'skip_unavailable_fragments': False,
2830 'keep_fragments': False,
2831 'overwrites': True,
2832 '_no_ytdl_file': True,
2833 }
2834 else:
2835 params = self.params
2836 fd = get_suitable_downloader(info, params, to_stdout=(name == '-'))(self, params)
2837 if not test:
2838 for ph in self._progress_hooks:
2839 fd.add_progress_hook(ph)
2840 urls = '", "'.join(
2841 (f['url'].split(',')[0] + ',<data>' if f['url'].startswith('data:') else f['url'])
2842 for f in info.get('requested_formats', []) or [info])
2843 self.write_debug(f'Invoking {fd.FD_NAME} downloader on "{urls}"')
2844
2845 # Note: Ideally info should be a deep-copied so that hooks cannot modify it.
2846 # But it may contain objects that are not deep-copyable
2847 new_info = self._copy_infodict(info)
2848 if new_info.get('http_headers') is None:
2849 new_info['http_headers'] = self._calc_headers(new_info)
2850 return fd.download(name, new_info, subtitle)
2851
2852 def existing_file(self, filepaths, *, default_overwrite=True):
2853 existing_files = list(filter(os.path.exists, orderedSet(filepaths)))
2854 if existing_files and not self.params.get('overwrites', default_overwrite):
2855 return existing_files[0]
2856
2857 for file in existing_files:
2858 self.report_file_delete(file)
2859 os.remove(file)
2860 return None
2861
2862 def process_info(self, info_dict):
2863 """Process a single resolved IE result. (Modifies it in-place)"""
2864
2865 assert info_dict.get('_type', 'video') == 'video'
2866 original_infodict = info_dict
2867
2868 if 'format' not in info_dict and 'ext' in info_dict:
2869 info_dict['format'] = info_dict['ext']
2870
2871 # This is mostly just for backward compatibility of process_info
2872 # As a side-effect, this allows for format-specific filters
2873 if self._match_entry(info_dict) is not None:
2874 info_dict['__write_download_archive'] = 'ignore'
2875 return
2876
2877 # Does nothing under normal operation - for backward compatibility of process_info
2878 self.post_extract(info_dict)
2879 self._num_downloads += 1
2880
2881 # info_dict['_filename'] needs to be set for backward compatibility
2882 info_dict['_filename'] = full_filename = self.prepare_filename(info_dict, warn=True)
2883 temp_filename = self.prepare_filename(info_dict, 'temp')
2884 files_to_move = {}
2885
2886 # Forced printings
2887 self.__forced_printings(info_dict, full_filename, incomplete=('format' not in info_dict))
2888
2889 def check_max_downloads():
2890 if self._num_downloads >= float(self.params.get('max_downloads') or 'inf'):
2891 raise MaxDownloadsReached()
2892
2893 if self.params.get('simulate'):
2894 info_dict['__write_download_archive'] = self.params.get('force_write_download_archive')
2895 check_max_downloads()
2896 return
2897
2898 if full_filename is None:
2899 return
2900 if not self._ensure_dir_exists(encodeFilename(full_filename)):
2901 return
2902 if not self._ensure_dir_exists(encodeFilename(temp_filename)):
2903 return
2904
2905 if self._write_description('video', info_dict,
2906 self.prepare_filename(info_dict, 'description')) is None:
2907 return
2908
2909 sub_files = self._write_subtitles(info_dict, temp_filename)
2910 if sub_files is None:
2911 return
2912 files_to_move.update(dict(sub_files))
2913
2914 thumb_files = self._write_thumbnails(
2915 'video', info_dict, temp_filename, self.prepare_filename(info_dict, 'thumbnail'))
2916 if thumb_files is None:
2917 return
2918 files_to_move.update(dict(thumb_files))
2919
2920 infofn = self.prepare_filename(info_dict, 'infojson')
2921 _infojson_written = self._write_info_json('video', info_dict, infofn)
2922 if _infojson_written:
2923 info_dict['infojson_filename'] = infofn
2924 # For backward compatibility, even though it was a private field
2925 info_dict['__infojson_filename'] = infofn
2926 elif _infojson_written is None:
2927 return
2928
2929 # Note: Annotations are deprecated
2930 annofn = None
2931 if self.params.get('writeannotations', False):
2932 annofn = self.prepare_filename(info_dict, 'annotation')
2933 if annofn:
2934 if not self._ensure_dir_exists(encodeFilename(annofn)):
2935 return
2936 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(annofn)):
2937 self.to_screen('[info] Video annotations are already present')
2938 elif not info_dict.get('annotations'):
2939 self.report_warning('There are no annotations to write.')
2940 else:
2941 try:
2942 self.to_screen('[info] Writing video annotations to: ' + annofn)
2943 with open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
2944 annofile.write(info_dict['annotations'])
2945 except (KeyError, TypeError):
2946 self.report_warning('There are no annotations to write.')
