]> jfr.im git - yt-dlp.git/blame - yt_dlp/YoutubeDL.py
[cleanup] Misc
[yt-dlp.git] / yt_dlp / YoutubeDL.py
CommitLineData
26e63931 1import collections
31bd3925 2import contextlib
9d2ecdbc 3import datetime
c1c9a79c 4import errno
31bd3925 5import fileinput
b5ae35ee 6import functools
8222d8de 7import io
b82f815f 8import itertools
8694c600 9import json
62fec3b2 10import locale
083c9df9 11import operator
8222d8de 12import os
f8271158 13import random
8222d8de
JMF
14import re
15import shutil
6f2287cb 16import string
dca08720 17import subprocess
8222d8de 18import sys
21cd8fae 19import tempfile
8222d8de 20import time
67134eab 21import tokenize
8222d8de 22import traceback
524e2e4f 23import unicodedata
961ea474 24
f8271158 25from .cache import Cache
3f66b6fe 26from .compat import urllib # isort: split
14f25df2 27from .compat import compat_os_name, compat_shlex_quote
982ee69a 28from .cookies import load_cookies
f8271158 29from .downloader import FFmpegFD, get_suitable_downloader, shorten_protocol_name
30from .downloader.rtmp import rtmpdump_version
f8271158 31from .extractor import gen_extractor_classes, get_info_extractor
fe7866d0 32from .extractor.common import UnsupportedURLIE
f8271158 33from .extractor.openload import PhantomJSwrapper
34from .minicurses import format_text
8e40b9d1 35from .plugins import directories as plugin_directories
e756f45b 36from .postprocessor import _PLUGIN_CLASSES as plugin_pps
f8271158 37from .postprocessor import (
38 EmbedThumbnailPP,
39 FFmpegFixupDuplicateMoovPP,
40 FFmpegFixupDurationPP,
41 FFmpegFixupM3u8PP,
42 FFmpegFixupM4aPP,
43 FFmpegFixupStretchedPP,
44 FFmpegFixupTimestampPP,
45 FFmpegMergerPP,
46 FFmpegPostProcessor,
ca9def71 47 FFmpegVideoConvertorPP,
f8271158 48 MoveFilesAfterDownloadPP,
49 get_postprocessor,
50)
ca9def71 51from .postprocessor.ffmpeg import resolve_mapping as resolve_recode_mapping
b5e7a2e6 52from .update import REPOSITORY, current_git_head, detect_variant
8c25f81b 53from .utils import (
f8271158 54 DEFAULT_OUTTMPL,
7b2c3f47 55 IDENTITY,
f8271158 56 LINK_TEMPLATES,
8dc59305 57 MEDIA_EXTENSIONS,
f8271158 58 NO_DEFAULT,
1d485a1a 59 NUMBER_RE,
f8271158 60 OUTTMPL_TYPES,
61 POSTPROCESS_WHEN,
62 STR_FORMAT_RE_TMPL,
63 STR_FORMAT_TYPES,
64 ContentTooShortError,
65 DateRange,
66 DownloadCancelled,
67 DownloadError,
68 EntryNotInPlaylist,
69 ExistingVideoReached,
70 ExtractorError,
784320c9 71 FormatSorter,
f8271158 72 GeoRestrictedError,
73 HEADRequest,
f8271158 74 ISO3166Utils,
75 LazyList,
76 MaxDownloadsReached,
19a03940 77 Namespace,
f8271158 78 PagedList,
79 PerRequestProxyHandler,
7e88d7d7 80 PlaylistEntries,
f8271158 81 Popen,
82 PostProcessingError,
83 ReExtractInfo,
84 RejectedVideoReached,
85 SameFileError,
86 UnavailableVideoError,
693f0600 87 UserNotLive,
f8271158 88 YoutubeDLCookieProcessor,
89 YoutubeDLHandler,
90 YoutubeDLRedirectHandler,
eedb7ba5
S
91 age_restricted,
92 args_to_str,
cb794ee0 93 bug_reports_message,
ce02ed60 94 date_from_str,
da4db748 95 deprecation_warning,
ce02ed60 96 determine_ext,
b5559424 97 determine_protocol,
c0384f22 98 encode_compat_str,
ce02ed60 99 encodeFilename,
a06916d9 100 error_to_compat_str,
47cdc68e 101 escapeHTML,
590bc6f6 102 expand_path,
90137ca4 103 filter_dict,
e29663c6 104 float_or_none,
02dbf93f 105 format_bytes,
e0fd9573 106 format_decimal_suffix,
f8271158 107 format_field,
525ef922 108 formatSeconds,
fc61aff4 109 get_compatible_ext,
0bb322b9 110 get_domain,
c9969434 111 int_or_none,
732044af 112 iri_to_uri,
941e881e 113 is_path_like,
34921b43 114 join_nonempty,
ce02ed60 115 locked_file,
0647d925 116 make_archive_id,
0202b52a 117 make_dir,
dca08720 118 make_HTTPS_handler,
8b7539d2 119 merge_headers,
3158150c 120 network_exceptions,
ec11a9f4 121 number_of_digits,
cd6fc19e 122 orderedSet,
5314b521 123 orderedSet_from_options,
083c9df9 124 parse_filesize,
ce02ed60 125 preferredencoding,
eedb7ba5 126 prepend_extension,
3efb96a6 127 remove_terminal_sequences,
cfb56d1a 128 render_table,
eedb7ba5 129 replace_extension,
ce02ed60 130 sanitize_filename,
1bb5c511 131 sanitize_path,
dcf77cf1 132 sanitize_url,
67dda517 133 sanitized_Request,
e5660ee6 134 std_headers,
1211bb6d 135 str_or_none,
e29663c6 136 strftime_or_none,
ce02ed60 137 subtitles_filename,
819e0531 138 supports_terminal_sequences,
b1f94422 139 system_identifier,
f2ebc5c7 140 timetuple_from_msec,
732044af 141 to_high_limit_path,
324ad820 142 traverse_obj,
fc61aff4 143 try_call,
6033d980 144 try_get,
29eb5174 145 url_basename,
7d1eb38a 146 variadic,
58b1f00d 147 version_tuple,
53973b4d 148 windows_enable_vt_mode,
ce02ed60
PH
149 write_json_file,
150 write_string,
4f026faf 151)
29cb20bd 152from .version import CHANNEL, RELEASE_GIT_HEAD, VARIANT, __version__
8222d8de 153
e9c0cdd3
YCH
154if compat_os_name == 'nt':
155 import ctypes
156
2459b6e1 157
86e5f3ed 158class YoutubeDL:
8222d8de
JMF
159 """YoutubeDL class.
160
161 YoutubeDL objects are the ones responsible of downloading the
162 actual video file and writing it to disk if the user has requested
163 it, among some other tasks. In most cases there should be one per
164 program. As, given a video URL, the downloader doesn't know how to
165 extract all the needed information, task that InfoExtractors do, it
166 has to pass the URL to one of them.
167
168 For this, YoutubeDL objects have a method that allows
169 InfoExtractors to be registered in a given order. When it is passed
170 a URL, the YoutubeDL object handles it to the first InfoExtractor it
171 finds that reports being able to handle it. The InfoExtractor extracts
172 all the information about the video or videos the URL refers to, and
173 YoutubeDL process the extracted information, possibly using a File
174 Downloader to download the video.
175
176 YoutubeDL objects accept a lot of parameters. In order not to saturate
177 the object constructor with arguments, it receives a dictionary of
178 options instead. These options are available through the params
179 attribute for the InfoExtractors to use. The YoutubeDL also
180 registers itself as the downloader in charge for the InfoExtractors
181 that are added to it, so this is a "mutual registration".
182
183 Available options:
184
185 username: Username for authentication purposes.
186 password: Password for authentication purposes.
180940e0 187 videopassword: Password for accessing a video.
1da50aa3
S
188 ap_mso: Adobe Pass multiple-system operator identifier.
189 ap_username: Multiple-system operator account username.
190 ap_password: Multiple-system operator account password.
8222d8de 191 usenetrc: Use netrc for authentication instead.
c8bc203f 192 netrc_location: Location of the netrc file. Defaults to ~/.netrc.
db3ad8a6 193 netrc_cmd: Use a shell command to get credentials
8222d8de
JMF
194 verbose: Print additional info to stdout.
195 quiet: Do not print messages to stdout.
ad8915b7 196 no_warnings: Do not print out anything for warnings.
bb66c247 197 forceprint: A dict with keys WHEN mapped to a list of templates to
198 print to stdout. The allowed keys are video or any of the
199 items in utils.POSTPROCESS_WHEN.
ca30f449 200 For compatibility, a single list is also accepted
bb66c247 201 print_to_file: A dict with keys WHEN (same as forceprint) mapped to
202 a list of tuples with (template, filename)
8694c600 203 forcejson: Force printing info_dict as JSON.
63e0be34
PH
204 dump_single_json: Force printing the info_dict of the whole playlist
205 (or video) as a single JSON line.
c25228e5 206 force_write_download_archive: Force writing download archive regardless
207 of 'skip_download' or 'simulate'.
b7b04c78 208 simulate: Do not download the video files. If unset (or None),
209 simulate only if listsubtitles, listformats or list_thumbnails is used
eb8a4433 210 format: Video format code. see "FORMAT SELECTION" for more details.
093a1710 211 You can also pass a function. The function takes 'ctx' as
212 argument and returns the formats to download.
213 See "build_format_selector" for an implementation
63ad4d43 214 allow_unplayable_formats: Allow unplayable formats to be extracted and downloaded.
b7da73eb 215 ignore_no_formats_error: Ignore "No video formats" error. Usefull for
216 extracting metadata even if the video is not actually
217 available for download (experimental)
0930b11f 218 format_sort: A list of fields by which to sort the video formats.
219 See "Sorting Formats" for more details.
c25228e5 220 format_sort_force: Force the given format_sort. see "Sorting Formats"
221 for more details.
08d30158 222 prefer_free_formats: Whether to prefer video formats with free containers
223 over non-free ones of same quality.
c25228e5 224 allow_multiple_video_streams: Allow multiple video streams to be merged
225 into a single file
226 allow_multiple_audio_streams: Allow multiple audio streams to be merged
227 into a single file
0ba692ac 228 check_formats Whether to test if the formats are downloadable.
9f1a1c36 229 Can be True (check all), False (check none),
230 'selected' (check selected formats),
0ba692ac 231 or None (check only if requested by extractor)
4524baf0 232 paths: Dictionary of output paths. The allowed keys are 'home'
233 'temp' and the keys of OUTTMPL_TYPES (in utils.py)
de6000d9 234 outtmpl: Dictionary of templates for output names. Allowed keys
4524baf0 235 are 'default' and the keys of OUTTMPL_TYPES (in utils.py).
34488702 236 For compatibility with youtube-dl, a single string can also be used
a820dc72
RA
237 outtmpl_na_placeholder: Placeholder for unavailable meta fields.
238 restrictfilenames: Do not allow "&" and spaces in file names
239 trim_file_name: Limit length of filename (extension excluded)
4524baf0 240 windowsfilenames: Force the filenames to be windows compatible
b1940459 241 ignoreerrors: Do not stop on download/postprocessing errors.
242 Can be 'only_download' to ignore only download errors.
243 Default is 'only_download' for CLI, but False for API
26e2805c 244 skip_playlist_after_errors: Number of allowed failures until the rest of
245 the playlist is skipped
fe7866d0 246 allowed_extractors: List of regexes to match against extractor names that are allowed
0c3d0f51 247 overwrites: Overwrite all video and metadata files if True,
248 overwrite only non-video files if None
249 and don't overwrite any file if False
34488702 250 For compatibility with youtube-dl,
251 "nooverwrites" may also be used instead
c14e88f0 252 playlist_items: Specific indices of playlist to download.
75822ca7 253 playlistrandom: Download playlist items in random order.
7e9a6125 254 lazy_playlist: Process playlist entries as they are received.
8222d8de
JMF
255 matchtitle: Download only matching titles.
256 rejecttitle: Reject downloads for matching titles.
8bf9319e 257 logger: Log messages to a logging.Logger instance.
17ffed18 258 logtostderr: Print everything to stderr instead of stdout.
259 consoletitle: Display progress in console window's titlebar.
8222d8de
JMF
260 writedescription: Write the video description to a .description file
261 writeinfojson: Write the video description to a .info.json file
ad54c913 262 clean_infojson: Remove internal metadata from the infojson
34488702 263 getcomments: Extract video comments. This will not be written to disk
06167fbb 264 unless writeinfojson is also given
1fb07d10 265 writeannotations: Write the video annotations to a .annotations.xml file
8222d8de 266 writethumbnail: Write the thumbnail image to a file
c25228e5 267 allow_playlist_files: Whether to write playlists' description, infojson etc
268 also to disk when using the 'write*' options
ec82d85a 269 write_all_thumbnails: Write all thumbnail formats to files
732044af 270 writelink: Write an internet shortcut file, depending on the
271 current platform (.url/.webloc/.desktop)
272 writeurllink: Write a Windows internet shortcut file (.url)
273 writewebloclink: Write a macOS internet shortcut file (.webloc)
274 writedesktoplink: Write a Linux internet shortcut file (.desktop)
8222d8de 275 writesubtitles: Write the video subtitles to a file
741dd8ea 276 writeautomaticsub: Write the automatically generated subtitles to a file
8222d8de 277 listsubtitles: Lists all available subtitles for the video
a504ced0 278 subtitlesformat: The format code for subtitles
c32b0aab 279 subtitleslangs: List of languages of the subtitles to download (can be regex).
280 The list may contain "all" to refer to all the available
281 subtitles. The language can be prefixed with a "-" to
62b58c09 282 exclude it from the requested languages, e.g. ['all', '-live_chat']
8222d8de 283 keepvideo: Keep the video file after post-processing
46f1370e 284 daterange: A utils.DateRange object, download only if the upload_date is in the range.
8222d8de 285 skip_download: Skip the actual download of the video file
c35f9e72 286 cachedir: Location of the cache files in the filesystem.
a0e07d31 287 False to disable filesystem cache.
47192f92 288 noplaylist: Download single video instead of a playlist if in doubt.
8dbe9899
PH
289 age_limit: An integer representing the user's age in years.
290 Unsuitable videos for the given age are skipped.
5fe18bdb
PH
291 min_views: An integer representing the minimum view count the video
292 must have in order to not be skipped.
293 Videos without view count information are always
294 downloaded. None for no limit.
295 max_views: An integer representing the maximum view count.
296 Videos that are more popular than that are not
297 downloaded.
298 Videos without view count information are always
299 downloaded. None for no limit.
ae103564 300 download_archive: A set, or the name of a file where all downloads are recorded.
301 Videos already present in the file are not downloaded again.
8a51f564 302 break_on_existing: Stop the download process after attempting to download a
303 file that is in the archive.
b222c271 304 break_per_url: Whether break_on_reject and break_on_existing
305 should act on each input URL as opposed to for the entire queue
d76fa1f3 306 cookiefile: File name or text stream from where cookies should be read and dumped to
f59f5ef8 307 cookiesfrombrowser: A tuple containing the name of the browser, the profile
9bd13fe5 308 name/path from where cookies are loaded, the name of the keyring,
309 and the container name, e.g. ('chrome', ) or
310 ('vivaldi', 'default', 'BASICTEXT') or ('firefox', 'default', None, 'Meta')
f81c62a6 311 legacyserverconnect: Explicitly allow HTTPS connection to servers that do not
312 support RFC 5746 secure renegotiation
f59f5ef8 313 nocheckcertificate: Do not verify SSL certificates
bb58c9ed 314 client_certificate: Path to client certificate file in PEM format. May include the private key
315 client_certificate_key: Path to private key file for client certificate
316 client_certificate_password: Password for client certificate private key, if encrypted.
317 If not provided and the key is encrypted, yt-dlp will ask interactively
7e8c0af0 318 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
c6e07cf1 319 (Only supported by some extractors)
8300774c 320 enable_file_urls: Enable file:// URLs. This is disabled by default for security reasons.
8b7539d2 321 http_headers: A dictionary of custom headers to be used for all requests
a1ee09e8 322 proxy: URL of the proxy server to use
38cce791 323 geo_verification_proxy: URL of the proxy to use for IP address verification
504f20dd 324 on geo-restricted sites.
e344693b 325 socket_timeout: Time to wait for unresponsive hosts, in seconds
0783b09b
PH
326 bidi_workaround: Work around buggy terminals without bidirectional text
327 support, using fridibi
a0ddb8a2 328 debug_printtraffic:Print out sent and received HTTP traffic
04b4d394
PH
329 default_search: Prepend this string if an input url is not valid.
330 'auto' for elaborate guessing
62fec3b2 331 encoding: Use this encoding instead of the system-specified.
134c913c 332 extract_flat: Whether to resolve and process url_results further
46f1370e 333 * False: Always process. Default for API
134c913c 334 * True: Never process
335 * 'in_playlist': Do not process inside playlist/multi_video
336 * 'discard': Always process, but don't return the result
337 from inside playlist/multi_video
338 * 'discard_in_playlist': Same as "discard", but only for
46f1370e 339 playlists (not multi_video). Default for CLI
f2ebc5c7 340 wait_for_video: If given, wait for scheduled streams to become available.
341 The value should be a tuple containing the range
342 (min_secs, max_secs) to wait between retries
4f026faf 343 postprocessors: A list of dictionaries, each with an entry
71b640cc 344 * key: The name of the postprocessor. See
7a5c1cfe 345 yt_dlp/postprocessor/__init__.py for a list.
bb66c247 346 * when: When to run the postprocessor. Allowed values are
347 the entries of utils.POSTPROCESS_WHEN
56d868db 348 Assumed to be 'post_process' if not given
71b640cc
PH
349 progress_hooks: A list of functions that get called on download
350 progress, with a dictionary with the entries
5cda4eda 351 * status: One of "downloading", "error", or "finished".
ee69b99a 352 Check this first and ignore unknown values.
3ba7740d 353 * info_dict: The extracted info_dict
71b640cc 354
5cda4eda 355 If status is one of "downloading", or "finished", the
ee69b99a
PH
356 following properties may also be present:
357 * filename: The final filename (always present)
5cda4eda 358 * tmpfilename: The filename we're currently writing to
71b640cc
PH
359 * downloaded_bytes: Bytes on disk
360 * total_bytes: Size of the whole file, None if unknown
5cda4eda
PH
361 * total_bytes_estimate: Guess of the eventual file size,
362 None if unavailable.
363 * elapsed: The number of seconds since download started.
71b640cc
PH
364 * eta: The estimated time in seconds, None if unknown
365 * speed: The download speed in bytes/second, None if
366 unknown
5cda4eda
PH
367 * fragment_index: The counter of the currently
368 downloaded video fragment.
369 * fragment_count: The number of fragments (= individual
370 files that will be merged)
71b640cc
PH
371
372 Progress hooks are guaranteed to be called at least once
373 (with status "finished") if the download is successful.
819e0531 374 postprocessor_hooks: A list of functions that get called on postprocessing
375 progress, with a dictionary with the entries
376 * status: One of "started", "processing", or "finished".
377 Check this first and ignore unknown values.
378 * postprocessor: Name of the postprocessor
379 * info_dict: The extracted info_dict
380
381 Progress hooks are guaranteed to be called at least twice
382 (with status "started" and "finished") if the processing is successful.
fc61aff4 383 merge_output_format: "/" separated list of extensions to use when merging formats.
6b591b29 384 final_ext: Expected final extension; used to detect when the file was
59a7a13e 385 already downloaded and converted
6271f1ca
PH
386 fixup: Automatically correct known faults of the file.
387 One of:
388 - "never": do nothing
389 - "warn": only emit a warning
390 - "detect_or_warn": check whether we can do anything
62cd676c 391 about it, warn otherwise (default)
504f20dd 392 source_address: Client-side IP address to bind to.
1cf376f5 393 sleep_interval_requests: Number of seconds to sleep between requests
394 during extraction
7aa589a5
S
395 sleep_interval: Number of seconds to sleep before each download when
396 used alone or a lower bound of a range for randomized
397 sleep before each download (minimum possible number
398 of seconds to sleep) when used along with
399 max_sleep_interval.
400 max_sleep_interval:Upper bound of a range for randomized sleep before each
401 download (maximum possible number of seconds to sleep).
402 Must only be used along with sleep_interval.
403 Actual sleep time will be a random float from range
404 [sleep_interval; max_sleep_interval].
1cf376f5 405 sleep_interval_subtitles: Number of seconds to sleep before each subtitle download
cfb56d1a
PH
406 listformats: Print an overview of available video formats and exit.
407 list_thumbnails: Print a table of all thumbnails and exit.
0a41f331 408 match_filter: A function that gets called for every video with the signature
409 (info_dict, *, incomplete: bool) -> Optional[str]
410 For backward compatibility with youtube-dl, the signature
411 (info_dict) -> Optional[str] is also allowed.
412 - If it returns a message, the video is ignored.
413 - If it returns None, the video is downloaded.
414 - If it returns utils.NO_DEFAULT, the user is interactively
415 asked whether to download the video.
fe2ce85a 416 - Raise utils.DownloadCancelled(msg) to abort remaining
417 downloads when a video is rejected.
347de493 418 match_filter_func in utils.py is one example for this.
8417f26b
SS
419 color: A Dictionary with output stream names as keys
420 and their respective color policy as values.
421 Can also just be a single color policy,
422 in which case it applies to all outputs.
423 Valid stream names are 'stdout' and 'stderr'.
424 Valid color policies are one of 'always', 'auto', 'no_color' or 'never'.
0a840f58 425 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
504f20dd 426 HTTP header
0a840f58 427 geo_bypass_country:
773f291d
S
428 Two-letter ISO 3166-2 country code that will be used for
429 explicit geographic restriction bypassing via faking
504f20dd 430 X-Forwarded-For HTTP header
5f95927a
S
431 geo_bypass_ip_block:
432 IP range in CIDR notation that will be used similarly to
504f20dd 433 geo_bypass_country
52a8a1e1 434 external_downloader: A dictionary of protocol keys and the executable of the
435 external downloader to use for it. The allowed protocols
436 are default|http|ftp|m3u8|dash|rtsp|rtmp|mms.
437 Set the value to 'native' to use the native downloader
53ed7066 438 compat_opts: Compatibility options. See "Differences in default behavior".
3acf6d38 439 The following options do not work when used through the API:
b5ae35ee 440 filename, abort-on-error, multistreams, no-live-chat, format-sort
dac5df5a 441 no-clean-infojson, no-playlist-metafiles, no-keep-subs, no-attach-info-json.
e4f02757 442 Refer __init__.py for their implementation
819e0531 443 progress_template: Dictionary of templates for progress outputs.
444 Allowed keys are 'download', 'postprocess',
445 'download-title' (console title) and 'postprocess-title'.
446 The template is mapped on a dictionary with keys 'progress' and 'info'
23326151 447 retry_sleep_functions: Dictionary of functions that takes the number of attempts
448 as argument and returns the time to sleep in seconds.
449 Allowed keys are 'http', 'fragment', 'file_access'
0f446365
SW
450 download_ranges: A callback function that gets called for every video with
451 the signature (info_dict, ydl) -> Iterable[Section].
452 Only the returned sections will be downloaded.
453 Each Section is a dict with the following keys:
5ec1b6b7 454 * start_time: Start time of the section in seconds
455 * end_time: End time of the section in seconds
456 * title: Section title (Optional)
457 * index: Section number (Optional)
0f446365 458 force_keyframes_at_cuts: Re-encode the video when downloading ranges to get precise cuts
a7dc6a89 459 noprogress: Do not print the progress bar
a831c2ea 460 live_from_start: Whether to download livestreams videos from the start
fe7e0c98 461
8222d8de 462 The following parameters are not used by YoutubeDL itself, they are used by
7a5c1cfe 463 the downloader (see yt_dlp/downloader/common.py):
51d9739f 464 nopart, updatetime, buffersize, ratelimit, throttledratelimit, min_filesize,
205a0654 465 max_filesize, test, noresizebuffer, retries, file_access_retries, fragment_retries,
a7dc6a89 466 continuedl, xattr_set_filesize, hls_use_mpegts, http_chunk_size,
59a7a13e 467 external_downloader_args, concurrent_fragment_downloads.
76b1bd67
JMF
468
469 The following options are used by the post processors:
c0b7d117
S
470 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
471 to the binary or its containing directory.
43820c03 472 postprocessor_args: A dictionary of postprocessor/executable keys (in lower case)
34488702 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
e409895f 479
480 The following options are used by the extractors:
46f1370e 481 extractor_retries: Number of times to retry for known errors (default: 3)
62bff2c1 482 dynamic_mpd: Whether to process dynamic DASH manifests (default: True)
e409895f 483 hls_split_discontinuity: Split HLS playlists to different formats at
62bff2c1 484 discontinuities such as ad breaks (default: False)
5d3a0e79 485 extractor_args: A dictionary of arguments to be passed to the extractors.
486 See "EXTRACTOR ARGUMENTS" for details.
62b58c09 487 E.g. {'youtube': {'skip': ['dash', 'hls']}}
88f23a18 488 mark_watched: Mark videos watched (even with --simulate). Only for YouTube
1890fc63 489
490 The following options are deprecated and may be removed in the future:
491
fe2ce85a 492 break_on_reject: Stop the download process when encountering a video that
493 has been filtered out.
494 - `raise DownloadCancelled(msg)` in match_filter instead
fe7866d0 495 force_generic_extractor: Force downloader to use the generic extractor
496 - Use allowed_extractors = ['generic', 'default']
7e9a6125 497 playliststart: - Use playlist_items
498 Playlist item to start at.
499 playlistend: - Use playlist_items
500 Playlist item to end at.
501 playlistreverse: - Use playlist_items
502 Download playlist items in reverse order.
1890fc63 503 forceurl: - Use forceprint
504 Force printing final URL.
505 forcetitle: - Use forceprint
506 Force printing title.
507 forceid: - Use forceprint
508 Force printing ID.
509 forcethumbnail: - Use forceprint
510 Force printing thumbnail URL.
511 forcedescription: - Use forceprint
512 Force printing description.
513 forcefilename: - Use forceprint
514 Force printing final filename.
515 forceduration: - Use forceprint
516 Force printing duration.
517 allsubtitles: - Use subtitleslangs = ['all']
518 Downloads all the subtitles of the video
519 (requires writesubtitles or writeautomaticsub)
520 include_ads: - Doesn't work
521 Download ads as well
522 call_home: - Not implemented
523 Boolean, true iff we are allowed to contact the
524 yt-dlp servers for debugging.
