]> jfr.im git - yt-dlp.git/blame - yt_dlp/YoutubeDL.py
[core] Include build origin in verbose output
[yt-dlp.git] / yt_dlp / YoutubeDL.py
CommitLineData
26e63931 1import collections
31bd3925 2import contextlib
31215122 3import copy
9d2ecdbc 4import datetime
c1c9a79c 5import errno
31bd3925 6import fileinput
31215122 7import http.cookiejar
8222d8de 8import io
b82f815f 9import itertools
8694c600 10import json
62fec3b2 11import locale
083c9df9 12import operator
8222d8de 13import os
f8271158 14import random
8222d8de
JMF
15import re
16import shutil
6f2287cb 17import string
dca08720 18import subprocess
8222d8de 19import sys
21cd8fae 20import tempfile
8222d8de 21import time
67134eab 22import tokenize
8222d8de 23import traceback
524e2e4f 24import unicodedata
961ea474 25
f8271158 26from .cache import Cache
227bf1a3 27from .compat import functools, urllib # isort: split
28from .compat import compat_os_name, compat_shlex_quote, urllib_req_to_req
31215122 29from .cookies import LenientSimpleCookie, load_cookies
f8271158 30from .downloader import FFmpegFD, get_suitable_downloader, shorten_protocol_name
31from .downloader.rtmp import rtmpdump_version
f8271158 32from .extractor import gen_extractor_classes, get_info_extractor
fe7866d0 33from .extractor.common import UnsupportedURLIE
f8271158 34from .extractor.openload import PhantomJSwrapper
35from .minicurses import format_text
3d2623a8 36from .networking import HEADRequest, Request, RequestDirector
db7b054a 37from .networking.common import _REQUEST_HANDLERS, _RH_PREFERENCES
227bf1a3 38from .networking.exceptions import (
39 HTTPError,
40 NoSupportingHandlers,
41 RequestError,
42 SSLError,
43 _CompatHTTPError,
3d2623a8 44 network_exceptions,
227bf1a3 45)
8e40b9d1 46from .plugins import directories as plugin_directories
e756f45b 47from .postprocessor import _PLUGIN_CLASSES as plugin_pps
f8271158 48from .postprocessor import (
49 EmbedThumbnailPP,
50 FFmpegFixupDuplicateMoovPP,
51 FFmpegFixupDurationPP,
52 FFmpegFixupM3u8PP,
53 FFmpegFixupM4aPP,
54 FFmpegFixupStretchedPP,
55 FFmpegFixupTimestampPP,
56 FFmpegMergerPP,
57 FFmpegPostProcessor,
ca9def71 58 FFmpegVideoConvertorPP,
f8271158 59 MoveFilesAfterDownloadPP,
60 get_postprocessor,
61)
ca9def71 62from .postprocessor.ffmpeg import resolve_mapping as resolve_recode_mapping
61bdf15f 63from .update import REPOSITORY, _get_system_deprecation, current_git_head, detect_variant
8c25f81b 64from .utils import (
f8271158 65 DEFAULT_OUTTMPL,
7b2c3f47 66 IDENTITY,
f8271158 67 LINK_TEMPLATES,
8dc59305 68 MEDIA_EXTENSIONS,
f8271158 69 NO_DEFAULT,
1d485a1a 70 NUMBER_RE,
f8271158 71 OUTTMPL_TYPES,
72 POSTPROCESS_WHEN,
73 STR_FORMAT_RE_TMPL,
74 STR_FORMAT_TYPES,
75 ContentTooShortError,
76 DateRange,
77 DownloadCancelled,
78 DownloadError,
79 EntryNotInPlaylist,
80 ExistingVideoReached,
81 ExtractorError,
784320c9 82 FormatSorter,
f8271158 83 GeoRestrictedError,
f8271158 84 ISO3166Utils,
85 LazyList,
86 MaxDownloadsReached,
19a03940 87 Namespace,
f8271158 88 PagedList,
7e88d7d7 89 PlaylistEntries,
f8271158 90 Popen,
91 PostProcessingError,
92 ReExtractInfo,
93 RejectedVideoReached,
94 SameFileError,
95 UnavailableVideoError,
693f0600 96 UserNotLive,
eedb7ba5
S
97 age_restricted,
98 args_to_str,
cb794ee0 99 bug_reports_message,
ce02ed60 100 date_from_str,
da4db748 101 deprecation_warning,
ce02ed60 102 determine_ext,
b5559424 103 determine_protocol,
c0384f22 104 encode_compat_str,
ce02ed60 105 encodeFilename,
a06916d9 106 error_to_compat_str,
47cdc68e 107 escapeHTML,
590bc6f6 108 expand_path,
227bf1a3 109 extract_basic_auth,
90137ca4 110 filter_dict,
e29663c6 111 float_or_none,
02dbf93f 112 format_bytes,
e0fd9573 113 format_decimal_suffix,
f8271158 114 format_field,
525ef922 115 formatSeconds,
fc61aff4 116 get_compatible_ext,
0bb322b9 117 get_domain,
c9969434 118 int_or_none,
732044af 119 iri_to_uri,
941e881e 120 is_path_like,
34921b43 121 join_nonempty,
ce02ed60 122 locked_file,
0647d925 123 make_archive_id,
0202b52a 124 make_dir,
ec11a9f4 125 number_of_digits,
cd6fc19e 126 orderedSet,
5314b521 127 orderedSet_from_options,
083c9df9 128 parse_filesize,
ce02ed60 129 preferredencoding,
eedb7ba5 130 prepend_extension,
3efb96a6 131 remove_terminal_sequences,
cfb56d1a 132 render_table,
eedb7ba5 133 replace_extension,
ce02ed60 134 sanitize_filename,
1bb5c511 135 sanitize_path,
dcf77cf1 136 sanitize_url,
1211bb6d 137 str_or_none,
e29663c6 138 strftime_or_none,
ce02ed60 139 subtitles_filename,
819e0531 140 supports_terminal_sequences,
b1f94422 141 system_identifier,
f2ebc5c7 142 timetuple_from_msec,
732044af 143 to_high_limit_path,
324ad820 144 traverse_obj,
fc61aff4 145 try_call,
6033d980 146 try_get,
29eb5174 147 url_basename,
7d1eb38a 148 variadic,
58b1f00d 149 version_tuple,
53973b4d 150 windows_enable_vt_mode,
ce02ed60
PH
151 write_json_file,
152 write_string,
4f026faf 153)
227bf1a3 154from .utils._utils import _YDLLogger
155from .utils.networking import (
156 HTTPHeaderDict,
157 clean_headers,
158 clean_proxies,
3d2623a8 159 std_headers,
227bf1a3 160)
20314dd4 161from .version import CHANNEL, ORIGIN, RELEASE_GIT_HEAD, VARIANT, __version__
8222d8de 162
e9c0cdd3
YCH
163if compat_os_name == 'nt':
164 import ctypes
165
2459b6e1 166
86e5f3ed 167class YoutubeDL:
8222d8de
JMF
168 """YoutubeDL class.
169
170 YoutubeDL objects are the ones responsible of downloading the
171 actual video file and writing it to disk if the user has requested
172 it, among some other tasks. In most cases there should be one per
173 program. As, given a video URL, the downloader doesn't know how to
174 extract all the needed information, task that InfoExtractors do, it
175 has to pass the URL to one of them.
176
177 For this, YoutubeDL objects have a method that allows
178 InfoExtractors to be registered in a given order. When it is passed
179 a URL, the YoutubeDL object handles it to the first InfoExtractor it
180 finds that reports being able to handle it. The InfoExtractor extracts
181 all the information about the video or videos the URL refers to, and
182 YoutubeDL process the extracted information, possibly using a File
183 Downloader to download the video.
184
185 YoutubeDL objects accept a lot of parameters. In order not to saturate
186 the object constructor with arguments, it receives a dictionary of
187 options instead. These options are available through the params
188 attribute for the InfoExtractors to use. The YoutubeDL also
189 registers itself as the downloader in charge for the InfoExtractors
190 that are added to it, so this is a "mutual registration".
191
192 Available options:
193
194 username: Username for authentication purposes.
195 password: Password for authentication purposes.
180940e0 196 videopassword: Password for accessing a video.
1da50aa3
S
197 ap_mso: Adobe Pass multiple-system operator identifier.
198 ap_username: Multiple-system operator account username.
199 ap_password: Multiple-system operator account password.
8222d8de 200 usenetrc: Use netrc for authentication instead.
c8bc203f 201 netrc_location: Location of the netrc file. Defaults to ~/.netrc.
db3ad8a6 202 netrc_cmd: Use a shell command to get credentials
8222d8de
JMF
203 verbose: Print additional info to stdout.
204 quiet: Do not print messages to stdout.
ad8915b7 205 no_warnings: Do not print out anything for warnings.
bb66c247 206 forceprint: A dict with keys WHEN mapped to a list of templates to
207 print to stdout. The allowed keys are video or any of the
208 items in utils.POSTPROCESS_WHEN.
ca30f449 209 For compatibility, a single list is also accepted
bb66c247 210 print_to_file: A dict with keys WHEN (same as forceprint) mapped to
211 a list of tuples with (template, filename)
8694c600 212 forcejson: Force printing info_dict as JSON.
63e0be34
PH
213 dump_single_json: Force printing the info_dict of the whole playlist
214 (or video) as a single JSON line.
c25228e5 215 force_write_download_archive: Force writing download archive regardless
216 of 'skip_download' or 'simulate'.
b7b04c78 217 simulate: Do not download the video files. If unset (or None),
218 simulate only if listsubtitles, listformats or list_thumbnails is used
eb8a4433 219 format: Video format code. see "FORMAT SELECTION" for more details.
093a1710 220 You can also pass a function. The function takes 'ctx' as
221 argument and returns the formats to download.
222 See "build_format_selector" for an implementation
63ad4d43 223 allow_unplayable_formats: Allow unplayable formats to be extracted and downloaded.
b7da73eb 224 ignore_no_formats_error: Ignore "No video formats" error. Usefull for
225 extracting metadata even if the video is not actually
226 available for download (experimental)
0930b11f 227 format_sort: A list of fields by which to sort the video formats.
228 See "Sorting Formats" for more details.
c25228e5 229 format_sort_force: Force the given format_sort. see "Sorting Formats"
230 for more details.
08d30158 231 prefer_free_formats: Whether to prefer video formats with free containers
232 over non-free ones of same quality.
c25228e5 233 allow_multiple_video_streams: Allow multiple video streams to be merged
234 into a single file
235 allow_multiple_audio_streams: Allow multiple audio streams to be merged
236 into a single file
0ba692ac 237 check_formats Whether to test if the formats are downloadable.
9f1a1c36 238 Can be True (check all), False (check none),
239 'selected' (check selected formats),
0ba692ac 240 or None (check only if requested by extractor)
4524baf0 241 paths: Dictionary of output paths. The allowed keys are 'home'
5ca095cb 242 'temp' and the keys of OUTTMPL_TYPES (in utils/_utils.py)
de6000d9 243 outtmpl: Dictionary of templates for output names. Allowed keys
5ca095cb 244 are 'default' and the keys of OUTTMPL_TYPES (in utils/_utils.py).
34488702 245 For compatibility with youtube-dl, a single string can also be used
a820dc72
RA
246 outtmpl_na_placeholder: Placeholder for unavailable meta fields.
247 restrictfilenames: Do not allow "&" and spaces in file names
248 trim_file_name: Limit length of filename (extension excluded)
4524baf0 249 windowsfilenames: Force the filenames to be windows compatible
b1940459 250 ignoreerrors: Do not stop on download/postprocessing errors.
251 Can be 'only_download' to ignore only download errors.
252 Default is 'only_download' for CLI, but False for API
26e2805c 253 skip_playlist_after_errors: Number of allowed failures until the rest of
254 the playlist is skipped
fe7866d0 255 allowed_extractors: List of regexes to match against extractor names that are allowed
0c3d0f51 256 overwrites: Overwrite all video and metadata files if True,
257 overwrite only non-video files if None
258 and don't overwrite any file if False
c14e88f0 259 playlist_items: Specific indices of playlist to download.
75822ca7 260 playlistrandom: Download playlist items in random order.
7e9a6125 261 lazy_playlist: Process playlist entries as they are received.
8222d8de
JMF
262 matchtitle: Download only matching titles.
263 rejecttitle: Reject downloads for matching titles.
8bf9319e 264 logger: Log messages to a logging.Logger instance.
17ffed18 265 logtostderr: Print everything to stderr instead of stdout.
266 consoletitle: Display progress in console window's titlebar.
8222d8de
JMF
267 writedescription: Write the video description to a .description file
268 writeinfojson: Write the video description to a .info.json file
ad54c913 269 clean_infojson: Remove internal metadata from the infojson
34488702 270 getcomments: Extract video comments. This will not be written to disk
06167fbb 271 unless writeinfojson is also given
1fb07d10 272 writeannotations: Write the video annotations to a .annotations.xml file
8222d8de 273 writethumbnail: Write the thumbnail image to a file
c25228e5 274 allow_playlist_files: Whether to write playlists' description, infojson etc
275 also to disk when using the 'write*' options
ec82d85a 276 write_all_thumbnails: Write all thumbnail formats to files
732044af 277 writelink: Write an internet shortcut file, depending on the
278 current platform (.url/.webloc/.desktop)
279 writeurllink: Write a Windows internet shortcut file (.url)
280 writewebloclink: Write a macOS internet shortcut file (.webloc)
281 writedesktoplink: Write a Linux internet shortcut file (.desktop)
8222d8de 282 writesubtitles: Write the video subtitles to a file
741dd8ea 283 writeautomaticsub: Write the automatically generated subtitles to a file
8222d8de 284 listsubtitles: Lists all available subtitles for the video
a504ced0 285 subtitlesformat: The format code for subtitles
c32b0aab 286 subtitleslangs: List of languages of the subtitles to download (can be regex).
287 The list may contain "all" to refer to all the available
288 subtitles. The language can be prefixed with a "-" to
62b58c09 289 exclude it from the requested languages, e.g. ['all', '-live_chat']
8222d8de 290 keepvideo: Keep the video file after post-processing
46f1370e 291 daterange: A utils.DateRange object, download only if the upload_date is in the range.
8222d8de 292 skip_download: Skip the actual download of the video file
c35f9e72 293 cachedir: Location of the cache files in the filesystem.
a0e07d31 294 False to disable filesystem cache.
47192f92 295 noplaylist: Download single video instead of a playlist if in doubt.
8dbe9899
PH
296 age_limit: An integer representing the user's age in years.
297 Unsuitable videos for the given age are skipped.
5fe18bdb
PH
298 min_views: An integer representing the minimum view count the video
299 must have in order to not be skipped.
300 Videos without view count information are always
301 downloaded. None for no limit.
302 max_views: An integer representing the maximum view count.
303 Videos that are more popular than that are not
304 downloaded.
305 Videos without view count information are always
306 downloaded. None for no limit.
ae103564 307 download_archive: A set, or the name of a file where all downloads are recorded.
308 Videos already present in the file are not downloaded again.
8a51f564 309 break_on_existing: Stop the download process after attempting to download a
310 file that is in the archive.
b222c271 311 break_per_url: Whether break_on_reject and break_on_existing
312 should act on each input URL as opposed to for the entire queue
d76fa1f3 313 cookiefile: File name or text stream from where cookies should be read and dumped to
f59f5ef8 314 cookiesfrombrowser: A tuple containing the name of the browser, the profile
9bd13fe5 315 name/path from where cookies are loaded, the name of the keyring,
316 and the container name, e.g. ('chrome', ) or
317 ('vivaldi', 'default', 'BASICTEXT') or ('firefox', 'default', None, 'Meta')
f81c62a6 318 legacyserverconnect: Explicitly allow HTTPS connection to servers that do not
319 support RFC 5746 secure renegotiation
f59f5ef8 320 nocheckcertificate: Do not verify SSL certificates
bb58c9ed 321 client_certificate: Path to client certificate file in PEM format. May include the private key
322 client_certificate_key: Path to private key file for client certificate
323 client_certificate_password: Password for client certificate private key, if encrypted.
324 If not provided and the key is encrypted, yt-dlp will ask interactively
7e8c0af0 325 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
c6e07cf1 326 (Only supported by some extractors)
8300774c 327 enable_file_urls: Enable file:// URLs. This is disabled by default for security reasons.
8b7539d2 328 http_headers: A dictionary of custom headers to be used for all requests
a1ee09e8 329 proxy: URL of the proxy server to use
38cce791 330 geo_verification_proxy: URL of the proxy to use for IP address verification
504f20dd 331 on geo-restricted sites.
e344693b 332 socket_timeout: Time to wait for unresponsive hosts, in seconds
0783b09b
PH
333 bidi_workaround: Work around buggy terminals without bidirectional text
334 support, using fridibi
a0ddb8a2 335 debug_printtraffic:Print out sent and received HTTP traffic
04b4d394
PH
336 default_search: Prepend this string if an input url is not valid.