2947 except OSError:
2948 self.report_error('Cannot write annotations file: ' + annofn)
2949 return
2950
2951 # Write internet shortcut files
2952 def _write_link_file(link_type):
2953 url = try_get(info_dict['webpage_url'], iri_to_uri)
2954 if not url:
2955 self.report_warning(
2956 f'Cannot write internet shortcut file because the actual URL of "{info_dict["webpage_url"]}" is unknown')
2957 return True
2958 linkfn = replace_extension(self.prepare_filename(info_dict, 'link'), link_type, info_dict.get('ext'))
2959 if not self._ensure_dir_exists(encodeFilename(linkfn)):
2960 return False
2961 if self.params.get('overwrites', True) and os.path.exists(encodeFilename(linkfn)):
2962 self.to_screen(f'[info] Internet shortcut (.{link_type}) is already present')
2963 return True
2964 try:
2965 self.to_screen(f'[info] Writing internet shortcut (.{link_type}) to: {linkfn}')
2966 with open(encodeFilename(to_high_limit_path(linkfn)), 'w', encoding='utf-8',
2967 newline='\r\n' if link_type == 'url' else '\n') as linkfile:
2968 template_vars = {'url': url}
2969 if link_type == 'desktop':
2970 template_vars['filename'] = linkfn[:-(len(link_type) + 1)]
2971 linkfile.write(LINK_TEMPLATES[link_type] % template_vars)
2972 except OSError:
2973 self.report_error(f'Cannot write internet shortcut {linkfn}')
2974 return False
2975 return True
2976
2977 write_links = {
2978 'url': self.params.get('writeurllink'),
2979 'webloc': self.params.get('writewebloclink'),
2980 'desktop': self.params.get('writedesktoplink'),
2981 }
2982 if self.params.get('writelink'):
2983 link_type = ('webloc' if sys.platform == 'darwin'
2984 else 'desktop' if sys.platform.startswith('linux')
2985 else 'url')
2986 write_links[link_type] = True
2987
2988 if any(should_write and not _write_link_file(link_type)
2989 for link_type, should_write in write_links.items()):
2990 return
2991
2992 def replace_info_dict(new_info):
2993 nonlocal info_dict
2994 if new_info == info_dict:
2995 return
2996 info_dict.clear()
2997 info_dict.update(new_info)
2998
2999 new_info, files_to_move = self.pre_process(info_dict, 'before_dl', files_to_move)
3000 replace_info_dict(new_info)
3001
3002 if self.params.get('skip_download'):
3003 info_dict['filepath'] = temp_filename
3004 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
3005 info_dict['__files_to_move'] = files_to_move
3006 replace_info_dict(self.run_pp(MoveFilesAfterDownloadPP(self, False), info_dict))
3007 info_dict['__write_download_archive'] = self.params.get('force_write_download_archive')
3008 else:
3009 # Download
3010 info_dict.setdefault('__postprocessors', [])
3011 try:
3012
3013 def existing_video_file(*filepaths):
3014 ext = info_dict.get('ext')
3015 converted = lambda file: replace_extension(file, self.params.get('final_ext') or ext, ext)
3016 file = self.existing_file(itertools.chain(*zip(map(converted, filepaths), filepaths)),
3017 default_overwrite=False)
3018 if file:
3019 info_dict['ext'] = os.path.splitext(file)[1][1:]
3020 return file
3021
3022 success = True
3023 if info_dict.get('requested_formats') is not None:
3024
3025 def compatible_formats(formats):
3026 # TODO: some formats actually allow this (mkv, webm, ogg, mp4), but not all of them.
3027 video_formats = [format for format in formats if format.get('vcodec') != 'none']
3028 audio_formats = [format for format in formats if format.get('acodec') != 'none']
3029 if len(video_formats) > 2 or len(audio_formats) > 2:
3030 return False
3031
3032 # Check extension
3033 exts = {format.get('ext') for format in formats}
3034 COMPATIBLE_EXTS = (
3035 {'mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma'},
3036 {'webm'},
3037 )
3038 for ext_sets in COMPATIBLE_EXTS:
3039 if ext_sets.issuperset(exts):
3040 return True
3041 # TODO: Check acodec/vcodec
3042 return False
3043
3044 requested_formats = info_dict['requested_formats']
3045 old_ext = info_dict['ext']
3046 if self.params.get('merge_output_format') is None:
3047 if not compatible_formats(requested_formats):
3048 info_dict['ext'] = 'mkv'
3049 self.report_warning(
3050 'Requested formats are incompatible for merge and will be merged into mkv')
3051 if (info_dict['ext'] == 'webm'
3052 and info_dict.get('thumbnails')
3053 # check with type instead of pp_key, __name__, or isinstance
3054 # since we dont want any custom PPs to trigger this
3055 and any(type(pp) == EmbedThumbnailPP for pp in self._pps['post_process'])): # noqa: E721
3056 info_dict['ext'] = 'mkv'
3057 self.report_warning(
3058 'webm doesn\'t support embedding a thumbnail, mkv will be used')
3059 new_ext = info_dict['ext']
3060
3061 def correct_ext(filename, ext=new_ext):
3062 if filename == '-':
3063 return filename
3064 filename_real_ext = os.path.splitext(filename)[1][1:]
3065 filename_wo_ext = (
3066 os.path.splitext(filename)[0]
3067 if filename_real_ext in (old_ext, new_ext)
3068 else filename)
3069 return f'{filename_wo_ext}.{ext}'
3070
3071 # Ensure filename always has a correct extension for successful merge
3072 full_filename = correct_ext(full_filename)
3073 temp_filename = correct_ext(temp_filename)
3074 dl_filename = existing_video_file(full_filename, temp_filename)
3075 info_dict['__real_download'] = False
3076
3077 downloaded = []
3078 merger = FFmpegMergerPP(self)
3079
3080 fd = get_suitable_downloader(info_dict, self.params, to_stdout=temp_filename == '-')
3081 if dl_filename is not None:
3082 self.report_file_already_downloaded(dl_filename)
3083 elif fd:
3084 for f in requested_formats if fd != FFmpegFD else []:
3085 f['filepath'] = fname = prepend_extension(
3086 correct_ext(temp_filename, info_dict['ext']),
3087 'f%s' % f['format_id'], info_dict['ext'])
3088 downloaded.append(fname)
3089 info_dict['url'] = '\n'.join(f['url'] for f in requested_formats)
3090 success, real_download = self.dl(temp_filename, info_dict)
3091 info_dict['__real_download'] = real_download
3092 else:
3093 if self.params.get('allow_unplayable_formats'):
3094 self.report_warning(
3095 'You have requested merging of multiple formats '
3096 'while also allowing unplayable formats to be downloaded. '
3097 'The formats won\'t be merged to prevent data corruption.')