525 post_hooks: - Register a custom postprocessor
526 A list of functions that get called as the final step
527 for each video file, after all postprocessors have been
528 called. The filename will be passed as the only argument.
529 hls_prefer_native: - Use external_downloader = {'m3u8': 'native'} or {'m3u8': 'ffmpeg'}.
530 Use the native HLS downloader instead of ffmpeg/avconv
531 if True, otherwise use ffmpeg/avconv if False, otherwise
532 use downloader suggested by extractor if None.
533 prefer_ffmpeg: - avconv support is deprecated
534 If False, use avconv instead of ffmpeg if both are available,
535 otherwise prefer ffmpeg.
536 youtube_include_dash_manifest: - Use extractor_args
5d3a0e79 537 If True (default), DASH manifests and related
62bff2c1 538 data will be downloaded and processed by extractor.
539 You can reduce network I/O by disabling it if you don't
540 care about DASH. (only for youtube)
1890fc63 541 youtube_include_hls_manifest: - Use extractor_args
5d3a0e79 542 If True (default), HLS manifests and related
62bff2c1 543 data will be downloaded and processed by extractor.
544 You can reduce network I/O by disabling it if you don't
545 care about HLS. (only for youtube)
8417f26b 546 no_color: Same as `color='no_color'`
8222d8de
JMF
547 """
548
86e5f3ed 549 _NUMERIC_FIELDS = {
b8ed0f15 550 'width', 'height', 'asr', 'audio_channels', 'fps',
551 'tbr', 'abr', 'vbr', 'filesize', 'filesize_approx',
e6f21b3d 552 'timestamp', 'release_timestamp',
c9969434
S
553 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
554 'average_rating', 'comment_count', 'age_limit',
555 'start_time', 'end_time',
556 'chapter_number', 'season_number', 'episode_number',
557 'track_number', 'disc_number', 'release_year',
86e5f3ed 558 }
c9969434 559
6db9c4d5 560 _format_fields = {
561 # NB: Keep in sync with the docstring of extractor/common.py
a44ca5a4 562 'url', 'manifest_url', 'manifest_stream_number', 'ext', 'format', 'format_id', 'format_note',
105bfd90 563 'width', 'height', 'aspect_ratio', 'resolution', 'dynamic_range', 'tbr', 'abr', 'acodec', 'asr', 'audio_channels',
d5d1df8a 564 'vbr', 'fps', 'vcodec', 'container', 'filesize', 'filesize_approx', 'rows', 'columns',
6db9c4d5 565 'player_url', 'protocol', 'fragment_base_url', 'fragments', 'is_from_start',
566 'preference', 'language', 'language_preference', 'quality', 'source_preference',
7e68567e 567 'http_headers', 'stretched_ratio', 'no_resume', 'has_drm', 'extra_param_to_segment_url', 'hls_aes', 'downloader_options',
6db9c4d5 568 'page_url', 'app', 'play_path', 'tc_url', 'flash_version', 'rtmp_live', 'rtmp_conn', 'rtmp_protocol', 'rtmp_real_time'
569 }
48ee10ee 570 _format_selection_exts = {
8dc59305 571 'audio': set(MEDIA_EXTENSIONS.common_audio),
572 'video': set(MEDIA_EXTENSIONS.common_video + ('3gp', )),
573 'storyboards': set(MEDIA_EXTENSIONS.storyboards),
48ee10ee 574 }
575
3511266b 576 def __init__(self, params=None, auto_init=True):
883d4b1e 577 """Create a FileDownloader object with the given options.
578 @param auto_init Whether to load the default extractors and print header (if verbose).
49a57e70 579 Set to 'no_verbose_header' to not print the header
883d4b1e 580 """
e9f9a10f
JMF
581 if params is None:
582 params = {}
592b7485 583 self.params = params
8b7491c8 584 self._ies = {}
56c73665 585 self._ies_instances = {}
1e43a6f7 586 self._pps = {k: [] for k in POSTPROCESS_WHEN}
b35496d8 587 self._printed_messages = set()
1cf376f5 588 self._first_webpage_request = True
ab8e5e51 589 self._post_hooks = []
933605d7 590 self._progress_hooks = []
819e0531 591 self._postprocessor_hooks = []
8222d8de
JMF
592 self._download_retcode = 0
593 self._num_downloads = 0
9c906919 594 self._num_videos = 0
592b7485 595 self._playlist_level = 0
596 self._playlist_urls = set()
a0e07d31 597 self.cache = Cache(self)
34308b30 598
591bb9d3 599 stdout = sys.stderr if self.params.get('logtostderr') else sys.stdout
600 self._out_files = Namespace(
601 out=stdout,
602 error=sys.stderr,
603 screen=sys.stderr if self.params.get('quiet') else stdout,
604 console=None if compat_os_name == 'nt' else next(
cf4f42cb 605 filter(supports_terminal_sequences, (sys.stderr, sys.stdout)), None)
591bb9d3 606 )
f0795149 607
608 try:
609 windows_enable_vt_mode()
610 except Exception as e:
611 self.write_debug(f'Failed to enable VT mode: {e}')
612
8417f26b
SS
613 if self.params.get('no_color'):
614 if self.params.get('color') is not None:
615 self.report_warning('Overwriting params from "color" with "no_color"')
616 self.params['color'] = 'no_color'
617
618 term_allow_color = os.environ.get('TERM', '').lower() != 'dumb'
619
620 def process_color_policy(stream):
621 stream_name = {sys.stdout: 'stdout', sys.stderr: 'stderr'}[stream]
622 policy = traverse_obj(self.params, ('color', (stream_name, None), {str}), get_all=False)
623 if policy in ('auto', None):
624 return term_allow_color and supports_terminal_sequences(stream)
625 assert policy in ('always', 'never', 'no_color')
626 return {'always': True, 'never': False}.get(policy, policy)
627
591bb9d3 628 self._allow_colors = Namespace(**{
8417f26b
SS
629 name: process_color_policy(stream)
630 for name, stream in self._out_files.items_ if name != 'console'
591bb9d3 631 })
819e0531 632
6929b41a 633 # The code is left like this to be reused for future deprecations
634 MIN_SUPPORTED, MIN_RECOMMENDED = (3, 7), (3, 7)
eff42759 635 current_version = sys.version_info[:2]
636 if current_version < MIN_RECOMMENDED:
9d339c41 637 msg = ('Support for Python version %d.%d has been deprecated. '
24093d52 638 'See https://github.com/yt-dlp/yt-dlp/issues/3764 for more details.'
c6e07cf1 639 '\n You will no longer receive updates on this version')
eff42759 640 if current_version < MIN_SUPPORTED:
641 msg = 'Python version %d.%d is no longer supported'
5b28cef7 642 self.deprecated_feature(
eff42759 643 f'{msg}! Please update to Python %d.%d or above' % (*current_version, *MIN_RECOMMENDED))
a61f4b28 644
88acdbc2 645 if self.params.get('allow_unplayable_formats'):
646 self.report_warning(
ec11a9f4 647 f'You have asked for {self._format_err("UNPLAYABLE", self.Styles.EMPHASIS)} formats to be listed/downloaded. '
819e0531 648 'This is a developer option intended for debugging. \n'
649 ' If you experience any issues while using this option, '
ec11a9f4 650 f'{self._format_err("DO NOT", self.Styles.ERROR)} open a bug report')
88acdbc2 651
497074f0 652 if self.params.get('bidi_workaround', False):
653 try:
654 import pty
655 master, slave = pty.openpty()
656 width = shutil.get_terminal_size().columns
657 width_args = [] if width is None else ['-w', str(width)]
658 sp_kwargs = {'stdin': subprocess.PIPE, 'stdout': slave, 'stderr': self._out_files.error}
659 try:
660 self._output_process = Popen(['bidiv'] + width_args, **sp_kwargs)
661 except OSError:
662 self._output_process = Popen(['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
663 self._output_channel = os.fdopen(master, 'rb')
664 except OSError as ose:
665 if ose.errno == errno.ENOENT:
666 self.report_warning(
667 'Could not find fribidi executable, ignoring --bidi-workaround. '
668 'Make sure that fribidi is an executable file in one of the directories in your $PATH.')
669 else:
670 raise
671
672 self.params['compat_opts'] = set(self.params.get('compat_opts', ()))
673 if auto_init and auto_init != 'no_verbose_header':
674 self.print_debug_header()
675
be5df5ee
S
676 def check_deprecated(param, option, suggestion):
677 if self.params.get(param) is not None:
86e5f3ed 678 self.report_warning(f'{option} is deprecated. Use {suggestion} instead')
be5df5ee
S
679 return True
680 return False
681
682 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
38cce791
YCH
683 if self.params.get('geo_verification_proxy') is None:
684 self.params['geo_verification_proxy'] = self.params['cn_verification_proxy']
685
0d1bb027 686 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
687 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
53ed7066 688 check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"')
0d1bb027 689
49a57e70 690 for msg in self.params.get('_warnings', []):
0d1bb027 691 self.report_warning(msg)
ee8dd27a 692 for msg in self.params.get('_deprecation_warnings', []):
da4db748 693 self.deprecated_feature(msg)
0d1bb027 694
8a82af35 695 if 'list-formats' in self.params['compat_opts']:
ec11a9f4 696 self.params['listformats_table'] = False
697
b5ae35ee 698 if 'overwrites' not in self.params and self.params.get('nooverwrites') is not None:
b868936c 699 # nooverwrites was unnecessarily changed to overwrites
700 # in 0c3d0f51778b153f65c21906031c2e091fcfb641
701 # This ensures compatibility with both keys
702 self.params['overwrites'] = not self.params['nooverwrites']
b5ae35ee 703 elif self.params.get('overwrites') is None:
704 self.params.pop('overwrites', None)
b868936c 705 else:
706 self.params['nooverwrites'] = not self.params['overwrites']
b9d973be 707
e4221b70 708 if self.params.get('simulate') is None and any((
709 self.params.get('list_thumbnails'),
710 self.params.get('listformats'),
711 self.params.get('listsubtitles'),
712 )):
713 self.params['simulate'] = 'list_only'
714
455a15e2 715 self.params.setdefault('forceprint', {})
716 self.params.setdefault('print_to_file', {})
bb66c247 717
718 # Compatibility with older syntax
ca30f449 719 if not isinstance(params['forceprint'], dict):
455a15e2 720 self.params['forceprint'] = {'video': params['forceprint']}
ca30f449 721
97ec5bc5 722 if auto_init:
97ec5bc5 723 self.add_default_info_extractors()
724
3089bc74
S
725 if (sys.platform != 'win32'
726 and sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
455a15e2 727 and not self.params.get('restrictfilenames', False)):
e9137224 728 # Unicode filesystem API will throw errors (#1474, #13027)
34308b30 729 self.report_warning(
6febd1c1 730 'Assuming --restrict-filenames since file system encoding '
1b725173 731 'cannot encode all characters. '
6febd1c1 732 'Set the LC_ALL environment variable to fix this.')
4a98cdbf 733 self.params['restrictfilenames'] = True
34308b30 734
bf1824b3 735 self._parse_outtmpl()
486dd09e 736
187986a8 737 # Creating format selector here allows us to catch syntax errors before the extraction
738 self.format_selector = (
fa9f30b8 739 self.params.get('format') if self.params.get('format') in (None, '-')
093a1710 740 else self.params['format'] if callable(self.params['format'])
187986a8 741 else self.build_format_selector(self.params['format']))
742
8b7539d2 743 # Set http_headers defaults according to std_headers
744 self.params['http_headers'] = merge_headers(std_headers, self.params.get('http_headers', {}))
745
013b50b7 746 hooks = {
747 'post_hooks': self.add_post_hook,
748 'progress_hooks': self.add_progress_hook,
749 'postprocessor_hooks': self.add_postprocessor_hook,
750 }
751 for opt, fn in hooks.items():
752 for ph in self.params.get(opt, []):
753 fn(ph)
71b640cc 754
5bfc8bee 755 for pp_def_raw in self.params.get('postprocessors', []):
756 pp_def = dict(pp_def_raw)
757 when = pp_def.pop('when', 'post_process')
758 self.add_post_processor(
f9934b96 759 get_postprocessor(pp_def.pop('key'))(self, **pp_def),
5bfc8bee 760 when=when)
761
97ec5bc5 762 self._setup_opener()
51fb4995 763
ed39cac5 764 def preload_download_archive(fn):
765 """Preload the archive, if any is specified"""
ae103564 766 archive = set()
ed39cac5 767 if fn is None:
ae103564 768 return archive
941e881e 769 elif not is_path_like(fn):
ae103564 770 return fn
771
49a57e70 772 self.write_debug(f'Loading archive file {fn!r}')
ed39cac5 773 try:
774 with locked_file(fn, 'r', encoding='utf-8') as archive_file:
775 for line in archive_file:
ae103564 776 archive.add(line.strip())
86e5f3ed 777 except OSError as ioe:
ed39cac5 778 if ioe.errno != errno.ENOENT:
779 raise
ae103564 780 return archive
ed39cac5 781
ae103564 782 self.archive = preload_download_archive(self.params.get('download_archive'))
ed39cac5 783
7d4111ed
PH
784 def warn_if_short_id(self, argv):
785 # short YouTube ID starting with dash?
786 idxs = [
787 i for i, a in enumerate(argv)
788 if re.match(r'^-[0-9A-Za-z_-]{10}$', a)]
789 if idxs:
790 correct_argv = (
7a5c1cfe 791 ['yt-dlp']
3089bc74
S
792 + [a for i, a in enumerate(argv) if i not in idxs]
793 + ['--'] + [argv[i] for i in idxs]
7d4111ed
PH
794 )
795 self.report_warning(
796 'Long argument string detected. '
49a57e70 797 'Use -- to separate parameters and URLs, like this:\n%s' %
7d4111ed
PH
798 args_to_str(correct_argv))
799
8222d8de
JMF
800 def add_info_extractor(self, ie):
801 """Add an InfoExtractor object to the end of the list."""
8b7491c8 802 ie_key = ie.ie_key()
803 self._ies[ie_key] = ie
e52d7f85 804 if not isinstance(ie, type):
8b7491c8 805 self._ies_instances[ie_key] = ie
e52d7f85 806 ie.set_downloader(self)
8222d8de 807
56c73665
JMF
808 def get_info_extractor(self, ie_key):
809 """
810 Get an instance of an IE with name ie_key, it will try to get one from
811 the _ies list, if there's no instance it will create a new one and add
812 it to the extractor list.
813 """
814 ie = self._ies_instances.get(ie_key)
815 if ie is None:
816 ie = get_info_extractor(ie_key)()
817 self.add_info_extractor(ie)
818 return ie
819
023fa8c4
JMF
820 def add_default_info_extractors(self):
821 """
822 Add the InfoExtractors returned by gen_extractors to the end of the list
823 """
fe7866d0 824 all_ies = {ie.IE_NAME.lower(): ie for ie in gen_extractor_classes()}
825 all_ies['end'] = UnsupportedURLIE()
826 try:
827 ie_names = orderedSet_from_options(
828 self.params.get('allowed_extractors', ['default']), {
829 'all': list(all_ies),
830 'default': [name for name, ie in all_ies.items() if ie._ENABLED],
831 }, use_regex=True)
832 except re.error as e:
833 raise ValueError(f'Wrong regex for allowed_extractors: {e.pattern}')
834 for name in ie_names:
835 self.add_info_extractor(all_ies[name])
836 self.write_debug(f'Loaded {len(ie_names)} extractors')
023fa8c4 837
56d868db 838 def add_post_processor(self, pp, when='post_process'):
8222d8de 839 """Add a PostProcessor object to the end of the chain."""
8aa0e7cd 840 assert when in POSTPROCESS_WHEN, f'Invalid when={when}'
5bfa4862 841 self._pps[when].append(pp)
8222d8de
JMF
842 pp.set_downloader(self)
843
ab8e5e51
AM
844 def add_post_hook(self, ph):
845 """Add the post hook"""
846 self._post_hooks.append(ph)
847
933605d7 848 def add_progress_hook(self, ph):
819e0531 849 """Add the download progress hook"""
933605d7 850 self._progress_hooks.append(ph)
8ab470f1 851
819e0531 852 def add_postprocessor_hook(self, ph):
853 """Add the postprocessing progress hook"""
854 self._postprocessor_hooks.append(ph)
5bfc8bee 855 for pps in self._pps.values():
856 for pp in pps:
857 pp.add_progress_hook(ph)
819e0531 858
1c088fa8 859 def _bidi_workaround(self, message):
5d681e96 860 if not hasattr(self, '_output_channel'):
1c088fa8
PH
861 return message
862
5d681e96 863 assert hasattr(self, '_output_process')
14f25df2 864 assert isinstance(message, str)
6febd1c1 865 line_count = message.count('\n') + 1
0f06bcd7 866 self._output_process.stdin.write((message + '\n').encode())
5d681e96 867 self._output_process.stdin.flush()
0f06bcd7 868 res = ''.join(self._output_channel.readline().decode()
9e1a5b84 869 for _ in range(line_count))
6febd1c1 870 return res[:-len('\n')]
1c088fa8 871
b35496d8 872 def _write_string(self, message, out=None, only_once=False):
873 if only_once:
874 if message in self._printed_messages:
875 return
876 self._printed_messages.add(message)
877 write_string(message, out=out, encoding=self.params.get('encoding'))
734f90bb 878
cf4f42cb 879 def to_stdout(self, message, skip_eol=False, quiet=None):
0760b0a7 880 """Print message to stdout"""
cf4f42cb 881 if quiet is not None:
da4db748 882 self.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument quiet. '
883 'Use "YoutubeDL.to_screen" instead')
8a82af35 884 if skip_eol is not False:
da4db748 885 self.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument skip_eol. '
886 'Use "YoutubeDL.to_screen" instead')
0bf9dc1e 887 self._write_string(f'{self._bidi_workaround(message)}\n', self._out_files.out)
cf4f42cb 888
dfea94f8 889 def to_screen(self, message, skip_eol=False, quiet=None, only_once=False):
cf4f42cb 890 """Print message to screen if not in quiet mode"""
8bf9319e 891 if self.params.get('logger'):
43afe285 892 self.params['logger'].debug(message)
cf4f42cb 893 return
894 if (self.params.get('quiet') if quiet is None else quiet) and not self.params.get('verbose'):
895 return
896 self._write_string(
897 '%s%s' % (self._bidi_workaround(message), ('' if skip_eol else '\n')),
dfea94f8 898 self._out_files.screen, only_once=only_once)
8222d8de 899
b35496d8 900 def to_stderr(self, message, only_once=False):
0760b0a7 901 """Print message to stderr"""
14f25df2 902 assert isinstance(message, str)
8bf9319e 903 if self.params.get('logger'):
43afe285
IB
904 self.params['logger'].error(message)
905 else:
5792c950 906 self._write_string(f'{self._bidi_workaround(message)}\n', self._out_files.error, only_once=only_once)
cf4f42cb 907
908 def _send_console_code(self, code):
591bb9d3 909 if compat_os_name == 'nt' or not self._out_files.console:
cf4f42cb 910 return
591bb9d3 911 self._write_string(code, self._out_files.console)
8222d8de 912
1e5b9a95
PH
913 def to_console_title(self, message):
914 if not self.params.get('consoletitle', False):
915 return
3efb96a6 916 message = remove_terminal_sequences(message)
4bede0d8
C
917 if compat_os_name == 'nt':
918 if ctypes.windll.kernel32.GetConsoleWindow():
919 # c_wchar_p() might not be necessary if `message` is
920 # already of type unicode()
921 ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
cf4f42cb 922 else:
923 self._send_console_code(f'\033]0;{message}\007')
1e5b9a95 924
bdde425c 925 def save_console_title(self):
cf4f42cb 926 if not self.params.get('consoletitle') or self.params.get('simulate'):
bdde425c 927 return
592b7485 928 self._send_console_code('\033[22;0t') # Save the title on stack
bdde425c
PH
929
930 def restore_console_title(self):
cf4f42cb 931 if not self.params.get('consoletitle') or self.params.get('simulate'):
bdde425c 932 return
592b7485 933 self._send_console_code('\033[23;0t') # Restore the title from stack
bdde425c
PH
934
935 def __enter__(self):
936 self.save_console_title()
937 return self
938
939 def __exit__(self, *args):
940 self.restore_console_title()
f89197d7 941
dca08720 942 if self.params.get('cookiefile') is not None:
1bab3437 943 self.cookiejar.save(ignore_discard=True, ignore_expires=True)
bdde425c 944
fa9f30b8 945 def trouble(self, message=None, tb=None, is_error=True):
8222d8de
JMF
946 """Determine action to take when a download problem appears.
947
948 Depending on if the downloader has been configured to ignore
949 download errors or not, this method may throw an exception or
950 not when errors are found, after printing the message.
951
fa9f30b8 952 @param tb If given, is additional traceback information
953 @param is_error Whether to raise error according to ignorerrors
8222d8de
JMF
954 """
955 if message is not None:
956 self.to_stderr(message)
957 if self.params.get('verbose'):
958 if tb is None:
959 if sys.exc_info()[0]: # if .trouble has been called from an except block
6febd1c1 960 tb = ''
8222d8de 961 if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
6febd1c1 962 tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
c0384f22 963 tb += encode_compat_str(traceback.format_exc())
8222d8de
JMF
964 else:
965 tb_data = traceback.format_list(traceback.extract_stack())
6febd1c1 966 tb = ''.join(tb_data)
c19bc311 967 if tb:
968 self.to_stderr(tb)
fa9f30b8 969 if not is_error:
970 return
b1940459 971 if not self.params.get('ignoreerrors'):
8222d8de
JMF
972 if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
973 exc_info = sys.exc_info()[1].exc_info
974 else:
975 exc_info = sys.exc_info()
976 raise DownloadError(message, exc_info)
977 self._download_retcode = 1
978
19a03940 979 Styles = Namespace(
980 HEADERS='yellow',
981 EMPHASIS='light blue',
492272fe 982 FILENAME='green',
19a03940 983 ID='green',
984 DELIM='blue',
985 ERROR='red',
986 WARNING='yellow',
987 SUPPRESS='light black',
988 )
ec11a9f4 989
7578d77d 990 def _format_text(self, handle, allow_colors, text, f, fallback=None, *, test_encoding=False):
e5a998f3 991 text = str(text)
ec11a9f4 992 if test_encoding:
993 original_text = text
5c104538 994 # handle.encoding can be None. See https://github.com/yt-dlp/yt-dlp/issues/2711
995 encoding = self.params.get('encoding') or getattr(handle, 'encoding', None) or 'ascii'
ec11a9f4 996 text = text.encode(encoding, 'ignore').decode(encoding)
997 if fallback is not None and text != original_text:
998 text = fallback
8417f26b 999 return format_text(text, f) if allow_colors is True else text if fallback is None else fallback
ec11a9f4 1000
591bb9d3 1001 def _format_out(self, *args, **kwargs):
1002 return self._format_text(self._out_files.out, self._allow_colors.out, *args, **kwargs)
1003
ec11a9f4 1004 def _format_screen(self, *args, **kwargs):
591bb9d3 1005 return self._format_text(self._out_files.screen, self._allow_colors.screen, *args, **kwargs)
ec11a9f4 1006
1007 def _format_err(self, *args, **kwargs):
591bb9d3 1008 return self._format_text(self._out_files.error, self._allow_colors.error, *args, **kwargs)
819e0531 1009
c84aeac6 1010 def report_warning(self, message, only_once=False):
8222d8de
JMF
1011 '''
1012 Print the message to stderr, it will be prefixed with 'WARNING:'
1013 If stderr is a tty file the 'WARNING:' will be colored
1014 '''
6d07ce01
JMF
1015 if self.params.get('logger') is not None:
1016 self.params['logger'].warning(message)
8222d8de 1017 else:
ad8915b7
PH
1018 if self.params.get('no_warnings'):
1019 return
ec11a9f4 1020 self.to_stderr(f'{self._format_err("WARNING:", self.Styles.WARNING)} {message}', only_once)
8222d8de 1021
da4db748 1022 def deprecation_warning(self, message, *, stacklevel=0):
1023 deprecation_warning(
1024 message, stacklevel=stacklevel + 1, printer=self.report_error, is_error=False)
1025
1026 def deprecated_feature(self, message):
ee8dd27a 1027 if self.params.get('logger') is not None:
da4db748 1028 self.params['logger'].warning(f'Deprecated Feature: {message}')
1029 self.to_stderr(f'{self._format_err("Deprecated Feature:", self.Styles.ERROR)} {message}', True)
ee8dd27a 1030
fa9f30b8 1031 def report_error(self, message, *args, **kwargs):
8222d8de
JMF
1032 '''
1033 Do the same as trouble, but prefixes the message with 'ERROR:', colored
1034 in red if stderr is a tty file.
1035 '''
fa9f30b8 1036 self.trouble(f'{self._format_err("ERROR:", self.Styles.ERROR)} {message}', *args, **kwargs)
8222d8de 1037
b35496d8 1038 def write_debug(self, message, only_once=False):
0760b0a7 1039 '''Log debug message or Print message to stderr'''
1040 if not self.params.get('verbose', False):
1041 return
8a82af35 1042 message = f'[debug] {message}'
0760b0a7 1043 if self.params.get('logger'):
1044 self.params['logger'].debug(message)
1045 else:
b35496d8 1046 self.to_stderr(message, only_once)
0760b0a7 1047
8222d8de
JMF
1048 def report_file_already_downloaded(self, file_name):
1049 """Report file has already been fully downloaded."""
1050 try:
6febd1c1 1051 self.to_screen('[download] %s has already been downloaded' % file_name)
ce02ed60 1052 except UnicodeEncodeError:
6febd1c1 1053 self.to_screen('[download] The file has already been downloaded')
8222d8de 1054
0c3d0f51 1055 def report_file_delete(self, file_name):
1056 """Report that existing file will be deleted."""
1057 try:
c25228e5 1058 self.to_screen('Deleting existing file %s' % file_name)
0c3d0f51 1059 except UnicodeEncodeError:
c25228e5 1060 self.to_screen('Deleting existing file')
0c3d0f51 1061
319b6059 1062 def raise_no_formats(self, info, forced=False, *, msg=None):
0a5a191a 1063 has_drm = info.get('_has_drm')
319b6059 1064 ignored, expected = self.params.get('ignore_no_formats_error'), bool(msg)
1065 msg = msg or has_drm and 'This video is DRM protected' or 'No video formats found!'