337 'auto' for elaborate guessing
62fec3b2 338 encoding: Use this encoding instead of the system-specified.
134c913c 339 extract_flat: Whether to resolve and process url_results further
46f1370e 340 * False: Always process. Default for API
134c913c 341 * True: Never process
342 * 'in_playlist': Do not process inside playlist/multi_video
343 * 'discard': Always process, but don't return the result
344 from inside playlist/multi_video
345 * 'discard_in_playlist': Same as "discard", but only for
46f1370e 346 playlists (not multi_video). Default for CLI
f2ebc5c7 347 wait_for_video: If given, wait for scheduled streams to become available.
348 The value should be a tuple containing the range
349 (min_secs, max_secs) to wait between retries
4f026faf 350 postprocessors: A list of dictionaries, each with an entry
71b640cc 351 * key: The name of the postprocessor. See
7a5c1cfe 352 yt_dlp/postprocessor/__init__.py for a list.
bb66c247 353 * when: When to run the postprocessor. Allowed values are
354 the entries of utils.POSTPROCESS_WHEN
56d868db 355 Assumed to be 'post_process' if not given
71b640cc
PH
356 progress_hooks: A list of functions that get called on download
357 progress, with a dictionary with the entries
5cda4eda 358 * status: One of "downloading", "error", or "finished".
ee69b99a 359 Check this first and ignore unknown values.
3ba7740d 360 * info_dict: The extracted info_dict
71b640cc 361
5cda4eda 362 If status is one of "downloading", or "finished", the
ee69b99a
PH
363 following properties may also be present:
364 * filename: The final filename (always present)
5cda4eda 365 * tmpfilename: The filename we're currently writing to
71b640cc
PH
366 * downloaded_bytes: Bytes on disk
367 * total_bytes: Size of the whole file, None if unknown
5cda4eda
PH
368 * total_bytes_estimate: Guess of the eventual file size,
369 None if unavailable.
370 * elapsed: The number of seconds since download started.
71b640cc
PH
371 * eta: The estimated time in seconds, None if unknown
372 * speed: The download speed in bytes/second, None if
373 unknown
5cda4eda
PH
374 * fragment_index: The counter of the currently
375 downloaded video fragment.
376 * fragment_count: The number of fragments (= individual
377 files that will be merged)
71b640cc
PH
378
379 Progress hooks are guaranteed to be called at least once
380 (with status "finished") if the download is successful.
819e0531 381 postprocessor_hooks: A list of functions that get called on postprocessing
382 progress, with a dictionary with the entries
383 * status: One of "started", "processing", or "finished".
384 Check this first and ignore unknown values.
385 * postprocessor: Name of the postprocessor
386 * info_dict: The extracted info_dict
387
388 Progress hooks are guaranteed to be called at least twice
389 (with status "started" and "finished") if the processing is successful.
fc61aff4 390 merge_output_format: "/" separated list of extensions to use when merging formats.
6b591b29 391 final_ext: Expected final extension; used to detect when the file was
59a7a13e 392 already downloaded and converted
6271f1ca
PH
393 fixup: Automatically correct known faults of the file.
394 One of:
395 - "never": do nothing
396 - "warn": only emit a warning
397 - "detect_or_warn": check whether we can do anything
62cd676c 398 about it, warn otherwise (default)
504f20dd 399 source_address: Client-side IP address to bind to.
1cf376f5 400 sleep_interval_requests: Number of seconds to sleep between requests
401 during extraction
7aa589a5
S
402 sleep_interval: Number of seconds to sleep before each download when
403 used alone or a lower bound of a range for randomized
404 sleep before each download (minimum possible number
405 of seconds to sleep) when used along with
406 max_sleep_interval.
407 max_sleep_interval:Upper bound of a range for randomized sleep before each
408 download (maximum possible number of seconds to sleep).
409 Must only be used along with sleep_interval.
410 Actual sleep time will be a random float from range
411 [sleep_interval; max_sleep_interval].
1cf376f5 412 sleep_interval_subtitles: Number of seconds to sleep before each subtitle download
cfb56d1a
PH
413 listformats: Print an overview of available video formats and exit.
414 list_thumbnails: Print a table of all thumbnails and exit.
0a41f331 415 match_filter: A function that gets called for every video with the signature
416 (info_dict, *, incomplete: bool) -> Optional[str]
417 For backward compatibility with youtube-dl, the signature
418 (info_dict) -> Optional[str] is also allowed.
419 - If it returns a message, the video is ignored.
420 - If it returns None, the video is downloaded.
421 - If it returns utils.NO_DEFAULT, the user is interactively
422 asked whether to download the video.
fe2ce85a 423 - Raise utils.DownloadCancelled(msg) to abort remaining
424 downloads when a video is rejected.
5ca095cb 425 match_filter_func in utils/_utils.py is one example for this.
8417f26b
SS
426 color: A Dictionary with output stream names as keys
427 and their respective color policy as values.
428 Can also just be a single color policy,
429 in which case it applies to all outputs.
430 Valid stream names are 'stdout' and 'stderr'.
431 Valid color policies are one of 'always', 'auto', 'no_color' or 'never'.
0a840f58 432 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
504f20dd 433 HTTP header
0a840f58 434 geo_bypass_country:
773f291d
S
435 Two-letter ISO 3166-2 country code that will be used for
436 explicit geographic restriction bypassing via faking
504f20dd 437 X-Forwarded-For HTTP header
5f95927a
S
438 geo_bypass_ip_block:
439 IP range in CIDR notation that will be used similarly to
504f20dd 440 geo_bypass_country
52a8a1e1 441 external_downloader: A dictionary of protocol keys and the executable of the
442 external downloader to use for it. The allowed protocols
443 are default|http|ftp|m3u8|dash|rtsp|rtmp|mms.
444 Set the value to 'native' to use the native downloader
53ed7066 445 compat_opts: Compatibility options. See "Differences in default behavior".
3acf6d38 446 The following options do not work when used through the API:
b5ae35ee 447 filename, abort-on-error, multistreams, no-live-chat, format-sort
dac5df5a 448 no-clean-infojson, no-playlist-metafiles, no-keep-subs, no-attach-info-json.
e4f02757 449 Refer __init__.py for their implementation
819e0531 450 progress_template: Dictionary of templates for progress outputs.
451 Allowed keys are 'download', 'postprocess',
452 'download-title' (console title) and 'postprocess-title'.
453 The template is mapped on a dictionary with keys 'progress' and 'info'
23326151 454 retry_sleep_functions: Dictionary of functions that takes the number of attempts
455 as argument and returns the time to sleep in seconds.
456 Allowed keys are 'http', 'fragment', 'file_access'
0f446365
SW
457 download_ranges: A callback function that gets called for every video with
458 the signature (info_dict, ydl) -> Iterable[Section].
459 Only the returned sections will be downloaded.
460 Each Section is a dict with the following keys:
5ec1b6b7 461 * start_time: Start time of the section in seconds
462 * end_time: End time of the section in seconds
463 * title: Section title (Optional)
464 * index: Section number (Optional)
0f446365 465 force_keyframes_at_cuts: Re-encode the video when downloading ranges to get precise cuts
a7dc6a89 466 noprogress: Do not print the progress bar
a831c2ea 467 live_from_start: Whether to download livestreams videos from the start
fe7e0c98 468
8222d8de 469 The following parameters are not used by YoutubeDL itself, they are used by
7a5c1cfe 470 the downloader (see yt_dlp/downloader/common.py):
51d9739f 471 nopart, updatetime, buffersize, ratelimit, throttledratelimit, min_filesize,
205a0654 472 max_filesize, test, noresizebuffer, retries, file_access_retries, fragment_retries,
a7dc6a89 473 continuedl, xattr_set_filesize, hls_use_mpegts, http_chunk_size,
59a7a13e 474 external_downloader_args, concurrent_fragment_downloads.
76b1bd67
JMF
475
476 The following options are used by the post processors:
c0b7d117
S
477 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
478 to the binary or its containing directory.
43820c03 479 postprocessor_args: A dictionary of postprocessor/executable keys (in lower case)
34488702 480 and a list of additional command-line arguments for the
481 postprocessor/executable. The dict can also have "PP+EXE" keys
482 which are used when the given exe is used by the given PP.
483 Use 'default' as the name for arguments to passed to all PP
484 For compatibility with youtube-dl, a single list of args
485 can also be used
e409895f 486
487 The following options are used by the extractors:
46f1370e 488 extractor_retries: Number of times to retry for known errors (default: 3)
62bff2c1 489 dynamic_mpd: Whether to process dynamic DASH manifests (default: True)
e409895f 490 hls_split_discontinuity: Split HLS playlists to different formats at
62bff2c1 491 discontinuities such as ad breaks (default: False)
5d3a0e79 492 extractor_args: A dictionary of arguments to be passed to the extractors.
493 See "EXTRACTOR ARGUMENTS" for details.
62b58c09 494 E.g. {'youtube': {'skip': ['dash', 'hls']}}
88f23a18 495 mark_watched: Mark videos watched (even with --simulate). Only for YouTube
1890fc63 496
497 The following options are deprecated and may be removed in the future:
498
fe2ce85a 499 break_on_reject: Stop the download process when encountering a video that
500 has been filtered out.
501 - `raise DownloadCancelled(msg)` in match_filter instead
fe7866d0 502 force_generic_extractor: Force downloader to use the generic extractor
503 - Use allowed_extractors = ['generic', 'default']
7e9a6125 504 playliststart: - Use playlist_items
505 Playlist item to start at.
506 playlistend: - Use playlist_items
507 Playlist item to end at.
508 playlistreverse: - Use playlist_items
509 Download playlist items in reverse order.
1890fc63 510 forceurl: - Use forceprint
511 Force printing final URL.
512 forcetitle: - Use forceprint
513 Force printing title.
514 forceid: - Use forceprint
515 Force printing ID.
516 forcethumbnail: - Use forceprint
517 Force printing thumbnail URL.
518 forcedescription: - Use forceprint
519 Force printing description.
520 forcefilename: - Use forceprint
521 Force printing final filename.
522 forceduration: - Use forceprint
523 Force printing duration.
524 allsubtitles: - Use subtitleslangs = ['all']
525 Downloads all the subtitles of the video
526 (requires writesubtitles or writeautomaticsub)
527 include_ads: - Doesn't work
528 Download ads as well
529 call_home: - Not implemented
530 Boolean, true iff we are allowed to contact the
531 yt-dlp servers for debugging.
532 post_hooks: - Register a custom postprocessor
533 A list of functions that get called as the final step
534 for each video file, after all postprocessors have been
535 called. The filename will be passed as the only argument.
536 hls_prefer_native: - Use external_downloader = {'m3u8': 'native'} or {'m3u8': 'ffmpeg'}.
537 Use the native HLS downloader instead of ffmpeg/avconv
538 if True, otherwise use ffmpeg/avconv if False, otherwise
539 use downloader suggested by extractor if None.
540 prefer_ffmpeg: - avconv support is deprecated
541 If False, use avconv instead of ffmpeg if both are available,
542 otherwise prefer ffmpeg.
543 youtube_include_dash_manifest: - Use extractor_args
5d3a0e79 544 If True (default), DASH manifests and related
62bff2c1 545 data will be downloaded and processed by extractor.
546 You can reduce network I/O by disabling it if you don't
547 care about DASH. (only for youtube)
1890fc63 548 youtube_include_hls_manifest: - Use extractor_args
5d3a0e79 549 If True (default), HLS manifests and related
62bff2c1 550 data will be downloaded and processed by extractor.
551 You can reduce network I/O by disabling it if you don't
552 care about HLS. (only for youtube)
8417f26b 553 no_color: Same as `color='no_color'`
6148833f 554 no_overwrites: Same as `overwrites=False`
8222d8de
JMF
555 """
556
86e5f3ed 557 _NUMERIC_FIELDS = {
b8ed0f15 558 'width', 'height', 'asr', 'audio_channels', 'fps',
559 'tbr', 'abr', 'vbr', 'filesize', 'filesize_approx',
e6f21b3d 560 'timestamp', 'release_timestamp',
c9969434
S
561 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
562 'average_rating', 'comment_count', 'age_limit',
563 'start_time', 'end_time',
564 'chapter_number', 'season_number', 'episode_number',
565 'track_number', 'disc_number', 'release_year',
86e5f3ed 566 }
c9969434 567
6db9c4d5 568 _format_fields = {
569 # NB: Keep in sync with the docstring of extractor/common.py
a44ca5a4 570 'url', 'manifest_url', 'manifest_stream_number', 'ext', 'format', 'format_id', 'format_note',
105bfd90 571 'width', 'height', 'aspect_ratio', 'resolution', 'dynamic_range', 'tbr', 'abr', 'acodec', 'asr', 'audio_channels',
d5d1df8a 572 'vbr', 'fps', 'vcodec', 'container', 'filesize', 'filesize_approx', 'rows', 'columns',
6db9c4d5 573 'player_url', 'protocol', 'fragment_base_url', 'fragments', 'is_from_start',
62b5c94c 574 'preference', 'language', 'language_preference', 'quality', 'source_preference', 'cookies',
7e68567e 575 'http_headers', 'stretched_ratio', 'no_resume', 'has_drm', 'extra_param_to_segment_url', 'hls_aes', 'downloader_options',
6db9c4d5 576 'page_url', 'app', 'play_path', 'tc_url', 'flash_version', 'rtmp_live', 'rtmp_conn', 'rtmp_protocol', 'rtmp_real_time'
577 }
48ee10ee 578 _format_selection_exts = {
8dc59305 579 'audio': set(MEDIA_EXTENSIONS.common_audio),
580 'video': set(MEDIA_EXTENSIONS.common_video + ('3gp', )),
581 'storyboards': set(MEDIA_EXTENSIONS.storyboards),
48ee10ee 582 }
583
3511266b 584 def __init__(self, params=None, auto_init=True):
883d4b1e 585 """Create a FileDownloader object with the given options.
586 @param auto_init Whether to load the default extractors and print header (if verbose).
49a57e70 587 Set to 'no_verbose_header' to not print the header
883d4b1e 588 """
e9f9a10f
JMF
589 if params is None:
590 params = {}
592b7485 591 self.params = params
8b7491c8 592 self._ies = {}
56c73665 593 self._ies_instances = {}
1e43a6f7 594 self._pps = {k: [] for k in POSTPROCESS_WHEN}
b35496d8 595 self._printed_messages = set()
1cf376f5 596 self._first_webpage_request = True
ab8e5e51 597 self._post_hooks = []
933605d7 598 self._progress_hooks = []
819e0531 599 self._postprocessor_hooks = []
8222d8de
JMF
600 self._download_retcode = 0
601 self._num_downloads = 0
9c906919 602 self._num_videos = 0
592b7485 603 self._playlist_level = 0
604 self._playlist_urls = set()
a0e07d31 605 self.cache = Cache(self)
6148833f 606 self.__header_cookies = []
34308b30 607
591bb9d3 608 stdout = sys.stderr if self.params.get('logtostderr') else sys.stdout
609 self._out_files = Namespace(
610 out=stdout,
611 error=sys.stderr,
612 screen=sys.stderr if self.params.get('quiet') else stdout,
613 console=None if compat_os_name == 'nt' else next(
cf4f42cb 614 filter(supports_terminal_sequences, (sys.stderr, sys.stdout)), None)
591bb9d3 615 )
f0795149 616
617 try:
618 windows_enable_vt_mode()
619 except Exception as e:
620 self.write_debug(f'Failed to enable VT mode: {e}')
621
8417f26b
SS
622 if self.params.get('no_color'):
623 if self.params.get('color') is not None:
62b5c94c 624 self.params.setdefault('_warnings', []).append(
625 'Overwriting params from "color" with "no_color"')
8417f26b
SS
626 self.params['color'] = 'no_color'
627
628 term_allow_color = os.environ.get('TERM', '').lower() != 'dumb'
629
630 def process_color_policy(stream):
631 stream_name = {sys.stdout: 'stdout', sys.stderr: 'stderr'}[stream]
632 policy = traverse_obj(self.params, ('color', (stream_name, None), {str}), get_all=False)
633 if policy in ('auto', None):
634 return term_allow_color and supports_terminal_sequences(stream)
6148833f 635 assert policy in ('always', 'never', 'no_color'), policy
8417f26b
SS
636 return {'always': True, 'never': False}.get(policy, policy)
637
591bb9d3 638 self._allow_colors = Namespace(**{
8417f26b
SS
639 name: process_color_policy(stream)
640 for name, stream in self._out_files.items_ if name != 'console'
591bb9d3 641 })
819e0531 642
61bdf15f
SS
643 system_deprecation = _get_system_deprecation()
644 if system_deprecation:
645 self.deprecated_feature(system_deprecation.replace('\n', '\n '))
a61f4b28 646
88acdbc2 647 if self.params.get('allow_unplayable_formats'):
648 self.report_warning(
ec11a9f4 649 f'You have asked for {self._format_err("UNPLAYABLE", self.Styles.EMPHASIS)} formats to be listed/downloaded. '
819e0531 650 'This is a developer option intended for debugging. \n'
651 ' If you experience any issues while using this option, '
ec11a9f4 652 f'{self._format_err("DO NOT", self.Styles.ERROR)} open a bug report')
88acdbc2 653
497074f0 654 if self.params.get('bidi_workaround', False):
655 try:
656 import pty
657 master, slave = pty.openpty()
658 width = shutil.get_terminal_size().columns
659 width_args = [] if width is None else ['-w', str(width)]
660 sp_kwargs = {'stdin': subprocess.PIPE, 'stdout': slave, 'stderr': self._out_files.error}
661 try:
662 self._output_process = Popen(['bidiv'] + width_args, **sp_kwargs)
663 except OSError:
664 self._output_process = Popen(['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
665 self._output_channel = os.fdopen(master, 'rb')
666 except OSError as ose:
667 if ose.errno == errno.ENOENT:
668 self.report_warning(
669 'Could not find fribidi executable, ignoring --bidi-workaround. '
670 'Make sure that fribidi is an executable file in one of the directories in your $PATH.')