3098 elif not merger.available:
3099 msg = 'You have requested merging of multiple formats but ffmpeg is not installed'
3100 if not self.params.get('ignoreerrors'):
3101 self.report_error(f'{msg}. Aborting due to --abort-on-error')
3102 return
3103 self.report_warning(f'{msg}. The formats won\'t be merged')
3104
3105 if temp_filename == '-':
3106 reason = ('using a downloader other than ffmpeg' if FFmpegFD.can_merge_formats(info_dict, self.params)
3107 else 'but the formats are incompatible for simultaneous download' if merger.available
3108 else 'but ffmpeg is not installed')
3109 self.report_warning(
3110 f'You have requested downloading multiple formats to stdout {reason}. '
3111 'The formats will be streamed one after the other')
3112 fname = temp_filename
3113 for f in requested_formats:
3114 new_info = dict(info_dict)
3115 del new_info['requested_formats']
3116 new_info.update(f)
3117 if temp_filename != '-':
3118 fname = prepend_extension(
3119 correct_ext(temp_filename, new_info['ext']),
3120 'f%s' % f['format_id'], new_info['ext'])
3121 if not self._ensure_dir_exists(fname):
3122 return
3123 f['filepath'] = fname
3124 downloaded.append(fname)
3125 partial_success, real_download = self.dl(fname, new_info)
3126 info_dict['__real_download'] = info_dict['__real_download'] or real_download
3127 success = success and partial_success
3128
3129 if downloaded and merger.available and not self.params.get('allow_unplayable_formats'):
3130 info_dict['__postprocessors'].append(merger)
3131 info_dict['__files_to_merge'] = downloaded
3132 # Even if there were no downloads, it is being merged only now
3133 info_dict['__real_download'] = True
3134 else:
3135 for file in downloaded:
3136 files_to_move[file] = None
3137 else:
3138 # Just a single file
3139 dl_filename = existing_video_file(full_filename, temp_filename)
3140 if dl_filename is None or dl_filename == temp_filename:
3141 # dl_filename == temp_filename could mean that the file was partially downloaded with --no-part.
3142 # So we should try to resume the download
3143 success, real_download = self.dl(temp_filename, info_dict)
3144 info_dict['__real_download'] = real_download
3145 else:
3146 self.report_file_already_downloaded(dl_filename)
3147
3148 dl_filename = dl_filename or temp_filename
3149 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
3150
3151 except network_exceptions as err:
3152 self.report_error('unable to download video data: %s' % error_to_compat_str(err))
3153 return
3154 except OSError as err:
3155 raise UnavailableVideoError(err)
3156 except (ContentTooShortError, ) as err:
3157 self.report_error(f'content too short (expected {err.expected} bytes and served {err.downloaded})')
3158 return
3159
3160 self._raise_pending_errors(info_dict)
3161 if success and full_filename != '-':
3162
3163 def fixup():
3164 do_fixup = True
3165 fixup_policy = self.params.get('fixup')
3166 vid = info_dict['id']
3167
3168 if fixup_policy in ('ignore', 'never'):
3169 return
3170 elif fixup_policy == 'warn':
3171 do_fixup = 'warn'
3172 elif fixup_policy != 'force':
3173 assert fixup_policy in ('detect_or_warn', None)
3174 if not info_dict.get('__real_download'):
3175 do_fixup = False
3176
3177 def ffmpeg_fixup(cndn, msg, cls):
3178 if not (do_fixup and cndn):
3179 return
3180 elif do_fixup == 'warn':
3181 self.report_warning(f'{vid}: {msg}')
3182 return
3183 pp = cls(self)
3184 if pp.available:
3185 info_dict['__postprocessors'].append(pp)
3186 else:
3187 self.report_warning(f'{vid}: {msg}. Install ffmpeg to fix this automatically')
3188
3189 stretched_ratio = info_dict.get('stretched_ratio')
3190 ffmpeg_fixup(
3191 stretched_ratio not in (1, None),
3192 f'Non-uniform pixel ratio {stretched_ratio}',
3193 FFmpegFixupStretchedPP)
3194
3195 ffmpeg_fixup(
3196 (info_dict.get('requested_formats') is None
3197 and info_dict.get('container') == 'm4a_dash'
3198 and info_dict.get('ext') == 'm4a'),
3199 'writing DASH m4a. Only some players support this container',
3200 FFmpegFixupM4aPP)
3201
3202 downloader = get_suitable_downloader(info_dict, self.params) if 'protocol' in info_dict else None
3203 downloader = downloader.FD_NAME if downloader else None
3204
3205 if info_dict.get('requested_formats') is None: # Not necessary if doing merger
3206 ffmpeg_fixup(downloader == 'hlsnative' and not self.params.get('hls_use_mpegts')
3207 or info_dict.get('is_live') and self.params.get('hls_use_mpegts') is None,
3208 'Possible MPEG-TS in MP4 container or malformed AAC timestamps',
3209 FFmpegFixupM3u8PP)
3210 ffmpeg_fixup(info_dict.get('is_live') and downloader == 'DashSegmentsFD',
3211 'Possible duplicate MOOV atoms', FFmpegFixupDuplicateMoovPP)
3212
3213 ffmpeg_fixup(downloader == 'web_socket_fragment', 'Malformed timestamps detected', FFmpegFixupTimestampPP)
3214 ffmpeg_fixup(downloader == 'web_socket_fragment', 'Malformed duration detected', FFmpegFixupDurationPP)
3215
3216 fixup()
3217 try:
3218 replace_info_dict(self.post_process(dl_filename, info_dict, files_to_move))
3219 except PostProcessingError as err:
3220 self.report_error('Postprocessing: %s' % str(err))
3221 return
3222 try:
3223 for ph in self._post_hooks:
3224 ph(info_dict['filepath'])
3225 except Exception as err:
3226 self.report_error('post hooks: %s' % str(err))
3227 return
3228 info_dict['__write_download_archive'] = True
3229
3230 assert info_dict is original_infodict # Make sure the info_dict was modified in-place
3231 if self.params.get('force_write_download_archive'):
3232 info_dict['__write_download_archive'] = True
3233 check_max_downloads()
3234
3235 def __download_wrapper(self, func):
3236 @functools.wraps(func)
3237 def wrapper(*args, **kwargs):
3238 try:
3239 res = func(*args, **kwargs)
3240 except UnavailableVideoError as e:
3241 self.report_error(e)
3242 except DownloadCancelled as e:
3243 self.to_screen(f'[info] {e}')
3244 if not self.params.get('break_per_url'):
3245 raise
3246 else:
3247 if self.params.get('dump_single_json', False):
3248 self.post_extract(res)
3249 self.to_stdout(json.dumps(self.sanitize_info(res)))
3250 return wrapper
3251
3252 def download(self, url_list):
3253 """Download a given list of URLs."""