1066 if forced or not ignored:
1151c407 1067 raise ExtractorError(msg, video_id=info['id'], ie=info['extractor'],
319b6059 1068 expected=has_drm or ignored or expected)
88acdbc2 1069 else:
1070 self.report_warning(msg)
1071
de6000d9 1072 def parse_outtmpl(self):
bf1824b3 1073 self.deprecation_warning('"YoutubeDL.parse_outtmpl" is deprecated and may be removed in a future version')
1074 self._parse_outtmpl()
1075 return self.params['outtmpl']
1076
1077 def _parse_outtmpl(self):
7b2c3f47 1078 sanitize = IDENTITY
bf1824b3 1079 if self.params.get('restrictfilenames'): # Remove spaces in the default template
71ce444a 1080 sanitize = lambda x: x.replace(' - ', ' ').replace(' ', '-')
bf1824b3 1081
1082 outtmpl = self.params.setdefault('outtmpl', {})
1083 if not isinstance(outtmpl, dict):
1084 self.params['outtmpl'] = outtmpl = {'default': outtmpl}
1085 outtmpl.update({k: sanitize(v) for k, v in DEFAULT_OUTTMPL.items() if outtmpl.get(k) is None})
de6000d9 1086
21cd8fae 1087 def get_output_path(self, dir_type='', filename=None):
1088 paths = self.params.get('paths', {})
d2c8aadf 1089 assert isinstance(paths, dict), '"paths" parameter must be a dictionary'
21cd8fae 1090 path = os.path.join(
1091 expand_path(paths.get('home', '').strip()),
1092 expand_path(paths.get(dir_type, '').strip()) if dir_type else '',
1093 filename or '')
21cd8fae 1094 return sanitize_path(path, force=self.params.get('windowsfilenames'))
1095
76a264ac 1096 @staticmethod
901130bb 1097 def _outtmpl_expandpath(outtmpl):
1098 # expand_path translates '%%' into '%' and '$$' into '$'
1099 # correspondingly that is not what we want since we need to keep
1100 # '%%' intact for template dict substitution step. Working around
1101 # with boundary-alike separator hack.
6f2287cb 1102 sep = ''.join(random.choices(string.ascii_letters, k=32))
86e5f3ed 1103 outtmpl = outtmpl.replace('%%', f'%{sep}%').replace('$$', f'${sep}$')
901130bb 1104
1105 # outtmpl should be expand_path'ed before template dict substitution
1106 # because meta fields may contain env variables we don't want to
62b58c09 1107 # be expanded. E.g. for outtmpl "%(title)s.%(ext)s" and
901130bb 1108 # title "Hello $PATH", we don't want `$PATH` to be expanded.
1109 return expand_path(outtmpl).replace(sep, '')
1110
1111 @staticmethod
1112 def escape_outtmpl(outtmpl):
1113 ''' Escape any remaining strings like %s, %abc% etc. '''
1114 return re.sub(
1115 STR_FORMAT_RE_TMPL.format('', '(?![%(\0])'),
1116 lambda mobj: ('' if mobj.group('has_key') else '%') + mobj.group(0),
1117 outtmpl)
1118
1119 @classmethod
1120 def validate_outtmpl(cls, outtmpl):
76a264ac 1121 ''' @return None or Exception object '''
7d1eb38a 1122 outtmpl = re.sub(
47cdc68e 1123 STR_FORMAT_RE_TMPL.format('[^)]*', '[ljhqBUDS]'),
7d1eb38a 1124 lambda mobj: f'{mobj.group(0)[:-1]}s',
1125 cls._outtmpl_expandpath(outtmpl))
76a264ac 1126 try:
7d1eb38a 1127 cls.escape_outtmpl(outtmpl) % collections.defaultdict(int)
76a264ac 1128 return None
1129 except ValueError as err:
1130 return err
1131
03b4de72 1132 @staticmethod
1133 def _copy_infodict(info_dict):
1134 info_dict = dict(info_dict)
09b49e1f 1135 info_dict.pop('__postprocessors', None)
415f8d51 1136 info_dict.pop('__pending_error', None)
03b4de72 1137 return info_dict
1138
e0fd9573 1139 def prepare_outtmpl(self, outtmpl, info_dict, sanitize=False):
1140 """ Make the outtmpl and info_dict suitable for substitution: ydl.escape_outtmpl(outtmpl) % info_dict
1141 @param sanitize Whether to sanitize the output as a filename.
1142 For backward compatibility, a function can also be passed
1143 """
1144
6e84b215 1145 info_dict.setdefault('epoch', int(time.time())) # keep epoch consistent once set
143db31d 1146
03b4de72 1147 info_dict = self._copy_infodict(info_dict)
752cda38 1148 info_dict['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
53c18592 1149 formatSeconds(info_dict['duration'], '-' if sanitize else ':')
143db31d 1150 if info_dict.get('duration', None) is not None
1151 else None)
1d485a1a 1152 info_dict['autonumber'] = int(self.params.get('autonumber_start', 1) - 1 + self._num_downloads)
9c906919 1153 info_dict['video_autonumber'] = self._num_videos
752cda38 1154 if info_dict.get('resolution') is None:
1155 info_dict['resolution'] = self.format_resolution(info_dict, default=None)
143db31d 1156
e6f21b3d 1157 # For fields playlist_index, playlist_autonumber and autonumber convert all occurrences
143db31d 1158 # of %(field)s to %(field)0Nd for backward compatibility
1159 field_size_compat_map = {
0a5a191a 1160 'playlist_index': number_of_digits(info_dict.get('__last_playlist_index') or 0),
ec11a9f4 1161 'playlist_autonumber': number_of_digits(info_dict.get('n_entries') or 0),
752cda38 1162 'autonumber': self.params.get('autonumber_size') or 5,
143db31d 1163 }
752cda38 1164
385a27fa 1165 TMPL_DICT = {}
47cdc68e 1166 EXTERNAL_FORMAT_RE = re.compile(STR_FORMAT_RE_TMPL.format('[^)]*', f'[{STR_FORMAT_TYPES}ljhqBUDS]'))
385a27fa 1167 MATH_FUNCTIONS = {
1168 '+': float.__add__,
1169 '-': float.__sub__,
1170 }
e625be0d 1171 # Field is of the form key1.key2...
07a1250e 1172 # where keys (except first) can be string, int, slice or "{field, ...}"
1173 FIELD_INNER_RE = r'(?:\w+|%(num)s|%(num)s?(?::%(num)s?){1,2})' % {'num': r'(?:-?\d+)'}
1174 FIELD_RE = r'\w*(?:\.(?:%(inner)s|{%(field)s(?:,%(field)s)*}))*' % {
1175 'inner': FIELD_INNER_RE,
1176 'field': rf'\w*(?:\.{FIELD_INNER_RE})*'
1177 }
1d485a1a 1178 MATH_FIELD_RE = rf'(?:{FIELD_RE}|-?{NUMBER_RE})'
385a27fa 1179 MATH_OPERATORS_RE = r'(?:%s)' % '|'.join(map(re.escape, MATH_FUNCTIONS.keys()))
78fde6e3 1180 INTERNAL_FORMAT_RE = re.compile(rf'''(?xs)
e625be0d 1181 (?P<negate>-)?
1d485a1a 1182 (?P<fields>{FIELD_RE})
1183 (?P<maths>(?:{MATH_OPERATORS_RE}{MATH_FIELD_RE})*)
e625be0d 1184 (?:>(?P<strf_format>.+?))?
34baa9fd 1185 (?P<remaining>
1186 (?P<alternate>(?<!\\),[^|&)]+)?
1187 (?:&(?P<replacement>.*?))?
1188 (?:\|(?P<default>.*?))?
1d485a1a 1189 )$''')
752cda38 1190
07a1250e 1191 def _traverse_infodict(fields):
1192 fields = [f for x in re.split(r'\.({.+?})\.?', fields)
1193 for f in ([x] if x.startswith('{') else x.split('.'))]
1194 for i in (0, -1):
1195 if fields and not fields[i]:
1196 fields.pop(i)
1197
1198 for i, f in enumerate(fields):
1199 if not f.startswith('{'):
1200 continue
1201 assert f.endswith('}'), f'No closing brace for {f} in {fields}'
1202 fields[i] = {k: k.split('.') for k in f[1:-1].split(',')}
1203
1204 return traverse_obj(info_dict, fields, is_user_input=True, traverse_string=True)
76a264ac 1205
752cda38 1206 def get_value(mdict):
1207 # Object traversal
2b8a2973 1208 value = _traverse_infodict(mdict['fields'])
752cda38 1209 # Negative
1210 if mdict['negate']:
1211 value = float_or_none(value)
1212 if value is not None:
1213 value *= -1
1214 # Do maths
385a27fa 1215 offset_key = mdict['maths']
1216 if offset_key:
752cda38 1217 value = float_or_none(value)
1218 operator = None
385a27fa 1219 while offset_key:
1220 item = re.match(
1221 MATH_FIELD_RE if operator else MATH_OPERATORS_RE,
1222 offset_key).group(0)
1223 offset_key = offset_key[len(item):]
1224 if operator is None:
752cda38 1225 operator = MATH_FUNCTIONS[item]
385a27fa 1226 continue
1227 item, multiplier = (item[1:], -1) if item[0] == '-' else (item, 1)
1228 offset = float_or_none(item)
1229 if offset is None:
2b8a2973 1230 offset = float_or_none(_traverse_infodict(item))
385a27fa 1231 try:
1232 value = operator(value, multiplier * offset)
1233 except (TypeError, ZeroDivisionError):
1234 return None
1235 operator = None
752cda38 1236 # Datetime formatting
1237 if mdict['strf_format']:
7c37ff97 1238 value = strftime_or_none(value, mdict['strf_format'].replace('\\,', ','))
752cda38 1239
a6bcaf71 1240 # XXX: Workaround for https://github.com/yt-dlp/yt-dlp/issues/4485
1241 if sanitize and value == '':
1242 value = None
752cda38 1243 return value
1244
b868936c 1245 na = self.params.get('outtmpl_na_placeholder', 'NA')
1246
e0fd9573 1247 def filename_sanitizer(key, value, restricted=self.params.get('restrictfilenames')):
5c3895ff 1248 return sanitize_filename(str(value), restricted=restricted, is_id=(
1249 bool(re.search(r'(^|[_.])id(\.|$)', key))
8a82af35 1250 if 'filename-sanitization' in self.params['compat_opts']
5c3895ff 1251 else NO_DEFAULT))
e0fd9573 1252
1253 sanitizer = sanitize if callable(sanitize) else filename_sanitizer
1254 sanitize = bool(sanitize)
1255
6e84b215 1256 def _dumpjson_default(obj):
1257 if isinstance(obj, (set, LazyList)):
1258 return list(obj)
adbc4ec4 1259 return repr(obj)
6e84b215 1260
6f2287cb 1261 class _ReplacementFormatter(string.Formatter):
ec9311c4 1262 def get_field(self, field_name, args, kwargs):
1263 if field_name.isdigit():
1264 return args[0], -1
1265 raise ValueError('Unsupported field')
1266
1267 replacement_formatter = _ReplacementFormatter()
1268
752cda38 1269 def create_key(outer_mobj):
1270 if not outer_mobj.group('has_key'):
b836dc94 1271 return outer_mobj.group(0)
752cda38 1272 key = outer_mobj.group('key')
752cda38 1273 mobj = re.match(INTERNAL_FORMAT_RE, key)
e0fd9573 1274 initial_field = mobj.group('fields') if mobj else ''
e978789f 1275 value, replacement, default = None, None, na
7c37ff97 1276 while mobj:
e625be0d 1277 mobj = mobj.groupdict()
7c37ff97 1278 default = mobj['default'] if mobj['default'] is not None else default
752cda38 1279 value = get_value(mobj)
e978789f 1280 replacement = mobj['replacement']
7c37ff97 1281 if value is None and mobj['alternate']:
34baa9fd 1282 mobj = re.match(INTERNAL_FORMAT_RE, mobj['remaining'][1:])
7c37ff97 1283 else:
1284 break
752cda38 1285
b868936c 1286 fmt = outer_mobj.group('format')
752cda38 1287 if fmt == 's' and value is not None and key in field_size_compat_map.keys():
86e5f3ed 1288 fmt = f'0{field_size_compat_map[key]:d}d'
752cda38 1289
ebe1b4e3 1290 if None not in (value, replacement):
ec9311c4 1291 try:
1292 value = replacement_formatter.format(replacement, value)
1293 except ValueError:
ebe1b4e3 1294 value, default = None, na
752cda38 1295
4476d2c7 1296 flags = outer_mobj.group('conversion') or ''
7d1eb38a 1297 str_fmt = f'{fmt[:-1]}s'
ebe1b4e3 1298 if value is None:
1299 value, fmt = default, 's'
1300 elif fmt[-1] == 'l': # list
4476d2c7 1301 delim = '\n' if '#' in flags else ', '
9e907ebd 1302 value, fmt = delim.join(map(str, variadic(value, allowed_types=(str, bytes)))), str_fmt
524e2e4f 1303 elif fmt[-1] == 'j': # json
deae7c17 1304 value, fmt = json.dumps(
1305 value, default=_dumpjson_default,
9b9dad11 1306 indent=4 if '#' in flags else None, ensure_ascii='+' not in flags), str_fmt
47cdc68e 1307 elif fmt[-1] == 'h': # html
deae7c17 1308 value, fmt = escapeHTML(str(value)), str_fmt
524e2e4f 1309 elif fmt[-1] == 'q': # quoted
4476d2c7 1310 value = map(str, variadic(value) if '#' in flags else [value])
1311 value, fmt = ' '.join(map(compat_shlex_quote, value)), str_fmt
524e2e4f 1312 elif fmt[-1] == 'B': # bytes
0f06bcd7 1313 value = f'%{str_fmt}'.encode() % str(value).encode()
f5aa5cfb 1314 value, fmt = value.decode('utf-8', 'ignore'), 's'
524e2e4f 1315 elif fmt[-1] == 'U': # unicode normalized
524e2e4f 1316 value, fmt = unicodedata.normalize(
1317 # "+" = compatibility equivalence, "#" = NFD
4476d2c7 1318 'NF%s%s' % ('K' if '+' in flags else '', 'D' if '#' in flags else 'C'),
524e2e4f 1319 value), str_fmt
e0fd9573 1320 elif fmt[-1] == 'D': # decimal suffix
abbeeebc 1321 num_fmt, fmt = fmt[:-1].replace('#', ''), 's'
1322 value = format_decimal_suffix(value, f'%{num_fmt}f%s' if num_fmt else '%d%s',
1323 factor=1024 if '#' in flags else 1000)
37893bb0 1324 elif fmt[-1] == 'S': # filename sanitization
e0fd9573 1325 value, fmt = filename_sanitizer(initial_field, value, restricted='#' in flags), str_fmt
7d1eb38a 1326 elif fmt[-1] == 'c':
524e2e4f 1327 if value:
1328 value = str(value)[0]
76a264ac 1329 else:
524e2e4f 1330 fmt = str_fmt
1619ab3e 1331 elif fmt[-1] not in 'rsa': # numeric
a439a3a4 1332 value = float_or_none(value)
752cda38 1333 if value is None:
1334 value, fmt = default, 's'
901130bb 1335
752cda38 1336 if sanitize:
1619ab3e 1337 # If value is an object, sanitize might convert it to a string
1338 # So we convert it to repr first
752cda38 1339 if fmt[-1] == 'r':
7d1eb38a 1340 value, fmt = repr(value), str_fmt
1619ab3e 1341 elif fmt[-1] == 'a':
1342 value, fmt = ascii(value), str_fmt
1343 if fmt[-1] in 'csra':
e0fd9573 1344 value = sanitizer(initial_field, value)
901130bb 1345
b868936c 1346 key = '%s\0%s' % (key.replace('%', '%\0'), outer_mobj.group('format'))
385a27fa 1347 TMPL_DICT[key] = value
b868936c 1348 return '{prefix}%({key}){fmt}'.format(key=key, fmt=fmt, prefix=outer_mobj.group('prefix'))
752cda38 1349
385a27fa 1350 return EXTERNAL_FORMAT_RE.sub(create_key, outtmpl), TMPL_DICT
143db31d 1351
819e0531 1352 def evaluate_outtmpl(self, outtmpl, info_dict, *args, **kwargs):
1353 outtmpl, info_dict = self.prepare_outtmpl(outtmpl, info_dict, *args, **kwargs)
1354 return self.escape_outtmpl(outtmpl) % info_dict
1355
5127e92a 1356 def _prepare_filename(self, info_dict, *, outtmpl=None, tmpl_type=None):
1357 assert None in (outtmpl, tmpl_type), 'outtmpl and tmpl_type are mutually exclusive'
1358 if outtmpl is None:
bf1824b3 1359 outtmpl = self.params['outtmpl'].get(tmpl_type or 'default', self.params['outtmpl']['default'])
8222d8de 1360 try:
5127e92a 1361 outtmpl = self._outtmpl_expandpath(outtmpl)
e0fd9573 1362 filename = self.evaluate_outtmpl(outtmpl, info_dict, True)
6a0546e3 1363 if not filename:
1364 return None
15da37c7 1365
5127e92a 1366 if tmpl_type in ('', 'temp'):
6a0546e3 1367 final_ext, ext = self.params.get('final_ext'), info_dict.get('ext')
1368 if final_ext and ext and final_ext != ext and filename.endswith(f'.{final_ext}'):
1369 filename = replace_extension(filename, ext, final_ext)
5127e92a 1370 elif tmpl_type:
6a0546e3 1371 force_ext = OUTTMPL_TYPES[tmpl_type]
1372 if force_ext:
1373 filename = replace_extension(filename, force_ext, info_dict.get('ext'))
de6000d9 1374
bdc3fd2f
U
1375 # https://github.com/blackjack4494/youtube-dlc/issues/85
1376 trim_file_name = self.params.get('trim_file_name', False)
1377 if trim_file_name:
5c22c63d 1378 no_ext, *ext = filename.rsplit('.', 2)
1379 filename = join_nonempty(no_ext[:trim_file_name], *ext, delim='.')
bdc3fd2f 1380
0202b52a 1381 return filename
8222d8de 1382 except ValueError as err:
6febd1c1 1383 self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')')
8222d8de
JMF
1384 return None
1385
5127e92a 1386 def prepare_filename(self, info_dict, dir_type='', *, outtmpl=None, warn=False):
1387 """Generate the output filename"""
1388 if outtmpl:
1389 assert not dir_type, 'outtmpl and dir_type are mutually exclusive'
1390 dir_type = None
1391 filename = self._prepare_filename(info_dict, tmpl_type=dir_type, outtmpl=outtmpl)
80c03fa9 1392 if not filename and dir_type not in ('', 'temp'):
1393 return ''
de6000d9 1394
c84aeac6 1395 if warn:
21cd8fae 1396 if not self.params.get('paths'):
de6000d9 1397 pass
1398 elif filename == '-':
c84aeac6 1399 self.report_warning('--paths is ignored when an outputting to stdout', only_once=True)
de6000d9 1400 elif os.path.isabs(filename):
c84aeac6 1401 self.report_warning('--paths is ignored since an absolute path is given in output template', only_once=True)
de6000d9 1402 if filename == '-' or not filename:
1403 return filename
1404
21cd8fae 1405 return self.get_output_path(dir_type, filename)
0202b52a 1406
120fe513 1407 def _match_entry(self, info_dict, incomplete=False, silent=False):
6368e2e6 1408 """Returns None if the file should be downloaded"""
93b39cdb 1409 _type = 'video' if 'playlist-match-filter' in self.params['compat_opts'] else info_dict.get('_type', 'video')
d7b460d0 1410 assert incomplete or _type == 'video', 'Only video result can be considered complete'
8222d8de 1411
3bec830a 1412 video_title = info_dict.get('title', info_dict.get('id', 'entry'))
c77495e3 1413
8b0d7497 1414 def check_filter():
d7b460d0 1415 if _type in ('playlist', 'multi_video'):
1416 return
1417 elif _type in ('url', 'url_transparent') and not try_call(
1418 lambda: self.get_info_extractor(info_dict['ie_key']).is_single_video(info_dict['url'])):
1419 return
1420
8b0d7497 1421 if 'title' in info_dict:
1422 # This can happen when we're just evaluating the playlist
1423 title = info_dict['title']
1424 matchtitle = self.params.get('matchtitle', False)
1425 if matchtitle:
1426 if not re.search(matchtitle, title, re.IGNORECASE):
1427 return '"' + title + '" title did not match pattern "' + matchtitle + '"'
1428 rejecttitle = self.params.get('rejecttitle', False)
1429 if rejecttitle:
1430 if re.search(rejecttitle, title, re.IGNORECASE):
1431 return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
6368e2e6 1432
8b0d7497 1433 date = info_dict.get('upload_date')
1434 if date is not None:
1435 dateRange = self.params.get('daterange', DateRange())
1436 if date not in dateRange:
86e5f3ed 1437 return f'{date_from_str(date).isoformat()} upload date is not in range {dateRange}'
8b0d7497 1438 view_count = info_dict.get('view_count')
1439 if view_count is not None:
1440 min_views = self.params.get('min_views')
1441 if min_views is not None and view_count < min_views:
1442 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
1443 max_views = self.params.get('max_views')
1444 if max_views is not None and view_count > max_views:
1445 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
1446 if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')):
1447 return 'Skipping "%s" because it is age restricted' % video_title
8b0d7497 1448
8f18aca8 1449 match_filter = self.params.get('match_filter')
fe2ce85a 1450 if match_filter is None:
1451 return None
1452
1453 cancelled = None
1454 try:
8f18aca8 1455 try:
1456 ret = match_filter(info_dict, incomplete=incomplete)
1457 except TypeError:
1458 # For backward compatibility
1459 ret = None if incomplete else match_filter(info_dict)
fe2ce85a 1460 except DownloadCancelled as err:
1461 if err.msg is not NO_DEFAULT:
1462 raise
1463 ret, cancelled = err.msg, err
1464
1465 if ret is NO_DEFAULT:
1466 while True:
1467 filename = self._format_screen(self.prepare_filename(info_dict), self.Styles.FILENAME)
1468 reply = input(self._format_screen(
1469 f'Download "{filename}"? (Y/n): ', self.Styles.EMPHASIS)).lower().strip()
1470 if reply in {'y', ''}:
1471 return None
1472 elif reply == 'n':
1473 if cancelled:
1474 raise type(cancelled)(f'Skipping {video_title}')
1475 return f'Skipping {video_title}'
1476 return ret
8b0d7497 1477
c77495e3 1478 if self.in_download_archive(info_dict):
1479 reason = '%s has already been recorded in the archive' % video_title
1480 break_opt, break_err = 'break_on_existing', ExistingVideoReached
1481 else:
fe2ce85a 1482 try:
1483 reason = check_filter()
1484 except DownloadCancelled as e:
1485 reason, break_opt, break_err = e.msg, 'match_filter', type(e)
1486 else:
1487 break_opt, break_err = 'break_on_reject', RejectedVideoReached
8b0d7497 1488 if reason is not None:
120fe513 1489 if not silent:
1490 self.to_screen('[download] ' + reason)
c77495e3 1491 if self.params.get(break_opt, False):
1492 raise break_err()
8b0d7497 1493 return reason
fe7e0c98 1494
b6c45014
JMF
1495 @staticmethod
1496 def add_extra_info(info_dict, extra_info):
1497 '''Set the keys from extra_info in info dict if they are missing'''
1498 for key, value in extra_info.items():
1499 info_dict.setdefault(key, value)
1500
409e1828 1501 def extract_info(self, url, download=True, ie_key=None, extra_info=None,
61aa5ba3 1502 process=True, force_generic_extractor=False):
41d1cca3 1503 """
17ffed18 1504 Extract and return the information dictionary of the URL
41d1cca3 1505
1506 Arguments:
17ffed18 1507 @param url URL to extract
41d1cca3 1508
1509 Keyword arguments:
17ffed18 1510 @param download Whether to download videos
1511 @param process Whether to resolve all unresolved references (URLs, playlist items).
1512 Must be True for download to work
1513 @param ie_key Use only the extractor with this key
1514
1515 @param extra_info Dictionary containing the extra values to add to the info (For internal use only)
1516 @force_generic_extractor Force using the generic extractor (Deprecated; use ie_key='Generic')
41d1cca3 1517 """
fe7e0c98 1518
409e1828 1519 if extra_info is None:
1520 extra_info = {}
1521
61aa5ba3 1522 if not ie_key and force_generic_extractor:
d22dec74
S
1523 ie_key = 'Generic'
1524
8222d8de 1525 if ie_key:
fe7866d0 1526 ies = {ie_key: self._ies[ie_key]} if ie_key in self._ies else {}
8222d8de
JMF
1527 else:
1528 ies = self._ies
1529
fe7866d0 1530 for key, ie in ies.items():
8222d8de
JMF
1531 if not ie.suitable(url):
1532 continue
1533
1534 if not ie.working():
6febd1c1
PH
1535 self.report_warning('The program functionality for this site has been marked as broken, '
1536 'and will probably not work.')