671 else:
672 raise
673
674 self.params['compat_opts'] = set(self.params.get('compat_opts', ()))
227bf1a3 675 self.params['http_headers'] = HTTPHeaderDict(std_headers, self.params.get('http_headers'))
6c5211ce 676 self._load_cookies(self.params['http_headers'].get('Cookie')) # compat
677 self.params['http_headers'].pop('Cookie', None)
db7b054a 678 self._request_director = self.build_request_director(_REQUEST_HANDLERS.values(), _RH_PREFERENCES)
6c5211ce 679
497074f0 680 if auto_init and auto_init != 'no_verbose_header':
681 self.print_debug_header()
682
be5df5ee
S
683 def check_deprecated(param, option, suggestion):
684 if self.params.get(param) is not None:
86e5f3ed 685 self.report_warning(f'{option} is deprecated. Use {suggestion} instead')
be5df5ee
S
686 return True
687 return False
688
689 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
38cce791
YCH
690 if self.params.get('geo_verification_proxy') is None:
691 self.params['geo_verification_proxy'] = self.params['cn_verification_proxy']
692
0d1bb027 693 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
694 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
53ed7066 695 check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"')
0d1bb027 696
49a57e70 697 for msg in self.params.get('_warnings', []):
0d1bb027 698 self.report_warning(msg)
ee8dd27a 699 for msg in self.params.get('_deprecation_warnings', []):
da4db748 700 self.deprecated_feature(msg)
0d1bb027 701
8a82af35 702 if 'list-formats' in self.params['compat_opts']:
ec11a9f4 703 self.params['listformats_table'] = False
704
b5ae35ee 705 if 'overwrites' not in self.params and self.params.get('nooverwrites') is not None:
b868936c 706 # nooverwrites was unnecessarily changed to overwrites
707 # in 0c3d0f51778b153f65c21906031c2e091fcfb641
708 # This ensures compatibility with both keys
709 self.params['overwrites'] = not self.params['nooverwrites']
b5ae35ee 710 elif self.params.get('overwrites') is None:
711 self.params.pop('overwrites', None)
b868936c 712 else:
713 self.params['nooverwrites'] = not self.params['overwrites']
b9d973be 714
e4221b70 715 if self.params.get('simulate') is None and any((
716 self.params.get('list_thumbnails'),
717 self.params.get('listformats'),
718 self.params.get('listsubtitles'),
719 )):
720 self.params['simulate'] = 'list_only'
721
455a15e2 722 self.params.setdefault('forceprint', {})
723 self.params.setdefault('print_to_file', {})
bb66c247 724
725 # Compatibility with older syntax
ca30f449 726 if not isinstance(params['forceprint'], dict):
455a15e2 727 self.params['forceprint'] = {'video': params['forceprint']}
ca30f449 728
97ec5bc5 729 if auto_init:
97ec5bc5 730 self.add_default_info_extractors()
731
3089bc74
S
732 if (sys.platform != 'win32'
733 and sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
455a15e2 734 and not self.params.get('restrictfilenames', False)):
e9137224 735 # Unicode filesystem API will throw errors (#1474, #13027)
34308b30 736 self.report_warning(
6febd1c1 737 'Assuming --restrict-filenames since file system encoding '
1b725173 738 'cannot encode all characters. '
6febd1c1 739 'Set the LC_ALL environment variable to fix this.')
4a98cdbf 740 self.params['restrictfilenames'] = True
34308b30 741
bf1824b3 742 self._parse_outtmpl()
486dd09e 743
187986a8 744 # Creating format selector here allows us to catch syntax errors before the extraction
745 self.format_selector = (
fa9f30b8 746 self.params.get('format') if self.params.get('format') in (None, '-')
093a1710 747 else self.params['format'] if callable(self.params['format'])
187986a8 748 else self.build_format_selector(self.params['format']))
749
013b50b7 750 hooks = {
751 'post_hooks': self.add_post_hook,
752 'progress_hooks': self.add_progress_hook,
753 'postprocessor_hooks': self.add_postprocessor_hook,
754 }
755 for opt, fn in hooks.items():
756 for ph in self.params.get(opt, []):
757 fn(ph)
71b640cc 758
5bfc8bee 759 for pp_def_raw in self.params.get('postprocessors', []):
760 pp_def = dict(pp_def_raw)
761 when = pp_def.pop('when', 'post_process')
762 self.add_post_processor(
f9934b96 763 get_postprocessor(pp_def.pop('key'))(self, **pp_def),
5bfc8bee 764 when=when)
765
ed39cac5 766 def preload_download_archive(fn):
767 """Preload the archive, if any is specified"""
ae103564 768 archive = set()
ed39cac5 769 if fn is None:
ae103564 770 return archive
941e881e 771 elif not is_path_like(fn):
ae103564 772 return fn
773
49a57e70 774 self.write_debug(f'Loading archive file {fn!r}')
ed39cac5 775 try:
776 with locked_file(fn, 'r', encoding='utf-8') as archive_file:
777 for line in archive_file:
ae103564 778 archive.add(line.strip())
86e5f3ed 779 except OSError as ioe:
ed39cac5 780 if ioe.errno != errno.ENOENT:
781 raise
ae103564 782 return archive
ed39cac5 783
ae103564 784 self.archive = preload_download_archive(self.params.get('download_archive'))
ed39cac5 785
7d4111ed
PH
786 def warn_if_short_id(self, argv):
787 # short YouTube ID starting with dash?
788 idxs = [
789 i for i, a in enumerate(argv)
790 if re.match(r'^-[0-9A-Za-z_-]{10}$', a)]
791 if idxs:
792 correct_argv = (
7a5c1cfe 793 ['yt-dlp']
3089bc74
S
794 + [a for i, a in enumerate(argv) if i not in idxs]
795 + ['--'] + [argv[i] for i in idxs]
7d4111ed
PH
796 )
797 self.report_warning(
798 'Long argument string detected. '
49a57e70 799 'Use -- to separate parameters and URLs, like this:\n%s' %
7d4111ed
PH
800 args_to_str(correct_argv))
801
8222d8de
JMF
802 def add_info_extractor(self, ie):
803 """Add an InfoExtractor object to the end of the list."""
8b7491c8 804 ie_key = ie.ie_key()
805 self._ies[ie_key] = ie
e52d7f85 806 if not isinstance(ie, type):
8b7491c8 807 self._ies_instances[ie_key] = ie
e52d7f85 808 ie.set_downloader(self)
8222d8de 809
56c73665
JMF
810 def get_info_extractor(self, ie_key):
811 """
812 Get an instance of an IE with name ie_key, it will try to get one from
813 the _ies list, if there's no instance it will create a new one and add
814 it to the extractor list.
815 """
816 ie = self._ies_instances.get(ie_key)
817 if ie is None:
818 ie = get_info_extractor(ie_key)()
819 self.add_info_extractor(ie)
820 return ie
821
023fa8c4
JMF
822 def add_default_info_extractors(self):
823 """
824 Add the InfoExtractors returned by gen_extractors to the end of the list
825 """
fe7866d0 826 all_ies = {ie.IE_NAME.lower(): ie for ie in gen_extractor_classes()}
827 all_ies['end'] = UnsupportedURLIE()
828 try:
829 ie_names = orderedSet_from_options(
830 self.params.get('allowed_extractors', ['default']), {
831 'all': list(all_ies),
832 'default': [name for name, ie in all_ies.items() if ie._ENABLED],
833 }, use_regex=True)
834 except re.error as e:
835 raise ValueError(f'Wrong regex for allowed_extractors: {e.pattern}')
836 for name in ie_names:
837 self.add_info_extractor(all_ies[name])
838 self.write_debug(f'Loaded {len(ie_names)} extractors')
023fa8c4 839
56d868db 840 def add_post_processor(self, pp, when='post_process'):
8222d8de 841 """Add a PostProcessor object to the end of the chain."""
8aa0e7cd 842 assert when in POSTPROCESS_WHEN, f'Invalid when={when}'
5bfa4862 843 self._pps[when].append(pp)
8222d8de
JMF
844 pp.set_downloader(self)
845
ab8e5e51
AM
846 def add_post_hook(self, ph):
847 """Add the post hook"""
848 self._post_hooks.append(ph)
849
933605d7 850 def add_progress_hook(self, ph):
819e0531 851 """Add the download progress hook"""
933605d7 852 self._progress_hooks.append(ph)
8ab470f1 853
819e0531 854 def add_postprocessor_hook(self, ph):
855 """Add the postprocessing progress hook"""
856 self._postprocessor_hooks.append(ph)
5bfc8bee 857 for pps in self._pps.values():
858 for pp in pps:
859 pp.add_progress_hook(ph)
819e0531 860
1c088fa8 861 def _bidi_workaround(self, message):
5d681e96 862 if not hasattr(self, '_output_channel'):
1c088fa8
PH
863 return message
864
5d681e96 865 assert hasattr(self, '_output_process')
14f25df2 866 assert isinstance(message, str)
6febd1c1 867 line_count = message.count('\n') + 1
0f06bcd7 868 self._output_process.stdin.write((message + '\n').encode())
5d681e96 869 self._output_process.stdin.flush()
0f06bcd7 870 res = ''.join(self._output_channel.readline().decode()
9e1a5b84 871 for _ in range(line_count))
6febd1c1 872 return res[:-len('\n')]
1c088fa8 873
b35496d8 874 def _write_string(self, message, out=None, only_once=False):
875 if only_once:
876 if message in self._printed_messages:
877 return
878 self._printed_messages.add(message)
879 write_string(message, out=out, encoding=self.params.get('encoding'))
734f90bb 880
cf4f42cb 881 def to_stdout(self, message, skip_eol=False, quiet=None):
0760b0a7 882 """Print message to stdout"""
cf4f42cb 883 if quiet is not None:
da4db748 884 self.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument quiet. '
885 'Use "YoutubeDL.to_screen" instead')
8a82af35 886 if skip_eol is not False:
da4db748 887 self.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument skip_eol. '
888 'Use "YoutubeDL.to_screen" instead')
0bf9dc1e 889 self._write_string(f'{self._bidi_workaround(message)}\n', self._out_files.out)
cf4f42cb 890
dfea94f8 891 def to_screen(self, message, skip_eol=False, quiet=None, only_once=False):
cf4f42cb 892 """Print message to screen if not in quiet mode"""
8bf9319e 893 if self.params.get('logger'):
43afe285 894 self.params['logger'].debug(message)
cf4f42cb 895 return
896 if (self.params.get('quiet') if quiet is None else quiet) and not self.params.get('verbose'):
897 return
898 self._write_string(
899 '%s%s' % (self._bidi_workaround(message), ('' if skip_eol else '\n')),
dfea94f8 900 self._out_files.screen, only_once=only_once)
8222d8de 901
b35496d8 902 def to_stderr(self, message, only_once=False):
0760b0a7 903 """Print message to stderr"""
14f25df2 904 assert isinstance(message, str)
8bf9319e 905 if self.params.get('logger'):
43afe285
IB
906 self.params['logger'].error(message)
907 else:
5792c950 908 self._write_string(f'{self._bidi_workaround(message)}\n', self._out_files.error, only_once=only_once)
cf4f42cb 909
910 def _send_console_code(self, code):
591bb9d3 911 if compat_os_name == 'nt' or not self._out_files.console:
cf4f42cb 912 return
591bb9d3 913 self._write_string(code, self._out_files.console)
8222d8de 914
1e5b9a95
PH
915 def to_console_title(self, message):
916 if not self.params.get('consoletitle', False):
917 return
3efb96a6 918 message = remove_terminal_sequences(message)
4bede0d8
C
919 if compat_os_name == 'nt':
920 if ctypes.windll.kernel32.GetConsoleWindow():
921 # c_wchar_p() might not be necessary if `message` is
922 # already of type unicode()
923 ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
cf4f42cb 924 else:
925 self._send_console_code(f'\033]0;{message}\007')
1e5b9a95 926
bdde425c 927 def save_console_title(self):
cf4f42cb 928 if not self.params.get('consoletitle') or self.params.get('simulate'):
bdde425c 929 return
592b7485 930 self._send_console_code('\033[22;0t') # Save the title on stack
bdde425c
PH
931
932 def restore_console_title(self):
cf4f42cb 933 if not self.params.get('consoletitle') or self.params.get('simulate'):
bdde425c 934 return
592b7485 935 self._send_console_code('\033[23;0t') # Restore the title from stack
bdde425c
PH
936
937 def __enter__(self):
938 self.save_console_title()
939 return self
940
c365dba8 941 def save_cookies(self):
dca08720 942 if self.params.get('cookiefile') is not None:
62b5c94c 943 self.cookiejar.save()
bdde425c 944
c365dba8 945 def __exit__(self, *args):
946 self.restore_console_title()
227bf1a3 947 self.close()
948
949 def close(self):
c365dba8 950 self.save_cookies()
227bf1a3 951 self._request_director.close()
c365dba8 952
fa9f30b8 953 def trouble(self, message=None, tb=None, is_error=True):
8222d8de
JMF
954 """Determine action to take when a download problem appears.
955
956 Depending on if the downloader has been configured to ignore
957 download errors or not, this method may throw an exception or
958 not when errors are found, after printing the message.