3254 url_list = variadic(url_list) # Passing a single URL is a common mistake
3255 outtmpl = self.outtmpl_dict['default']
3256 if (len(url_list) > 1
3257 and outtmpl != '-'
3258 and '%' not in outtmpl
3259 and self.params.get('max_downloads') != 1):
3260 raise SameFileError(outtmpl)
3261
3262 for url in url_list:
3263 self.__download_wrapper(self.extract_info)(
3264 url, force_generic_extractor=self.params.get('force_generic_extractor', False))
3265
3266 return self._download_retcode
3267
3268 def download_with_info_file(self, info_filename):
3269 with contextlib.closing(fileinput.FileInput(
3270 [info_filename], mode='r',
3271 openhook=fileinput.hook_encoded('utf-8'))) as f:
3272 # FileInput doesn't have a read method, we can't call json.load
3273 info = self.sanitize_info(json.loads('\n'.join(f)), self.params.get('clean_infojson', True))
3274 try:
3275 self.__download_wrapper(self.process_ie_result)(info, download=True)
3276 except (DownloadError, EntryNotInPlaylist, ReExtractInfo) as e:
3277 if not isinstance(e, EntryNotInPlaylist):
3278 self.to_stderr('\r')
3279 webpage_url = info.get('webpage_url')
3280 if webpage_url is not None:
3281 self.report_warning(f'The info failed to download: {e}; trying with URL {webpage_url}')
3282 return self.download([webpage_url])
3283 else:
3284 raise
3285 return self._download_retcode
3286
3287 @staticmethod
3288 def sanitize_info(info_dict, remove_private_keys=False):
3289 ''' Sanitize the infodict for converting to json '''
3290 if info_dict is None:
3291 return info_dict
3292 info_dict.setdefault('epoch', int(time.time()))
3293 info_dict.setdefault('_type', 'video')
3294
3295 if remove_private_keys:
3296 reject = lambda k, v: v is None or k.startswith('__') or k in {
3297 'requested_downloads', 'requested_formats', 'requested_subtitles', 'requested_entries',
3298 'entries', 'filepath', '_filename', 'infojson_filename', 'original_url', 'playlist_autonumber',
3299 }
3300 else:
3301 reject = lambda k, v: False
3302
3303 def filter_fn(obj):
3304 if isinstance(obj, dict):
3305 return {k: filter_fn(v) for k, v in obj.items() if not reject(k, v)}
3306 elif isinstance(obj, (list, tuple, set, LazyList)):
3307 return list(map(filter_fn, obj))
3308 elif obj is None or isinstance(obj, (str, int, float, bool)):
3309 return obj
3310 else:
3311 return repr(obj)
3312
3313 return filter_fn(info_dict)
3314
3315 @staticmethod
3316 def filter_requested_info(info_dict, actually_filter=True):
3317 ''' Alias of sanitize_info for backward compatibility '''
3318 return YoutubeDL.sanitize_info(info_dict, actually_filter)
3319
3320 def _delete_downloaded_files(self, *files_to_delete, info={}, msg=None):
3321 for filename in set(filter(None, files_to_delete)):
3322 if msg:
3323 self.to_screen(msg % filename)
3324 try:
3325 os.remove(filename)
3326 except OSError:
3327 self.report_warning(f'Unable to delete file {filename}')
3328 if filename in info.get('__files_to_move', []): # NB: Delete even if None
3329 del info['__files_to_move'][filename]
3330
3331 @staticmethod
3332 def post_extract(info_dict):
3333 def actual_post_extract(info_dict):
3334 if info_dict.get('_type') in ('playlist', 'multi_video'):
3335 for video_dict in info_dict.get('entries', {}):
3336 actual_post_extract(video_dict or {})
3337 return
3338
3339 post_extractor = info_dict.pop('__post_extractor', None) or (lambda: {})
3340 info_dict.update(post_extractor())
3341
3342 actual_post_extract(info_dict or {})
3343
3344 def run_pp(self, pp, infodict):
3345 files_to_delete = []
3346 if '__files_to_move' not in infodict:
3347 infodict['__files_to_move'] = {}
3348 try:
3349 files_to_delete, infodict = pp.run(infodict)
3350 except PostProcessingError as e:
3351 # Must be True and not 'only_download'
3352 if self.params.get('ignoreerrors') is True:
3353 self.report_error(e)
3354 return infodict
3355 raise
3356
3357 if not files_to_delete:
3358 return infodict
3359 if self.params.get('keepvideo', False):
3360 for f in files_to_delete:
3361 infodict['__files_to_move'].setdefault(f, '')
3362 else:
3363 self._delete_downloaded_files(
3364 *files_to_delete, info=infodict, msg='Deleting original file %s (pass -k to keep)')
3365 return infodict
3366
3367 def run_all_pps(self, key, info, *, additional_pps=None):
3368 self._forceprint(key, info)
3369 for pp in (additional_pps or []) + self._pps[key]:
3370 info = self.run_pp(pp, info)
3371 return info
3372
3373 def pre_process(self, ie_info, key='pre_process', files_to_move=None):
3374 info = dict(ie_info)
3375 info['__files_to_move'] = files_to_move or {}
3376 try:
3377 info = self.run_all_pps(key, info)
3378 except PostProcessingError as err:
3379 msg = f'Preprocessing: {err}'
3380 info.setdefault('__pending_error', msg)
3381 self.report_error(msg, is_error=False)
3382 return info, info.pop('__files_to_move', None)
3383
3384 def post_process(self, filename, info, files_to_move=None):
3385 """Run all the postprocessors on the given file."""