8222d8de 1537
1151c407 1538 temp_id = ie.get_temp_id(url)
fe7866d0 1539 if temp_id is not None and self.in_download_archive({'id': temp_id, 'ie_key': key}):
1540 self.to_screen(f'[{key}] {temp_id}: has already been recorded in the archive')
5e5be0c0 1541 if self.params.get('break_on_existing', False):
1542 raise ExistingVideoReached()
a0566bbf 1543 break
fe7866d0 1544 return self.__extract_info(url, self.get_info_extractor(key), download, extra_info, process)
a0566bbf 1545 else:
fe7866d0 1546 extractors_restricted = self.params.get('allowed_extractors') not in (None, ['default'])
1547 self.report_error(f'No suitable extractor{format_field(ie_key, None, " (%s)")} found for URL {url}',
1548 tb=False if extractors_restricted else None)
a0566bbf 1549
7e88d7d7 1550 def _handle_extraction_exceptions(func):
b5ae35ee 1551 @functools.wraps(func)
a0566bbf 1552 def wrapper(self, *args, **kwargs):
6da22e7d 1553 while True:
1554 try:
1555 return func(self, *args, **kwargs)
1556 except (DownloadCancelled, LazyList.IndexError, PagedList.IndexError):
8222d8de 1557 raise
6da22e7d 1558 except ReExtractInfo as e:
1559 if e.expected:
1560 self.to_screen(f'{e}; Re-extracting data')
1561 else:
1562 self.to_stderr('\r')
1563 self.report_warning(f'{e}; Re-extracting data')
1564 continue
1565 except GeoRestrictedError as e:
1566 msg = e.msg
1567 if e.countries:
1568 msg += '\nThis video is available in %s.' % ', '.join(
1569 map(ISO3166Utils.short2full, e.countries))
1570 msg += '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1571 self.report_error(msg)
1572 except ExtractorError as e: # An error we somewhat expected
1573 self.report_error(str(e), e.format_traceback())
1574 except Exception as e:
1575 if self.params.get('ignoreerrors'):
1576 self.report_error(str(e), tb=encode_compat_str(traceback.format_exc()))
1577 else:
1578 raise
1579 break
a0566bbf 1580 return wrapper
1581
693f0600 1582 def _wait_for_video(self, ie_result={}):
f2ebc5c7 1583 if (not self.params.get('wait_for_video')
1584 or ie_result.get('_type', 'video') != 'video'
1585 or ie_result.get('formats') or ie_result.get('url')):
1586 return
1587
1588 format_dur = lambda dur: '%02d:%02d:%02d' % timetuple_from_msec(dur * 1000)[:-1]
1589 last_msg = ''
1590
1591 def progress(msg):
1592 nonlocal last_msg
a7dc6a89 1593 full_msg = f'{msg}\n'
1594 if not self.params.get('noprogress'):
1595 full_msg = msg + ' ' * (len(last_msg) - len(msg)) + '\r'
1596 elif last_msg:
1597 return
1598 self.to_screen(full_msg, skip_eol=True)
f2ebc5c7 1599 last_msg = msg
1600
1601 min_wait, max_wait = self.params.get('wait_for_video')
1602 diff = try_get(ie_result, lambda x: x['release_timestamp'] - time.time())
1603 if diff is None and ie_result.get('live_status') == 'is_upcoming':
16c620bc 1604 diff = round(random.uniform(min_wait, max_wait) if (max_wait and min_wait) else (max_wait or min_wait), 0)
f2ebc5c7 1605 self.report_warning('Release time of video is not known')
693f0600 1606 elif ie_result and (diff or 0) <= 0:
f2ebc5c7 1607 self.report_warning('Video should already be available according to extracted info')
38d79fd1 1608 diff = min(max(diff or 0, min_wait or 0), max_wait or float('inf'))
f2ebc5c7 1609 self.to_screen(f'[wait] Waiting for {format_dur(diff)} - Press Ctrl+C to try now')
1610
1611 wait_till = time.time() + diff
1612 try:
1613 while True:
1614 diff = wait_till - time.time()
1615 if diff <= 0:
1616 progress('')
1617 raise ReExtractInfo('[wait] Wait period ended', expected=True)
1618 progress(f'[wait] Remaining time until next attempt: {self._format_screen(format_dur(diff), self.Styles.EMPHASIS)}')
1619 time.sleep(1)
1620 except KeyboardInterrupt:
1621 progress('')
1622 raise ReExtractInfo('[wait] Interrupted by user', expected=True)
1623 except BaseException as e:
1624 if not isinstance(e, ReExtractInfo):
1625 self.to_screen('')
1626 raise
1627
7e88d7d7 1628 @_handle_extraction_exceptions
58f197b7 1629 def __extract_info(self, url, ie, download, extra_info, process):
693f0600 1630 try:
1631 ie_result = ie.extract(url)
1632 except UserNotLive as e:
1633 if process:
1634 if self.params.get('wait_for_video'):
1635 self.report_warning(e)
1636 self._wait_for_video()
1637 raise
a0566bbf 1638 if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
cb794ee0 1639 self.report_warning(f'Extractor {ie.IE_NAME} returned nothing{bug_reports_message()}')
a0566bbf 1640 return
1641 if isinstance(ie_result, list):
1642 # Backwards compatibility: old IE result format
1643 ie_result = {
1644 '_type': 'compat_list',
1645 'entries': ie_result,
1646 }
e37d0efb 1647 if extra_info.get('original_url'):
1648 ie_result.setdefault('original_url', extra_info['original_url'])
a0566bbf 1649 self.add_default_extra_info(ie_result, ie, url)
1650 if process:
f2ebc5c7 1651 self._wait_for_video(ie_result)
a0566bbf 1652 return self.process_ie_result(ie_result, download, extra_info)
8222d8de 1653 else:
a0566bbf 1654 return ie_result
fe7e0c98 1655
ea38e55f 1656 def add_default_extra_info(self, ie_result, ie, url):
6033d980 1657 if url is not None:
1658 self.add_extra_info(ie_result, {
1659 'webpage_url': url,
1660 'original_url': url,
57ebfca3 1661 })
1662 webpage_url = ie_result.get('webpage_url')
1663 if webpage_url:
1664 self.add_extra_info(ie_result, {
1665 'webpage_url_basename': url_basename(webpage_url),
1666 'webpage_url_domain': get_domain(webpage_url),
6033d980 1667 })
1668 if ie is not None:
1669 self.add_extra_info(ie_result, {
1670 'extractor': ie.IE_NAME,
1671 'extractor_key': ie.ie_key(),
1672 })
ea38e55f 1673
58adec46 1674 def process_ie_result(self, ie_result, download=True, extra_info=None):
8222d8de
JMF
1675 """
1676 Take the result of the ie(may be modified) and resolve all unresolved
1677 references (URLs, playlist items).
1678
1679 It will also download the videos if 'download'.
1680 Returns the resolved ie_result.
1681 """
58adec46 1682 if extra_info is None:
1683 extra_info = {}
e8ee972c
PH
1684 result_type = ie_result.get('_type', 'video')
1685
057a5206 1686 if result_type in ('url', 'url_transparent'):
8f97a15d 1687 ie_result['url'] = sanitize_url(
1688 ie_result['url'], scheme='http' if self.params.get('prefer_insecure') else 'https')
8791e78c 1689 if ie_result.get('original_url') and not extra_info.get('original_url'):
1690 extra_info = {'original_url': ie_result['original_url'], **extra_info}
e37d0efb 1691
057a5206 1692 extract_flat = self.params.get('extract_flat', False)
3089bc74
S
1693 if ((extract_flat == 'in_playlist' and 'playlist' in extra_info)
1694 or extract_flat is True):
ecb54191 1695 info_copy = ie_result.copy()
6033d980 1696 ie = try_get(ie_result.get('ie_key'), self.get_info_extractor)
360167b9 1697 if ie and not ie_result.get('id'):
4614bc22 1698 info_copy['id'] = ie.get_temp_id(ie_result['url'])
6033d980 1699 self.add_default_extra_info(info_copy, ie, ie_result['url'])
4614bc22 1700 self.add_extra_info(info_copy, extra_info)
b5475f11 1701 info_copy, _ = self.pre_process(info_copy)
94dc8604 1702 self._fill_common_fields(info_copy, False)
17060584 1703 self.__forced_printings(info_copy)
415f8d51 1704 self._raise_pending_errors(info_copy)
4614bc22 1705 if self.params.get('force_write_download_archive', False):
1706 self.record_download_archive(info_copy)
e8ee972c
PH
1707 return ie_result
1708
8222d8de 1709 if result_type == 'video':
b6c45014 1710 self.add_extra_info(ie_result, extra_info)
9c2b75b5 1711 ie_result = self.process_video_result(ie_result, download=download)
415f8d51 1712 self._raise_pending_errors(ie_result)
28b0eb0f 1713 additional_urls = (ie_result or {}).get('additional_urls')
9c2b75b5 1714 if additional_urls:
e9f4ccd1 1715 # TODO: Improve MetadataParserPP to allow setting a list
14f25df2 1716 if isinstance(additional_urls, str):
9c2b75b5 1717 additional_urls = [additional_urls]
1718 self.to_screen(
1719 '[info] %s: %d additional URL(s) requested' % (ie_result['id'], len(additional_urls)))
1720 self.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls))
1721 ie_result['additional_entries'] = [
1722 self.extract_info(
b69fd25c 1723 url, download, extra_info=extra_info,
9c2b75b5 1724 force_generic_extractor=self.params.get('force_generic_extractor'))
1725 for url in additional_urls
1726 ]
1727 return ie_result
8222d8de
JMF
1728 elif result_type == 'url':
1729 # We have to add extra_info to the results because it may be
1730 # contained in a playlist
07cce701 1731 return self.extract_info(
1732 ie_result['url'], download,
1733 ie_key=ie_result.get('ie_key'),
1734 extra_info=extra_info)
7fc3fa05
PH
1735 elif result_type == 'url_transparent':
1736 # Use the information from the embedding page
1737 info = self.extract_info(
1738 ie_result['url'], ie_key=ie_result.get('ie_key'),
1739 extra_info=extra_info, download=False, process=False)
1740
1640eb09
S
1741 # extract_info may return None when ignoreerrors is enabled and
1742 # extraction failed with an error, don't crash and return early
1743 # in this case
1744 if not info:
1745 return info
1746
3975b4d2 1747 exempted_fields = {'_type', 'url', 'ie_key'}
1748 if not ie_result.get('section_end') and ie_result.get('section_start') is None:
1749 # For video clips, the id etc of the clip extractor should be used
1750 exempted_fields |= {'id', 'extractor', 'extractor_key'}
1751
412c617d 1752 new_result = info.copy()
3975b4d2 1753 new_result.update(filter_dict(ie_result, lambda k, v: v is not None and k not in exempted_fields))
7fc3fa05 1754
0563f7ac
S
1755 # Extracted info may not be a video result (i.e.
1756 # info.get('_type', 'video') != video) but rather an url or
1757 # url_transparent. In such cases outer metadata (from ie_result)
1758 # should be propagated to inner one (info). For this to happen
1759 # _type of info should be overridden with url_transparent. This
067aa17e 1760 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
0563f7ac
S
1761 if new_result.get('_type') == 'url':
1762 new_result['_type'] = 'url_transparent'
7fc3fa05
PH
1763
1764 return self.process_ie_result(
1765 new_result, download=download, extra_info=extra_info)
40fcba5e 1766 elif result_type in ('playlist', 'multi_video'):
30a074c2 1767 # Protect from infinite recursion due to recursively nested playlists
1768 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
0bd5a039 1769 webpage_url = ie_result.get('webpage_url') # Playlists maynot have webpage_url
1770 if webpage_url and webpage_url in self._playlist_urls:
7e85e872 1771 self.to_screen(
30a074c2 1772 '[download] Skipping already downloaded playlist: %s'
1773 % ie_result.get('title') or ie_result.get('id'))
1774 return
7e85e872 1775
30a074c2 1776 self._playlist_level += 1
1777 self._playlist_urls.add(webpage_url)
03f83004 1778 self._fill_common_fields(ie_result, False)
bc516a3f 1779 self._sanitize_thumbnails(ie_result)
30a074c2 1780 try:
1781 return self.__process_playlist(ie_result, download)
1782 finally:
1783 self._playlist_level -= 1
1784 if not self._playlist_level:
1785 self._playlist_urls.clear()
8222d8de 1786 elif result_type == 'compat_list':
c9bf4114
PH
1787 self.report_warning(
1788 'Extractor %s returned a compat_list result. '
1789 'It needs to be updated.' % ie_result.get('extractor'))
5f6a1245 1790
8222d8de 1791 def _fixup(r):
b868936c 1792 self.add_extra_info(r, {
1793 'extractor': ie_result['extractor'],
1794 'webpage_url': ie_result['webpage_url'],
1795 'webpage_url_basename': url_basename(ie_result['webpage_url']),
0bb322b9 1796 'webpage_url_domain': get_domain(ie_result['webpage_url']),
b868936c 1797 'extractor_key': ie_result['extractor_key'],
1798 })
8222d8de
JMF
1799 return r
1800 ie_result['entries'] = [
b6c45014 1801 self.process_ie_result(_fixup(r), download, extra_info)
8222d8de
JMF
1802 for r in ie_result['entries']
1803 ]
1804 return ie_result
1805 else:
1806 raise Exception('Invalid result type: %s' % result_type)
1807
e92caff5 1808 def _ensure_dir_exists(self, path):
1809 return make_dir(path, self.report_error)
1810
3b603dbd 1811 @staticmethod
3bec830a 1812 def _playlist_infodict(ie_result, strict=False, **kwargs):
1813 info = {
1814 'playlist_count': ie_result.get('playlist_count'),
3b603dbd 1815 'playlist': ie_result.get('title') or ie_result.get('id'),
1816 'playlist_id': ie_result.get('id'),
1817 'playlist_title': ie_result.get('title'),
1818 'playlist_uploader': ie_result.get('uploader'),
1819 'playlist_uploader_id': ie_result.get('uploader_id'),
3b603dbd 1820 **kwargs,
1821 }
3bec830a 1822 if strict:
1823 return info
0bd5a039 1824 if ie_result.get('webpage_url'):
1825 info.update({
1826 'webpage_url': ie_result['webpage_url'],
1827 'webpage_url_basename': url_basename(ie_result['webpage_url']),
1828 'webpage_url_domain': get_domain(ie_result['webpage_url']),
1829 })
3bec830a 1830 return {
1831 **info,
1832 'playlist_index': 0,
59d7de0d 1833 '__last_playlist_index': max(ie_result.get('requested_entries') or (0, 0)),
3bec830a 1834 'extractor': ie_result['extractor'],
3bec830a 1835 'extractor_key': ie_result['extractor_key'],
1836 }
3b603dbd 1837
30a074c2 1838 def __process_playlist(self, ie_result, download):
7e88d7d7 1839 """Process each entry in the playlist"""
f5ea4748 1840 assert ie_result['_type'] in ('playlist', 'multi_video')
1841
3bec830a 1842 common_info = self._playlist_infodict(ie_result, strict=True)
3955b207 1843 title = common_info.get('playlist') or '<Untitled>'
3bec830a 1844 if self._match_entry(common_info, incomplete=True) is not None:
1845 return
c6e07cf1 1846 self.to_screen(f'[download] Downloading {ie_result["_type"]}: {title}')
f0d785d3 1847
7e88d7d7 1848 all_entries = PlaylistEntries(self, ie_result)
7e9a6125 1849 entries = orderedSet(all_entries.get_requested_items(), lazy=True)
1850
1851 lazy = self.params.get('lazy_playlist')
1852 if lazy:
1853 resolved_entries, n_entries = [], 'N/A'
1854 ie_result['requested_entries'], ie_result['entries'] = None, None
1855 else:
1856 entries = resolved_entries = list(entries)
1857 n_entries = len(resolved_entries)
1858 ie_result['requested_entries'], ie_result['entries'] = tuple(zip(*resolved_entries)) or ([], [])
1859 if not ie_result.get('playlist_count'):
1860 # Better to do this after potentially exhausting entries
1861 ie_result['playlist_count'] = all_entries.get_full_count()
498f5606 1862
0647d925 1863 extra = self._playlist_infodict(ie_result, n_entries=int_or_none(n_entries))
1864 ie_copy = collections.ChainMap(ie_result, extra)
3bec830a 1865
e08a85d8 1866 _infojson_written = False
0bfc53d0 1867 write_playlist_files = self.params.get('allow_playlist_files', True)
1868 if write_playlist_files and self.params.get('list_thumbnails'):
1869 self.list_thumbnails(ie_result)
1870 if write_playlist_files and not self.params.get('simulate'):
e08a85d8 1871 _infojson_written = self._write_info_json(
1872 'playlist', ie_result, self.prepare_filename(ie_copy, 'pl_infojson'))
1873 if _infojson_written is None:
80c03fa9 1874 return
1875 if self._write_description('playlist', ie_result,
1876 self.prepare_filename(ie_copy, 'pl_description')) is None:
1877 return
681de68e 1878 # TODO: This should be passed to ThumbnailsConvertor if necessary
3bec830a 1879 self._write_thumbnails('playlist', ie_result, self.prepare_filename(ie_copy, 'pl_thumbnail'))
30a074c2 1880
7e9a6125 1881 if lazy:
1882 if self.params.get('playlistreverse') or self.params.get('playlistrandom'):
1883 self.report_warning('playlistreverse and playlistrandom are not supported with lazy_playlist', only_once=True)
1884 elif self.params.get('playlistreverse'):
1885 entries.reverse()
1886 elif self.params.get('playlistrandom'):
30a074c2 1887 random.shuffle(entries)
1888
bc5c2f8a 1889 self.to_screen(f'[{ie_result["extractor"]}] Playlist {title}: Downloading {n_entries} items'
7e88d7d7 1890 f'{format_field(ie_result, "playlist_count", " of %s")}')
30a074c2 1891
134c913c 1892 keep_resolved_entries = self.params.get('extract_flat') != 'discard'
1893 if self.params.get('extract_flat') == 'discard_in_playlist':
1894 keep_resolved_entries = ie_result['_type'] != 'playlist'
1895 if keep_resolved_entries:
1896 self.write_debug('The information of all playlist entries will be held in memory')
1897
26e2805c 1898 failures = 0
1899 max_failures = self.params.get('skip_playlist_after_errors') or float('inf')
7e9a6125 1900 for i, (playlist_index, entry) in enumerate(entries):
1901 if lazy:
1902 resolved_entries.append((playlist_index, entry))
3bec830a 1903 if not entry:
7e88d7d7 1904 continue
1905
7e88d7d7 1906 entry['__x_forwarded_for_ip'] = ie_result.get('__x_forwarded_for_ip')
ad54c913 1907 if not lazy and 'playlist-index' in self.params['compat_opts']:
7e9a6125 1908 playlist_index = ie_result['requested_entries'][i]
1909
0647d925 1910 entry_copy = collections.ChainMap(entry, {
3bec830a 1911 **common_info,
3955b207 1912 'n_entries': int_or_none(n_entries),
71729754 1913 'playlist_index': playlist_index,
7e9a6125 1914 'playlist_autonumber': i + 1,
0647d925 1915 })
3bec830a 1916
0647d925 1917 if self._match_entry(entry_copy, incomplete=True) is not None:
f0ad6f8c 1918 # For compatabilty with youtube-dl. See https://github.com/yt-dlp/yt-dlp/issues/4369
1919 resolved_entries[i] = (playlist_index, NO_DEFAULT)
3bec830a 1920 continue
1921
bc5c2f8a 1922 self.to_screen('[download] Downloading item %s of %s' % (
3bec830a 1923 self._format_screen(i + 1, self.Styles.ID), self._format_screen(n_entries, self.Styles.EMPHASIS)))
1924
ec54bd43 1925 entry_result = self.__process_iterable_entry(entry, download, collections.ChainMap({
a6ca61d4 1926 'playlist_index': playlist_index,
1927 'playlist_autonumber': i + 1,
ec54bd43 1928 }, extra))
26e2805c 1929 if not entry_result:
1930 failures += 1
1931 if failures >= max_failures:
1932 self.report_error(
7e88d7d7 1933 f'Skipping the remaining entries in playlist "{title}" since {failures} items failed extraction')
26e2805c 1934 break
134c913c 1935 if keep_resolved_entries:
1936 resolved_entries[i] = (playlist_index, entry_result)
7e88d7d7 1937
1938 # Update with processed data
f0ad6f8c 1939 ie_result['entries'] = [e for _, e in resolved_entries if e is not NO_DEFAULT]
bc5c2f8a 1940 ie_result['requested_entries'] = [i for i, e in resolved_entries if e is not NO_DEFAULT]
1941 if ie_result['requested_entries'] == try_call(lambda: list(range(1, ie_result['playlist_count'] + 1))):
1942 # Do not set for full playlist
1943 ie_result.pop('requested_entries')
e08a85d8 1944
1945 # Write the updated info to json
cb96c5be 1946 if _infojson_written is True and self._write_info_json(
e08a85d8 1947 'updated playlist', ie_result,
1948 self.prepare_filename(ie_copy, 'pl_infojson'), overwrite=True) is None:
1949 return
ca30f449 1950
ed5835b4 1951 ie_result = self.run_all_pps('playlist', ie_result)
7e88d7d7 1952 self.to_screen(f'[download] Finished downloading playlist: {title}')
30a074c2 1953 return ie_result
1954
7e88d7d7 1955 @_handle_extraction_exceptions
a0566bbf 1956 def __process_iterable_entry(self, entry, download, extra_info):
1957 return self.process_ie_result(
1958 entry, download=download, extra_info=extra_info)
1959
67134eab
JMF
1960 def _build_format_filter(self, filter_spec):
1961 " Returns a function to filter the formats according to the filter_spec "
083c9df9
PH
1962
1963 OPERATORS = {
1964 '<': operator.lt,
1965 '<=': operator.le,
1966 '>': operator.gt,
1967 '>=': operator.ge,
1968 '=': operator.eq,
1969 '!=': operator.ne,
1970 }
67134eab 1971 operator_rex = re.compile(r'''(?x)\s*
c3f624ef 1972 (?P<key>[\w.-]+)\s*
187986a8 1973 (?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1974 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)\s*
083c9df9 1975 ''' % '|'.join(map(re.escape, OPERATORS.keys())))
187986a8 1976 m = operator_rex.fullmatch(filter_spec)
9ddb6925
S
1977 if m:
1978 try:
1979 comparison_value = int(m.group('value'))
1980 except ValueError:
1981 comparison_value = parse_filesize(m.group('value'))
1982 if comparison_value is None:
1983 comparison_value = parse_filesize(m.group('value') + 'B')
1984 if comparison_value is None:
1985 raise ValueError(
1986 'Invalid value %r in format specification %r' % (
67134eab 1987 m.group('value'), filter_spec))
9ddb6925
S
1988 op = OPERATORS[m.group('op')]
1989
083c9df9 1990 if not m:
9ddb6925
S
1991 STR_OPERATORS = {
1992 '=': operator.eq,
10d33b34
YCH
1993 '^=': lambda attr, value: attr.startswith(value),
1994 '$=': lambda attr, value: attr.endswith(value),
1995 '*=': lambda attr, value: value in attr,
1ce9a3cb 1996 '~=': lambda attr, value: value.search(attr) is not None
9ddb6925 1997 }
187986a8 1998 str_operator_rex = re.compile(r'''(?x)\s*
1999 (?P<key>[a-zA-Z0-9._-]+)\s*
1ce9a3cb
LF
2000 (?P<negation>!\s*)?(?P<op>%s)\s*(?P<none_inclusive>\?\s*)?
2001 (?P<quote>["'])?
2002 (?P<value>(?(quote)(?:(?!(?P=quote))[^\\]|\\.)+|[\w.-]+))
2003 (?(quote)(?P=quote))\s*
9ddb6925 2004 ''' % '|'.join(map(re.escape, STR_OPERATORS.keys())))
187986a8 2005 m = str_operator_rex.fullmatch(filter_spec)
9ddb6925 2006 if m:
1ce9a3cb
LF
2007 if m.group('op') == '~=':
2008 comparison_value = re.compile(m.group('value'))
2009 else:
2010 comparison_value = re.sub(r'''\\([\\"'])''', r'\1', m.group('value'))
2cc779f4
S
2011 str_op = STR_OPERATORS[m.group('op')]
2012 if m.group('negation'):
e118a879 2013 op = lambda attr, value: not str_op(attr, value)
2cc779f4
S
2014 else:
2015 op = str_op
083c9df9 2016
9ddb6925 2017 if not m:
187986a8 2018 raise SyntaxError('Invalid filter specification %r' % filter_spec)
083c9df9
PH
2019
2020 def _filter(f):
2021 actual_value = f.get(m.group('key'))
2022 if actual_value is None:
2023 return m.group('none_inclusive')
2024 return op(actual_value, comparison_value)
67134eab
JMF
2025 return _filter
2026
9f1a1c36 2027 def _check_formats(self, formats):
2028 for f in formats:
2029 self.to_screen('[info] Testing format %s' % f['format_id'])
75689fe5 2030 path = self.get_output_path('temp')
2031 if not self._ensure_dir_exists(f'{path}/'):
2032 continue
2033 temp_file = tempfile.NamedTemporaryFile(suffix='.tmp', delete=False, dir=path or None)
9f1a1c36 2034 temp_file.close()
2035 try:
2036 success, _ = self.dl(temp_file.name, f, test=True)
8a82af35 2037 except (DownloadError, OSError, ValueError) + network_exceptions:
9f1a1c36 2038 success = False
2039 finally:
2040 if os.path.exists(temp_file.name):
2041 try:
2042 os.remove(temp_file.name)
2043 except OSError:
2044 self.report_warning('Unable to delete temporary file "%s"' % temp_file.name)
2045 if success:
2046 yield f
2047 else:
2048 self.to_screen('[info] Unable to download format %s. Skipping...' % f['format_id'])
2049
0017d9ad 2050 def _default_format_spec(self, info_dict, download=True):
0017d9ad 2051
af0f7428
S
2052 def can_merge():
2053 merger = FFmpegMergerPP(self)
2054 return merger.available and merger.can_merge()
2055
91ebc640 2056 prefer_best = (
b7b04c78 2057 not self.params.get('simulate')
91ebc640 2058 and download
2059 and (
2060 not can_merge()
21633673 2061 or info_dict.get('is_live') and not self.params.get('live_from_start')
bf1824b3 2062 or self.params['outtmpl']['default'] == '-'))
53ed7066 2063 compat = (
2064 prefer_best
2065 or self.params.get('allow_multiple_audio_streams', False)
8a82af35 2066 or 'format-spec' in self.params['compat_opts'])
91ebc640 2067
2068 return (
53ed7066 2069 'best/bestvideo+bestaudio' if prefer_best
2070 else 'bestvideo*+bestaudio/best' if not compat
91ebc640 2071 else 'bestvideo+bestaudio/best')
0017d9ad 2072
67134eab
JMF
2073 def build_format_selector(self, format_spec):
2074 def syntax_error(note, start):
2075 message = (
2076 'Invalid format specification: '
86e5f3ed 2077 '{}\n\t{}\n\t{}^'.format(note, format_spec, ' ' * start[1]))
67134eab
JMF
2078 return SyntaxError(message)
2079
2080 PICKFIRST = 'PICKFIRST'
2081 MERGE = 'MERGE'
2082 SINGLE = 'SINGLE'
0130afb7 2083 GROUP = 'GROUP'
67134eab
JMF
2084 FormatSelector = collections.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
2085
91ebc640 2086 allow_multiple_streams = {'audio': self.params.get('allow_multiple_audio_streams', False),
2087 'video': self.params.get('allow_multiple_video_streams', False)}
909d24dd 2088
9f1a1c36 2089 check_formats = self.params.get('check_formats') == 'selected'
e8e73840 2090
67134eab
JMF
2091 def _parse_filter(tokens):
2092 filter_parts = []
6f2287cb 2093 for type, string_, start, _, _ in tokens:
2094 if type == tokenize.OP and string_ == ']':
67134eab
JMF
2095 return ''.join(filter_parts)
2096 else:
6f2287cb 2097 filter_parts.append(string_)
67134eab 2098
232541df 2099 def _remove_unused_ops(tokens):
62b58c09
L
2100 # Remove operators that we don't use and join them with the surrounding strings.