959
fa9f30b8 960 @param tb If given, is additional traceback information
961 @param is_error Whether to raise error according to ignorerrors
8222d8de
JMF
962 """
963 if message is not None:
964 self.to_stderr(message)
965 if self.params.get('verbose'):
966 if tb is None:
967 if sys.exc_info()[0]: # if .trouble has been called from an except block
6febd1c1 968 tb = ''
8222d8de 969 if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
6febd1c1 970 tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
c0384f22 971 tb += encode_compat_str(traceback.format_exc())
8222d8de
JMF
972 else:
973 tb_data = traceback.format_list(traceback.extract_stack())
6febd1c1 974 tb = ''.join(tb_data)
c19bc311 975 if tb:
976 self.to_stderr(tb)
fa9f30b8 977 if not is_error:
978 return
b1940459 979 if not self.params.get('ignoreerrors'):
8222d8de
JMF
980 if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
981 exc_info = sys.exc_info()[1].exc_info
982 else:
983 exc_info = sys.exc_info()
984 raise DownloadError(message, exc_info)
985 self._download_retcode = 1
986
19a03940 987 Styles = Namespace(
988 HEADERS='yellow',
989 EMPHASIS='light blue',
492272fe 990 FILENAME='green',
19a03940 991 ID='green',
992 DELIM='blue',
993 ERROR='red',
bc344cd4 994 BAD_FORMAT='light red',
19a03940 995 WARNING='yellow',
996 SUPPRESS='light black',
997 )
ec11a9f4 998
7578d77d 999 def _format_text(self, handle, allow_colors, text, f, fallback=None, *, test_encoding=False):
e5a998f3 1000 text = str(text)
ec11a9f4 1001 if test_encoding:
1002 original_text = text
5c104538 1003 # handle.encoding can be None. See https://github.com/yt-dlp/yt-dlp/issues/2711
1004 encoding = self.params.get('encoding') or getattr(handle, 'encoding', None) or 'ascii'
ec11a9f4 1005 text = text.encode(encoding, 'ignore').decode(encoding)
1006 if fallback is not None and text != original_text:
1007 text = fallback
8417f26b 1008 return format_text(text, f) if allow_colors is True else text if fallback is None else fallback
ec11a9f4 1009
591bb9d3 1010 def _format_out(self, *args, **kwargs):
1011 return self._format_text(self._out_files.out, self._allow_colors.out, *args, **kwargs)
1012
ec11a9f4 1013 def _format_screen(self, *args, **kwargs):
591bb9d3 1014 return self._format_text(self._out_files.screen, self._allow_colors.screen, *args, **kwargs)
ec11a9f4 1015
1016 def _format_err(self, *args, **kwargs):
591bb9d3 1017 return self._format_text(self._out_files.error, self._allow_colors.error, *args, **kwargs)
819e0531 1018
c84aeac6 1019 def report_warning(self, message, only_once=False):
8222d8de
JMF
1020 '''
1021 Print the message to stderr, it will be prefixed with 'WARNING:'
1022 If stderr is a tty file the 'WARNING:' will be colored
1023 '''
6d07ce01
JMF
1024 if self.params.get('logger') is not None:
1025 self.params['logger'].warning(message)
8222d8de 1026 else:
ad8915b7
PH
1027 if self.params.get('no_warnings'):
1028 return
ec11a9f4 1029 self.to_stderr(f'{self._format_err("WARNING:", self.Styles.WARNING)} {message}', only_once)
8222d8de 1030
da4db748 1031 def deprecation_warning(self, message, *, stacklevel=0):
1032 deprecation_warning(
1033 message, stacklevel=stacklevel + 1, printer=self.report_error, is_error=False)
1034
1035 def deprecated_feature(self, message):
ee8dd27a 1036 if self.params.get('logger') is not None:
da4db748 1037 self.params['logger'].warning(f'Deprecated Feature: {message}')
1038 self.to_stderr(f'{self._format_err("Deprecated Feature:", self.Styles.ERROR)} {message}', True)
ee8dd27a 1039
fa9f30b8 1040 def report_error(self, message, *args, **kwargs):
8222d8de
JMF
1041 '''
1042 Do the same as trouble, but prefixes the message with 'ERROR:', colored
1043 in red if stderr is a tty file.
1044 '''
fa9f30b8 1045 self.trouble(f'{self._format_err("ERROR:", self.Styles.ERROR)} {message}', *args, **kwargs)
8222d8de 1046
b35496d8 1047 def write_debug(self, message, only_once=False):
0760b0a7 1048 '''Log debug message or Print message to stderr'''
1049 if not self.params.get('verbose', False):
1050 return
8a82af35 1051 message = f'[debug] {message}'
0760b0a7 1052 if self.params.get('logger'):
1053 self.params['logger'].debug(message)
1054 else:
b35496d8 1055 self.to_stderr(message, only_once)
0760b0a7 1056
8222d8de
JMF
1057 def report_file_already_downloaded(self, file_name):
1058 """Report file has already been fully downloaded."""
1059 try:
6febd1c1 1060 self.to_screen('[download] %s has already been downloaded' % file_name)
ce02ed60 1061 except UnicodeEncodeError:
6febd1c1 1062 self.to_screen('[download] The file has already been downloaded')
8222d8de 1063
0c3d0f51 1064 def report_file_delete(self, file_name):
1065 """Report that existing file will be deleted."""
1066 try:
c25228e5 1067 self.to_screen('Deleting existing file %s' % file_name)
0c3d0f51 1068 except UnicodeEncodeError:
c25228e5 1069 self.to_screen('Deleting existing file')
0c3d0f51 1070
319b6059 1071 def raise_no_formats(self, info, forced=False, *, msg=None):
0a5a191a 1072 has_drm = info.get('_has_drm')
319b6059 1073 ignored, expected = self.params.get('ignore_no_formats_error'), bool(msg)
1074 msg = msg or has_drm and 'This video is DRM protected' or 'No video formats found!'
1075 if forced or not ignored:
1151c407 1076 raise ExtractorError(msg, video_id=info['id'], ie=info['extractor'],
319b6059 1077 expected=has_drm or ignored or expected)
88acdbc2 1078 else:
1079 self.report_warning(msg)
1080
de6000d9 1081 def parse_outtmpl(self):
bf1824b3 1082 self.deprecation_warning('"YoutubeDL.parse_outtmpl" is deprecated and may be removed in a future version')
1083 self._parse_outtmpl()
1084 return self.params['outtmpl']
1085
1086 def _parse_outtmpl(self):
7b2c3f47 1087 sanitize = IDENTITY
bf1824b3 1088 if self.params.get('restrictfilenames'): # Remove spaces in the default template
71ce444a 1089 sanitize = lambda x: x.replace(' - ', ' ').replace(' ', '-')
bf1824b3 1090
1091 outtmpl = self.params.setdefault('outtmpl', {})
1092 if not isinstance(outtmpl, dict):
1093 self.params['outtmpl'] = outtmpl = {'default': outtmpl}
1094 outtmpl.update({k: sanitize(v) for k, v in DEFAULT_OUTTMPL.items() if outtmpl.get(k) is None})
de6000d9 1095
21cd8fae 1096 def get_output_path(self, dir_type='', filename=None):
1097 paths = self.params.get('paths', {})
d2c8aadf 1098 assert isinstance(paths, dict), '"paths" parameter must be a dictionary'
21cd8fae 1099 path = os.path.join(
1100 expand_path(paths.get('home', '').strip()),
1101 expand_path(paths.get(dir_type, '').strip()) if dir_type else '',
1102 filename or '')
21cd8fae 1103 return sanitize_path(path, force=self.params.get('windowsfilenames'))
1104
76a264ac 1105 @staticmethod
901130bb 1106 def _outtmpl_expandpath(outtmpl):
1107 # expand_path translates '%%' into '%' and '$$' into '$'
1108 # correspondingly that is not what we want since we need to keep
1109 # '%%' intact for template dict substitution step. Working around
1110 # with boundary-alike separator hack.
6f2287cb 1111 sep = ''.join(random.choices(string.ascii_letters, k=32))
86e5f3ed 1112 outtmpl = outtmpl.replace('%%', f'%{sep}%').replace('$$', f'${sep}$')
901130bb 1113
1114 # outtmpl should be expand_path'ed before template dict substitution
1115 # because meta fields may contain env variables we don't want to
62b58c09 1116 # be expanded. E.g. for outtmpl "%(title)s.%(ext)s" and
901130bb 1117 # title "Hello $PATH", we don't want `$PATH` to be expanded.
1118 return expand_path(outtmpl).replace(sep, '')
1119
1120 @staticmethod
1121 def escape_outtmpl(outtmpl):
1122 ''' Escape any remaining strings like %s, %abc% etc. '''
1123 return re.sub(
1124 STR_FORMAT_RE_TMPL.format('', '(?![%(\0])'),
1125 lambda mobj: ('' if mobj.group('has_key') else '%') + mobj.group(0),
1126 outtmpl)
1127
1128 @classmethod
1129 def validate_outtmpl(cls, outtmpl):
76a264ac 1130 ''' @return None or Exception object '''
7d1eb38a 1131 outtmpl = re.sub(
47cdc68e 1132 STR_FORMAT_RE_TMPL.format('[^)]*', '[ljhqBUDS]'),
7d1eb38a 1133 lambda mobj: f'{mobj.group(0)[:-1]}s',
1134 cls._outtmpl_expandpath(outtmpl))
76a264ac 1135 try:
7d1eb38a 1136 cls.escape_outtmpl(outtmpl) % collections.defaultdict(int)
76a264ac 1137 return None
1138 except ValueError as err:
1139 return err
1140
03b4de72 1141 @staticmethod
1142 def _copy_infodict(info_dict):
1143 info_dict = dict(info_dict)
09b49e1f 1144 info_dict.pop('__postprocessors', None)
415f8d51 1145 info_dict.pop('__pending_error', None)
03b4de72 1146 return info_dict
1147
e0fd9573 1148 def prepare_outtmpl(self, outtmpl, info_dict, sanitize=False):
1149 """ Make the outtmpl and info_dict suitable for substitution: ydl.escape_outtmpl(outtmpl) % info_dict
1150 @param sanitize Whether to sanitize the output as a filename.
1151 For backward compatibility, a function can also be passed
1152 """
1153
6e84b215 1154 info_dict.setdefault('epoch', int(time.time())) # keep epoch consistent once set
143db31d 1155
03b4de72 1156 info_dict = self._copy_infodict(info_dict)
752cda38 1157 info_dict['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
53c18592 1158 formatSeconds(info_dict['duration'], '-' if sanitize else ':')
143db31d 1159 if info_dict.get('duration', None) is not None
1160 else None)
1d485a1a 1161 info_dict['autonumber'] = int(self.params.get('autonumber_start', 1) - 1 + self._num_downloads)
9c906919 1162 info_dict['video_autonumber'] = self._num_videos
752cda38 1163 if info_dict.get('resolution') is None:
1164 info_dict['resolution'] = self.format_resolution(info_dict, default=None)
143db31d 1165
e6f21b3d 1166 # For fields playlist_index, playlist_autonumber and autonumber convert all occurrences
143db31d 1167 # of %(field)s to %(field)0Nd for backward compatibility
1168 field_size_compat_map = {
0a5a191a 1169 'playlist_index': number_of_digits(info_dict.get('__last_playlist_index') or 0),
ec11a9f4 1170 'playlist_autonumber': number_of_digits(info_dict.get('n_entries') or 0),
752cda38 1171 'autonumber': self.params.get('autonumber_size') or 5,
143db31d 1172 }
752cda38 1173
385a27fa 1174 TMPL_DICT = {}
47cdc68e 1175 EXTERNAL_FORMAT_RE = re.compile(STR_FORMAT_RE_TMPL.format('[^)]*', f'[{STR_FORMAT_TYPES}ljhqBUDS]'))
385a27fa 1176 MATH_FUNCTIONS = {
1177 '+': float.__add__,
1178 '-': float.__sub__,
1179 }
e625be0d 1180 # Field is of the form key1.key2...
07a1250e 1181 # where keys (except first) can be string, int, slice or "{field, ...}"
1182 FIELD_INNER_RE = r'(?:\w+|%(num)s|%(num)s?(?::%(num)s?){1,2})' % {'num': r'(?:-?\d+)'}
1183 FIELD_RE = r'\w*(?:\.(?:%(inner)s|{%(field)s(?:,%(field)s)*}))*' % {
1184 'inner': FIELD_INNER_RE,
1185 'field': rf'\w*(?:\.{FIELD_INNER_RE})*'
1186 }
1d485a1a 1187 MATH_FIELD_RE = rf'(?:{FIELD_RE}|-?{NUMBER_RE})'
385a27fa 1188 MATH_OPERATORS_RE = r'(?:%s)' % '|'.join(map(re.escape, MATH_FUNCTIONS.keys()))
78fde6e3 1189 INTERNAL_FORMAT_RE = re.compile(rf'''(?xs)
e625be0d 1190 (?P<negate>-)?
1d485a1a 1191 (?P<fields>{FIELD_RE})
1192 (?P<maths>(?:{MATH_OPERATORS_RE}{MATH_FIELD_RE})*)
e625be0d 1193 (?:>(?P<strf_format>.+?))?
34baa9fd 1194 (?P<remaining>
1195 (?P<alternate>(?<!\\),[^|&)]+)?
1196 (?:&(?P<replacement>.*?))?
1197 (?:\|(?P<default>.*?))?
1d485a1a 1198 )$''')
752cda38 1199
07a1250e 1200 def _traverse_infodict(fields):
1201 fields = [f for x in re.split(r'\.({.+?})\.?', fields)
1202 for f in ([x] if x.startswith('{') else x.split('.'))]