3386 info['filepath'] = filename
3387 info['__files_to_move'] = files_to_move or {}
3388 info = self.run_all_pps('post_process', info, additional_pps=info.get('__postprocessors'))
3389 info = self.run_pp(MoveFilesAfterDownloadPP(self), info)
3390 del info['__files_to_move']
3391 return self.run_all_pps('after_move', info)
3392
3393 def _make_archive_id(self, info_dict):
3394 video_id = info_dict.get('id')
3395 if not video_id:
3396 return
3397 # Future-proof against any change in case
3398 # and backwards compatibility with prior versions
3399 extractor = info_dict.get('extractor_key') or info_dict.get('ie_key') # key in a playlist
3400 if extractor is None:
3401 url = str_or_none(info_dict.get('url'))
3402 if not url:
3403 return
3404 # Try to find matching extractor for the URL and take its ie_key
3405 for ie_key, ie in self._ies.items():
3406 if ie.suitable(url):
3407 extractor = ie_key
3408 break
3409 else:
3410 return
3411 return f'{extractor.lower()} {video_id}'
3412
3413 def in_download_archive(self, info_dict):
3414 fn = self.params.get('download_archive')
3415 if fn is None:
3416 return False
3417
3418 vid_id = self._make_archive_id(info_dict)
3419 if not vid_id:
3420 return False # Incomplete video information
3421
3422 return vid_id in self.archive
3423
3424 def record_download_archive(self, info_dict):
3425 fn = self.params.get('download_archive')
3426 if fn is None:
3427 return
3428 vid_id = self._make_archive_id(info_dict)
3429 assert vid_id
3430 self.write_debug(f'Adding to archive: {vid_id}')
3431 with locked_file(fn, 'a', encoding='utf-8') as archive_file:
3432 archive_file.write(vid_id + '\n')
3433 self.archive.add(vid_id)
3434
3435 @staticmethod
3436 def format_resolution(format, default='unknown'):
3437 if format.get('vcodec') == 'none' and format.get('acodec') != 'none':
3438 return 'audio only'
3439 if format.get('resolution') is not None:
3440 return format['resolution']
3441 if format.get('width') and format.get('height'):
3442 return '%dx%d' % (format['width'], format['height'])
3443 elif format.get('height'):
3444 return '%sp' % format['height']
3445 elif format.get('width'):
3446 return '%dx?' % format['width']
3447 return default
3448
3449 def _list_format_headers(self, *headers):
3450 if self.params.get('listformats_table', True) is not False:
3451 return [self._format_out(header, self.Styles.HEADERS) for header in headers]
3452 return headers
3453
3454 def _format_note(self, fdict):
3455 res = ''
3456 if fdict.get('ext') in ['f4f', 'f4m']:
3457 res += '(unsupported)'
3458 if fdict.get('language'):
3459 if res:
3460 res += ' '
3461 res += '[%s]' % fdict['language']
3462 if fdict.get('format_note') is not None:
3463 if res:
3464 res += ' '
3465 res += fdict['format_note']
3466 if fdict.get('tbr') is not None:
3467 if res:
3468 res += ', '
3469 res += '%4dk' % fdict['tbr']
3470 if fdict.get('container') is not None:
3471 if res:
3472 res += ', '
3473 res += '%s container' % fdict['container']
3474 if (fdict.get('vcodec') is not None
3475 and fdict.get('vcodec') != 'none'):
3476 if res:
3477 res += ', '
3478 res += fdict['vcodec']
3479 if fdict.get('vbr') is not None:
3480 res += '@'
3481 elif fdict.get('vbr') is not None and fdict.get('abr') is not None:
3482 res += 'video@'
3483 if fdict.get('vbr') is not None:
3484 res += '%4dk' % fdict['vbr']
3485 if fdict.get('fps') is not None:
3486 if res:
3487 res += ', '
3488 res += '%sfps' % fdict['fps']
3489 if fdict.get('acodec') is not None:
3490 if res:
3491 res += ', '
3492 if fdict['acodec'] == 'none':
3493 res += 'video only'
3494 else:
3495 res += '%-5s' % fdict['acodec']
3496 elif fdict.get('abr') is not None:
3497 if res:
3498 res += ', '
3499 res += 'audio'
3500 if fdict.get('abr') is not None:
3501 res += '@%3dk' % fdict['abr']
3502 if fdict.get('asr') is not None:
3503 res += ' (%5dHz)' % fdict['asr']
3504 if fdict.get('filesize') is not None:
3505 if res:
3506 res += ', '
3507 res += format_bytes(fdict['filesize'])
3508 elif fdict.get('filesize_approx') is not None:
3509 if res:
3510 res += ', '
3511 res += '~' + format_bytes(fdict['filesize_approx'])
3512 return res
3513
3514 def render_formats_table(self, info_dict):
3515 if not info_dict.get('formats') and not info_dict.get('url'):
3516 return None
3517
3518 formats = info_dict.get('formats', [info_dict])
3519 if not self.params.get('listformats_table', True) is not False:
3520 table = [
3521 [
3522 format_field(f, 'format_id'),
3523 format_field(f, 'ext'),
3524 self.format_resolution(f),
3525 self._format_note(f)
3526 ] for f in formats if f.get('preference') is None or f['preference'] >= -1000]
3527 return render_table(['format code', 'extension', 'resolution', 'note'], table, extra_gap=1)
3528
3529 delim = self._format_out('\u2502', self.Styles.DELIM, '|', test_encoding=True)
3530 table = [
3531 [
3532 self._format_out(format_field(f, 'format_id'), self.Styles.ID),
3533 format_field(f, 'ext'),
3534 format_field(f, func=self.format_resolution, ignore=('audio only', 'images')),
3535 format_field(f, 'fps', '\t%d'),