2101 # E.g. 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
232541df
JMF
2102 ALLOWED_OPS = ('/', '+', ',', '(', ')')
2103 last_string, last_start, last_end, last_line = None, None, None, None
6f2287cb 2104 for type, string_, start, end, line in tokens:
2105 if type == tokenize.OP and string_ == '[':
232541df
JMF
2106 if last_string:
2107 yield tokenize.NAME, last_string, last_start, last_end, last_line
2108 last_string = None
6f2287cb 2109 yield type, string_, start, end, line
232541df 2110 # everything inside brackets will be handled by _parse_filter
6f2287cb 2111 for type, string_, start, end, line in tokens:
2112 yield type, string_, start, end, line
2113 if type == tokenize.OP and string_ == ']':
232541df 2114 break
6f2287cb 2115 elif type == tokenize.OP and string_ in ALLOWED_OPS:
232541df
JMF
2116 if last_string:
2117 yield tokenize.NAME, last_string, last_start, last_end, last_line
2118 last_string = None
6f2287cb 2119 yield type, string_, start, end, line
232541df
JMF
2120 elif type in [tokenize.NAME, tokenize.NUMBER, tokenize.OP]:
2121 if not last_string:
6f2287cb 2122 last_string = string_
232541df
JMF
2123 last_start = start
2124 last_end = end
2125 else:
6f2287cb 2126 last_string += string_
232541df
JMF
2127 if last_string:
2128 yield tokenize.NAME, last_string, last_start, last_end, last_line
2129
cf2ac6df 2130 def _parse_format_selection(tokens, inside_merge=False, inside_choice=False, inside_group=False):
67134eab
JMF
2131 selectors = []
2132 current_selector = None
6f2287cb 2133 for type, string_, start, _, _ in tokens:
67134eab
JMF
2134 # ENCODING is only defined in python 3.x
2135 if type == getattr(tokenize, 'ENCODING', None):
2136 continue
2137 elif type in [tokenize.NAME, tokenize.NUMBER]:
6f2287cb 2138 current_selector = FormatSelector(SINGLE, string_, [])
67134eab 2139 elif type == tokenize.OP:
6f2287cb 2140 if string_ == ')':
cf2ac6df
JMF
2141 if not inside_group:
2142 # ')' will be handled by the parentheses group
2143 tokens.restore_last_token()
67134eab 2144 break
6f2287cb 2145 elif inside_merge and string_ in ['/', ',']:
0130afb7
JMF
2146 tokens.restore_last_token()
2147 break
6f2287cb 2148 elif inside_choice and string_ == ',':
cf2ac6df
JMF
2149 tokens.restore_last_token()
2150 break
6f2287cb 2151 elif string_ == ',':
0a31a350
JMF
2152 if not current_selector:
2153 raise syntax_error('"," must follow a format selector', start)
67134eab
JMF
2154 selectors.append(current_selector)
2155 current_selector = None
6f2287cb 2156 elif string_ == '/':
d96d604e
JMF
2157 if not current_selector:
2158 raise syntax_error('"/" must follow a format selector', start)
67134eab 2159 first_choice = current_selector
cf2ac6df 2160 second_choice = _parse_format_selection(tokens, inside_choice=True)
f5f4a27a 2161 current_selector = FormatSelector(PICKFIRST, (first_choice, second_choice), [])
6f2287cb 2162 elif string_ == '[':
67134eab
JMF
2163 if not current_selector:
2164 current_selector = FormatSelector(SINGLE, 'best', [])
2165 format_filter = _parse_filter(tokens)
2166 current_selector.filters.append(format_filter)
6f2287cb 2167 elif string_ == '(':
0130afb7
JMF
2168 if current_selector:
2169 raise syntax_error('Unexpected "("', start)
cf2ac6df
JMF
2170 group = _parse_format_selection(tokens, inside_group=True)
2171 current_selector = FormatSelector(GROUP, group, [])
6f2287cb 2172 elif string_ == '+':
d03cfdce 2173 if not current_selector:
2174 raise syntax_error('Unexpected "+"', start)
2175 selector_1 = current_selector
2176 selector_2 = _parse_format_selection(tokens, inside_merge=True)
2177 if not selector_2:
2178 raise syntax_error('Expected a selector', start)
2179 current_selector = FormatSelector(MERGE, (selector_1, selector_2), [])
67134eab 2180 else:
6f2287cb 2181 raise syntax_error(f'Operator not recognized: "{string_}"', start)
67134eab
JMF
2182 elif type == tokenize.ENDMARKER:
2183 break
2184 if current_selector:
2185 selectors.append(current_selector)
2186 return selectors
2187
f8d4ad9a 2188 def _merge(formats_pair):
2189 format_1, format_2 = formats_pair
2190
2191 formats_info = []
2192 formats_info.extend(format_1.get('requested_formats', (format_1,)))
2193 formats_info.extend(format_2.get('requested_formats', (format_2,)))
2194
2195 if not allow_multiple_streams['video'] or not allow_multiple_streams['audio']:
551f9388 2196 get_no_more = {'video': False, 'audio': False}
f8d4ad9a 2197 for (i, fmt_info) in enumerate(formats_info):
551f9388 2198 if fmt_info.get('acodec') == fmt_info.get('vcodec') == 'none':
2199 formats_info.pop(i)
2200 continue
2201 for aud_vid in ['audio', 'video']:
f8d4ad9a 2202 if not allow_multiple_streams[aud_vid] and fmt_info.get(aud_vid[0] + 'codec') != 'none':
2203 if get_no_more[aud_vid]:
2204 formats_info.pop(i)
f5510afe 2205 break
f8d4ad9a 2206 get_no_more[aud_vid] = True
2207
2208 if len(formats_info) == 1:
2209 return formats_info[0]
2210
2211 video_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('vcodec') != 'none']
2212 audio_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('acodec') != 'none']
2213
2214 the_only_video = video_fmts[0] if len(video_fmts) == 1 else None
2215 the_only_audio = audio_fmts[0] if len(audio_fmts) == 1 else None
2216
fc61aff4
LL
2217 output_ext = get_compatible_ext(
2218 vcodecs=[f.get('vcodec') for f in video_fmts],
2219 acodecs=[f.get('acodec') for f in audio_fmts],
2220 vexts=[f['ext'] for f in video_fmts],
2221 aexts=[f['ext'] for f in audio_fmts],
2222 preferences=(try_call(lambda: self.params['merge_output_format'].split('/'))
2223 or self.params.get('prefer_free_formats') and ('webm', 'mkv')))
f8d4ad9a 2224
975a0d0d 2225 filtered = lambda *keys: filter(None, (traverse_obj(fmt, *keys) for fmt in formats_info))
2226
f8d4ad9a 2227 new_dict = {
2228 'requested_formats': formats_info,
975a0d0d 2229 'format': '+'.join(filtered('format')),
2230 'format_id': '+'.join(filtered('format_id')),
f8d4ad9a 2231 'ext': output_ext,
975a0d0d 2232 'protocol': '+'.join(map(determine_protocol, formats_info)),
093a1710 2233 'language': '+'.join(orderedSet(filtered('language'))) or None,
2234 'format_note': '+'.join(orderedSet(filtered('format_note'))) or None,
2235 'filesize_approx': sum(filtered('filesize', 'filesize_approx')) or None,
975a0d0d 2236 'tbr': sum(filtered('tbr', 'vbr', 'abr')),
f8d4ad9a 2237 }
2238
2239 if the_only_video:
2240 new_dict.update({
2241 'width': the_only_video.get('width'),
2242 'height': the_only_video.get('height'),
2243 'resolution': the_only_video.get('resolution') or self.format_resolution(the_only_video),
2244 'fps': the_only_video.get('fps'),
49a57e70 2245 'dynamic_range': the_only_video.get('dynamic_range'),
f8d4ad9a 2246 'vcodec': the_only_video.get('vcodec'),
2247 'vbr': the_only_video.get('vbr'),
2248 'stretched_ratio': the_only_video.get('stretched_ratio'),
105bfd90 2249 'aspect_ratio': the_only_video.get('aspect_ratio'),
f8d4ad9a 2250 })
2251
2252 if the_only_audio:
2253 new_dict.update({
2254 'acodec': the_only_audio.get('acodec'),
2255 'abr': the_only_audio.get('abr'),
975a0d0d 2256 'asr': the_only_audio.get('asr'),
b8ed0f15 2257 'audio_channels': the_only_audio.get('audio_channels')
f8d4ad9a 2258 })
2259
2260 return new_dict
2261
e8e73840 2262 def _check_formats(formats):
981052c9 2263 if not check_formats:
2264 yield from formats
b5ac45b1 2265 return
9f1a1c36 2266 yield from self._check_formats(formats)
e8e73840 2267
67134eab 2268 def _build_selector_function(selector):
909d24dd 2269 if isinstance(selector, list): # ,
67134eab
JMF
2270 fs = [_build_selector_function(s) for s in selector]
2271
317f7ab6 2272 def selector_function(ctx):
67134eab 2273 for f in fs:
981052c9 2274 yield from f(ctx)
67134eab 2275 return selector_function
909d24dd 2276
2277 elif selector.type == GROUP: # ()
0130afb7 2278 selector_function = _build_selector_function(selector.selector)
909d24dd 2279
2280 elif selector.type == PICKFIRST: # /
67134eab
JMF
2281 fs = [_build_selector_function(s) for s in selector.selector]
2282
317f7ab6 2283 def selector_function(ctx):
67134eab 2284 for f in fs:
317f7ab6 2285 picked_formats = list(f(ctx))
67134eab
JMF
2286 if picked_formats:
2287 return picked_formats
2288 return []
67134eab 2289
981052c9 2290 elif selector.type == MERGE: # +
2291 selector_1, selector_2 = map(_build_selector_function, selector.selector)
2292
2293 def selector_function(ctx):
adbc4ec4 2294 for pair in itertools.product(selector_1(ctx), selector_2(ctx)):
981052c9 2295 yield _merge(pair)
2296
909d24dd 2297 elif selector.type == SINGLE: # atom
598d185d 2298 format_spec = selector.selector or 'best'
909d24dd 2299
f8d4ad9a 2300 # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
909d24dd 2301 if format_spec == 'all':
2302 def selector_function(ctx):
9222c381 2303 yield from _check_formats(ctx['formats'][::-1])
f8d4ad9a 2304 elif format_spec == 'mergeall':
2305 def selector_function(ctx):
316f2650 2306 formats = list(_check_formats(
2307 f for f in ctx['formats'] if f.get('vcodec') != 'none' or f.get('acodec') != 'none'))
e01d6aa4 2308 if not formats:
2309 return
921b76ca 2310 merged_format = formats[-1]
2311 for f in formats[-2::-1]:
f8d4ad9a 2312 merged_format = _merge((merged_format, f))
2313 yield merged_format
909d24dd 2314
2315 else:
85e801a9 2316 format_fallback, seperate_fallback, format_reverse, format_idx = False, None, True, 1
eff63539 2317 mobj = re.match(
2318 r'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
2319 format_spec)
2320 if mobj is not None:
2321 format_idx = int_or_none(mobj.group('n'), default=1)
e8e73840 2322 format_reverse = mobj.group('bw')[0] == 'b'
eff63539 2323 format_type = (mobj.group('type') or [None])[0]
2324 not_format_type = {'v': 'a', 'a': 'v'}.get(format_type)
2325 format_modified = mobj.group('mod') is not None
909d24dd 2326
2327 format_fallback = not format_type and not format_modified # for b, w
8326b00a 2328 _filter_f = (
eff63539 2329 (lambda f: f.get('%scodec' % format_type) != 'none')
2330 if format_type and format_modified # bv*, ba*, wv*, wa*
2331 else (lambda f: f.get('%scodec' % not_format_type) == 'none')
2332 if format_type # bv, ba, wv, wa
2333 else (lambda f: f.get('vcodec') != 'none' and f.get('acodec') != 'none')
2334 if not format_modified # b, w
8326b00a 2335 else lambda f: True) # b*, w*
2336 filter_f = lambda f: _filter_f(f) and (
2337 f.get('vcodec') != 'none' or f.get('acodec') != 'none')
67134eab 2338 else:
48ee10ee 2339 if format_spec in self._format_selection_exts['audio']:
b11c04a8 2340 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none'
48ee10ee 2341 elif format_spec in self._format_selection_exts['video']:
b11c04a8 2342 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none' and f.get('vcodec') != 'none'
85e801a9 2343 seperate_fallback = lambda f: f.get('ext') == format_spec and f.get('vcodec') != 'none'
48ee10ee 2344 elif format_spec in self._format_selection_exts['storyboards']:
b11c04a8 2345 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') == 'none' and f.get('vcodec') == 'none'
2346 else:
b5ae35ee 2347 filter_f = lambda f: f.get('format_id') == format_spec # id
909d24dd 2348
2349 def selector_function(ctx):
2350 formats = list(ctx['formats'])
909d24dd 2351 matches = list(filter(filter_f, formats)) if filter_f is not None else formats
85e801a9 2352 if not matches:
2353 if format_fallback and ctx['incomplete_formats']:
2354 # for extractors with incomplete formats (audio only (soundcloud)
2355 # or video only (imgur)) best/worst will fallback to
2356 # best/worst {video,audio}-only format
2357 matches = formats
2358 elif seperate_fallback and not ctx['has_merged_format']:
2359 # for compatibility with youtube-dl when there is no pre-merged format
2360 matches = list(filter(seperate_fallback, formats))
981052c9 2361 matches = LazyList(_check_formats(matches[::-1 if format_reverse else 1]))
2362 try:
e8e73840 2363 yield matches[format_idx - 1]
4abea8ca 2364 except LazyList.IndexError:
981052c9 2365 return
083c9df9 2366
67134eab 2367 filters = [self._build_format_filter(f) for f in selector.filters]
083c9df9 2368
317f7ab6 2369 def final_selector(ctx):
adbc4ec4 2370 ctx_copy = dict(ctx)
67134eab 2371 for _filter in filters:
317f7ab6
S
2372 ctx_copy['formats'] = list(filter(_filter, ctx_copy['formats']))
2373 return selector_function(ctx_copy)
67134eab 2374 return final_selector
083c9df9 2375
0f06bcd7 2376 stream = io.BytesIO(format_spec.encode())
0130afb7 2377 try:
f9934b96 2378 tokens = list(_remove_unused_ops(tokenize.tokenize(stream.readline)))
0130afb7
JMF
2379 except tokenize.TokenError:
2380 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec)))
2381
86e5f3ed 2382 class TokenIterator:
0130afb7
JMF
2383 def __init__(self, tokens):
2384 self.tokens = tokens
2385 self.counter = 0
2386
2387 def __iter__(self):
2388 return self
2389
2390 def __next__(self):
2391 if self.counter >= len(self.tokens):
2392 raise StopIteration()
2393 value = self.tokens[self.counter]
2394 self.counter += 1
2395 return value
2396
2397 next = __next__
2398
2399 def restore_last_token(self):
2400 self.counter -= 1
2401
2402 parsed_selector = _parse_format_selection(iter(TokenIterator(tokens)))
67134eab 2403 return _build_selector_function(parsed_selector)
a9c58ad9 2404
e5660ee6 2405 def _calc_headers(self, info_dict):
8b7539d2 2406 res = merge_headers(self.params['http_headers'], info_dict.get('http_headers') or {})
955c8958 2407 if 'Youtubedl-No-Compression' in res: # deprecated
2408 res.pop('Youtubedl-No-Compression', None)
2409 res['Accept-Encoding'] = 'identity'
b87e01c1 2410 cookies = self.cookiejar.get_cookie_header(info_dict['url'])
e5660ee6
JMF
2411 if cookies:
2412 res['Cookie'] = cookies
2413
0016b84e
S
2414 if 'X-Forwarded-For' not in res:
2415 x_forwarded_for_ip = info_dict.get('__x_forwarded_for_ip')
2416 if x_forwarded_for_ip:
2417 res['X-Forwarded-For'] = x_forwarded_for_ip
2418
e5660ee6
JMF
2419 return res
2420
c487cf00 2421 def _calc_cookies(self, url):
b87e01c1 2422 self.deprecation_warning('"YoutubeDL._calc_cookies" is deprecated and may be removed in a future version')
2423 return self.cookiejar.get_cookie_header(url)
e5660ee6 2424
9f1a1c36 2425 def _sort_thumbnails(self, thumbnails):
2426 thumbnails.sort(key=lambda t: (
2427 t.get('preference') if t.get('preference') is not None else -1,
2428 t.get('width') if t.get('width') is not None else -1,
2429 t.get('height') if t.get('height') is not None else -1,
2430 t.get('id') if t.get('id') is not None else '',
2431 t.get('url')))
2432
b0249bca 2433 def _sanitize_thumbnails(self, info_dict):
bc516a3f 2434 thumbnails = info_dict.get('thumbnails')
2435 if thumbnails is None:
2436 thumbnail = info_dict.get('thumbnail')
2437 if thumbnail:
2438 info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}]
9f1a1c36 2439 if not thumbnails:
2440 return
2441
2442 def check_thumbnails(thumbnails):
2443 for t in thumbnails:
2444 self.to_screen(f'[info] Testing thumbnail {t["id"]}')
2445 try:
2446 self.urlopen(HEADRequest(t['url']))
2447 except network_exceptions as err:
2448 self.to_screen(f'[info] Unable to connect to thumbnail {t["id"]} URL {t["url"]!r} - {err}. Skipping...')
2449 continue
2450 yield t
2451
2452 self._sort_thumbnails(thumbnails)
2453 for i, t in enumerate(thumbnails):
2454 if t.get('id') is None:
2455 t['id'] = '%d' % i
2456 if t.get('width') and t.get('height'):
2457 t['resolution'] = '%dx%d' % (t['width'], t['height'])
2458 t['url'] = sanitize_url(t['url'])
2459
2460 if self.params.get('check_formats') is True:
282f5709 2461 info_dict['thumbnails'] = LazyList(check_thumbnails(thumbnails[::-1]), reverse=True)
9f1a1c36 2462 else:
2463 info_dict['thumbnails'] = thumbnails
bc516a3f 2464
94dc8604 2465 def _fill_common_fields(self, info_dict, final=True):
03f83004 2466 # TODO: move sanitization here
94dc8604 2467 if final:
7aefd19a 2468 title = info_dict['fulltitle'] = info_dict.get('title')
d4736fdb 2469 if not title:
2470 if title == '':
2471 self.write_debug('Extractor gave empty title. Creating a generic title')
2472 else:
2473 self.report_warning('Extractor failed to obtain "title". Creating a generic title instead')
1d485a1a 2474 info_dict['title'] = f'{info_dict["extractor"].replace(":", "-")} video #{info_dict["id"]}'
03f83004
LNO
2475
2476 if info_dict.get('duration') is not None:
2477 info_dict['duration_string'] = formatSeconds(info_dict['duration'])
2478
2479 for ts_key, date_key in (
2480 ('timestamp', 'upload_date'),
2481 ('release_timestamp', 'release_date'),
2482 ('modified_timestamp', 'modified_date'),
2483 ):
2484 if info_dict.get(date_key) is None and info_dict.get(ts_key) is not None:
2485 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
2486 # see http://bugs.python.org/issue1646728)
19a03940 2487 with contextlib.suppress(ValueError, OverflowError, OSError):
03f83004
LNO
2488 upload_date = datetime.datetime.utcfromtimestamp(info_dict[ts_key])
2489 info_dict[date_key] = upload_date.strftime('%Y%m%d')
03f83004
LNO
2490
2491 live_keys = ('is_live', 'was_live')
2492 live_status = info_dict.get('live_status')
2493 if live_status is None:
2494 for key in live_keys:
2495 if info_dict.get(key) is False:
2496 continue
2497 if info_dict.get(key):
2498 live_status = key
2499 break
2500 if all(info_dict.get(key) is False for key in live_keys):
2501 live_status = 'not_live'
2502 if live_status:
2503 info_dict['live_status'] = live_status
2504 for key in live_keys:
2505 if info_dict.get(key) is None:
2506 info_dict[key] = (live_status == key)
a057779d 2507 if live_status == 'post_live':
2508 info_dict['was_live'] = True
03f83004
LNO
2509
2510 # Auto generate title fields corresponding to the *_number fields when missing
2511 # in order to always have clean titles. This is very common for TV series.