1203 for i in (0, -1):
1204 if fields and not fields[i]:
1205 fields.pop(i)
1206
1207 for i, f in enumerate(fields):
1208 if not f.startswith('{'):
1209 continue
1210 assert f.endswith('}'), f'No closing brace for {f} in {fields}'
1211 fields[i] = {k: k.split('.') for k in f[1:-1].split(',')}
1212
1213 return traverse_obj(info_dict, fields, is_user_input=True, traverse_string=True)
76a264ac 1214
752cda38 1215 def get_value(mdict):
1216 # Object traversal
2b8a2973 1217 value = _traverse_infodict(mdict['fields'])
752cda38 1218 # Negative
1219 if mdict['negate']:
1220 value = float_or_none(value)
1221 if value is not None:
1222 value *= -1
1223 # Do maths
385a27fa 1224 offset_key = mdict['maths']
1225 if offset_key:
752cda38 1226 value = float_or_none(value)
1227 operator = None
385a27fa 1228 while offset_key:
1229 item = re.match(
1230 MATH_FIELD_RE if operator else MATH_OPERATORS_RE,
1231 offset_key).group(0)
1232 offset_key = offset_key[len(item):]
1233 if operator is None:
752cda38 1234 operator = MATH_FUNCTIONS[item]
385a27fa 1235 continue
1236 item, multiplier = (item[1:], -1) if item[0] == '-' else (item, 1)
1237 offset = float_or_none(item)
1238 if offset is None:
2b8a2973 1239 offset = float_or_none(_traverse_infodict(item))
385a27fa 1240 try:
1241 value = operator(value, multiplier * offset)
1242 except (TypeError, ZeroDivisionError):
1243 return None
1244 operator = None
752cda38 1245 # Datetime formatting
1246 if mdict['strf_format']:
7c37ff97 1247 value = strftime_or_none(value, mdict['strf_format'].replace('\\,', ','))
752cda38 1248
a6bcaf71 1249 # XXX: Workaround for https://github.com/yt-dlp/yt-dlp/issues/4485
1250 if sanitize and value == '':
1251 value = None
752cda38 1252 return value
1253
b868936c 1254 na = self.params.get('outtmpl_na_placeholder', 'NA')
1255
e0fd9573 1256 def filename_sanitizer(key, value, restricted=self.params.get('restrictfilenames')):
5c3895ff 1257 return sanitize_filename(str(value), restricted=restricted, is_id=(
1258 bool(re.search(r'(^|[_.])id(\.|$)', key))
8a82af35 1259 if 'filename-sanitization' in self.params['compat_opts']
5c3895ff 1260 else NO_DEFAULT))
e0fd9573 1261
1262 sanitizer = sanitize if callable(sanitize) else filename_sanitizer
1263 sanitize = bool(sanitize)
1264
6e84b215 1265 def _dumpjson_default(obj):
1266 if isinstance(obj, (set, LazyList)):
1267 return list(obj)
adbc4ec4 1268 return repr(obj)
6e84b215 1269
6f2287cb 1270 class _ReplacementFormatter(string.Formatter):
ec9311c4 1271 def get_field(self, field_name, args, kwargs):
1272 if field_name.isdigit():
1273 return args[0], -1
1274 raise ValueError('Unsupported field')
1275
1276 replacement_formatter = _ReplacementFormatter()
1277
752cda38 1278 def create_key(outer_mobj):
1279 if not outer_mobj.group('has_key'):
b836dc94 1280 return outer_mobj.group(0)
752cda38 1281 key = outer_mobj.group('key')
752cda38 1282 mobj = re.match(INTERNAL_FORMAT_RE, key)
47bcd437 1283 value, replacement, default, last_field = None, None, na, ''
7c37ff97 1284 while mobj:
e625be0d 1285 mobj = mobj.groupdict()
7c37ff97 1286 default = mobj['default'] if mobj['default'] is not None else default
752cda38 1287 value = get_value(mobj)
47bcd437 1288 last_field, replacement = mobj['fields'], mobj['replacement']
7c37ff97 1289 if value is None and mobj['alternate']:
34baa9fd 1290 mobj = re.match(INTERNAL_FORMAT_RE, mobj['remaining'][1:])
7c37ff97 1291 else:
1292 break
752cda38 1293
ebe1b4e3 1294 if None not in (value, replacement):
ec9311c4 1295 try:
1296 value = replacement_formatter.format(replacement, value)
1297 except ValueError:
ebe1b4e3 1298 value, default = None, na
752cda38 1299
a264433c 1300 fmt = outer_mobj.group('format')
1301 if fmt == 's' and last_field in field_size_compat_map.keys() and isinstance(value, int):
1302 fmt = f'0{field_size_compat_map[last_field]:d}d'
1303
4476d2c7 1304 flags = outer_mobj.group('conversion') or ''
7d1eb38a 1305 str_fmt = f'{fmt[:-1]}s'
ebe1b4e3 1306 if value is None:
1307 value, fmt = default, 's'
1308 elif fmt[-1] == 'l': # list
4476d2c7 1309 delim = '\n' if '#' in flags else ', '
9e907ebd 1310 value, fmt = delim.join(map(str, variadic(value, allowed_types=(str, bytes)))), str_fmt
524e2e4f 1311 elif fmt[-1] == 'j': # json
deae7c17 1312 value, fmt = json.dumps(
1313 value, default=_dumpjson_default,
9b9dad11 1314 indent=4 if '#' in flags else None, ensure_ascii='+' not in flags), str_fmt
47cdc68e 1315 elif fmt[-1] == 'h': # html
deae7c17 1316 value, fmt = escapeHTML(str(value)), str_fmt
524e2e4f 1317 elif fmt[-1] == 'q': # quoted
4476d2c7 1318 value = map(str, variadic(value) if '#' in flags else [value])
1319 value, fmt = ' '.join(map(compat_shlex_quote, value)), str_fmt
524e2e4f 1320 elif fmt[-1] == 'B': # bytes
0f06bcd7 1321 value = f'%{str_fmt}'.encode() % str(value).encode()
f5aa5cfb 1322 value, fmt = value.decode('utf-8', 'ignore'), 's'
524e2e4f 1323 elif fmt[-1] == 'U': # unicode normalized
524e2e4f 1324 value, fmt = unicodedata.normalize(
1325 # "+" = compatibility equivalence, "#" = NFD
4476d2c7 1326 'NF%s%s' % ('K' if '+' in flags else '', 'D' if '#' in flags else 'C'),
524e2e4f 1327 value), str_fmt
e0fd9573 1328 elif fmt[-1] == 'D': # decimal suffix
abbeeebc 1329 num_fmt, fmt = fmt[:-1].replace('#', ''), 's'
1330 value = format_decimal_suffix(value, f'%{num_fmt}f%s' if num_fmt else '%d%s',
1331 factor=1024 if '#' in flags else 1000)
37893bb0 1332 elif fmt[-1] == 'S': # filename sanitization
47bcd437 1333 value, fmt = filename_sanitizer(last_field, value, restricted='#' in flags), str_fmt
7d1eb38a 1334 elif fmt[-1] == 'c':
524e2e4f 1335 if value:
1336 value = str(value)[0]
76a264ac 1337 else:
524e2e4f 1338 fmt = str_fmt
1619ab3e 1339 elif fmt[-1] not in 'rsa': # numeric
a439a3a4 1340 value = float_or_none(value)
752cda38 1341 if value is None:
1342 value, fmt = default, 's'
901130bb 1343
752cda38 1344 if sanitize:
1619ab3e 1345 # If value is an object, sanitize might convert it to a string
1346 # So we convert it to repr first
752cda38 1347 if fmt[-1] == 'r':
7d1eb38a 1348 value, fmt = repr(value), str_fmt
1619ab3e 1349 elif fmt[-1] == 'a':
1350 value, fmt = ascii(value), str_fmt
1351 if fmt[-1] in 'csra':
47bcd437 1352 value = sanitizer(last_field, value)
901130bb 1353
b868936c 1354 key = '%s\0%s' % (key.replace('%', '%\0'), outer_mobj.group('format'))
385a27fa 1355 TMPL_DICT[key] = value
b868936c 1356 return '{prefix}%({key}){fmt}'.format(key=key, fmt=fmt, prefix=outer_mobj.group('prefix'))
752cda38 1357
385a27fa 1358 return EXTERNAL_FORMAT_RE.sub(create_key, outtmpl), TMPL_DICT
143db31d 1359
819e0531 1360 def evaluate_outtmpl(self, outtmpl, info_dict, *args, **kwargs):
1361 outtmpl, info_dict = self.prepare_outtmpl(outtmpl, info_dict, *args, **kwargs)
1362 return self.escape_outtmpl(outtmpl) % info_dict
1363
5127e92a 1364 def _prepare_filename(self, info_dict, *, outtmpl=None, tmpl_type=None):
1365 assert None in (outtmpl, tmpl_type), 'outtmpl and tmpl_type are mutually exclusive'
1366 if outtmpl is None:
bf1824b3 1367 outtmpl = self.params['outtmpl'].get(tmpl_type or 'default', self.params['outtmpl']['default'])
8222d8de 1368 try:
5127e92a 1369 outtmpl = self._outtmpl_expandpath(outtmpl)
e0fd9573 1370 filename = self.evaluate_outtmpl(outtmpl, info_dict, True)
6a0546e3 1371 if not filename:
1372 return None
15da37c7 1373
5127e92a 1374 if tmpl_type in ('', 'temp'):
6a0546e3 1375 final_ext, ext = self.params.get('final_ext'), info_dict.get('ext')
1376 if final_ext and ext and final_ext != ext and filename.endswith(f'.{final_ext}'):
1377 filename = replace_extension(filename, ext, final_ext)
5127e92a 1378 elif tmpl_type:
6a0546e3 1379 force_ext = OUTTMPL_TYPES[tmpl_type]
1380 if force_ext:
1381 filename = replace_extension(filename, force_ext, info_dict.get('ext'))
de6000d9 1382
bdc3fd2f
U
1383 # https://github.com/blackjack4494/youtube-dlc/issues/85
1384 trim_file_name = self.params.get('trim_file_name', False)
1385 if trim_file_name:
5c22c63d 1386 no_ext, *ext = filename.rsplit('.', 2)
1387 filename = join_nonempty(no_ext[:trim_file_name], *ext, delim='.')
bdc3fd2f 1388
0202b52a 1389 return filename
8222d8de 1390 except ValueError as err:
6febd1c1 1391 self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')')
8222d8de
JMF
1392 return None
1393
5127e92a 1394 def prepare_filename(self, info_dict, dir_type='', *, outtmpl=None, warn=False):
1395 """Generate the output filename"""
1396 if outtmpl:
1397 assert not dir_type, 'outtmpl and dir_type are mutually exclusive'
1398 dir_type = None
1399 filename = self._prepare_filename(info_dict, tmpl_type=dir_type, outtmpl=outtmpl)
80c03fa9 1400 if not filename and dir_type not in ('', 'temp'):
1401 return ''
de6000d9 1402
c84aeac6 1403 if warn:
21cd8fae 1404 if not self.params.get('paths'):
de6000d9 1405 pass
1406 elif filename == '-':
c84aeac6 1407 self.report_warning('--paths is ignored when an outputting to stdout', only_once=True)
de6000d9 1408 elif os.path.isabs(filename):
c84aeac6 1409 self.report_warning('--paths is ignored since an absolute path is given in output template', only_once=True)
de6000d9 1410 if filename == '-' or not filename:
1411 return filename
1412
21cd8fae 1413 return self.get_output_path(dir_type, filename)
0202b52a 1414
120fe513 1415 def _match_entry(self, info_dict, incomplete=False, silent=False):
6368e2e6 1416 """Returns None if the file should be downloaded"""
93b39cdb 1417 _type = 'video' if 'playlist-match-filter' in self.params['compat_opts'] else info_dict.get('_type', 'video')
d7b460d0 1418 assert incomplete or _type == 'video', 'Only video result can be considered complete'
8222d8de 1419
3bec830a 1420 video_title = info_dict.get('title', info_dict.get('id', 'entry'))
c77495e3 1421
8b0d7497 1422 def check_filter():
d7b460d0 1423 if _type in ('playlist', 'multi_video'):
1424 return
1425 elif _type in ('url', 'url_transparent') and not try_call(
1426 lambda: self.get_info_extractor(info_dict['ie_key']).is_single_video(info_dict['url'])):
1427 return
1428
8b0d7497 1429 if 'title' in info_dict:
1430 # This can happen when we're just evaluating the playlist
1431 title = info_dict['title']
1432 matchtitle = self.params.get('matchtitle', False)
1433 if matchtitle:
1434 if not re.search(matchtitle, title, re.IGNORECASE):
1435 return '"' + title + '" title did not match pattern "' + matchtitle + '"'
1436 rejecttitle = self.params.get('rejecttitle', False)
1437 if rejecttitle:
1438 if re.search(rejecttitle, title, re.IGNORECASE):
1439 return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
6368e2e6 1440
8b0d7497 1441 date = info_dict.get('upload_date')
1442 if date is not None:
1443 dateRange = self.params.get('daterange', DateRange())
1444 if date not in dateRange:
86e5f3ed 1445 return f'{date_from_str(date).isoformat()} upload date is not in range {dateRange}'
8b0d7497 1446 view_count = info_dict.get('view_count')
1447 if view_count is not None:
1448 min_views = self.params.get('min_views')
1449 if min_views is not None and view_count < min_views:
1450 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
1451 max_views = self.params.get('max_views')
1452 if max_views is not None and view_count > max_views:
1453 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
1454 if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')):
1455 return 'Skipping "%s" because it is age restricted' % video_title
8b0d7497 1456
8f18aca8 1457 match_filter = self.params.get('match_filter')
fe2ce85a 1458 if match_filter is None:
1459 return None
1460
1461 cancelled = None
1462 try:
8f18aca8 1463 try:
1464 ret = match_filter(info_dict, incomplete=incomplete)
1465 except TypeError:
1466 # For backward compatibility
1467 ret = None if incomplete else match_filter(info_dict)
fe2ce85a 1468 except DownloadCancelled as err:
1469 if err.msg is not NO_DEFAULT:
1470 raise
1471 ret, cancelled = err.msg, err
1472
1473 if ret is NO_DEFAULT:
1474 while True:
1475 filename = self._format_screen(self.prepare_filename(info_dict), self.Styles.FILENAME)
1476 reply = input(self._format_screen(
1477 f'Download "{filename}"? (Y/n): ', self.Styles.EMPHASIS)).lower().strip()
1478 if reply in {'y', ''}:
1479 return None
1480 elif reply == 'n':
1481 if cancelled:
1482 raise type(cancelled)(f'Skipping {video_title}')
1483 return f'Skipping {video_title}'
1484 return ret
8b0d7497 1485
c77495e3 1486 if self.in_download_archive(info_dict):
2b029ca0
AK
1487 reason = ''.join((
1488 format_field(info_dict, 'id', f'{self._format_screen("%s", self.Styles.ID)}: '),
1489 format_field(info_dict, 'title', f'{self._format_screen("%s", self.Styles.EMPHASIS)} '),
1490 'has already been recorded in the archive'))
c77495e3 1491 break_opt, break_err = 'break_on_existing', ExistingVideoReached
1492 else:
fe2ce85a 1493 try:
1494 reason = check_filter()
1495 except DownloadCancelled as e:
1496 reason, break_opt, break_err = e.msg, 'match_filter', type(e)
1497 else:
1498 break_opt, break_err = 'break_on_reject', RejectedVideoReached
8b0d7497 1499 if reason is not None:
120fe513 1500 if not silent:
1501 self.to_screen('[download] ' + reason)
c77495e3 1502 if self.params.get(break_opt, False):
1503 raise break_err()
8b0d7497 1504 return reason
fe7e0c98 1505
b6c45014
JMF
1506 @staticmethod
1507 def add_extra_info(info_dict, extra_info):
1508 '''Set the keys from extra_info in info dict if they are missing'''
1509 for key, value in extra_info.items():
1510 info_dict.setdefault(key, value)
1511
409e1828 1512 def extract_info(self, url, download=True, ie_key=None, extra_info=None,
61aa5ba3 1513 process=True, force_generic_extractor=False):
41d1cca3 1514 """
17ffed18 1515 Extract and return the information dictionary of the URL
41d1cca3 1516
1517 Arguments:
17ffed18 1518 @param url URL to extract
41d1cca3 1519
1520 Keyword arguments:
17ffed18 1521 @param download Whether to download videos
1522 @param process Whether to resolve all unresolved references (URLs, playlist items).
1523 Must be True for download to work
1524 @param ie_key Use only the extractor with this key
1525
1526 @param extra_info Dictionary containing the extra values to add to the info (For internal use only)
1527 @force_generic_extractor Force using the generic extractor (Deprecated; use ie_key='Generic')
41d1cca3 1528 """
fe7e0c98 1529
409e1828 1530 if extra_info is None:
1531 extra_info = {}
1532
61aa5ba3 1533 if not ie_key and force_generic_extractor:
d22dec74
S
1534 ie_key = 'Generic'
1535
8222d8de 1536 if ie_key:
fe7866d0 1537 ies = {ie_key: self._ies[ie_key]} if ie_key in self._ies else {}
8222d8de
JMF
1538 else:
1539 ies = self._ies
1540
fe7866d0 1541 for key, ie in ies.items():
8222d8de
JMF
1542 if not ie.suitable(url):
1543 continue
1544
1545 if not ie.working():
6febd1c1
PH
1546 self.report_warning('The program functionality for this site has been marked as broken, '
1547 'and will probably not work.')
8222d8de 1548
1151c407 1549 temp_id = ie.get_temp_id(url)
fe7866d0 1550 if temp_id is not None and self.in_download_archive({'id': temp_id, 'ie_key': key}):
2b029ca0
AK
1551 self.to_screen(f'[download] {self._format_screen(temp_id, self.Styles.ID)}: '
1552 'has already been recorded in the archive')
5e5be0c0 1553 if self.params.get('break_on_existing', False):
1554 raise ExistingVideoReached()
a0566bbf 1555 break
fe7866d0 1556 return self.__extract_info(url, self.get_info_extractor(key), download, extra_info, process)
a0566bbf 1557 else:
fe7866d0 1558 extractors_restricted = self.params.get('allowed_extractors') not in (None, ['default'])
1559 self.report_error(f'No suitable extractor{format_field(ie_key, None, " (%s)")} found for URL {url}',
1560 tb=False if extractors_restricted else None)
a0566bbf 1561
7e88d7d7 1562 def _handle_extraction_exceptions(func):
b5ae35ee 1563 @functools.wraps(func)
a0566bbf 1564 def wrapper(self, *args, **kwargs):
6da22e7d 1565 while True:
1566 try:
1567 return func(self, *args, **kwargs)
1568 except (DownloadCancelled, LazyList.IndexError, PagedList.IndexError):
8222d8de 1569 raise
6da22e7d 1570 except ReExtractInfo as e:
1571 if e.expected:
1572 self.to_screen(f'{e}; Re-extracting data')
1573 else:
1574 self.to_stderr('\r')
1575 self.report_warning(f'{e}; Re-extracting data')
1576 continue
1577 except GeoRestrictedError as e:
1578 msg = e.msg
1579 if e.countries:
1580 msg += '\nThis video is available in %s.' % ', '.join(
1581 map(ISO3166Utils.short2full, e.countries))
1582 msg += '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1583 self.report_error(msg)
1584 except ExtractorError as e: # An error we somewhat expected
1585 self.report_error(str(e), e.format_traceback())
1586 except Exception as e:
1587 if self.params.get('ignoreerrors'):
1588 self.report_error(str(e), tb=encode_compat_str(traceback.format_exc()))
1589 else:
1590 raise
1591 break
a0566bbf 1592 return wrapper
1593
693f0600 1594 def _wait_for_video(self, ie_result={}):
f2ebc5c7 1595 if (not self.params.get('wait_for_video')
1596 or ie_result.get('_type', 'video') != 'video'
1597 or ie_result.get('formats') or ie_result.get('url')):
1598 return
1599
1600 format_dur = lambda dur: '%02d:%02d:%02d' % timetuple_from_msec(dur * 1000)[:-1]
1601 last_msg = ''
1602
1603 def progress(msg):
1604 nonlocal last_msg
a7dc6a89 1605 full_msg = f'{msg}\n'
1606 if not self.params.get('noprogress'):
1607 full_msg = msg + ' ' * (len(last_msg) - len(msg)) + '\r'
1608 elif last_msg:
1609 return
1610 self.to_screen(full_msg, skip_eol=True)
f2ebc5c7 1611 last_msg = msg
1612
1613 min_wait, max_wait = self.params.get('wait_for_video')
1614 diff = try_get(ie_result, lambda x: x['release_timestamp'] - time.time())
1615 if diff is None and ie_result.get('live_status') == 'is_upcoming':
16c620bc 1616 diff = round(random.uniform(min_wait, max_wait) if (max_wait and min_wait) else (max_wait or min_wait), 0)
f2ebc5c7 1617 self.report_warning('Release time of video is not known')
693f0600 1618 elif ie_result and (diff or 0) <= 0:
f2ebc5c7 1619 self.report_warning('Video should already be available according to extracted info')
38d79fd1 1620 diff = min(max(diff or 0, min_wait or 0), max_wait or float('inf'))
f2ebc5c7 1621 self.to_screen(f'[wait] Waiting for {format_dur(diff)} - Press Ctrl+C to try now')
1622
1623 wait_till = time.time() + diff
1624 try:
1625 while True:
1626 diff = wait_till - time.time()
1627 if diff <= 0:
1628 progress('')
1629 raise ReExtractInfo('[wait] Wait period ended', expected=True)
1630 progress(f'[wait] Remaining time until next attempt: {self._format_screen(format_dur(diff), self.Styles.EMPHASIS)}')
1631 time.sleep(1)
1632 except KeyboardInterrupt:
1633 progress('')
1634 raise ReExtractInfo('[wait] Interrupted by user', expected=True)
1635 except BaseException as e:
1636 if not isinstance(e, ReExtractInfo):
1637 self.to_screen('')
1638 raise
1639
6c5211ce 1640 def _load_cookies(self, data, *, autoscope=True):
31215122
SS
1641 """Loads cookies from a `Cookie` header
1642
1643 This tries to work around the security vulnerability of passing cookies to every domain.
1644 See: https://github.com/yt-dlp/yt-dlp/security/advisories/GHSA-v8mc-9377-rwjj
31215122
SS
1645
1646 @param data The Cookie header as string to load the cookies from
6c5211ce 1647 @param autoscope If `False`, scope cookies using Set-Cookie syntax and error for cookie without domains
1648 If `True`, save cookies for later to be stored in the jar with a limited scope
1649 If a URL, save cookies in the jar with the domain of the URL
31215122
SS
1650 """
1651 for cookie in LenientSimpleCookie(data).values():
6c5211ce 1652 if autoscope and any(cookie.values()):
31215122
SS
1653 raise ValueError('Invalid syntax in Cookie Header')
1654
1655 domain = cookie.get('domain') or ''
1656 expiry = cookie.get('expires')
1657 if expiry == '': # 0 is valid
1658 expiry = None
1659 prepared_cookie = http.cookiejar.Cookie(
1660 cookie.get('version') or 0, cookie.key, cookie.value, None, False,
1661 domain, True, True, cookie.get('path') or '', bool(cookie.get('path')),
1662 cookie.get('secure') or False, expiry, False, None, None, {})
1663
1664 if domain:
1665 self.cookiejar.set_cookie(prepared_cookie)
6c5211ce 1666 elif autoscope is True:
31215122
SS
1667 self.deprecated_feature(
1668 'Passing cookies as a header is a potential security risk; '
1669 'they will be scoped to the domain of the downloaded urls. '
1670 'Please consider loading cookies from a file or browser instead.')