3536 format_field(f, 'dynamic_range', '%s', ignore=(None, 'SDR')).replace('HDR', ''),
3537 delim,
3538 format_field(f, 'filesize', ' \t%s', func=format_bytes) + format_field(f, 'filesize_approx', '~\t%s', func=format_bytes),
3539 format_field(f, 'tbr', '\t%dk'),
3540 shorten_protocol_name(f.get('protocol', '')),
3541 delim,
3542 format_field(f, 'vcodec', default='unknown').replace(
3543 'none', 'images' if f.get('acodec') == 'none'
3544 else self._format_out('audio only', self.Styles.SUPPRESS)),
3545 format_field(f, 'vbr', '\t%dk'),
3546 format_field(f, 'acodec', default='unknown').replace(
3547 'none', '' if f.get('vcodec') == 'none'
3548 else self._format_out('video only', self.Styles.SUPPRESS)),
3549 format_field(f, 'abr', '\t%dk'),
3550 format_field(f, 'asr', '\t%dHz'),
3551 join_nonempty(
3552 self._format_out('UNSUPPORTED', 'light red') if f.get('ext') in ('f4f', 'f4m') else None,
3553 format_field(f, 'language', '[%s]'),
3554 join_nonempty(format_field(f, 'format_note'),
3555 format_field(f, 'container', ignore=(None, f.get('ext'))),
3556 delim=', '),
3557 delim=' '),
3558 ] for f in formats if f.get('preference') is None or f['preference'] >= -1000]
3559 header_line = self._list_format_headers(
3560 'ID', 'EXT', 'RESOLUTION', '\tFPS', 'HDR', delim, '\tFILESIZE', '\tTBR', 'PROTO',
3561 delim, 'VCODEC', '\tVBR', 'ACODEC', '\tABR', '\tASR', 'MORE INFO')
3562
3563 return render_table(
3564 header_line, table, hide_empty=True,
3565 delim=self._format_out('\u2500', self.Styles.DELIM, '-', test_encoding=True))
3566
3567 def render_thumbnails_table(self, info_dict):
3568 thumbnails = list(info_dict.get('thumbnails') or [])
3569 if not thumbnails:
3570 return None
3571 return render_table(
3572 self._list_format_headers('ID', 'Width', 'Height', 'URL'),
3573 [[t.get('id'), t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails])
3574
3575 def render_subtitles_table(self, video_id, subtitles):
3576 def _row(lang, formats):
3577 exts, names = zip(*((f['ext'], f.get('name') or 'unknown') for f in reversed(formats)))
3578 if len(set(names)) == 1:
3579 names = [] if names[0] == 'unknown' else names[:1]
3580 return [lang, ', '.join(names), ', '.join(exts)]
3581
3582 if not subtitles:
3583 return None
3584 return render_table(
3585 self._list_format_headers('Language', 'Name', 'Formats'),
3586 [_row(lang, formats) for lang, formats in subtitles.items()],
3587 hide_empty=True)
3588
3589 def __list_table(self, video_id, name, func, *args):
3590 table = func(*args)
3591 if not table:
3592 self.to_screen(f'{video_id} has no {name}')
3593 return
3594 self.to_screen(f'[info] Available {name} for {video_id}:')
3595 self.to_stdout(table)
3596
3597 def list_formats(self, info_dict):
3598 self.__list_table(info_dict['id'], 'formats', self.render_formats_table, info_dict)
3599
3600 def list_thumbnails(self, info_dict):
3601 self.__list_table(info_dict['id'], 'thumbnails', self.render_thumbnails_table, info_dict)
3602
3603 def list_subtitles(self, video_id, subtitles, name='subtitles'):
3604 self.__list_table(video_id, name, self.render_subtitles_table, video_id, subtitles)
3605
3606 def urlopen(self, req):
3607 """ Start an HTTP download """
3608 if isinstance(req, str):
3609 req = sanitized_Request(req)
3610 return self._opener.open(req, timeout=self._socket_timeout)
3611
3612 def print_debug_header(self):
3613 if not self.params.get('verbose'):
3614 return
3615
3616 def get_encoding(stream):
3617 ret = str(getattr(stream, 'encoding', 'missing (%s)' % type(stream).__name__))
3618 if not supports_terminal_sequences(stream):
3619 from .utils import WINDOWS_VT_MODE # Must be imported locally
3620 ret += ' (No VT)' if WINDOWS_VT_MODE is False else ' (No ANSI)'
3621 return ret
3622
3623 encoding_str = 'Encodings: locale %s, fs %s, pref %s, %s' % (
3624 locale.getpreferredencoding(),
3625 sys.getfilesystemencoding(),
3626 self.get_encoding(),
3627 ', '.join(
3628 f'{key} {get_encoding(stream)}' for key, stream in self._out_files
3629 if stream is not None and key != 'console')
3630 )
3631
3632 logger = self.params.get('logger')
3633 if logger:
3634 write_debug = lambda msg: logger.debug(f'[debug] {msg}')
3635 write_debug(encoding_str)
3636 else:
3637 write_string(f'[debug] {encoding_str}\n', encoding=None)
3638 write_debug = lambda msg: self._write_string(f'[debug] {msg}\n')
3639
3640 source = detect_variant()
3641 write_debug(join_nonempty(
3642 'yt-dlp version', __version__,
3643 f'[{RELEASE_GIT_HEAD}]' if RELEASE_GIT_HEAD else '',
3644 '' if source == 'unknown' else f'({source})',
3645 delim=' '))
3646 if not _LAZY_LOADER:
3647 if os.environ.get('YTDLP_NO_LAZY_EXTRACTORS'):
3648 write_debug('Lazy loading extractors is forcibly disabled')
3649 else:
3650 write_debug('Lazy loading extractors is disabled')
3651 if plugin_extractors or plugin_postprocessors:
3652 write_debug('Plugins: %s' % [
3653 '%s%s' % (klass.__name__, '' if klass.__name__ == name else f' as {name}')
3654 for name, klass in itertools.chain(plugin_extractors.items(), plugin_postprocessors.items())])