2512 for field in ('chapter', 'season', 'episode'):
94dc8604 2513 if final and info_dict.get('%s_number' % field) is not None and not info_dict.get(field):
03f83004
LNO
2514 info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field])
2515
415f8d51 2516 def _raise_pending_errors(self, info):
2517 err = info.pop('__pending_error', None)
2518 if err:
2519 self.report_error(err, tb=False)
2520
784320c9 2521 def sort_formats(self, info_dict):
2522 formats = self._get_formats(info_dict)
784320c9 2523 formats.sort(key=FormatSorter(
c154302c 2524 self, info_dict.get('_format_sort_fields') or []).calculate_preference)
784320c9 2525
dd82ffea
JMF
2526 def process_video_result(self, info_dict, download=True):
2527 assert info_dict.get('_type', 'video') == 'video'
9c906919 2528 self._num_videos += 1
dd82ffea 2529
bec1fad2 2530 if 'id' not in info_dict:
fc08bdd6 2531 raise ExtractorError('Missing "id" field in extractor result', ie=info_dict['extractor'])
2532 elif not info_dict.get('id'):
2533 raise ExtractorError('Extractor failed to obtain "id"', ie=info_dict['extractor'])
455a15e2 2534
c9969434
S
2535 def report_force_conversion(field, field_not, conversion):
2536 self.report_warning(
2537 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
2538 % (field, field_not, conversion))
2539
2540 def sanitize_string_field(info, string_field):
2541 field = info.get(string_field)
14f25df2 2542 if field is None or isinstance(field, str):
c9969434
S
2543 return
2544 report_force_conversion(string_field, 'a string', 'string')
14f25df2 2545 info[string_field] = str(field)
c9969434
S
2546
2547 def sanitize_numeric_fields(info):
2548 for numeric_field in self._NUMERIC_FIELDS:
2549 field = info.get(numeric_field)
f9934b96 2550 if field is None or isinstance(field, (int, float)):
c9969434
S
2551 continue
2552 report_force_conversion(numeric_field, 'numeric', 'int')
2553 info[numeric_field] = int_or_none(field)
2554
2555 sanitize_string_field(info_dict, 'id')
2556 sanitize_numeric_fields(info_dict)
3975b4d2 2557 if info_dict.get('section_end') and info_dict.get('section_start') is not None:
2558 info_dict['duration'] = round(info_dict['section_end'] - info_dict['section_start'], 3)
4c3f8c3f 2559 if (info_dict.get('duration') or 0) <= 0 and info_dict.pop('duration', None):
50e93e03 2560 self.report_warning('"duration" field is negative, there is an error in extractor')
be6217b2 2561
9eef7c4e 2562 chapters = info_dict.get('chapters') or []
a3976e07 2563 if chapters and chapters[0].get('start_time'):
2564 chapters.insert(0, {'start_time': 0})
2565
9eef7c4e 2566 dummy_chapter = {'end_time': 0, 'start_time': info_dict.get('duration')}
a3976e07 2567 for idx, (prev, current, next_) in enumerate(zip(
2568 (dummy_chapter, *chapters), chapters, (*chapters[1:], dummy_chapter)), 1):
9eef7c4e 2569 if current.get('start_time') is None:
2570 current['start_time'] = prev.get('end_time')
2571 if not current.get('end_time'):
2572 current['end_time'] = next_.get('start_time')
a3976e07 2573 if not current.get('title'):
2574 current['title'] = f'<Untitled Chapter {idx}>'
9eef7c4e 2575
dd82ffea
JMF
2576 if 'playlist' not in info_dict:
2577 # It isn't part of a playlist
2578 info_dict['playlist'] = None
2579 info_dict['playlist_index'] = None
2580
bc516a3f 2581 self._sanitize_thumbnails(info_dict)
d5519808 2582
536a55da 2583 thumbnail = info_dict.get('thumbnail')
bc516a3f 2584 thumbnails = info_dict.get('thumbnails')
536a55da
S
2585 if thumbnail:
2586 info_dict['thumbnail'] = sanitize_url(thumbnail)
2587 elif thumbnails:
d5519808
PH
2588 info_dict['thumbnail'] = thumbnails[-1]['url']
2589
ae30b840 2590 if info_dict.get('display_id') is None and 'id' in info_dict:
0afef30b
PH
2591 info_dict['display_id'] = info_dict['id']
2592
03f83004 2593 self._fill_common_fields(info_dict)
33d2fc2f 2594
05108a49
S
2595 for cc_kind in ('subtitles', 'automatic_captions'):
2596 cc = info_dict.get(cc_kind)
2597 if cc:
2598 for _, subtitle in cc.items():
2599 for subtitle_format in subtitle:
2600 if subtitle_format.get('url'):
2601 subtitle_format['url'] = sanitize_url(subtitle_format['url'])
2602 if subtitle_format.get('ext') is None:
2603 subtitle_format['ext'] = determine_ext(subtitle_format['url']).lower()
2604
2605 automatic_captions = info_dict.get('automatic_captions')
4bba3716 2606 subtitles = info_dict.get('subtitles')
4bba3716 2607
360e1ca5 2608 info_dict['requested_subtitles'] = self.process_subtitles(
05108a49 2609 info_dict['id'], subtitles, automatic_captions)
a504ced0 2610
aebb4f4b 2611 formats = self._get_formats(info_dict)
dd82ffea 2612
c154302c 2613 # Backward compatibility with InfoExtractor._sort_formats
9ebac355 2614 field_preference = (formats or [{}])[0].pop('__sort_fields', None)
c154302c 2615 if field_preference:
2616 info_dict['_format_sort_fields'] = field_preference
2617
0a5a191a 2618 # or None ensures --clean-infojson removes it
2619 info_dict['_has_drm'] = any(f.get('has_drm') for f in formats) or None
88acdbc2 2620 if not self.params.get('allow_unplayable_formats'):
2621 formats = [f for f in formats if not f.get('has_drm')]
17ffed18 2622
2623 if formats and all(f.get('acodec') == f.get('vcodec') == 'none' for f in formats):
2624 self.report_warning(
2625 f'{"This video is DRM protected and " if info_dict["_has_drm"] else ""}'
2626 'only images are available for download. Use --list-formats to see them'.capitalize())
88acdbc2 2627
319b6059 2628 get_from_start = not info_dict.get('is_live') or bool(self.params.get('live_from_start'))
2629 if not get_from_start:
2630 info_dict['title'] += ' ' + datetime.datetime.now().strftime('%Y-%m-%d %H:%M')
2631 if info_dict.get('is_live') and formats:
adbc4ec4 2632 formats = [f for f in formats if bool(f.get('is_from_start')) == get_from_start]
319b6059 2633 if get_from_start and not formats:
a44ca5a4 2634 self.raise_no_formats(info_dict, msg=(
2635 '--live-from-start is passed, but there are no formats that can be downloaded from the start. '
2636 'If you want to download from the current time, use --no-live-from-start'))
adbc4ec4 2637
73af5cc8
S
2638 def is_wellformed(f):
2639 url = f.get('url')
a5ac0c47 2640 if not url:
73af5cc8
S
2641 self.report_warning(
2642 '"url" field is missing or empty - skipping format, '
2643 'there is an error in extractor')
a5ac0c47
S
2644 return False
2645 if isinstance(url, bytes):
2646 sanitize_string_field(f, 'url')
2647 return True
73af5cc8
S
2648
2649 # Filter out malformed formats for better extraction robustness
1ac7f461 2650 formats = list(filter(is_wellformed, formats or []))
2651
2652 if not formats:
2653 self.raise_no_formats(info_dict)
73af5cc8 2654
39f32f17 2655 for format in formats:
c9969434
S
2656 sanitize_string_field(format, 'format_id')
2657 sanitize_numeric_fields(format)
dcf77cf1 2658 format['url'] = sanitize_url(format['url'])
39f32f17 2659 if format.get('ext') is None:
2660 format['ext'] = determine_ext(format['url']).lower()
2661 if format.get('protocol') is None:
2662 format['protocol'] = determine_protocol(format)
2663 if format.get('resolution') is None:
2664 format['resolution'] = self.format_resolution(format, default=None)
2665 if format.get('dynamic_range') is None and format.get('vcodec') != 'none':
2666 format['dynamic_range'] = 'SDR'
2667 if format.get('aspect_ratio') is None:
2668 format['aspect_ratio'] = try_call(lambda: round(format['width'] / format['height'], 2))
0dff8e4d 2669 if (not format.get('manifest_url') # For fragmented formats, "tbr" is often max bitrate and not average
2670 and info_dict.get('duration') and format.get('tbr')
39f32f17 2671 and not format.get('filesize') and not format.get('filesize_approx')):
2672 format['filesize_approx'] = int(info_dict['duration'] * format['tbr'] * (1024 / 8))
2673 format['http_headers'] = self._calc_headers(collections.ChainMap(format, info_dict))
2674
2675 # This is copied to http_headers by the above _calc_headers and can now be removed
2676 if '__x_forwarded_for_ip' in info_dict:
2677 del info_dict['__x_forwarded_for_ip']
2678
c154302c 2679 self.sort_formats({
2680 'formats': formats,
2681 '_format_sort_fields': info_dict.get('_format_sort_fields')
2682 })
39f32f17 2683
2684 # Sanitize and group by format_id
2685 formats_dict = {}
2686 for i, format in enumerate(formats):
e74e3b63 2687 if not format.get('format_id'):
14f25df2 2688 format['format_id'] = str(i)
e2effb08
S
2689 else:
2690 # Sanitize format_id from characters used in format selector expression
ec85ded8 2691 format['format_id'] = re.sub(r'[\s,/+\[\]()]', '_', format['format_id'])
39f32f17 2692 formats_dict.setdefault(format['format_id'], []).append(format)
181c7053
S
2693
2694 # Make sure all formats have unique format_id
03b4de72 2695 common_exts = set(itertools.chain(*self._format_selection_exts.values()))
181c7053 2696 for format_id, ambiguous_formats in formats_dict.items():
48ee10ee 2697 ambigious_id = len(ambiguous_formats) > 1
2698 for i, format in enumerate(ambiguous_formats):
2699 if ambigious_id:
181c7053 2700 format['format_id'] = '%s-%d' % (format_id, i)
48ee10ee 2701 # Ensure there is no conflict between id and ext in format selection
2702 # See https://github.com/yt-dlp/yt-dlp/issues/1282
2703 if format['format_id'] != format['ext'] and format['format_id'] in common_exts:
2704 format['format_id'] = 'f%s' % format['format_id']
181c7053 2705
39f32f17 2706 if format.get('format') is None:
2707 format['format'] = '{id} - {res}{note}'.format(
2708 id=format['format_id'],
2709 res=self.format_resolution(format),
2710 note=format_field(format, 'format_note', ' (%s)'),
2711 )
dd82ffea 2712
9f1a1c36 2713 if self.params.get('check_formats') is True:
282f5709 2714 formats = LazyList(self._check_formats(formats[::-1]), reverse=True)
9f1a1c36 2715
88acdbc2 2716 if not formats or formats[0] is not info_dict:
b3d9ef88
JMF
2717 # only set the 'formats' fields if the original info_dict list them
2718 # otherwise we end up with a circular reference, the first (and unique)
f89197d7 2719 # element in the 'formats' field in info_dict is info_dict itself,
dfb1b146 2720 # which can't be exported to json
b3d9ef88 2721 info_dict['formats'] = formats
4ec82a72 2722
2723 info_dict, _ = self.pre_process(info_dict)
2724
6db9c4d5 2725 if self._match_entry(info_dict, incomplete=self._format_fields) is not None:
09b49e1f 2726 return info_dict
2727
2728 self.post_extract(info_dict)
2729 info_dict, _ = self.pre_process(info_dict, 'after_filter')
2730
093a1710 2731 # The pre-processors may have modified the formats
aebb4f4b 2732 formats = self._get_formats(info_dict)
093a1710 2733
e4221b70 2734 list_only = self.params.get('simulate') == 'list_only'
fa9f30b8 2735 interactive_format_selection = not list_only and self.format_selector == '-'
b7b04c78 2736 if self.params.get('list_thumbnails'):
2737 self.list_thumbnails(info_dict)
b7b04c78 2738 if self.params.get('listsubtitles'):
2739 if 'automatic_captions' in info_dict:
2740 self.list_subtitles(
2741 info_dict['id'], automatic_captions, 'automatic captions')
2742 self.list_subtitles(info_dict['id'], subtitles, 'subtitles')
fa9f30b8 2743 if self.params.get('listformats') or interactive_format_selection:
b69fd25c 2744 self.list_formats(info_dict)
169dbde9 2745 if list_only:
b7b04c78 2746 # Without this printing, -F --print-json will not work
17060584 2747 self.__forced_printings(info_dict)
c487cf00 2748 return info_dict
bfaae0a7 2749
187986a8 2750 format_selector = self.format_selector
fa9f30b8 2751 while True:
2752 if interactive_format_selection:
372a0f3b
IS
2753 req_format = input(self._format_screen('\nEnter format selector ', self.Styles.EMPHASIS)
2754 + '(Press ENTER for default, or Ctrl+C to quit)'
2755 + self._format_screen(': ', self.Styles.EMPHASIS))
fa9f30b8 2756 try:
372a0f3b 2757 format_selector = self.build_format_selector(req_format) if req_format else None
fa9f30b8 2758 except SyntaxError as err:
2759 self.report_error(err, tb=False, is_error=False)
2760 continue
2761
372a0f3b
IS
2762 if format_selector is None:
2763 req_format = self._default_format_spec(info_dict, download=download)
2764 self.write_debug(f'Default format spec: {req_format}')
2765 format_selector = self.build_format_selector(req_format)
2766
85e801a9 2767 formats_to_download = list(format_selector({
fa9f30b8 2768 'formats': formats,
85e801a9 2769 'has_merged_format': any('none' not in (f.get('acodec'), f.get('vcodec')) for f in formats),
2770 'incomplete_formats': (
2771 # All formats are video-only or
2772 all(f.get('vcodec') != 'none' and f.get('acodec') == 'none' for f in formats)
2773 # all formats are audio-only
2774 or all(f.get('vcodec') == 'none' and f.get('acodec') != 'none' for f in formats)),
2775 }))
fa9f30b8 2776 if interactive_format_selection and not formats_to_download:
2777 self.report_error('Requested format is not available', tb=False, is_error=False)
2778 continue
2779 break
317f7ab6 2780
dd82ffea 2781 if not formats_to_download:
b7da73eb 2782 if not self.params.get('ignore_no_formats_error'):
c0b6e5c7 2783 raise ExtractorError(
2784 'Requested format is not available. Use --list-formats for a list of available formats',
2785 expected=True, video_id=info_dict['id'], ie=info_dict['extractor'])
b62fa6d7 2786 self.report_warning('Requested format is not available')
2787 # Process what we can, even without any available formats.
2788 formats_to_download = [{}]
a13e6848 2789
0500ee3d 2790 requested_ranges = tuple(self.params.get('download_ranges', lambda *_: [{}])(info_dict, self))
5ec1b6b7 2791 best_format, downloaded_formats = formats_to_download[-1], []
b62fa6d7 2792 if download:
0500ee3d 2793 if best_format and requested_ranges:
5ec1b6b7 2794 def to_screen(*msg):
2795 self.to_screen(f'[info] {info_dict["id"]}: {" ".join(", ".join(variadic(m)) for m in msg)}')
2796
2797 to_screen(f'Downloading {len(formats_to_download)} format(s):',
2798 (f['format_id'] for f in formats_to_download))
0500ee3d 2799 if requested_ranges != ({}, ):
5ec1b6b7 2800 to_screen(f'Downloading {len(requested_ranges)} time ranges:',
fc2ba496 2801 (f'{c["start_time"]:.1f}-{c["end_time"]:.1f}' for c in requested_ranges))
a13e6848 2802 max_downloads_reached = False
5ec1b6b7 2803
0500ee3d 2804 for fmt, chapter in itertools.product(formats_to_download, requested_ranges):
5ec1b6b7 2805 new_info = self._copy_infodict(info_dict)
b7da73eb 2806 new_info.update(fmt)
3975b4d2 2807 offset, duration = info_dict.get('section_start') or 0, info_dict.get('duration') or float('inf')
fc2ba496 2808 end_time = offset + min(chapter.get('end_time', duration), duration)
b4e0d758 2809 # duration may not be accurate. So allow deviations <1sec
2810 if end_time == float('inf') or end_time > offset + duration + 1:
2811 end_time = None
3975b4d2 2812 if chapter or offset:
5ec1b6b7 2813 new_info.update({
3975b4d2 2814 'section_start': offset + chapter.get('start_time', 0),
b4e0d758 2815 'section_end': end_time,
5ec1b6b7 2816 'section_title': chapter.get('title'),
2817 'section_number': chapter.get('index'),
2818 })
2819 downloaded_formats.append(new_info)
a13e6848 2820 try:
2821 self.process_info(new_info)
2822 except MaxDownloadsReached:
2823 max_downloads_reached = True
415f8d51 2824 self._raise_pending_errors(new_info)
f46e2f9d 2825 # Remove copied info
2826 for key, val in tuple(new_info.items()):
2827 if info_dict.get(key) == val:
2828 new_info.pop(key)
a13e6848 2829 if max_downloads_reached:
2830 break
ebed8b37 2831
5ec1b6b7 2832 write_archive = {f.get('__write_download_archive', False) for f in downloaded_formats}
a13e6848 2833 assert write_archive.issubset({True, False, 'ignore'})
2834 if True in write_archive and False not in write_archive:
2835 self.record_download_archive(info_dict)
be72c624 2836
5ec1b6b7 2837 info_dict['requested_downloads'] = downloaded_formats
ed5835b4 2838 info_dict = self.run_all_pps('after_video', info_dict)
a13e6848 2839 if max_downloads_reached:
2840 raise MaxDownloadsReached()
ebed8b37 2841
49a57e70 2842 # We update the info dict with the selected best quality format (backwards compatibility)
be72c624 2843 info_dict.update(best_format)
dd82ffea
JMF
2844 return info_dict
2845
98c70d6f 2846 def process_subtitles(self, video_id, normal_subtitles, automatic_captions):
a504ced0 2847 """Select the requested subtitles and their format"""
d8a58ddc 2848 available_subs, normal_sub_langs = {}, []
98c70d6f
JMF
2849 if normal_subtitles and self.params.get('writesubtitles'):
2850 available_subs.update(normal_subtitles)
d8a58ddc 2851 normal_sub_langs = tuple(normal_subtitles.keys())
98c70d6f
JMF
2852 if automatic_captions and self.params.get('writeautomaticsub'):
2853 for lang, cap_info in automatic_captions.items():
360e1ca5
JMF
2854 if lang not in available_subs:
2855 available_subs[lang] = cap_info
2856
d2c8aadf 2857 if not available_subs or (
2858 not self.params.get('writesubtitles')
2859 and not self.params.get('writeautomaticsub')):
4d171848 2860 return None
a504ced0 2861
d8a58ddc 2862 all_sub_langs = tuple(available_subs.keys())
a504ced0 2863 if self.params.get('allsubtitles', False):
c32b0aab 2864 requested_langs = all_sub_langs
2865 elif self.params.get('subtitleslangs', False):
5314b521 2866 try:
2867 requested_langs = orderedSet_from_options(
2868 self.params.get('subtitleslangs'), {'all': all_sub_langs}, use_regex=True)
2869 except re.error as e:
2870 raise ValueError(f'Wrong regex for subtitlelangs: {e.pattern}')
a504ced0 2871 else:
376aa24b
SS
2872 requested_langs = LazyList(itertools.chain(
2873 ['en'] if 'en' in normal_sub_langs else [],
2874 filter(lambda f: f.startswith('en'), normal_sub_langs),
2875 ['en'] if 'en' in all_sub_langs else [],
2876 filter(lambda f: f.startswith('en'), all_sub_langs),
2877 normal_sub_langs, all_sub_langs,
2878 ))[:1]
ad3dc496 2879 if requested_langs:
d2c8aadf 2880 self.to_screen(f'[info] {video_id}: Downloading subtitles: {", ".join(requested_langs)}')
a504ced0
JMF
2881
2882 formats_query = self.params.get('subtitlesformat', 'best')
2883 formats_preference = formats_query.split('/') if formats_query else []
2884 subs = {}
2885 for lang in requested_langs:
2886 formats = available_subs.get(lang)
2887 if formats is None:
86e5f3ed 2888 self.report_warning(f'{lang} subtitles not available for {video_id}')
a504ced0 2889 continue
a504ced0
JMF
2890 for ext in formats_preference:
2891 if ext == 'best':
2892 f = formats[-1]
2893 break
2894 matches = list(filter(lambda f: f['ext'] == ext, formats))
2895 if matches:
2896 f = matches[-1]
2897 break
2898 else:
2899 f = formats[-1]
2900 self.report_warning(
2901 'No subtitle format found matching "%s" for language %s, '
2902 'using %s' % (formats_query, lang, f['ext']))
2903 subs[lang] = f
2904 return subs
2905
bb66c247 2906 def _forceprint(self, key, info_dict):
2907 if info_dict is None:
2908 return
2909 info_copy = info_dict.copy()
17060584 2910 info_copy.setdefault('filename', self.prepare_filename(info_dict))
2911 if info_dict.get('requested_formats') is not None:
2912 # For RTMP URLs, also include the playpath
2913 info_copy['urls'] = '\n'.join(f['url'] + f.get('play_path', '') for f in info_dict['requested_formats'])
2914 elif info_dict.get('url'):
2915 info_copy['urls'] = info_dict['url'] + info_dict.get('play_path', '')
bb66c247 2916 info_copy['formats_table'] = self.render_formats_table(info_dict)
2917 info_copy['thumbnails_table'] = self.render_thumbnails_table(info_dict)
2918 info_copy['subtitles_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('subtitles'))
2919 info_copy['automatic_captions_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('automatic_captions'))
2920
2921 def format_tmpl(tmpl):
48c8424b 2922 mobj = re.fullmatch(r'([\w.:,]|-\d|(?P<dict>{([\w.:,]|-\d)+}))+=?', tmpl)
07a1250e 2923 if not mobj:
2924 return tmpl
48c8424b 2925
2926 fmt = '%({})s'
2927 if tmpl.startswith('{'):
6f2287cb 2928 tmpl, fmt = f'.{tmpl}', '%({})j'
48c8424b 2929 if tmpl.endswith('='):
2930 tmpl, fmt = tmpl[:-1], '{0} = %({0})#j'
2931 return '\n'.join(map(fmt.format, [tmpl] if mobj.group('dict') else tmpl.split(',')))
8130779d 2932
bb66c247 2933 for tmpl in self.params['forceprint'].get(key, []):
2934 self.to_stdout(self.evaluate_outtmpl(format_tmpl(tmpl), info_copy))
2935
2936 for tmpl, file_tmpl in self.params['print_to_file'].get(key, []):
5127e92a 2937 filename = self.prepare_filename(info_dict, outtmpl=file_tmpl)
bb66c247 2938 tmpl = format_tmpl(tmpl)
2939 self.to_screen(f'[info] Writing {tmpl!r} to: {filename}')
8d93e69d 2940 if self._ensure_dir_exists(filename):
9874e82b 2941 with open(filename, 'a', encoding='utf-8', newline='') as f:
2942 f.write(self.evaluate_outtmpl(tmpl, info_copy) + os.linesep)
ca30f449 2943
17060584 2944 return info_copy
2945
2946 def __forced_printings(self, info_dict, filename=None, incomplete=True):
bb66c247 2947 if (self.params.get('forcejson')
2948 or self.params['forceprint'].get('video')
2949 or self.params['print_to_file'].get('video')):
2b8a2973 2950 self.post_extract(info_dict)
17060584 2951 if filename:
2952 info_dict['filename'] = filename
b5f61b69 2953 info_copy = self._forceprint('video', info_dict)
2954
2955 def print_field(field, actual_field=None, optional=False):
2956 if actual_field is None:
2957 actual_field = field
2958 if self.params.get(f'force{field}') and (
2959 info_copy.get(field) is not None or (not optional and not incomplete)):
2960 self.to_stdout(info_copy[actual_field])
2961
2962 print_field('title')
2963 print_field('id')
2964 print_field('url', 'urls')
2965 print_field('thumbnail', optional=True)
2966 print_field('description', optional=True)
ad54c913 2967 print_field('filename')
b5f61b69 2968 if self.params.get('forceduration') and info_copy.get('duration') is not None:
2969 self.to_stdout(formatSeconds(info_copy['duration']))
2970 print_field('format')
53c18592 2971
2b8a2973 2972 if self.params.get('forcejson'):
6e84b215 2973 self.to_stdout(json.dumps(self.sanitize_info(info_dict)))
d06daf23 2974
e8e73840 2975 def dl(self, name, info, subtitle=False, test=False):
88acdbc2 2976 if not info.get('url'):
1151c407 2977 self.raise_no_formats(info, True)
e8e73840 2978
2979 if test:
2980 verbose = self.params.get('verbose')
2981 params = {
2982 'test': True,
a169858f 2983 'quiet': self.params.get('quiet') or not verbose,
e8e73840 2984 'verbose': verbose,
2985 'noprogress': not verbose,
2986 'nopart': True,
2987 'skip_unavailable_fragments': False,
2988 'keep_fragments': False,
2989 'overwrites': True,
2990 '_no_ytdl_file': True,
2991 }
2992 else:
2993 params = self.params
96fccc10 2994 fd = get_suitable_downloader(info, params, to_stdout=(name == '-'))(self, params)
e8e73840 2995 if not test:
2996 for ph in self._progress_hooks:
2997 fd.add_progress_hook(ph)
42676437
M
2998 urls = '", "'.join(
2999 (f['url'].split(',')[0] + ',<data>' if f['url'].startswith('data:') else f['url'])
3000 for f in info.get('requested_formats', []) or [info])
3a408f9d 3001 self.write_debug(f'Invoking {fd.FD_NAME} downloader on "{urls}"')
03b4de72 3002
adbc4ec4
THD
3003 # Note: Ideally info should be a deep-copied so that hooks cannot modify it.
3004 # But it may contain objects that are not deep-copyable
3005 new_info = self._copy_infodict(info)
e8e73840 3006 if new_info.get('http_headers') is None:
3007 new_info['http_headers'] = self._calc_headers(new_info)
3008 return fd.download(name, new_info, subtitle)
3009
e04938ab 3010 def existing_file(self, filepaths, *, default_overwrite=True):
3011 existing_files = list(filter(os.path.exists, orderedSet(filepaths)))
3012 if existing_files and not self.params.get('overwrites', default_overwrite):
3013 return existing_files[0]
3014
3015 for file in existing_files:
3016 self.report_file_delete(file)
3017 os.remove(file)
3018 return None
3019
8222d8de 3020 def process_info(self, info_dict):
09b49e1f 3021 """Process a single resolved IE result. (Modifies it in-place)"""
8222d8de
JMF
3022
3023 assert info_dict.get('_type', 'video') == 'video'
f46e2f9d 3024 original_infodict = info_dict
fd288278 3025
4513a41a 3026 if 'format' not in info_dict and 'ext' in info_dict:
8222d8de
JMF
3027 info_dict['format'] = info_dict['ext']
3028
c77495e3 3029 if self._match_entry(info_dict) is not None:
9e907ebd 3030 info_dict['__write_download_archive'] = 'ignore'
8222d8de
JMF
3031 return
3032
09b49e1f 3033 # Does nothing under normal operation - for backward compatibility of process_info
277d6ff5 3034 self.post_extract(info_dict)
119e40ef 3035
3036 def replace_info_dict(new_info):
3037 nonlocal info_dict
3038 if new_info == info_dict:
3039 return
3040 info_dict.clear()
3041 info_dict.update(new_info)
3042
3043 new_info, _ = self.pre_process(info_dict, 'video')
3044 replace_info_dict(new_info)
0c14d66a 3045 self._num_downloads += 1
8222d8de 3046
dcf64d43 3047 # info_dict['_filename'] needs to be set for backward compatibility
de6000d9 3048 info_dict['_filename'] = full_filename = self.prepare_filename(info_dict, warn=True)
3049 temp_filename = self.prepare_filename(info_dict, 'temp')
0202b52a 3050 files_to_move = {}
8222d8de
JMF
3051
3052 # Forced printings
4513a41a 3053 self.__forced_printings(info_dict, full_filename, incomplete=('format' not in info_dict))
8222d8de 3054
ca6d59d2 3055 def check_max_downloads():
3056 if self._num_downloads >= float(self.params.get('max_downloads') or 'inf'):
3057 raise MaxDownloadsReached()
3058
b7b04c78 3059 if self.params.get('simulate'):
9e907ebd 3060 info_dict['__write_download_archive'] = self.params.get('force_write_download_archive')
ca6d59d2 3061 check_max_downloads()
8222d8de
JMF
3062 return
3063
de6000d9 3064 if full_filename is None:
8222d8de 3065 return
e92caff5 3066 if not self._ensure_dir_exists(encodeFilename(full_filename)):
0202b52a 3067 return
e92caff5 3068 if not self._ensure_dir_exists(encodeFilename(temp_filename)):
8222d8de
JMF
3069 return
3070
80c03fa9 3071 if self._write_description('video', info_dict,
3072 self.prepare_filename(info_dict, 'description')) is None:
3073 return
3074
3075 sub_files = self._write_subtitles(info_dict, temp_filename)
3076 if sub_files is None:
3077 return
3078 files_to_move.update(dict(sub_files))
3079
3080 thumb_files = self._write_thumbnails(
3081 'video', info_dict, temp_filename, self.prepare_filename(info_dict, 'thumbnail'))
3082 if thumb_files is None:
3083 return
3084 files_to_move.update(dict(thumb_files))
8222d8de 3085
80c03fa9 3086 infofn = self.prepare_filename(info_dict, 'infojson')
3087 _infojson_written = self._write_info_json('video', info_dict, infofn)
3088 if _infojson_written:
dac5df5a 3089 info_dict['infojson_filename'] = infofn
e75bb0d6 3090 # For backward compatibility, even though it was a private field
80c03fa9 3091 info_dict['__infojson_filename'] = infofn
3092 elif _infojson_written is None:
3093 return
3094
3095 # Note: Annotations are deprecated
3096 annofn = None
1fb07d10 3097 if self.params.get('writeannotations', False):
de6000d9 3098 annofn = self.prepare_filename(info_dict, 'annotation')
80c03fa9 3099 if annofn:
e92caff5 3100 if not self._ensure_dir_exists(encodeFilename(annofn)):
0202b52a 3101 return
0c3d0f51 3102 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(annofn)):
6febd1c1 3103 self.to_screen('[info] Video annotations are already present')
ffddb112
RA
3104 elif not info_dict.get('annotations'):
3105 self.report_warning('There are no annotations to write.')
7b6fefc9
PH
3106 else:
3107 try:
6febd1c1 3108 self.to_screen('[info] Writing video annotations to: ' + annofn)
86e5f3ed 3109 with open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
7b6fefc9
PH
3110 annofile.write(info_dict['annotations'])
3111 except (KeyError, TypeError):
6febd1c1 3112 self.report_warning('There are no annotations to write.')