1671 self.__header_cookies.append(prepared_cookie)
6c5211ce 1672 elif autoscope:
1673 self.report_warning(
1674 'The extractor result contains an unscoped cookie as an HTTP header. '
1675 f'If you are using yt-dlp with an input URL{bug_reports_message(before=",")}',
1676 only_once=True)
1677 self._apply_header_cookies(autoscope, [prepared_cookie])
31215122
SS
1678 else:
1679 self.report_error('Unscoped cookies are not allowed; please specify some sort of scoping',
1680 tb=False, is_error=False)
1681
6c5211ce 1682 def _apply_header_cookies(self, url, cookies=None):
31215122
SS
1683 """Applies stray header cookies to the provided url
1684
1685 This loads header cookies and scopes them to the domain provided in `url`.
1686 While this is not ideal, it helps reduce the risk of them being sent
1687 to an unintended destination while mostly maintaining compatibility.
1688 """
1689 parsed = urllib.parse.urlparse(url)
1690 if not parsed.hostname:
1691 return
1692
6c5211ce 1693 for cookie in map(copy.copy, cookies or self.__header_cookies):
31215122
SS
1694 cookie.domain = f'.{parsed.hostname}'
1695 self.cookiejar.set_cookie(cookie)
1696
7e88d7d7 1697 @_handle_extraction_exceptions
58f197b7 1698 def __extract_info(self, url, ie, download, extra_info, process):
31215122
SS
1699 self._apply_header_cookies(url)
1700
693f0600 1701 try:
1702 ie_result = ie.extract(url)
1703 except UserNotLive as e:
1704 if process:
1705 if self.params.get('wait_for_video'):
1706 self.report_warning(e)
1707 self._wait_for_video()
1708 raise
a0566bbf 1709 if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
cb794ee0 1710 self.report_warning(f'Extractor {ie.IE_NAME} returned nothing{bug_reports_message()}')
a0566bbf 1711 return
1712 if isinstance(ie_result, list):
1713 # Backwards compatibility: old IE result format
1714 ie_result = {
1715 '_type': 'compat_list',
1716 'entries': ie_result,
1717 }
e37d0efb 1718 if extra_info.get('original_url'):
1719 ie_result.setdefault('original_url', extra_info['original_url'])
a0566bbf 1720 self.add_default_extra_info(ie_result, ie, url)
1721 if process:
f2ebc5c7 1722 self._wait_for_video(ie_result)
a0566bbf 1723 return self.process_ie_result(ie_result, download, extra_info)
8222d8de 1724 else:
a0566bbf 1725 return ie_result
fe7e0c98 1726
ea38e55f 1727 def add_default_extra_info(self, ie_result, ie, url):
6033d980 1728 if url is not None:
1729 self.add_extra_info(ie_result, {
1730 'webpage_url': url,
1731 'original_url': url,
57ebfca3 1732 })
1733 webpage_url = ie_result.get('webpage_url')
1734 if webpage_url:
1735 self.add_extra_info(ie_result, {
1736 'webpage_url_basename': url_basename(webpage_url),
1737 'webpage_url_domain': get_domain(webpage_url),
6033d980 1738 })
1739 if ie is not None:
1740 self.add_extra_info(ie_result, {
1741 'extractor': ie.IE_NAME,
1742 'extractor_key': ie.ie_key(),
1743 })
ea38e55f 1744
58adec46 1745 def process_ie_result(self, ie_result, download=True, extra_info=None):
8222d8de
JMF
1746 """
1747 Take the result of the ie(may be modified) and resolve all unresolved
1748 references (URLs, playlist items).
1749
1750 It will also download the videos if 'download'.
1751 Returns the resolved ie_result.
1752 """
58adec46 1753 if extra_info is None:
1754 extra_info = {}
e8ee972c
PH
1755 result_type = ie_result.get('_type', 'video')
1756
057a5206 1757 if result_type in ('url', 'url_transparent'):
8f97a15d 1758 ie_result['url'] = sanitize_url(
1759 ie_result['url'], scheme='http' if self.params.get('prefer_insecure') else 'https')
8791e78c 1760 if ie_result.get('original_url') and not extra_info.get('original_url'):
1761 extra_info = {'original_url': ie_result['original_url'], **extra_info}
e37d0efb 1762
057a5206 1763 extract_flat = self.params.get('extract_flat', False)
3089bc74
S
1764 if ((extract_flat == 'in_playlist' and 'playlist' in extra_info)
1765 or extract_flat is True):
ecb54191 1766 info_copy = ie_result.copy()
6033d980 1767 ie = try_get(ie_result.get('ie_key'), self.get_info_extractor)
360167b9 1768 if ie and not ie_result.get('id'):
4614bc22 1769 info_copy['id'] = ie.get_temp_id(ie_result['url'])
6033d980 1770 self.add_default_extra_info(info_copy, ie, ie_result['url'])
4614bc22 1771 self.add_extra_info(info_copy, extra_info)
b5475f11 1772 info_copy, _ = self.pre_process(info_copy)
94dc8604 1773 self._fill_common_fields(info_copy, False)
17060584 1774 self.__forced_printings(info_copy)
415f8d51 1775 self._raise_pending_errors(info_copy)
4614bc22 1776 if self.params.get('force_write_download_archive', False):
1777 self.record_download_archive(info_copy)
e8ee972c
PH
1778 return ie_result
1779
8222d8de 1780 if result_type == 'video':
b6c45014 1781 self.add_extra_info(ie_result, extra_info)
9c2b75b5 1782 ie_result = self.process_video_result(ie_result, download=download)
415f8d51 1783 self._raise_pending_errors(ie_result)
28b0eb0f 1784 additional_urls = (ie_result or {}).get('additional_urls')
9c2b75b5 1785 if additional_urls:
e9f4ccd1 1786 # TODO: Improve MetadataParserPP to allow setting a list
14f25df2 1787 if isinstance(additional_urls, str):
9c2b75b5 1788 additional_urls = [additional_urls]
1789 self.to_screen(
1790 '[info] %s: %d additional URL(s) requested' % (ie_result['id'], len(additional_urls)))
1791 self.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls))
1792 ie_result['additional_entries'] = [
1793 self.extract_info(
b69fd25c 1794 url, download, extra_info=extra_info,
9c2b75b5 1795 force_generic_extractor=self.params.get('force_generic_extractor'))
1796 for url in additional_urls
1797 ]
1798 return ie_result
8222d8de
JMF
1799 elif result_type == 'url':
1800 # We have to add extra_info to the results because it may be
1801 # contained in a playlist
07cce701 1802 return self.extract_info(
1803 ie_result['url'], download,
1804 ie_key=ie_result.get('ie_key'),
1805 extra_info=extra_info)
7fc3fa05
PH
1806 elif result_type == 'url_transparent':
1807 # Use the information from the embedding page
1808 info = self.extract_info(
1809 ie_result['url'], ie_key=ie_result.get('ie_key'),
1810 extra_info=extra_info, download=False, process=False)
1811
1640eb09
S
1812 # extract_info may return None when ignoreerrors is enabled and
1813 # extraction failed with an error, don't crash and return early
1814 # in this case
1815 if not info:
1816 return info
1817
3975b4d2 1818 exempted_fields = {'_type', 'url', 'ie_key'}
1819 if not ie_result.get('section_end') and ie_result.get('section_start') is None:
1820 # For video clips, the id etc of the clip extractor should be used
1821 exempted_fields |= {'id', 'extractor', 'extractor_key'}
1822
412c617d 1823 new_result = info.copy()
3975b4d2 1824 new_result.update(filter_dict(ie_result, lambda k, v: v is not None and k not in exempted_fields))
7fc3fa05 1825
0563f7ac
S
1826 # Extracted info may not be a video result (i.e.
1827 # info.get('_type', 'video') != video) but rather an url or
1828 # url_transparent. In such cases outer metadata (from ie_result)
1829 # should be propagated to inner one (info). For this to happen
1830 # _type of info should be overridden with url_transparent. This
067aa17e 1831 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
0563f7ac
S
1832 if new_result.get('_type') == 'url':
1833 new_result['_type'] = 'url_transparent'
7fc3fa05
PH
1834
1835 return self.process_ie_result(
1836 new_result, download=download, extra_info=extra_info)
40fcba5e 1837 elif result_type in ('playlist', 'multi_video'):
30a074c2 1838 # Protect from infinite recursion due to recursively nested playlists
1839 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
0bd5a039 1840 webpage_url = ie_result.get('webpage_url') # Playlists maynot have webpage_url
1841 if webpage_url and webpage_url in self._playlist_urls:
7e85e872 1842 self.to_screen(
30a074c2 1843 '[download] Skipping already downloaded playlist: %s'
1844 % ie_result.get('title') or ie_result.get('id'))
1845 return
7e85e872 1846
30a074c2 1847 self._playlist_level += 1
1848 self._playlist_urls.add(webpage_url)
03f83004 1849 self._fill_common_fields(ie_result, False)
bc516a3f 1850 self._sanitize_thumbnails(ie_result)
30a074c2 1851 try:
1852 return self.__process_playlist(ie_result, download)
1853 finally:
1854 self._playlist_level -= 1
1855 if not self._playlist_level:
1856 self._playlist_urls.clear()
8222d8de 1857 elif result_type == 'compat_list':
c9bf4114
PH
1858 self.report_warning(
1859 'Extractor %s returned a compat_list result. '
1860 'It needs to be updated.' % ie_result.get('extractor'))
5f6a1245 1861
8222d8de 1862 def _fixup(r):
b868936c 1863 self.add_extra_info(r, {
1864 'extractor': ie_result['extractor'],
1865 'webpage_url': ie_result['webpage_url'],
1866 'webpage_url_basename': url_basename(ie_result['webpage_url']),
0bb322b9 1867 'webpage_url_domain': get_domain(ie_result['webpage_url']),
b868936c 1868 'extractor_key': ie_result['extractor_key'],
1869 })
8222d8de
JMF
1870 return r
1871 ie_result['entries'] = [
b6c45014 1872 self.process_ie_result(_fixup(r), download, extra_info)
8222d8de
JMF
1873 for r in ie_result['entries']
1874 ]
1875 return ie_result
1876 else:
1877 raise Exception('Invalid result type: %s' % result_type)
1878
e92caff5 1879 def _ensure_dir_exists(self, path):
1880 return make_dir(path, self.report_error)
1881
3b603dbd 1882 @staticmethod
3bec830a 1883 def _playlist_infodict(ie_result, strict=False, **kwargs):
1884 info = {
1885 'playlist_count': ie_result.get('playlist_count'),
3b603dbd 1886 'playlist': ie_result.get('title') or ie_result.get('id'),
1887 'playlist_id': ie_result.get('id'),
1888 'playlist_title': ie_result.get('title'),
1889 'playlist_uploader': ie_result.get('uploader'),
1890 'playlist_uploader_id': ie_result.get('uploader_id'),
3b603dbd 1891 **kwargs,
1892 }
3bec830a 1893 if strict:
1894 return info
0bd5a039 1895 if ie_result.get('webpage_url'):
1896 info.update({
1897 'webpage_url': ie_result['webpage_url'],
1898 'webpage_url_basename': url_basename(ie_result['webpage_url']),
1899 'webpage_url_domain': get_domain(ie_result['webpage_url']),
1900 })
3bec830a 1901 return {
1902 **info,
1903 'playlist_index': 0,
59d7de0d 1904 '__last_playlist_index': max(ie_result.get('requested_entries') or (0, 0)),
3bec830a 1905 'extractor': ie_result['extractor'],
3bec830a 1906 'extractor_key': ie_result['extractor_key'],
1907 }
3b603dbd 1908
30a074c2 1909 def __process_playlist(self, ie_result, download):
7e88d7d7 1910 """Process each entry in the playlist"""
f5ea4748 1911 assert ie_result['_type'] in ('playlist', 'multi_video')
1912
3bec830a 1913 common_info = self._playlist_infodict(ie_result, strict=True)
3955b207 1914 title = common_info.get('playlist') or '<Untitled>'
3bec830a 1915 if self._match_entry(common_info, incomplete=True) is not None:
1916 return
c6e07cf1 1917 self.to_screen(f'[download] Downloading {ie_result["_type"]}: {title}')
f0d785d3 1918
7e88d7d7 1919 all_entries = PlaylistEntries(self, ie_result)
7e9a6125 1920 entries = orderedSet(all_entries.get_requested_items(), lazy=True)
1921
1922 lazy = self.params.get('lazy_playlist')
1923 if lazy:
1924 resolved_entries, n_entries = [], 'N/A'
1925 ie_result['requested_entries'], ie_result['entries'] = None, None
1926 else:
1927 entries = resolved_entries = list(entries)
1928 n_entries = len(resolved_entries)
1929 ie_result['requested_entries'], ie_result['entries'] = tuple(zip(*resolved_entries)) or ([], [])
1930 if not ie_result.get('playlist_count'):
1931 # Better to do this after potentially exhausting entries
1932 ie_result['playlist_count'] = all_entries.get_full_count()
498f5606 1933
0647d925 1934 extra = self._playlist_infodict(ie_result, n_entries=int_or_none(n_entries))
1935 ie_copy = collections.ChainMap(ie_result, extra)
3bec830a 1936
e08a85d8 1937 _infojson_written = False
0bfc53d0 1938 write_playlist_files = self.params.get('allow_playlist_files', True)
1939 if write_playlist_files and self.params.get('list_thumbnails'):
1940 self.list_thumbnails(ie_result)
1941 if write_playlist_files and not self.params.get('simulate'):
e08a85d8 1942 _infojson_written = self._write_info_json(
1943 'playlist', ie_result, self.prepare_filename(ie_copy, 'pl_infojson'))
1944 if _infojson_written is None:
80c03fa9 1945 return
1946 if self._write_description('playlist', ie_result,
1947 self.prepare_filename(ie_copy, 'pl_description')) is None:
1948 return
681de68e 1949 # TODO: This should be passed to ThumbnailsConvertor if necessary
3bec830a 1950 self._write_thumbnails('playlist', ie_result, self.prepare_filename(ie_copy, 'pl_thumbnail'))
30a074c2 1951
7e9a6125 1952 if lazy:
1953 if self.params.get('playlistreverse') or self.params.get('playlistrandom'):
1954 self.report_warning('playlistreverse and playlistrandom are not supported with lazy_playlist', only_once=True)
1955 elif self.params.get('playlistreverse'):
1956 entries.reverse()
1957 elif self.params.get('playlistrandom'):
30a074c2 1958 random.shuffle(entries)
1959
bc5c2f8a 1960 self.to_screen(f'[{ie_result["extractor"]}] Playlist {title}: Downloading {n_entries} items'
7e88d7d7 1961 f'{format_field(ie_result, "playlist_count", " of %s")}')
30a074c2 1962
134c913c 1963 keep_resolved_entries = self.params.get('extract_flat') != 'discard'
1964 if self.params.get('extract_flat') == 'discard_in_playlist':
1965 keep_resolved_entries = ie_result['_type'] != 'playlist'
1966 if keep_resolved_entries:
1967 self.write_debug('The information of all playlist entries will be held in memory')
1968
26e2805c 1969 failures = 0
1970 max_failures = self.params.get('skip_playlist_after_errors') or float('inf')
7e9a6125 1971 for i, (playlist_index, entry) in enumerate(entries):
1972 if lazy:
1973 resolved_entries.append((playlist_index, entry))
3bec830a 1974 if not entry:
7e88d7d7 1975 continue
1976
7e88d7d7 1977 entry['__x_forwarded_for_ip'] = ie_result.get('__x_forwarded_for_ip')
ad54c913 1978 if not lazy and 'playlist-index' in self.params['compat_opts']:
7e9a6125 1979 playlist_index = ie_result['requested_entries'][i]
1980
0647d925 1981 entry_copy = collections.ChainMap(entry, {
3bec830a 1982 **common_info,
3955b207 1983 'n_entries': int_or_none(n_entries),
71729754 1984 'playlist_index': playlist_index,
7e9a6125 1985 'playlist_autonumber': i + 1,
0647d925 1986 })
3bec830a 1987
0647d925 1988 if self._match_entry(entry_copy, incomplete=True) is not None:
f0ad6f8c 1989 # For compatabilty with youtube-dl. See https://github.com/yt-dlp/yt-dlp/issues/4369
1990 resolved_entries[i] = (playlist_index, NO_DEFAULT)
3bec830a 1991 continue
1992
bc5c2f8a 1993 self.to_screen('[download] Downloading item %s of %s' % (
3bec830a 1994 self._format_screen(i + 1, self.Styles.ID), self._format_screen(n_entries, self.Styles.EMPHASIS)))
1995
ec54bd43 1996 entry_result = self.__process_iterable_entry(entry, download, collections.ChainMap({
a6ca61d4 1997 'playlist_index': playlist_index,
1998 'playlist_autonumber': i + 1,
ec54bd43 1999 }, extra))
26e2805c 2000 if not entry_result:
2001 failures += 1
2002 if failures >= max_failures:
2003 self.report_error(
7e88d7d7 2004 f'Skipping the remaining entries in playlist "{title}" since {failures} items failed extraction')
26e2805c 2005 break
134c913c 2006 if keep_resolved_entries:
2007 resolved_entries[i] = (playlist_index, entry_result)
7e88d7d7 2008
2009 # Update with processed data
f0ad6f8c 2010 ie_result['entries'] = [e for _, e in resolved_entries if e is not NO_DEFAULT]
bc5c2f8a 2011 ie_result['requested_entries'] = [i for i, e in resolved_entries if e is not NO_DEFAULT]
2012 if ie_result['requested_entries'] == try_call(lambda: list(range(1, ie_result['playlist_count'] + 1))):
2013 # Do not set for full playlist
2014 ie_result.pop('requested_entries')
e08a85d8 2015
2016 # Write the updated info to json
cb96c5be 2017 if _infojson_written is True and self._write_info_json(
e08a85d8 2018 'updated playlist', ie_result,
2019 self.prepare_filename(ie_copy, 'pl_infojson'), overwrite=True) is None:
2020 return
ca30f449 2021
ed5835b4 2022 ie_result = self.run_all_pps('playlist', ie_result)
7e88d7d7 2023 self.to_screen(f'[download] Finished downloading playlist: {title}')
30a074c2 2024 return ie_result
2025
7e88d7d7 2026 @_handle_extraction_exceptions
a0566bbf 2027 def __process_iterable_entry(self, entry, download, extra_info):
2028 return self.process_ie_result(
2029 entry, download=download, extra_info=extra_info)
2030
67134eab
JMF
2031 def _build_format_filter(self, filter_spec):
2032 " Returns a function to filter the formats according to the filter_spec "
083c9df9
PH
2033
2034 OPERATORS = {
2035 '<': operator.lt,
2036 '<=': operator.le,
2037 '>': operator.gt,
2038 '>=': operator.ge,
2039 '=': operator.eq,
2040 '!=': operator.ne,
2041 }
67134eab 2042 operator_rex = re.compile(r'''(?x)\s*
c3f624ef 2043 (?P<key>[\w.-]+)\s*
187986a8 2044 (?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
2045 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)\s*
083c9df9 2046 ''' % '|'.join(map(re.escape, OPERATORS.keys())))
187986a8 2047 m = operator_rex.fullmatch(filter_spec)
9ddb6925
S
2048 if m:
2049 try:
2050 comparison_value = int(m.group('value'))
2051 except ValueError:
2052 comparison_value = parse_filesize(m.group('value'))
2053 if comparison_value is None:
2054 comparison_value = parse_filesize(m.group('value') + 'B')
2055 if comparison_value is None:
2056 raise ValueError(
2057 'Invalid value %r in format specification %r' % (
67134eab 2058 m.group('value'), filter_spec))
9ddb6925
S
2059 op = OPERATORS[m.group('op')]
2060
083c9df9 2061 if not m:
9ddb6925
S
2062 STR_OPERATORS = {
2063 '=': operator.eq,
10d33b34
YCH
2064 '^=': lambda attr, value: attr.startswith(value),
2065 '$=': lambda attr, value: attr.endswith(value),
2066 '*=': lambda attr, value: value in attr,
1ce9a3cb 2067 '~=': lambda attr, value: value.search(attr) is not None
9ddb6925 2068 }
187986a8 2069 str_operator_rex = re.compile(r'''(?x)\s*
2070 (?P<key>[a-zA-Z0-9._-]+)\s*
1ce9a3cb
LF
2071 (?P<negation>!\s*)?(?P<op>%s)\s*(?P<none_inclusive>\?\s*)?
2072 (?P<quote>["'])?
2073 (?P<value>(?(quote)(?:(?!(?P=quote))[^\\]|\\.)+|[\w.-]+))
2074 (?(quote)(?P=quote))\s*
9ddb6925 2075 ''' % '|'.join(map(re.escape, STR_OPERATORS.keys())))
187986a8 2076 m = str_operator_rex.fullmatch(filter_spec)
9ddb6925 2077 if m:
1ce9a3cb
LF
2078 if m.group('op') == '~=':
2079 comparison_value = re.compile(m.group('value'))
2080 else:
2081 comparison_value = re.sub(r'''\\([\\"'])''', r'\1', m.group('value'))
2cc779f4
S
2082 str_op = STR_OPERATORS[m.group('op')]
2083 if m.group('negation'):
e118a879 2084 op = lambda attr, value: not str_op(attr, value)
2cc779f4
S
2085 else:
2086 op = str_op
083c9df9 2087
9ddb6925 2088 if not m:
187986a8 2089 raise SyntaxError('Invalid filter specification %r' % filter_spec)
083c9df9
PH
2090
2091 def _filter(f):
2092 actual_value = f.get(m.group('key'))
2093 if actual_value is None:
2094 return m.group('none_inclusive')
2095 return op(actual_value, comparison_value)
67134eab
JMF
2096 return _filter
2097
9f1a1c36 2098 def _check_formats(self, formats):
2099 for f in formats:
2100 self.to_screen('[info] Testing format %s' % f['format_id'])
75689fe5 2101 path = self.get_output_path('temp')
2102 if not self._ensure_dir_exists(f'{path}/'):
2103 continue
2104 temp_file = tempfile.NamedTemporaryFile(suffix='.tmp', delete=False, dir=path or None)
9f1a1c36 2105 temp_file.close()
2106 try:
2107 success, _ = self.dl(temp_file.name, f, test=True)
8a82af35 2108 except (DownloadError, OSError, ValueError) + network_exceptions:
9f1a1c36 2109 success = False
2110 finally:
2111 if os.path.exists(temp_file.name):
2112 try:
2113 os.remove(temp_file.name)
2114 except OSError:
2115 self.report_warning('Unable to delete temporary file "%s"' % temp_file.name)
2116 if success:
2117 yield f
2118 else:
2119 self.to_screen('[info] Unable to download format %s. Skipping...' % f['format_id'])
2120
0017d9ad 2121 def _default_format_spec(self, info_dict, download=True):
0017d9ad 2122
af0f7428
S
2123 def can_merge():
2124 merger = FFmpegMergerPP(self)
2125 return merger.available and merger.can_merge()
2126
91ebc640 2127 prefer_best = (
b7b04c78 2128 not self.params.get('simulate')
91ebc640 2129 and download
2130 and (
2131 not can_merge()
21633673 2132 or info_dict.get('is_live') and not self.params.get('live_from_start')
bf1824b3 2133 or self.params['outtmpl']['default'] == '-'))
53ed7066 2134 compat = (
2135 prefer_best
2136 or self.params.get('allow_multiple_audio_streams', False)
8a82af35 2137 or 'format-spec' in self.params['compat_opts'])
91ebc640 2138
2139 return (
53ed7066 2140 'best/bestvideo+bestaudio' if prefer_best
2141 else 'bestvideo*+bestaudio/best' if not compat
91ebc640 2142 else 'bestvideo+bestaudio/best')
0017d9ad 2143
67134eab
JMF
2144 def build_format_selector(self, format_spec):
2145 def syntax_error(note, start):
2146 message = (
2147 'Invalid format specification: '
86e5f3ed 2148 '{}\n\t{}\n\t{}^'.format(note, format_spec, ' ' * start[1]))
67134eab
JMF
2149 return SyntaxError(message)
2150
2151 PICKFIRST = 'PICKFIRST'
2152 MERGE = 'MERGE'
2153 SINGLE = 'SINGLE'
0130afb7 2154 GROUP = 'GROUP'
67134eab
JMF
2155 FormatSelector = collections.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
2156
91ebc640 2157 allow_multiple_streams = {'audio': self.params.get('allow_multiple_audio_streams', False),
2158 'video': self.params.get('allow_multiple_video_streams', False)}
909d24dd 2159
67134eab
JMF
2160 def _parse_filter(tokens):
2161 filter_parts = []
6f2287cb 2162 for type, string_, start, _, _ in tokens:
2163 if type == tokenize.OP and string_ == ']':
67134eab
JMF
2164 return ''.join(filter_parts)
2165 else:
6f2287cb 2166 filter_parts.append(string_)
67134eab 2167
232541df 2168 def _remove_unused_ops(tokens):
62b58c09
L
2169 # Remove operators that we don't use and join them with the surrounding strings.
2170 # E.g. 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
232541df
JMF
2171 ALLOWED_OPS = ('/', '+', ',', '(', ')')
2172 last_string, last_start, last_end, last_line = None, None, None, None
6f2287cb 2173 for type, string_, start, end, line in tokens:
2174 if type == tokenize.OP and string_ == '[':
232541df
JMF
2175 if last_string:
2176 yield tokenize.NAME, last_string, last_start, last_end, last_line
2177 last_string = None
6f2287cb 2178 yield type, string_, start, end, line
232541df 2179 # everything inside brackets will be handled by _parse_filter
6f2287cb 2180 for type, string_, start, end, line in tokens:
2181 yield type, string_, start, end, line
2182 if type == tokenize.OP and string_ == ']':
232541df 2183 break
6f2287cb 2184 elif type == tokenize.OP and string_ in ALLOWED_OPS:
232541df
JMF
2185 if last_string:
2186 yield tokenize.NAME, last_string, last_start, last_end, last_line
2187 last_string = None
6f2287cb 2188 yield type, string_, start, end, line
232541df
JMF
2189 elif type in [tokenize.NAME, tokenize.NUMBER, tokenize.OP]:
2190 if not last_string:
6f2287cb 2191 last_string = string_
232541df
JMF
2192 last_start = start
2193 last_end = end
2194 else:
6f2287cb 2195 last_string += string_
232541df
JMF
2196 if last_string:
2197 yield tokenize.NAME, last_string, last_start, last_end, last_line
2198
cf2ac6df 2199 def _parse_format_selection(tokens, inside_merge=False, inside_choice=False, inside_group=False):
67134eab
JMF
2200 selectors = []
2201 current_selector = None
6f2287cb 2202 for type, string_, start, _, _ in tokens:
67134eab
JMF
2203 # ENCODING is only defined in python 3.x
2204 if type == getattr(tokenize, 'ENCODING', None):
2205 continue
2206 elif type in [tokenize.NAME, tokenize.NUMBER]:
6f2287cb 2207 current_selector = FormatSelector(SINGLE, string_, [])
67134eab 2208 elif type == tokenize.OP:
6f2287cb 2209 if string_ == ')':
cf2ac6df
JMF
2210 if not inside_group:
2211 # ')' will be handled by the parentheses group
2212 tokens.restore_last_token()
67134eab 2213 break
6f2287cb 2214 elif inside_merge and string_ in ['/', ',']:
0130afb7
JMF
2215 tokens.restore_last_token()
2216 break
6f2287cb 2217 elif inside_choice and string_ == ',':
cf2ac6df
JMF
2218 tokens.restore_last_token()
2219 break
6f2287cb 2220 elif string_ == ',':
0a31a350
JMF
2221 if not current_selector:
2222 raise syntax_error('"," must follow a format selector', start)
67134eab
JMF
2223 selectors.append(current_selector)
2224 current_selector = None
6f2287cb 2225 elif string_ == '/':
d96d604e
JMF
2226 if not current_selector:
2227 raise syntax_error('"/" must follow a format selector', start)
67134eab 2228 first_choice = current_selector
cf2ac6df 2229 second_choice = _parse_format_selection(tokens, inside_choice=True)
f5f4a27a 2230 current_selector = FormatSelector(PICKFIRST, (first_choice, second_choice), [])
6f2287cb 2231 elif string_ == '[':
67134eab
JMF
2232 if not current_selector:
2233 current_selector = FormatSelector(SINGLE, 'best', [])
2234 format_filter = _parse_filter(tokens)
2235 current_selector.filters.append(format_filter)
6f2287cb 2236 elif string_ == '(':
0130afb7
JMF
2237 if current_selector:
2238 raise syntax_error('Unexpected "("', start)
cf2ac6df
JMF
2239 group = _parse_format_selection(tokens, inside_group=True)
2240 current_selector = FormatSelector(GROUP, group, [])
6f2287cb 2241 elif string_ == '+':
d03cfdce 2242 if not current_selector:
2243 raise syntax_error('Unexpected "+"', start)
2244 selector_1 = current_selector
2245 selector_2 = _parse_format_selection(tokens, inside_merge=True)
2246 if not selector_2:
2247 raise syntax_error('Expected a selector', start)
2248 current_selector = FormatSelector(MERGE, (selector_1, selector_2), [])
67134eab 2249 else:
6f2287cb 2250 raise syntax_error(f'Operator not recognized: "{string_}"', start)
67134eab
JMF
2251 elif type == tokenize.ENDMARKER:
2252 break
2253 if current_selector:
2254 selectors.append(current_selector)
2255 return selectors
2256
f8d4ad9a 2257 def _merge(formats_pair):
2258 format_1, format_2 = formats_pair
2259
2260 formats_info = []
2261 formats_info.extend(format_1.get('requested_formats', (format_1,)))
2262 formats_info.extend(format_2.get('requested_formats', (format_2,)))
2263
2264 if not allow_multiple_streams['video'] or not allow_multiple_streams['audio']:
551f9388 2265 get_no_more = {'video': False, 'audio': False}
f8d4ad9a 2266 for (i, fmt_info) in enumerate(formats_info):
551f9388 2267 if fmt_info.get('acodec') == fmt_info.get('vcodec') == 'none':
2268 formats_info.pop(i)
2269 continue
2270 for aud_vid in ['audio', 'video']:
f8d4ad9a 2271 if not allow_multiple_streams[aud_vid] and fmt_info.get(aud_vid[0] + 'codec') != 'none':
2272 if get_no_more[aud_vid]:
2273 formats_info.pop(i)
f5510afe 2274 break
f8d4ad9a 2275 get_no_more[aud_vid] = True
2276
2277 if len(formats_info) == 1:
2278 return formats_info[0]
2279
2280 video_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('vcodec') != 'none']
2281 audio_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('acodec') != 'none']
2282
2283 the_only_video = video_fmts[0] if len(video_fmts) == 1 else None
2284 the_only_audio = audio_fmts[0] if len(audio_fmts) == 1 else None
2285
fc61aff4
LL
2286 output_ext = get_compatible_ext(
2287 vcodecs=[f.get('vcodec') for f in video_fmts],
2288 acodecs=[f.get('acodec') for f in audio_fmts],
2289 vexts=[f['ext'] for f in video_fmts],
2290 aexts=[f['ext'] for f in audio_fmts],
2291 preferences=(try_call(lambda: self.params['merge_output_format'].split('/'))
2292 or self.params.get('prefer_free_formats') and ('webm', 'mkv')))
f8d4ad9a 2293
975a0d0d 2294 filtered = lambda *keys: filter(None, (traverse_obj(fmt, *keys) for fmt in formats_info))
2295
f8d4ad9a 2296 new_dict = {
2297 'requested_formats': formats_info,
975a0d0d 2298 'format': '+'.join(filtered('format')),
2299 'format_id': '+'.join(filtered('format_id')),
f8d4ad9a 2300 'ext': output_ext,