3655 if self.params.get('compat_opts'):
3656 write_debug('Compatibility options: %s' % ', '.join(self.params.get('compat_opts')))
3657
3658 if source == 'source':
3659 try:
3660 sp = Popen(
3661 ['git', 'rev-parse', '--short', 'HEAD'],
3662 stdout=subprocess.PIPE, stderr=subprocess.PIPE,
3663 cwd=os.path.dirname(os.path.abspath(__file__)))
3664 out, err = sp.communicate_or_kill()
3665 out = out.decode().strip()
3666 if re.match('[0-9a-f]+', out):
3667 write_debug('Git HEAD: %s' % out)
3668 except Exception:
3669 with contextlib.suppress(Exception):
3670 sys.exc_clear()
3671
3672 def python_implementation():
3673 impl_name = platform.python_implementation()
3674 if impl_name == 'PyPy' and hasattr(sys, 'pypy_version_info'):
3675 return impl_name + ' version %d.%d.%d' % sys.pypy_version_info[:3]
3676 return impl_name
3677
3678 write_debug('Python version %s (%s %s) - %s' % (
3679 platform.python_version(),
3680 python_implementation(),
3681 platform.architecture()[0],
3682 platform_name()))
3683
3684 exe_versions, ffmpeg_features = FFmpegPostProcessor.get_versions_and_features(self)
3685 ffmpeg_features = {key for key, val in ffmpeg_features.items() if val}
3686 if ffmpeg_features:
3687 exe_versions['ffmpeg'] += ' (%s)' % ','.join(sorted(ffmpeg_features))
3688
3689 exe_versions['rtmpdump'] = rtmpdump_version()
3690 exe_versions['phantomjs'] = PhantomJSwrapper._version()
3691 exe_str = ', '.join(
3692 f'{exe} {v}' for exe, v in sorted(exe_versions.items()) if v
3693 ) or 'none'
3694 write_debug('exe versions: %s' % exe_str)
3695
3696 from .compat.compat_utils import get_package_info
3697 from .dependencies import available_dependencies
3698
3699 write_debug('Optional libraries: %s' % (', '.join(sorted({
3700 join_nonempty(*get_package_info(m)) for m in available_dependencies.values()
3701 })) or 'none'))
3702
3703 self._setup_opener()
3704 proxy_map = {}
3705 for handler in self._opener.handlers:
3706 if hasattr(handler, 'proxies'):
3707 proxy_map.update(handler.proxies)
3708 write_debug(f'Proxy map: {proxy_map}')
3709
3710 # Not implemented
3711 if False and self.params.get('call_home'):
3712 ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode()
3713 write_debug('Public IP address: %s' % ipaddr)
3714 latest_version = self.urlopen(
3715 'https://yt-dl.org/latest/version').read().decode()
3716 if version_tuple(latest_version) > version_tuple(__version__):
3717 self.report_warning(
3718 'You are using an outdated version (newest version: %s)! '
3719 'See https://yt-dl.org/update if you need help updating.' %
3720 latest_version)
3721
3722 def _setup_opener(self):
3723 if hasattr(self, '_opener'):
3724 return
3725 timeout_val = self.params.get('socket_timeout')
3726 self._socket_timeout = 20 if timeout_val is None else float(timeout_val)
3727
3728 opts_cookiesfrombrowser = self.params.get('cookiesfrombrowser')
3729 opts_cookiefile = self.params.get('cookiefile')
3730 opts_proxy = self.params.get('proxy')
3731
3732 self.cookiejar = load_cookies(opts_cookiefile, opts_cookiesfrombrowser, self)
3733
3734 cookie_processor = YoutubeDLCookieProcessor(self.cookiejar)
3735 if opts_proxy is not None:
3736 if opts_proxy == '':
3737 proxies = {}
3738 else:
3739 proxies = {'http': opts_proxy, 'https': opts_proxy}
3740 else:
3741 proxies = compat_urllib_request.getproxies()
3742 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
3743 if 'http' in proxies and 'https' not in proxies:
3744 proxies['https'] = proxies['http']
3745 proxy_handler = PerRequestProxyHandler(proxies)
3746
3747 debuglevel = 1 if self.params.get('debug_printtraffic') else 0
3748 https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel)
3749 ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel)
3750 redirect_handler = YoutubeDLRedirectHandler()
3751 data_handler = urllib.request.DataHandler()
3752
3753 # When passing our own FileHandler instance, build_opener won't add the
3754 # default FileHandler and allows us to disable the file protocol, which
3755 # can be used for malicious purposes (see
3756 # https://github.com/ytdl-org/youtube-dl/issues/8227)
3757 file_handler = compat_urllib_request.FileHandler()
3758
3759 def file_open(*args, **kwargs):
3760 raise compat_urllib_error.URLError('file:// scheme is explicitly disabled in yt-dlp for security reasons')
3761 file_handler.file_open = file_open
3762
3763 opener = compat_urllib_request.build_opener(
3764 proxy_handler, https_handler, cookie_processor, ydlh, redirect_handler, data_handler, file_handler)
3765
3766 # Delete the default user-agent header, which would otherwise apply in
3767 # cases where our custom HTTP handler doesn't come into play
3768 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
3769 opener.addheaders = []
3770 self._opener = opener
3771
3772 def encode(self, s):
3773 if isinstance(s, bytes):
3774 return s # Already encoded
3775
3776 try:
3777 return s.encode(self.get_encoding())
3778 except UnicodeEncodeError as err:
3779 err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.'