86e5f3ed 3113 except OSError:
6febd1c1 3114 self.report_error('Cannot write annotations file: ' + annofn)
7b6fefc9 3115 return
1fb07d10 3116
732044af 3117 # Write internet shortcut files
08438d2c 3118 def _write_link_file(link_type):
60f3e995 3119 url = try_get(info_dict['webpage_url'], iri_to_uri)
3120 if not url:
3121 self.report_warning(
3122 f'Cannot write internet shortcut file because the actual URL of "{info_dict["webpage_url"]}" is unknown')
3123 return True
08438d2c 3124 linkfn = replace_extension(self.prepare_filename(info_dict, 'link'), link_type, info_dict.get('ext'))
0e6b018a
Z
3125 if not self._ensure_dir_exists(encodeFilename(linkfn)):
3126 return False
10e3742e 3127 if self.params.get('overwrites', True) and os.path.exists(encodeFilename(linkfn)):
08438d2c 3128 self.to_screen(f'[info] Internet shortcut (.{link_type}) is already present')
3129 return True
3130 try:
3131 self.to_screen(f'[info] Writing internet shortcut (.{link_type}) to: {linkfn}')
86e5f3ed 3132 with open(encodeFilename(to_high_limit_path(linkfn)), 'w', encoding='utf-8',
3133 newline='\r\n' if link_type == 'url' else '\n') as linkfile:
60f3e995 3134 template_vars = {'url': url}
08438d2c 3135 if link_type == 'desktop':
3136 template_vars['filename'] = linkfn[:-(len(link_type) + 1)]
3137 linkfile.write(LINK_TEMPLATES[link_type] % template_vars)
86e5f3ed 3138 except OSError:
08438d2c 3139 self.report_error(f'Cannot write internet shortcut {linkfn}')
3140 return False
732044af 3141 return True
3142
08438d2c 3143 write_links = {
3144 'url': self.params.get('writeurllink'),
3145 'webloc': self.params.get('writewebloclink'),
3146 'desktop': self.params.get('writedesktoplink'),
3147 }
3148 if self.params.get('writelink'):
3149 link_type = ('webloc' if sys.platform == 'darwin'
3150 else 'desktop' if sys.platform.startswith('linux')
3151 else 'url')
3152 write_links[link_type] = True
3153
3154 if any(should_write and not _write_link_file(link_type)
3155 for link_type, should_write in write_links.items()):
3156 return
732044af 3157
415f8d51 3158 new_info, files_to_move = self.pre_process(info_dict, 'before_dl', files_to_move)
3159 replace_info_dict(new_info)
56d868db 3160
a13e6848 3161 if self.params.get('skip_download'):
56d868db 3162 info_dict['filepath'] = temp_filename
3163 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
3164 info_dict['__files_to_move'] = files_to_move
f46e2f9d 3165 replace_info_dict(self.run_pp(MoveFilesAfterDownloadPP(self, False), info_dict))
9e907ebd 3166 info_dict['__write_download_archive'] = self.params.get('force_write_download_archive')
56d868db 3167 else:
3168 # Download
b868936c 3169 info_dict.setdefault('__postprocessors', [])
4340deca 3170 try:
0202b52a 3171
e04938ab 3172 def existing_video_file(*filepaths):
6b591b29 3173 ext = info_dict.get('ext')
e04938ab 3174 converted = lambda file: replace_extension(file, self.params.get('final_ext') or ext, ext)
3175 file = self.existing_file(itertools.chain(*zip(map(converted, filepaths), filepaths)),
3176 default_overwrite=False)
3177 if file:
3178 info_dict['ext'] = os.path.splitext(file)[1][1:]
3179 return file
0202b52a 3180
7b2c3f47 3181 fd, success = None, True
fccf90e7 3182 if info_dict.get('protocol') or info_dict.get('url'):
56ba69e4 3183 fd = get_suitable_downloader(info_dict, self.params, to_stdout=temp_filename == '-')
71df9b7f 3184 if fd is not FFmpegFD and 'no-direct-merge' not in self.params['compat_opts'] and (
56ba69e4 3185 info_dict.get('section_start') or info_dict.get('section_end')):
7b2c3f47 3186 msg = ('This format cannot be partially downloaded' if FFmpegFD.available()
56ba69e4 3187 else 'You have requested downloading the video partially, but ffmpeg is not installed')
3188 self.report_error(f'{msg}. Aborting')
5ec1b6b7 3189 return
5ec1b6b7 3190
4340deca 3191 if info_dict.get('requested_formats') is not None:
0202b52a 3192 old_ext = info_dict['ext']
4e3b637d 3193 if self.params.get('merge_output_format') is None:
4e3b637d 3194 if (info_dict['ext'] == 'webm'
3195 and info_dict.get('thumbnails')
3196 # check with type instead of pp_key, __name__, or isinstance
3197 # since we dont want any custom PPs to trigger this
c487cf00 3198 and any(type(pp) == EmbedThumbnailPP for pp in self._pps['post_process'])): # noqa: E721
4e3b637d 3199 info_dict['ext'] = 'mkv'
3200 self.report_warning(
3201 'webm doesn\'t support embedding a thumbnail, mkv will be used')
124bc071 3202 new_ext = info_dict['ext']
0202b52a 3203
124bc071 3204 def correct_ext(filename, ext=new_ext):
96fccc10 3205 if filename == '-':
3206 return filename
0202b52a 3207 filename_real_ext = os.path.splitext(filename)[1][1:]
3208 filename_wo_ext = (
3209 os.path.splitext(filename)[0]
124bc071 3210 if filename_real_ext in (old_ext, new_ext)
0202b52a 3211 else filename)
86e5f3ed 3212 return f'{filename_wo_ext}.{ext}'
0202b52a 3213
38c6902b 3214 # Ensure filename always has a correct extension for successful merge
0202b52a 3215 full_filename = correct_ext(full_filename)
3216 temp_filename = correct_ext(temp_filename)
e04938ab 3217 dl_filename = existing_video_file(full_filename, temp_filename)
ad54c913 3218
1ea24129 3219 info_dict['__real_download'] = False
84078a8b 3220 # NOTE: Copy so that original format dicts are not modified
3221 info_dict['requested_formats'] = list(map(dict, info_dict['requested_formats']))
18e674b4 3222
7b2c3f47 3223 merger = FFmpegMergerPP(self)
adbc4ec4 3224 downloaded = []
dbf5416a 3225 if dl_filename is not None:
6c7274ec 3226 self.report_file_already_downloaded(dl_filename)
adbc4ec4 3227 elif fd:
ad54c913 3228 for f in info_dict['requested_formats'] if fd != FFmpegFD else []:
adbc4ec4
THD
3229 f['filepath'] = fname = prepend_extension(
3230 correct_ext(temp_filename, info_dict['ext']),
3231 'f%s' % f['format_id'], info_dict['ext'])
3232 downloaded.append(fname)
ad54c913 3233 info_dict['url'] = '\n'.join(f['url'] for f in info_dict['requested_formats'])
dbf5416a 3234 success, real_download = self.dl(temp_filename, info_dict)
3235 info_dict['__real_download'] = real_download
18e674b4 3236 else:
18e674b4 3237 if self.params.get('allow_unplayable_formats'):
3238 self.report_warning(
3239 'You have requested merging of multiple formats '
3240 'while also allowing unplayable formats to be downloaded. '
3241 'The formats won\'t be merged to prevent data corruption.')
3242 elif not merger.available:
e8969bda 3243 msg = 'You have requested merging of multiple formats but ffmpeg is not installed'
3244 if not self.params.get('ignoreerrors'):
3245 self.report_error(f'{msg}. Aborting due to --abort-on-error')
3246 return
3247 self.report_warning(f'{msg}. The formats won\'t be merged')
18e674b4 3248
96fccc10 3249 if temp_filename == '-':
adbc4ec4 3250 reason = ('using a downloader other than ffmpeg' if FFmpegFD.can_merge_formats(info_dict, self.params)
96fccc10 3251 else 'but the formats are incompatible for simultaneous download' if merger.available
3252 else 'but ffmpeg is not installed')
3253 self.report_warning(
3254 f'You have requested downloading multiple formats to stdout {reason}. '
3255 'The formats will be streamed one after the other')
3256 fname = temp_filename
ad54c913 3257 for f in info_dict['requested_formats']:
dbf5416a 3258 new_info = dict(info_dict)
3259 del new_info['requested_formats']
3260 new_info.update(f)
96fccc10 3261 if temp_filename != '-':
124bc071 3262 fname = prepend_extension(
3263 correct_ext(temp_filename, new_info['ext']),
3264 'f%s' % f['format_id'], new_info['ext'])
96fccc10 3265 if not self._ensure_dir_exists(fname):
3266 return
a21e0ab1 3267 f['filepath'] = fname
96fccc10 3268 downloaded.append(fname)
dbf5416a 3269 partial_success, real_download = self.dl(fname, new_info)
3270 info_dict['__real_download'] = info_dict['__real_download'] or real_download
3271 success = success and partial_success
adbc4ec4
THD
3272
3273 if downloaded and merger.available and not self.params.get('allow_unplayable_formats'):
3274 info_dict['__postprocessors'].append(merger)
3275 info_dict['__files_to_merge'] = downloaded
3276 # Even if there were no downloads, it is being merged only now
3277 info_dict['__real_download'] = True
3278 else:
3279 for file in downloaded:
3280 files_to_move[file] = None
4340deca
P
3281 else:
3282 # Just a single file
e04938ab 3283 dl_filename = existing_video_file(full_filename, temp_filename)
6c7274ec 3284 if dl_filename is None or dl_filename == temp_filename:
3285 # dl_filename == temp_filename could mean that the file was partially downloaded with --no-part.
3286 # So we should try to resume the download
e8e73840 3287 success, real_download = self.dl(temp_filename, info_dict)
0202b52a 3288 info_dict['__real_download'] = real_download
6c7274ec 3289 else:
3290 self.report_file_already_downloaded(dl_filename)
0202b52a 3291
0202b52a 3292 dl_filename = dl_filename or temp_filename
c571435f 3293 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
0202b52a 3294
3158150c 3295 except network_exceptions as err:
7960b056 3296 self.report_error('unable to download video data: %s' % error_to_compat_str(err))
4340deca 3297 return
86e5f3ed 3298 except OSError as err:
4340deca
P
3299 raise UnavailableVideoError(err)
3300 except (ContentTooShortError, ) as err:
86e5f3ed 3301 self.report_error(f'content too short (expected {err.expected} bytes and served {err.downloaded})')
4340deca 3302 return
8222d8de 3303
415f8d51 3304 self._raise_pending_errors(info_dict)
de6000d9 3305 if success and full_filename != '-':
f17f8651 3306
fd7cfb64 3307 def fixup():
3308 do_fixup = True
3309 fixup_policy = self.params.get('fixup')
3310 vid = info_dict['id']
3311
3312 if fixup_policy in ('ignore', 'never'):
3313 return
3314 elif fixup_policy == 'warn':
3fe75fdc 3315 do_fixup = 'warn'
f89b3e2d 3316 elif fixup_policy != 'force':
3317 assert fixup_policy in ('detect_or_warn', None)
3318 if not info_dict.get('__real_download'):
3319 do_fixup = False
fd7cfb64 3320
3321 def ffmpeg_fixup(cndn, msg, cls):
3fe75fdc 3322 if not (do_fixup and cndn):
fd7cfb64 3323 return
3fe75fdc 3324 elif do_fixup == 'warn':
fd7cfb64 3325 self.report_warning(f'{vid}: {msg}')
3326 return
3327 pp = cls(self)
3328 if pp.available:
3329 info_dict['__postprocessors'].append(pp)
3330 else:
3331 self.report_warning(f'{vid}: {msg}. Install ffmpeg to fix this automatically')
3332
3333 stretched_ratio = info_dict.get('stretched_ratio')
ca9def71
LNO
3334 ffmpeg_fixup(stretched_ratio not in (1, None),
3335 f'Non-uniform pixel ratio {stretched_ratio}',
3336 FFmpegFixupStretchedPP)
fd7cfb64 3337
993191c0 3338 downloader = get_suitable_downloader(info_dict, self.params) if 'protocol' in info_dict else None
24146491 3339 downloader = downloader.FD_NAME if downloader else None
adbc4ec4 3340
ca9def71
LNO
3341 ext = info_dict.get('ext')
3342 postprocessed_by_ffmpeg = info_dict.get('requested_formats') or any((
3343 isinstance(pp, FFmpegVideoConvertorPP)
3344 and resolve_recode_mapping(ext, pp.mapping)[0] not in (ext, None)
3345 ) for pp in self._pps['post_process'])
3346
3347 if not postprocessed_by_ffmpeg:
3348 ffmpeg_fixup(ext == 'm4a' and info_dict.get('container') == 'm4a_dash',
f2df4071 3349 'writing DASH m4a. Only some players support this container',
3350 FFmpegFixupM4aPP)
24146491 3351 ffmpeg_fixup(downloader == 'hlsnative' and not self.params.get('hls_use_mpegts')
494f5230 3352 or info_dict.get('is_live') and self.params.get('hls_use_mpegts') is None,
adbc4ec4
THD
3353 'Possible MPEG-TS in MP4 container or malformed AAC timestamps',
3354 FFmpegFixupM3u8PP)
26010b5c 3355 ffmpeg_fixup(info_dict.get('is_live') and downloader == 'dashsegments',
adbc4ec4
THD
3356 'Possible duplicate MOOV atoms', FFmpegFixupDuplicateMoovPP)
3357
24146491 3358 ffmpeg_fixup(downloader == 'web_socket_fragment', 'Malformed timestamps detected', FFmpegFixupTimestampPP)
3359 ffmpeg_fixup(downloader == 'web_socket_fragment', 'Malformed duration detected', FFmpegFixupDurationPP)
fd7cfb64 3360
3361 fixup()
8222d8de 3362 try:
f46e2f9d 3363 replace_info_dict(self.post_process(dl_filename, info_dict, files_to_move))
af819c21 3364 except PostProcessingError as err:
3365 self.report_error('Postprocessing: %s' % str(err))
8222d8de 3366 return
ab8e5e51
AM
3367 try:
3368 for ph in self._post_hooks:
23c1a667 3369 ph(info_dict['filepath'])
ab8e5e51
AM
3370 except Exception as err:
3371 self.report_error('post hooks: %s' % str(err))
3372 return
9e907ebd 3373 info_dict['__write_download_archive'] = True
2d30509f 3374
c487cf00 3375 assert info_dict is original_infodict # Make sure the info_dict was modified in-place
a13e6848 3376 if self.params.get('force_write_download_archive'):
9e907ebd 3377 info_dict['__write_download_archive'] = True
ca6d59d2 3378 check_max_downloads()
8222d8de 3379
aa9369a2 3380 def __download_wrapper(self, func):
3381 @functools.wraps(func)
3382 def wrapper(*args, **kwargs):
3383 try:
3384 res = func(*args, **kwargs)
3385 except UnavailableVideoError as e:
3386 self.report_error(e)
b222c271 3387 except DownloadCancelled as e:
3388 self.to_screen(f'[info] {e}')
3389 if not self.params.get('break_per_url'):
3390 raise
fd404bec 3391 self._num_downloads = 0
aa9369a2 3392 else:
3393 if self.params.get('dump_single_json', False):
3394 self.post_extract(res)
3395 self.to_stdout(json.dumps(self.sanitize_info(res)))
3396 return wrapper
3397
8222d8de
JMF
3398 def download(self, url_list):
3399 """Download a given list of URLs."""
aa9369a2 3400 url_list = variadic(url_list) # Passing a single URL is a common mistake
bf1824b3 3401 outtmpl = self.params['outtmpl']['default']
3089bc74
S
3402 if (len(url_list) > 1
3403 and outtmpl != '-'
3404 and '%' not in outtmpl
3405 and self.params.get('max_downloads') != 1):
acd69589 3406 raise SameFileError(outtmpl)
8222d8de
JMF
3407
3408 for url in url_list:
aa9369a2 3409 self.__download_wrapper(self.extract_info)(
3410 url, force_generic_extractor=self.params.get('force_generic_extractor', False))
8222d8de
JMF
3411
3412 return self._download_retcode
3413
1dcc4c0c 3414 def download_with_info_file(self, info_filename):
31bd3925
JMF
3415 with contextlib.closing(fileinput.FileInput(
3416 [info_filename], mode='r',
3417 openhook=fileinput.hook_encoded('utf-8'))) as f:
3418 # FileInput doesn't have a read method, we can't call json.load
ab1de9cb 3419 infos = [self.sanitize_info(info, self.params.get('clean_infojson', True))
3420 for info in variadic(json.loads('\n'.join(f)))]
3421 for info in infos:
3422 try:
3423 self.__download_wrapper(self.process_ie_result)(info, download=True)
3424 except (DownloadError, EntryNotInPlaylist, ReExtractInfo) as e:
3425 if not isinstance(e, EntryNotInPlaylist):
3426 self.to_stderr('\r')
3427 webpage_url = info.get('webpage_url')
3428 if webpage_url is None:
3429 raise
aa9369a2 3430 self.report_warning(f'The info failed to download: {e}; trying with URL {webpage_url}')
ab1de9cb 3431 self.download([webpage_url])
d4943898 3432 return self._download_retcode
1dcc4c0c 3433
cb202fd2 3434 @staticmethod
8012d892 3435 def sanitize_info(info_dict, remove_private_keys=False):
3436 ''' Sanitize the infodict for converting to json '''
3ad56b42 3437 if info_dict is None:
3438 return info_dict
6e84b215 3439 info_dict.setdefault('epoch', int(time.time()))
6a5a30f9 3440 info_dict.setdefault('_type', 'video')
b5e7a2e6 3441 info_dict.setdefault('_version', {
3442 'version': __version__,
3443 'current_git_head': current_git_head(),
3444 'release_git_head': RELEASE_GIT_HEAD,
3445 'repository': REPOSITORY,
3446 })
09b49e1f 3447
8012d892 3448 if remove_private_keys:
0a5a191a 3449 reject = lambda k, v: v is None or k.startswith('__') or k in {
f46e2f9d 3450 'requested_downloads', 'requested_formats', 'requested_subtitles', 'requested_entries',
6f2287cb 3451 'entries', 'filepath', '_filename', 'filename', 'infojson_filename', 'original_url',
3452 'playlist_autonumber', '_format_sort_fields',
6e84b215 3453 }
ae8f99e6 3454 else:
09b49e1f 3455 reject = lambda k, v: False
adbc4ec4
THD
3456
3457 def filter_fn(obj):
3458 if isinstance(obj, dict):
3459 return {k: filter_fn(v) for k, v in obj.items() if not reject(k, v)}
3460 elif isinstance(obj, (list, tuple, set, LazyList)):
3461 return list(map(filter_fn, obj))
3462 elif obj is None or isinstance(obj, (str, int, float, bool)):
3463 return obj
3464 else:
3465 return repr(obj)
3466
5226731e 3467 return filter_fn(info_dict)
cb202fd2 3468
8012d892 3469 @staticmethod
3470 def filter_requested_info(info_dict, actually_filter=True):
3471 ''' Alias of sanitize_info for backward compatibility '''
3472 return YoutubeDL.sanitize_info(info_dict, actually_filter)
3473
43d7f5a5 3474 def _delete_downloaded_files(self, *files_to_delete, info={}, msg=None):
3475 for filename in set(filter(None, files_to_delete)):
3476 if msg:
3477 self.to_screen(msg % filename)
3478 try:
3479 os.remove(filename)
3480 except OSError:
3481 self.report_warning(f'Unable to delete file {filename}')
3482 if filename in info.get('__files_to_move', []): # NB: Delete even if None
3483 del info['__files_to_move'][filename]
3484
ed5835b4 3485 @staticmethod
3486 def post_extract(info_dict):
3487 def actual_post_extract(info_dict):
3488 if info_dict.get('_type') in ('playlist', 'multi_video'):
3489 for video_dict in info_dict.get('entries', {}):
3490 actual_post_extract(video_dict or {})
3491 return
3492
09b49e1f 3493 post_extractor = info_dict.pop('__post_extractor', None) or (lambda: {})
3494 info_dict.update(post_extractor())
ed5835b4 3495
3496 actual_post_extract(info_dict or {})
3497
dcf64d43 3498 def run_pp(self, pp, infodict):
5bfa4862 3499 files_to_delete = []
dcf64d43 3500 if '__files_to_move' not in infodict:
3501 infodict['__files_to_move'] = {}
b1940459 3502 try:
3503 files_to_delete, infodict = pp.run(infodict)
3504 except PostProcessingError as e:
3505 # Must be True and not 'only_download'
3506 if self.params.get('ignoreerrors') is True:
3507 self.report_error(e)
3508 return infodict
3509 raise
3510
5bfa4862 3511 if not files_to_delete:
dcf64d43 3512 return infodict
5bfa4862 3513 if self.params.get('keepvideo', False):
3514 for f in files_to_delete:
dcf64d43 3515 infodict['__files_to_move'].setdefault(f, '')
5bfa4862 3516 else:
43d7f5a5 3517 self._delete_downloaded_files(
3518 *files_to_delete, info=infodict, msg='Deleting original file %s (pass -k to keep)')
dcf64d43 3519 return infodict
5bfa4862 3520
6f2287cb 3521 def run_all_pps(self, key, info, *, additional_pps=None):
17ba4343 3522 if key != 'video':
3523 self._forceprint(key, info)
3524 for pp in (additional_pps or []) + self._pps[key]:
3525 info = self.run_pp(pp, info)
ed5835b4 3526 return info
277d6ff5 3527
56d868db 3528 def pre_process(self, ie_info, key='pre_process', files_to_move=None):
5bfa4862 3529 info = dict(ie_info)
56d868db 3530 info['__files_to_move'] = files_to_move or {}
415f8d51 3531 try:
3532 info = self.run_all_pps(key, info)
3533 except PostProcessingError as err:
3534 msg = f'Preprocessing: {err}'
3535 info.setdefault('__pending_error', msg)
3536 self.report_error(msg, is_error=False)
56d868db 3537 return info, info.pop('__files_to_move', None)
5bfa4862 3538
f46e2f9d 3539 def post_process(self, filename, info, files_to_move=None):
8222d8de 3540 """Run all the postprocessors on the given file."""
8222d8de 3541 info['filepath'] = filename
dcf64d43 3542 info['__files_to_move'] = files_to_move or {}
ed5835b4 3543 info = self.run_all_pps('post_process', info, additional_pps=info.get('__postprocessors'))
dcf64d43 3544 info = self.run_pp(MoveFilesAfterDownloadPP(self), info)
3545 del info['__files_to_move']
ed5835b4 3546 return self.run_all_pps('after_move', info)
c1c9a79c 3547
5db07df6 3548 def _make_archive_id(self, info_dict):
e9fef7ee
S
3549 video_id = info_dict.get('id')
3550 if not video_id:
3551 return
5db07df6
PH
3552 # Future-proof against any change in case
3553 # and backwards compatibility with prior versions
e9fef7ee 3554 extractor = info_dict.get('extractor_key') or info_dict.get('ie_key') # key in a playlist
7012b23c 3555 if extractor is None:
1211bb6d
S
3556 url = str_or_none(info_dict.get('url'))
3557 if not url:
3558 return
e9fef7ee 3559 # Try to find matching extractor for the URL and take its ie_key
8b7491c8 3560 for ie_key, ie in self._ies.items():
1211bb6d 3561 if ie.suitable(url):
8b7491c8 3562 extractor = ie_key
e9fef7ee
S
3563 break
3564 else:
3565 return
0647d925 3566 return make_archive_id(extractor, video_id)
5db07df6
PH
3567
3568 def in_download_archive(self, info_dict):
ae103564 3569 if not self.archive:
5db07df6
PH
3570 return False
3571
1e8fe57e 3572 vid_ids = [self._make_archive_id(info_dict)]
c200096c 3573 vid_ids.extend(info_dict.get('_old_archive_ids') or [])
1e8fe57e 3574 return any(id_ in self.archive for id_ in vid_ids)
c1c9a79c
PH
3575
3576 def record_download_archive(self, info_dict):
3577 fn = self.params.get('download_archive')
3578 if fn is None:
3579 return
5db07df6
PH
3580 vid_id = self._make_archive_id(info_dict)
3581 assert vid_id
ae103564 3582
a13e6848 3583 self.write_debug(f'Adding to archive: {vid_id}')
9c935fbc 3584 if is_path_like(fn):
ae103564 3585 with locked_file(fn, 'a', encoding='utf-8') as archive_file:
3586 archive_file.write(vid_id + '\n')
a45e8619 3587 self.archive.add(vid_id)
dd82ffea 3588
8c51aa65 3589 @staticmethod
8abeeb94 3590 def format_resolution(format, default='unknown'):
9359f3d4 3591 if format.get('vcodec') == 'none' and format.get('acodec') != 'none':
fb04e403 3592 return 'audio only'
f49d89ee
PH
3593 if format.get('resolution') is not None:
3594 return format['resolution']
35615307 3595 if format.get('width') and format.get('height'):
ff51ed58 3596 return '%dx%d' % (format['width'], format['height'])
35615307 3597 elif format.get('height'):
ff51ed58 3598 return '%sp' % format['height']
35615307 3599 elif format.get('width'):
ff51ed58 3600 return '%dx?' % format['width']
3601 return default
8c51aa65 3602
8130779d 3603 def _list_format_headers(self, *headers):
3604 if self.params.get('listformats_table', True) is not False:
591bb9d3 3605 return [self._format_out(header, self.Styles.HEADERS) for header in headers]
8130779d 3606 return headers
3607
c57f7757
PH
3608 def _format_note(self, fdict):
3609 res = ''
3610 if fdict.get('ext') in ['f4f', 'f4m']:
f304da8a 3611 res += '(unsupported)'
32f90364
PH
3612 if fdict.get('language'):
3613 if res:
3614 res += ' '
f304da8a 3615 res += '[%s]' % fdict['language']
c57f7757 3616 if fdict.get('format_note') is not None:
f304da8a 3617 if res:
3618 res += ' '
3619 res += fdict['format_note']
c57f7757 3620 if fdict.get('tbr') is not None:
f304da8a 3621 if res:
3622 res += ', '
3623 res += '%4dk' % fdict['tbr']
c57f7757
PH
3624 if fdict.get('container') is not None:
3625 if res:
3626 res += ', '
3627 res += '%s container' % fdict['container']
3089bc74
S
3628 if (fdict.get('vcodec') is not None
3629 and fdict.get('vcodec') != 'none'):
c57f7757
PH
3630 if res:
3631 res += ', '
3632 res += fdict['vcodec']
91c7271a 3633 if fdict.get('vbr') is not None:
c57f7757
PH
3634 res += '@'
3635 elif fdict.get('vbr') is not None and fdict.get('abr') is not None:
3636 res += 'video@'
3637 if fdict.get('vbr') is not None:
3638 res += '%4dk' % fdict['vbr']
fbb21cf5 3639 if fdict.get('fps') is not None:
5d583bdf
S
3640 if res:
3641 res += ', '
3642 res += '%sfps' % fdict['fps']
c57f7757
PH
3643 if fdict.get('acodec') is not None:
3644 if res:
3645 res += ', '
3646 if fdict['acodec'] == 'none':
3647 res += 'video only'
3648 else:
3649 res += '%-5s' % fdict['acodec']
3650 elif fdict.get('abr') is not None:
3651 if res:
3652 res += ', '
3653 res += 'audio'
3654 if fdict.get('abr') is not None:
3655 res += '@%3dk' % fdict['abr']
3656 if fdict.get('asr') is not None:
3657 res += ' (%5dHz)' % fdict['asr']
3658 if fdict.get('filesize') is not None:
3659 if res:
3660 res += ', '
3661 res += format_bytes(fdict['filesize'])
9732d77e
PH
3662 elif fdict.get('filesize_approx') is not None:
3663 if res:
3664 res += ', '
3665 res += '~' + format_bytes(fdict['filesize_approx'])
c57f7757 3666 return res
91c7271a 3667
aebb4f4b 3668 def _get_formats(self, info_dict):
3669 if info_dict.get('formats') is None:
3670 if info_dict.get('url') and info_dict.get('_type', 'video') == 'video':
3671 return [info_dict]
3672 return []
3673 return info_dict['formats']
b69fd25c 3674
aebb4f4b 3675 def render_formats_table(self, info_dict):
3676 formats = self._get_formats(info_dict)
3677 if not formats:
3678 return
8130779d 3679 if not self.params.get('listformats_table', True) is not False:
76d321f6 3680 table = [
3681 [
3682 format_field(f, 'format_id'),
3683 format_field(f, 'ext'),
3684 self.format_resolution(f),
8130779d 3685 self._format_note(f)
d5d1df8a 3686 ] for f in formats if (f.get('preference') or 0) >= -1000]
8130779d 3687 return render_table(['format code', 'extension', 'resolution', 'note'], table, extra_gap=1)
3688
d816f61f 3689 def simplified_codec(f, field):
3690 assert field in ('acodec', 'vcodec')
3691 codec = f.get(field, 'unknown')
f5ea4748 3692 if not codec:
3693 return 'unknown'
3694 elif codec != 'none':
d816f61f 3695 return '.'.join(codec.split('.')[:4])
3696
3697 if field == 'vcodec' and f.get('acodec') == 'none':
3698 return 'images'
3699 elif field == 'acodec' and f.get('vcodec') == 'none':
3700 return ''
3701 return self._format_out('audio only' if field == 'vcodec' else 'video only',
3702 self.Styles.SUPPRESS)
3703
591bb9d3 3704 delim = self._format_out('\u2502', self.Styles.DELIM, '|', test_encoding=True)
8130779d 3705 table = [
3706 [
591bb9d3 3707 self._format_out(format_field(f, 'format_id'), self.Styles.ID),
8130779d 3708 format_field(f, 'ext'),
3709 format_field(f, func=self.format_resolution, ignore=('audio only', 'images')),
563e0bf8 3710 format_field(f, 'fps', '\t%d', func=round),
8130779d 3711 format_field(f, 'dynamic_range', '%s', ignore=(None, 'SDR')).replace('HDR', ''),
b8ed0f15 3712 format_field(f, 'audio_channels', '\t%s'),
0dff8e4d 3713 delim, (
3714 format_field(f, 'filesize', ' \t%s', func=format_bytes)
3715 or format_field(f, 'filesize_approx', '≈\t%s', func=format_bytes)
3716 or format_field(try_call(lambda: format_bytes(int(info_dict['duration'] * f['tbr'] * (1024 / 8)))),
3717 None, self._format_out('~\t%s', self.Styles.SUPPRESS))),
563e0bf8 3718 format_field(f, 'tbr', '\t%dk', func=round),
8130779d 3719 shorten_protocol_name(f.get('protocol', '')),
3720 delim,
d816f61f 3721 simplified_codec(f, 'vcodec'),
563e0bf8 3722 format_field(f, 'vbr', '\t%dk', func=round),
d816f61f 3723 simplified_codec(f, 'acodec'),
563e0bf8 3724 format_field(f, 'abr', '\t%dk', func=round),
ae61d108 3725 format_field(f, 'asr', '\t%s', func=format_decimal_suffix),
8130779d 3726 join_nonempty(
591bb9d3 3727 self._format_out('UNSUPPORTED', 'light red') if f.get('ext') in ('f4f', 'f4m') else None,
a5387729 3728 self._format_out('DRM', 'light red') if f.get('has_drm') else None,
8130779d 3729 format_field(f, 'language', '[%s]'),
3730 join_nonempty(format_field(f, 'format_note'),
3731 format_field(f, 'container', ignore=(None, f.get('ext'))),
3732 delim=', '),
3733 delim=' '),
3734 ] for f in formats if f.get('preference') is None or f['preference'] >= -1000]
3735 header_line = self._list_format_headers(
b8ed0f15 3736 'ID', 'EXT', 'RESOLUTION', '\tFPS', 'HDR', 'CH', delim, '\tFILESIZE', '\tTBR', 'PROTO',
8130779d 3737 delim, 'VCODEC', '\tVBR', 'ACODEC', '\tABR', '\tASR', 'MORE INFO')
3738
3739 return render_table(
3740 header_line, table, hide_empty=True,
591bb9d3 3741 delim=self._format_out('\u2500', self.Styles.DELIM, '-', test_encoding=True))
8130779d 3742
3743 def render_thumbnails_table(self, info_dict):
88f23a18 3744 thumbnails = list(info_dict.get('thumbnails') or [])
cfb56d1a 3745 if not thumbnails:
8130779d 3746 return None
3747 return render_table(
ec11a9f4 3748 self._list_format_headers('ID', 'Width', 'Height', 'URL'),
177662e0 3749 [[t.get('id'), t.get('width') or 'unknown', t.get('height') or 'unknown', t['url']] for t in thumbnails])
2412044c 3750
8130779d 3751 def render_subtitles_table(self, video_id, subtitles):
2412044c 3752 def _row(lang, formats):
49c258e1 3753 exts, names = zip(*((f['ext'], f.get('name') or 'unknown') for f in reversed(formats)))
2412044c 3754 if len(set(names)) == 1:
7aee40c1 3755 names = [] if names[0] == 'unknown' else names[:1]
2412044c 3756 return [lang, ', '.join(names), ', '.join(exts)]
3757
8130779d 3758 if not subtitles:
3759 return None
3760 return render_table(
ec11a9f4 3761 self._list_format_headers('Language', 'Name', 'Formats'),
2412044c 3762 [_row(lang, formats) for lang, formats in subtitles.items()],
8130779d 3763 hide_empty=True)
3764
3765 def __list_table(self, video_id, name, func, *args):
3766 table = func(*args)
3767 if not table:
3768 self.to_screen(f'{video_id} has no {name}')
3769 return
3770 self.to_screen(f'[info] Available {name} for {video_id}:')
3771 self.to_stdout(table)
3772
3773 def list_formats(self, info_dict):
3774 self.__list_table(info_dict['id'], 'formats', self.render_formats_table, info_dict)
3775
3776 def list_thumbnails(self, info_dict):
3777 self.__list_table(info_dict['id'], 'thumbnails', self.render_thumbnails_table, info_dict)
3778
3779 def list_subtitles(self, video_id, subtitles, name='subtitles'):
3780 self.__list_table(video_id, name, self.render_subtitles_table, video_id, subtitles)
a504ced0 3781
dca08720
PH
3782 def urlopen(self, req):
3783 """ Start an HTTP download """
f9934b96 3784 if isinstance(req, str):
67dda517 3785 req = sanitized_Request(req)
19a41fc6 3786 return self._opener.open(req, timeout=self._socket_timeout)
dca08720
PH
3787
3788 def print_debug_header(self):
3789 if not self.params.get('verbose'):
3790 return
49a57e70 3791
a057779d 3792 from . import _IN_CLI # Must be delayed import
3793
560738f3 3794 # These imports can be slow. So import them only as needed
3795 from .extractor.extractors import _LAZY_LOADER
e756f45b
M
3796 from .extractor.extractors import (
3797 _PLUGIN_CLASSES as plugin_ies,
3798 _PLUGIN_OVERRIDES as plugin_ie_overrides
3799 )
560738f3 3800
49a57e70 3801 def get_encoding(stream):
2a938746 3802 ret = str(getattr(stream, 'encoding', 'missing (%s)' % type(stream).__name__))
8417f26b
SS
3803 additional_info = []
3804 if os.environ.get('TERM', '').lower() == 'dumb':
3805 additional_info.append('dumb')
49a57e70 3806 if not supports_terminal_sequences(stream):
53973b4d 3807 from .utils import WINDOWS_VT_MODE # Must be imported locally
8417f26b
SS
3808 additional_info.append('No VT' if WINDOWS_VT_MODE is False else 'No ANSI')
3809 if additional_info:
3810 ret = f'{ret} ({",".join(additional_info)})'
49a57e70 3811 return ret
3812
591bb9d3 3813 encoding_str = 'Encodings: locale %s, fs %s, pref %s, %s' % (
49a57e70 3814 locale.getpreferredencoding(),
3815 sys.getfilesystemencoding(),
591bb9d3 3816 self.get_encoding(),
3817 ', '.join(
64fa820c 3818 f'{key} {get_encoding(stream)}' for key, stream in self._out_files.items_
591bb9d3 3819 if stream is not None and key != 'console')
3820 )
883d4b1e 3821
3822 logger = self.params.get('logger')
3823 if logger:
3824 write_debug = lambda msg: logger.debug(f'[debug] {msg}')
3825 write_debug(encoding_str)
3826 else:
96565c7e 3827 write_string(f'[debug] {encoding_str}\n', encoding=None)
49a57e70 3828 write_debug = lambda msg: self._write_string(f'[debug] {msg}\n')
734f90bb 3829
4c88ff87 3830 source = detect_variant()
70b23409 3831 if VARIANT not in (None, 'pip'):
3832 source += '*'
a5387729 3833 klass = type(self)
36eaf303 3834 write_debug(join_nonempty(
b5e7a2e6 3835 f'{"yt-dlp" if REPOSITORY == "yt-dlp/yt-dlp" else REPOSITORY} version',
392389b7 3836 f'{CHANNEL}@{__version__}',
29cb20bd 3837 f'[{RELEASE_GIT_HEAD[:9]}]' if RELEASE_GIT_HEAD else '',
36eaf303 3838 '' if source == 'unknown' else f'({source})',
a5387729 3839 '' if _IN_CLI else 'API' if klass == YoutubeDL else f'API:{self.__module__}.{klass.__qualname__}',
36eaf303 3840 delim=' '))
497074f0 3841
3842 if not _IN_CLI:
3843 write_debug(f'params: {self.params}')
3844
6e21fdd2 3845 if not _LAZY_LOADER:
3846 if os.environ.get('YTDLP_NO_LAZY_EXTRACTORS'):
49a57e70 3847 write_debug('Lazy loading extractors is forcibly disabled')
6e21fdd2 3848 else:
49a57e70 3849 write_debug('Lazy loading extractors is disabled')
8a82af35 3850 if self.params['compat_opts']:
3851 write_debug('Compatibility options: %s' % ', '.join(self.params['compat_opts']))
36eaf303 3852
b5e7a2e6 3853 if current_git_head():
3854 write_debug(f'Git HEAD: {current_git_head()}')
b1f94422 3855 write_debug(system_identifier())
d28b5171 3856
8913ef74 3857 exe_versions, ffmpeg_features = FFmpegPostProcessor.get_versions_and_features(self)
3858 ffmpeg_features = {key for key, val in ffmpeg_features.items() if val}
3859 if ffmpeg_features:
19a03940 3860 exe_versions['ffmpeg'] += ' (%s)' % ','.join(sorted(ffmpeg_features))
8913ef74 3861
4c83c967 3862 exe_versions['rtmpdump'] = rtmpdump_version()
feee8d32 3863 exe_versions['phantomjs'] = PhantomJSwrapper._version()
d28b5171 3864 exe_str = ', '.join(
2831b468 3865 f'{exe} {v}' for exe, v in sorted(exe_versions.items()) if v
3866 ) or 'none'
49a57e70 3867 write_debug('exe versions: %s' % exe_str)
dca08720 3868
1d485a1a 3869 from .compat.compat_utils import get_package_info
9b8ee23b 3870 from .dependencies import available_dependencies
3871
3872 write_debug('Optional libraries: %s' % (', '.join(sorted({
1d485a1a 3873 join_nonempty(*get_package_info(m)) for m in available_dependencies.values()
9b8ee23b 3874 })) or 'none'))
2831b468 3875
97ec5bc5 3876 self._setup_opener()
dca08720
PH
3877 proxy_map = {}
3878 for handler in self._opener.handlers:
3879 if hasattr(handler, 'proxies'):
3880 proxy_map.update(handler.proxies)
49a57e70 3881 write_debug(f'Proxy map: {proxy_map}')
dca08720 3882
e756f45b
M
3883 for plugin_type, plugins in {'Extractor': plugin_ies, 'Post-Processor': plugin_pps}.items():
3884 display_list = ['%s%s' % (
8e40b9d1 3885 klass.__name__, '' if klass.__name__ == name else f' as {name}')
e756f45b
M
3886 for name, klass in plugins.items()]
3887 if plugin_type == 'Extractor':
3888 display_list.extend(f'{plugins[-1].IE_NAME.partition("+")[2]} ({parent.__name__})'
3889 for parent, plugins in plugin_ie_overrides.items())
3890 if not display_list:
3891 continue
3892 write_debug(f'{plugin_type} Plugins: {", ".join(sorted(display_list))}')
3893
8e40b9d1
M
3894 plugin_dirs = plugin_directories()
3895 if plugin_dirs:
3896 write_debug(f'Plugin directories: {plugin_dirs}')
3897
49a57e70 3898 # Not implemented
3899 if False and self.params.get('call_home'):
0f06bcd7 3900 ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode()
49a57e70 3901 write_debug('Public IP address: %s' % ipaddr)
58b1f00d 3902 latest_version = self.urlopen(
0f06bcd7 3903 'https://yt-dl.org/latest/version').read().decode()
58b1f00d
PH
3904 if version_tuple(latest_version) > version_tuple(__version__):
3905 self.report_warning(
3906 'You are using an outdated version (newest version: %s)! '
3907 'See https://yt-dl.org/update if you need help updating.' %
3908 latest_version)
3909
e344693b 3910 def _setup_opener(self):
97ec5bc5 3911 if hasattr(self, '_opener'):
3912 return
6ad14cab 3913 timeout_val = self.params.get('socket_timeout')
17bddf3e 3914 self._socket_timeout = 20 if timeout_val is None else float(timeout_val)
6ad14cab 3915
982ee69a 3916 opts_cookiesfrombrowser = self.params.get('cookiesfrombrowser')
dca08720
PH
3917 opts_cookiefile = self.params.get('cookiefile')
3918 opts_proxy = self.params.get('proxy')
3919
982ee69a 3920 self.cookiejar = load_cookies(opts_cookiefile, opts_cookiesfrombrowser, self)
dca08720 3921
6a3f4c3f 3922 cookie_processor = YoutubeDLCookieProcessor(self.cookiejar)
dca08720
PH
3923 if opts_proxy is not None:
3924 if opts_proxy == '':
3925 proxies = {}
3926 else:
3927 proxies = {'http': opts_proxy, 'https': opts_proxy}
3928 else:
ac668111 3929 proxies = urllib.request.getproxies()
067aa17e 3930 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
dca08720
PH
3931 if 'http' in proxies and 'https' not in proxies:
3932 proxies['https'] = proxies['http']
91410c9b 3933 proxy_handler = PerRequestProxyHandler(proxies)
a0ddb8a2
PH
3934
3935 debuglevel = 1 if self.params.get('debug_printtraffic') else 0
be4a824d
PH
3936 https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel)
3937 ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel)
fca6dba8 3938 redirect_handler = YoutubeDLRedirectHandler()
f9934b96 3939 data_handler = urllib.request.DataHandler()
6240b0a2
JMF
3940
3941 # When passing our own FileHandler instance, build_opener won't add the
3942 # default FileHandler and allows us to disable the file protocol, which
3943 # can be used for malicious purposes (see
067aa17e 3944 # https://github.com/ytdl-org/youtube-dl/issues/8227)
ac668111 3945 file_handler = urllib.request.FileHandler()
6240b0a2 3946
8300774c
M
3947 if not self.params.get('enable_file_urls'):
3948 def file_open(*args, **kwargs):
3949 raise urllib.error.URLError(
3950 'file:// URLs are explicitly disabled in yt-dlp for security reasons. '
3951 'Use --enable-file-urls to enable at your own risk.')
3952 file_handler.file_open = file_open
6240b0a2 3953
ac668111 3954 opener = urllib.request.build_opener(
fca6dba8 3955 proxy_handler, https_handler, cookie_processor, ydlh, redirect_handler, data_handler, file_handler)
2461f79d 3956
dca08720
PH
3957 # Delete the default user-agent header, which would otherwise apply in
3958 # cases where our custom HTTP handler doesn't come into play
067aa17e 3959 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
dca08720
PH
3960 opener.addheaders = []
3961 self._opener = opener
62fec3b2
PH
3962
3963 def encode(self, s):
3964 if isinstance(s, bytes):
3965 return s # Already encoded
3966
3967 try:
3968 return s.encode(self.get_encoding())
3969 except UnicodeEncodeError as err:
3970 err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.'
3971 raise
3972
3973 def get_encoding(self):
3974 encoding = self.params.get('encoding')
3975 if encoding is None:
3976 encoding = preferredencoding()
3977 return encoding
ec82d85a 3978
e08a85d8 3979 def _write_info_json(self, label, ie_result, infofn, overwrite=None):
cb96c5be 3980 ''' Write infojson and returns True = written, 'exists' = Already exists, False = skip, None = error '''
e08a85d8 3981 if overwrite is None:
3982 overwrite = self.params.get('overwrites', True)
80c03fa9 3983 if not self.params.get('writeinfojson'):
3984 return False
3985 elif not infofn:
3986 self.write_debug(f'Skipping writing {label} infojson')
3987 return False
3988 elif not self._ensure_dir_exists(infofn):
3989 return None
e08a85d8 3990 elif not overwrite and os.path.exists(infofn):
80c03fa9 3991 self.to_screen(f'[info] {label.title()} metadata is already present')
cb96c5be 3992 return 'exists'
3993
3994 self.to_screen(f'[info] Writing {label} metadata as JSON to: {infofn}')
3995 try:
3996 write_json_file(self.sanitize_info(ie_result, self.params.get('clean_infojson', True)), infofn)
3997 return True
86e5f3ed 3998 except OSError:
cb96c5be 3999 self.report_error(f'Cannot write {label} metadata to JSON file {infofn}')
4000 return None
80c03fa9 4001
4002 def _write_description(self, label, ie_result, descfn):
4003 ''' Write description and returns True = written, False = skip, None = error '''
4004 if not self.params.get('writedescription'):
4005 return False
4006 elif not descfn:
4007 self.write_debug(f'Skipping writing {label} description')
4008 return False
4009 elif not self._ensure_dir_exists(descfn):
4010 return None
4011 elif not self.params.get('overwrites', True) and os.path.exists(descfn):
4012 self.to_screen(f'[info] {label.title()} description is already present')
4013 elif ie_result.get('description') is None:
88fb9425 4014 self.to_screen(f'[info] There\'s no {label} description to write')
80c03fa9 4015 return False
4016 else:
4017 try:
4018 self.to_screen(f'[info] Writing {label} description to: {descfn}')
86e5f3ed 4019 with open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
80c03fa9 4020 descfile.write(ie_result['description'])
86e5f3ed 4021 except OSError:
80c03fa9 4022 self.report_error(f'Cannot write {label} description file {descfn}')
4023 return None
4024 return True
4025
4026 def _write_subtitles(self, info_dict, filename):
4027 ''' Write subtitles to file and return list of (sub_filename, final_sub_filename); or None if error'''
4028 ret = []
4029 subtitles = info_dict.get('requested_subtitles')
88fb9425 4030 if not (self.params.get('writesubtitles') or self.params.get('writeautomaticsub')):
80c03fa9 4031 # subtitles download errors are already managed as troubles in relevant IE
4032 # that way it will silently go on when used with unsupporting IE
4033 return ret
88fb9425 4034 elif not subtitles:
c8bc203f 4035 self.to_screen('[info] There are no subtitles for the requested languages')
88fb9425 4036 return ret
80c03fa9 4037 sub_filename_base = self.prepare_filename(info_dict, 'subtitle')
4038 if not sub_filename_base:
4039 self.to_screen('[info] Skipping writing video subtitles')
4040 return ret
88fb9425 4041
80c03fa9 4042 for sub_lang, sub_info in subtitles.items():
4043 sub_format = sub_info['ext']
4044 sub_filename = subtitles_filename(filename, sub_lang, sub_format, info_dict.get('ext'))
4045 sub_filename_final = subtitles_filename(sub_filename_base, sub_lang, sub_format, info_dict.get('ext'))
e04938ab 4046 existing_sub = self.existing_file((sub_filename_final, sub_filename))
4047 if existing_sub:
80c03fa9 4048 self.to_screen(f'[info] Video subtitle {sub_lang}.{sub_format} is already present')
e04938ab 4049 sub_info['filepath'] = existing_sub
4050 ret.append((existing_sub, sub_filename_final))
80c03fa9 4051 continue
4052
4053 self.to_screen(f'[info] Writing video subtitles to: {sub_filename}')
4054 if sub_info.get('data') is not None:
4055 try:
4056 # Use newline='' to prevent conversion of newline characters
4057 # See https://github.com/ytdl-org/youtube-dl/issues/10268
86e5f3ed 4058 with open(sub_filename, 'w', encoding='utf-8', newline='') as subfile:
80c03fa9 4059 subfile.write(sub_info['data'])
4060 sub_info['filepath'] = sub_filename
4061 ret.append((sub_filename, sub_filename_final))
4062 continue
86e5f3ed 4063 except OSError:
80c03fa9 4064 self.report_error(f'Cannot write video subtitles file {sub_filename}')
4065 return None
4066
4067 try:
4068 sub_copy = sub_info.copy()
4069 sub_copy.setdefault('http_headers', info_dict.get('http_headers'))
4070 self.dl(sub_filename, sub_copy, subtitle=True)
4071 sub_info['filepath'] = sub_filename
4072 ret.append((sub_filename, sub_filename_final))
6020e05d 4073 except (DownloadError, ExtractorError, IOError, OSError, ValueError) + network_exceptions as err:
c70c418d 4074 msg = f'Unable to download video subtitles for {sub_lang!r}: {err}'
6020e05d 4075 if self.params.get('ignoreerrors') is not True: # False or 'only_download'
c70c418d 4076 if not self.params.get('ignoreerrors'):
4077 self.report_error(msg)
4078 raise DownloadError(msg)
4079 self.report_warning(msg)
519804a9 4080 return ret
80c03fa9 4081
4082 def _write_thumbnails(self, label, info_dict, filename, thumb_filename_base=None):
4083 ''' Write thumbnails to file and return list of (thumb_filename, final_thumb_filename) '''
6c4fd172 4084 write_all = self.params.get('write_all_thumbnails', False)
80c03fa9 4085 thumbnails, ret = [], []
6c4fd172 4086 if write_all or self.params.get('writethumbnail', False):
0202b52a 4087 thumbnails = info_dict.get('thumbnails') or []
88fb9425 4088 if not thumbnails:
c8bc203f 4089 self.to_screen(f'[info] There are no {label} thumbnails to download')
88fb9425 4090 return ret
6c4fd172 4091 multiple = write_all and len(thumbnails) > 1
ec82d85a 4092
80c03fa9 4093 if thumb_filename_base is None:
4094 thumb_filename_base = filename
4095 if thumbnails and not thumb_filename_base:
4096 self.write_debug(f'Skipping writing {label} thumbnail')
4097 return ret
4098
dd0228ce 4099 for idx, t in list(enumerate(thumbnails))[::-1]:
80c03fa9 4100 thumb_ext = (f'{t["id"]}.' if multiple else '') + determine_ext(t['url'], 'jpg')
aa9369a2 4101 thumb_display_id = f'{label} thumbnail {t["id"]}'
80c03fa9 4102 thumb_filename = replace_extension(filename, thumb_ext, info_dict.get('ext'))
4103 thumb_filename_final = replace_extension(thumb_filename_base, thumb_ext, info_dict.get('ext'))
ec82d85a 4104
e04938ab 4105 existing_thumb = self.existing_file((thumb_filename_final, thumb_filename))
4106 if existing_thumb:
aa9369a2 4107 self.to_screen('[info] %s is already present' % (
4108 thumb_display_id if multiple else f'{label} thumbnail').capitalize())
e04938ab 4109 t['filepath'] = existing_thumb
4110 ret.append((existing_thumb, thumb_filename_final))
ec82d85a 4111 else:
80c03fa9 4112 self.to_screen(f'[info] Downloading {thumb_display_id} ...')
ec82d85a 4113 try:
297e9952 4114 uf = self.urlopen(sanitized_Request(t['url'], headers=t.get('http_headers', {})))
80c03fa9 4115 self.to_screen(f'[info] Writing {thumb_display_id} to: {thumb_filename}')
d3d89c32 4116 with open(encodeFilename(thumb_filename), 'wb') as thumbf:
ec82d85a 4117 shutil.copyfileobj(uf, thumbf)
80c03fa9 4118 ret.append((thumb_filename, thumb_filename_final))
885cc0b7 4119 t['filepath'] = thumb_filename
3158150c 4120 except network_exceptions as err:
ad54c913 4121 if isinstance(err, urllib.error.HTTPError) and err.code == 404:
4122 self.to_screen(f'[info] {thumb_display_id.title()} does not exist')
4123 else:
4124 self.report_warning(f'Unable to download {thumb_display_id}: {err}')
dd0228ce 4125 thumbnails.pop(idx)
6c4fd172 4126 if ret and not write_all:
4127 break
0202b52a 4128 return ret