3780 raise
3781
3782 def get_encoding(self):
3783 encoding = self.params.get('encoding')
3784 if encoding is None:
3785 encoding = preferredencoding()
3786 return encoding
3787
3788 def _write_info_json(self, label, ie_result, infofn, overwrite=None):
3789 ''' Write infojson and returns True = written, 'exists' = Already exists, False = skip, None = error '''
3790 if overwrite is None:
3791 overwrite = self.params.get('overwrites', True)
3792 if not self.params.get('writeinfojson'):
3793 return False
3794 elif not infofn:
3795 self.write_debug(f'Skipping writing {label} infojson')
3796 return False
3797 elif not self._ensure_dir_exists(infofn):
3798 return None
3799 elif not overwrite and os.path.exists(infofn):
3800 self.to_screen(f'[info] {label.title()} metadata is already present')
3801 return 'exists'
3802
3803 self.to_screen(f'[info] Writing {label} metadata as JSON to: {infofn}')
3804 try:
3805 write_json_file(self.sanitize_info(ie_result, self.params.get('clean_infojson', True)), infofn)
3806 return True
3807 except OSError:
3808 self.report_error(f'Cannot write {label} metadata to JSON file {infofn}')
3809 return None
3810
3811 def _write_description(self, label, ie_result, descfn):
3812 ''' Write description and returns True = written, False = skip, None = error '''
3813 if not self.params.get('writedescription'):
3814 return False
3815 elif not descfn:
3816 self.write_debug(f'Skipping writing {label} description')
3817 return False
3818 elif not self._ensure_dir_exists(descfn):
3819 return None
3820 elif not self.params.get('overwrites', True) and os.path.exists(descfn):
3821 self.to_screen(f'[info] {label.title()} description is already present')
3822 elif ie_result.get('description') is None:
3823 self.report_warning(f'There\'s no {label} description to write')
3824 return False
3825 else:
3826 try:
3827 self.to_screen(f'[info] Writing {label} description to: {descfn}')
3828 with open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
3829 descfile.write(ie_result['description'])
3830 except OSError:
3831 self.report_error(f'Cannot write {label} description file {descfn}')
3832 return None
3833 return True
3834
3835 def _write_subtitles(self, info_dict, filename):
3836 ''' Write subtitles to file and return list of (sub_filename, final_sub_filename); or None if error'''
3837 ret = []
3838 subtitles = info_dict.get('requested_subtitles')
3839 if not subtitles or not (self.params.get('writesubtitles') or self.params.get('writeautomaticsub')):
3840 # subtitles download errors are already managed as troubles in relevant IE
3841 # that way it will silently go on when used with unsupporting IE
3842 return ret
3843
3844 sub_filename_base = self.prepare_filename(info_dict, 'subtitle')
3845 if not sub_filename_base:
3846 self.to_screen('[info] Skipping writing video subtitles')
3847 return ret
3848 for sub_lang, sub_info in subtitles.items():
3849 sub_format = sub_info['ext']
3850 sub_filename = subtitles_filename(filename, sub_lang, sub_format, info_dict.get('ext'))
3851 sub_filename_final = subtitles_filename(sub_filename_base, sub_lang, sub_format, info_dict.get('ext'))
3852 existing_sub = self.existing_file((sub_filename_final, sub_filename))
3853 if existing_sub:
3854 self.to_screen(f'[info] Video subtitle {sub_lang}.{sub_format} is already present')
3855 sub_info['filepath'] = existing_sub
3856 ret.append((existing_sub, sub_filename_final))
3857 continue
3858
3859 self.to_screen(f'[info] Writing video subtitles to: {sub_filename}')
3860 if sub_info.get('data') is not None:
3861 try:
3862 # Use newline='' to prevent conversion of newline characters
3863 # See https://github.com/ytdl-org/youtube-dl/issues/10268
3864 with open(sub_filename, 'w', encoding='utf-8', newline='') as subfile:
3865 subfile.write(sub_info['data'])
3866 sub_info['filepath'] = sub_filename
3867 ret.append((sub_filename, sub_filename_final))
3868 continue
3869 except OSError:
3870 self.report_error(f'Cannot write video subtitles file {sub_filename}')
3871 return None
3872
3873 try:
3874 sub_copy = sub_info.copy()
3875 sub_copy.setdefault('http_headers', info_dict.get('http_headers'))
3876 self.dl(sub_filename, sub_copy, subtitle=True)
3877 sub_info['filepath'] = sub_filename
3878 ret.append((sub_filename, sub_filename_final))
3879 except (DownloadError, ExtractorError, IOError, OSError, ValueError) + network_exceptions as err:
3880 msg = f'Unable to download video subtitles for {sub_lang!r}: {err}'
3881 if self.params.get('ignoreerrors') is not True: # False or 'only_download'
3882 if not self.params.get('ignoreerrors'):
3883 self.report_error(msg)
3884 raise DownloadError(msg)
3885 self.report_warning(msg)
3886 return ret
3887
3888 def _write_thumbnails(self, label, info_dict, filename, thumb_filename_base=None):
3889 ''' Write thumbnails to file and return list of (thumb_filename, final_thumb_filename) '''
3890 write_all = self.params.get('write_all_thumbnails', False)
3891 thumbnails, ret = [], []
3892 if write_all or self.params.get('writethumbnail', False):
3893 thumbnails = info_dict.get('thumbnails') or []
3894 multiple = write_all and len(thumbnails) > 1
3895
3896 if thumb_filename_base is None:
3897 thumb_filename_base = filename
3898 if thumbnails and not thumb_filename_base:
3899 self.write_debug(f'Skipping writing {label} thumbnail')
3900 return ret
3901
3902 for idx, t in list(enumerate(thumbnails))[::-1]:
3903 thumb_ext = (f'{t["id"]}.' if multiple else '') + determine_ext(t['url'], 'jpg')
3904 thumb_display_id = f'{label} thumbnail {t["id"]}'
3905 thumb_filename = replace_extension(filename, thumb_ext, info_dict.get('ext'))
3906 thumb_filename_final = replace_extension(thumb_filename_base, thumb_ext, info_dict.get('ext'))
3907
3908 existing_thumb = self.existing_file((thumb_filename_final, thumb_filename))
3909 if existing_thumb:
3910 self.to_screen('[info] %s is already present' % (
3911 thumb_display_id if multiple else f'{label} thumbnail').capitalize())
3912 t['filepath'] = existing_thumb
3913 ret.append((existing_thumb, thumb_filename_final))
3914 else:
3915 self.to_screen(f'[info] Downloading {thumb_display_id} ...')
3916 try:
3917 uf = self.urlopen(sanitized_Request(t['url'], headers=t.get('http_headers', {})))
3918 self.to_screen(f'[info] Writing {thumb_display_id} to: {thumb_filename}')
3919 with open(encodeFilename(thumb_filename), 'wb') as thumbf:
3920 shutil.copyfileobj(uf, thumbf)
3921 ret.append((thumb_filename, thumb_filename_final))
3922 t['filepath'] = thumb_filename
3923 except network_exceptions as err:
3924 thumbnails.pop(idx)
3925 self.report_warning(f'Unable to download {thumb_display_id}: {err}')
3926 if ret and not write_all:
3927 break
3928 return ret