975a0d0d 2301 'protocol': '+'.join(map(determine_protocol, formats_info)),
093a1710 2302 'language': '+'.join(orderedSet(filtered('language'))) or None,
2303 'format_note': '+'.join(orderedSet(filtered('format_note'))) or None,
2304 'filesize_approx': sum(filtered('filesize', 'filesize_approx')) or None,
975a0d0d 2305 'tbr': sum(filtered('tbr', 'vbr', 'abr')),
f8d4ad9a 2306 }
2307
2308 if the_only_video:
2309 new_dict.update({
2310 'width': the_only_video.get('width'),
2311 'height': the_only_video.get('height'),
2312 'resolution': the_only_video.get('resolution') or self.format_resolution(the_only_video),
2313 'fps': the_only_video.get('fps'),
49a57e70 2314 'dynamic_range': the_only_video.get('dynamic_range'),
f8d4ad9a 2315 'vcodec': the_only_video.get('vcodec'),
2316 'vbr': the_only_video.get('vbr'),
2317 'stretched_ratio': the_only_video.get('stretched_ratio'),
105bfd90 2318 'aspect_ratio': the_only_video.get('aspect_ratio'),
f8d4ad9a 2319 })
2320
2321 if the_only_audio:
2322 new_dict.update({
2323 'acodec': the_only_audio.get('acodec'),
2324 'abr': the_only_audio.get('abr'),
975a0d0d 2325 'asr': the_only_audio.get('asr'),
b8ed0f15 2326 'audio_channels': the_only_audio.get('audio_channels')
f8d4ad9a 2327 })
2328
2329 return new_dict
2330
e8e73840 2331 def _check_formats(formats):
8cb7fc44 2332 if self.params.get('check_formats') == 'selected':
2333 yield from self._check_formats(formats)
2334 return
2335 elif (self.params.get('check_formats') is not None
bc344cd4 2336 or self.params.get('allow_unplayable_formats')):
981052c9 2337 yield from formats
b5ac45b1 2338 return
bc344cd4 2339
2340 for f in formats:
ef79d20d 2341 if f.get('has_drm') or f.get('__needs_testing'):
bc344cd4 2342 yield from self._check_formats([f])
2343 else:
2344 yield f
e8e73840 2345
67134eab 2346 def _build_selector_function(selector):
909d24dd 2347 if isinstance(selector, list): # ,
67134eab
JMF
2348 fs = [_build_selector_function(s) for s in selector]
2349
317f7ab6 2350 def selector_function(ctx):
67134eab 2351 for f in fs:
981052c9 2352 yield from f(ctx)
67134eab 2353 return selector_function
909d24dd 2354
2355 elif selector.type == GROUP: # ()
0130afb7 2356 selector_function = _build_selector_function(selector.selector)
909d24dd 2357
2358 elif selector.type == PICKFIRST: # /
67134eab
JMF
2359 fs = [_build_selector_function(s) for s in selector.selector]
2360
317f7ab6 2361 def selector_function(ctx):
67134eab 2362 for f in fs:
317f7ab6 2363 picked_formats = list(f(ctx))
67134eab
JMF
2364 if picked_formats:
2365 return picked_formats
2366 return []
67134eab 2367
981052c9 2368 elif selector.type == MERGE: # +
2369 selector_1, selector_2 = map(_build_selector_function, selector.selector)
2370
2371 def selector_function(ctx):
adbc4ec4 2372 for pair in itertools.product(selector_1(ctx), selector_2(ctx)):
981052c9 2373 yield _merge(pair)
2374
909d24dd 2375 elif selector.type == SINGLE: # atom
598d185d 2376 format_spec = selector.selector or 'best'
909d24dd 2377
f8d4ad9a 2378 # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
909d24dd 2379 if format_spec == 'all':
2380 def selector_function(ctx):
9222c381 2381 yield from _check_formats(ctx['formats'][::-1])
f8d4ad9a 2382 elif format_spec == 'mergeall':
2383 def selector_function(ctx):
316f2650 2384 formats = list(_check_formats(
2385 f for f in ctx['formats'] if f.get('vcodec') != 'none' or f.get('acodec') != 'none'))
e01d6aa4 2386 if not formats:
2387 return
921b76ca 2388 merged_format = formats[-1]
2389 for f in formats[-2::-1]:
f8d4ad9a 2390 merged_format = _merge((merged_format, f))
2391 yield merged_format
909d24dd 2392
2393 else:
85e801a9 2394 format_fallback, seperate_fallback, format_reverse, format_idx = False, None, True, 1
eff63539 2395 mobj = re.match(
2396 r'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
2397 format_spec)
2398 if mobj is not None:
2399 format_idx = int_or_none(mobj.group('n'), default=1)
e8e73840 2400 format_reverse = mobj.group('bw')[0] == 'b'
eff63539 2401 format_type = (mobj.group('type') or [None])[0]
2402 not_format_type = {'v': 'a', 'a': 'v'}.get(format_type)
2403 format_modified = mobj.group('mod') is not None
909d24dd 2404
2405 format_fallback = not format_type and not format_modified # for b, w
8326b00a 2406 _filter_f = (
eff63539 2407 (lambda f: f.get('%scodec' % format_type) != 'none')
2408 if format_type and format_modified # bv*, ba*, wv*, wa*
2409 else (lambda f: f.get('%scodec' % not_format_type) == 'none')
2410 if format_type # bv, ba, wv, wa
2411 else (lambda f: f.get('vcodec') != 'none' and f.get('acodec') != 'none')
2412 if not format_modified # b, w
8326b00a 2413 else lambda f: True) # b*, w*
2414 filter_f = lambda f: _filter_f(f) and (
2415 f.get('vcodec') != 'none' or f.get('acodec') != 'none')
67134eab 2416 else:
48ee10ee 2417 if format_spec in self._format_selection_exts['audio']:
b11c04a8 2418 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none'
48ee10ee 2419 elif format_spec in self._format_selection_exts['video']:
b11c04a8 2420 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none' and f.get('vcodec') != 'none'
85e801a9 2421 seperate_fallback = lambda f: f.get('ext') == format_spec and f.get('vcodec') != 'none'
48ee10ee 2422 elif format_spec in self._format_selection_exts['storyboards']:
b11c04a8 2423 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') == 'none' and f.get('vcodec') == 'none'
2424 else:
b5ae35ee 2425 filter_f = lambda f: f.get('format_id') == format_spec # id
909d24dd 2426
2427 def selector_function(ctx):
2428 formats = list(ctx['formats'])
909d24dd 2429 matches = list(filter(filter_f, formats)) if filter_f is not None else formats
85e801a9 2430 if not matches:
2431 if format_fallback and ctx['incomplete_formats']:
2432 # for extractors with incomplete formats (audio only (soundcloud)
2433 # or video only (imgur)) best/worst will fallback to
2434 # best/worst {video,audio}-only format
2435 matches = formats
2436 elif seperate_fallback and not ctx['has_merged_format']:
2437 # for compatibility with youtube-dl when there is no pre-merged format
2438 matches = list(filter(seperate_fallback, formats))
981052c9 2439 matches = LazyList(_check_formats(matches[::-1 if format_reverse else 1]))
2440 try:
e8e73840 2441 yield matches[format_idx - 1]
4abea8ca 2442 except LazyList.IndexError:
981052c9 2443 return
083c9df9 2444
67134eab 2445 filters = [self._build_format_filter(f) for f in selector.filters]
083c9df9 2446
317f7ab6 2447 def final_selector(ctx):
adbc4ec4 2448 ctx_copy = dict(ctx)
67134eab 2449 for _filter in filters:
317f7ab6
S
2450 ctx_copy['formats'] = list(filter(_filter, ctx_copy['formats']))
2451 return selector_function(ctx_copy)
67134eab 2452 return final_selector
083c9df9 2453
0f06bcd7 2454 stream = io.BytesIO(format_spec.encode())
0130afb7 2455 try:
f9934b96 2456 tokens = list(_remove_unused_ops(tokenize.tokenize(stream.readline)))
0130afb7
JMF
2457 except tokenize.TokenError:
2458 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec)))
2459
86e5f3ed 2460 class TokenIterator:
0130afb7
JMF
2461 def __init__(self, tokens):
2462 self.tokens = tokens
2463 self.counter = 0
2464
2465 def __iter__(self):
2466 return self
2467
2468 def __next__(self):
2469 if self.counter >= len(self.tokens):
2470 raise StopIteration()
2471 value = self.tokens[self.counter]
2472 self.counter += 1
2473 return value
2474
2475 next = __next__
2476
2477 def restore_last_token(self):
2478 self.counter -= 1
2479
2480 parsed_selector = _parse_format_selection(iter(TokenIterator(tokens)))
67134eab 2481 return _build_selector_function(parsed_selector)
a9c58ad9 2482
6c5211ce 2483 def _calc_headers(self, info_dict, load_cookies=False):
227bf1a3 2484 res = HTTPHeaderDict(self.params['http_headers'], info_dict.get('http_headers'))
c365dba8 2485 clean_headers(res)
6c5211ce 2486
2487 if load_cookies: # For --load-info-json
2488 self._load_cookies(res.get('Cookie'), autoscope=info_dict['url']) # compat
2489 self._load_cookies(info_dict.get('cookies'), autoscope=False)
2490 # The `Cookie` header is removed to prevent leaks and unscoped cookies.
2491 # See: https://github.com/yt-dlp/yt-dlp/security/advisories/GHSA-v8mc-9377-rwjj
2492 res.pop('Cookie', None)
31215122 2493 cookies = self.cookiejar.get_cookies_for_url(info_dict['url'])
e5660ee6 2494 if cookies:
31215122
SS
2495 encoder = LenientSimpleCookie()
2496 values = []
2497 for cookie in cookies:
2498 _, value = encoder.value_encode(cookie.value)
2499 values.append(f'{cookie.name}={value}')
2500 if cookie.domain:
2501 values.append(f'Domain={cookie.domain}')
2502 if cookie.path:
2503 values.append(f'Path={cookie.path}')
2504 if cookie.secure:
2505 values.append('Secure')
2506 if cookie.expires:
2507 values.append(f'Expires={cookie.expires}')
2508 if cookie.version:
2509 values.append(f'Version={cookie.version}')
2510 info_dict['cookies'] = '; '.join(values)
e5660ee6 2511
0016b84e
S
2512 if 'X-Forwarded-For' not in res:
2513 x_forwarded_for_ip = info_dict.get('__x_forwarded_for_ip')
2514 if x_forwarded_for_ip:
2515 res['X-Forwarded-For'] = x_forwarded_for_ip
2516
e5660ee6
JMF
2517 return res
2518
c487cf00 2519 def _calc_cookies(self, url):
b87e01c1 2520 self.deprecation_warning('"YoutubeDL._calc_cookies" is deprecated and may be removed in a future version')
2521 return self.cookiejar.get_cookie_header(url)
e5660ee6 2522
9f1a1c36 2523 def _sort_thumbnails(self, thumbnails):
2524 thumbnails.sort(key=lambda t: (
2525 t.get('preference') if t.get('preference') is not None else -1,
2526 t.get('width') if t.get('width') is not None else -1,
2527 t.get('height') if t.get('height') is not None else -1,
2528 t.get('id') if t.get('id') is not None else '',
2529 t.get('url')))
2530
b0249bca 2531 def _sanitize_thumbnails(self, info_dict):
bc516a3f 2532 thumbnails = info_dict.get('thumbnails')
2533 if thumbnails is None:
2534 thumbnail = info_dict.get('thumbnail')
2535 if thumbnail:
2536 info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}]
9f1a1c36 2537 if not thumbnails:
2538 return
2539
2540 def check_thumbnails(thumbnails):
2541 for t in thumbnails:
2542 self.to_screen(f'[info] Testing thumbnail {t["id"]}')
2543 try:
2544 self.urlopen(HEADRequest(t['url']))
2545 except network_exceptions as err:
2546 self.to_screen(f'[info] Unable to connect to thumbnail {t["id"]} URL {t["url"]!r} - {err}. Skipping...')
2547 continue
2548 yield t
2549
2550 self._sort_thumbnails(thumbnails)
2551 for i, t in enumerate(thumbnails):
2552 if t.get('id') is None:
2553 t['id'] = '%d' % i
2554 if t.get('width') and t.get('height'):
2555 t['resolution'] = '%dx%d' % (t['width'], t['height'])
2556 t['url'] = sanitize_url(t['url'])
2557
2558 if self.params.get('check_formats') is True:
282f5709 2559 info_dict['thumbnails'] = LazyList(check_thumbnails(thumbnails[::-1]), reverse=True)
9f1a1c36 2560 else:
2561 info_dict['thumbnails'] = thumbnails
bc516a3f 2562
94dc8604 2563 def _fill_common_fields(self, info_dict, final=True):
03f83004 2564 # TODO: move sanitization here
94dc8604 2565 if final:
7aefd19a 2566 title = info_dict['fulltitle'] = info_dict.get('title')
d4736fdb 2567 if not title:
2568 if title == '':
2569 self.write_debug('Extractor gave empty title. Creating a generic title')
2570 else:
2571 self.report_warning('Extractor failed to obtain "title". Creating a generic title instead')
1d485a1a 2572 info_dict['title'] = f'{info_dict["extractor"].replace(":", "-")} video #{info_dict["id"]}'
03f83004
LNO
2573
2574 if info_dict.get('duration') is not None:
2575 info_dict['duration_string'] = formatSeconds(info_dict['duration'])
2576
2577 for ts_key, date_key in (
2578 ('timestamp', 'upload_date'),
2579 ('release_timestamp', 'release_date'),
2580 ('modified_timestamp', 'modified_date'),
2581 ):
2582 if info_dict.get(date_key) is None and info_dict.get(ts_key) is not None:
2583 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
2584 # see http://bugs.python.org/issue1646728)
19a03940 2585 with contextlib.suppress(ValueError, OverflowError, OSError):
836e06d2 2586 upload_date = datetime.datetime.fromtimestamp(info_dict[ts_key], datetime.timezone.utc)
03f83004 2587 info_dict[date_key] = upload_date.strftime('%Y%m%d')
03f83004
LNO
2588
2589 live_keys = ('is_live', 'was_live')
2590 live_status = info_dict.get('live_status')
2591 if live_status is None:
2592 for key in live_keys:
2593 if info_dict.get(key) is False:
2594 continue
2595 if info_dict.get(key):
2596 live_status = key
2597 break
2598 if all(info_dict.get(key) is False for key in live_keys):
2599 live_status = 'not_live'
2600 if live_status:
2601 info_dict['live_status'] = live_status
2602 for key in live_keys:
2603 if info_dict.get(key) is None:
2604 info_dict[key] = (live_status == key)
a057779d 2605 if live_status == 'post_live':
2606 info_dict['was_live'] = True
03f83004
LNO
2607
2608 # Auto generate title fields corresponding to the *_number fields when missing
2609 # in order to always have clean titles. This is very common for TV series.
2610 for field in ('chapter', 'season', 'episode'):
94dc8604 2611 if final and info_dict.get('%s_number' % field) is not None and not info_dict.get(field):
03f83004
LNO
2612 info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field])
2613
415f8d51 2614 def _raise_pending_errors(self, info):
2615 err = info.pop('__pending_error', None)
2616 if err:
2617 self.report_error(err, tb=False)
2618
784320c9 2619 def sort_formats(self, info_dict):
2620 formats = self._get_formats(info_dict)
784320c9 2621 formats.sort(key=FormatSorter(
c154302c 2622 self, info_dict.get('_format_sort_fields') or []).calculate_preference)
784320c9 2623
dd82ffea
JMF
2624 def process_video_result(self, info_dict, download=True):
2625 assert info_dict.get('_type', 'video') == 'video'
9c906919 2626 self._num_videos += 1
dd82ffea 2627
bec1fad2 2628 if 'id' not in info_dict:
fc08bdd6 2629 raise ExtractorError('Missing "id" field in extractor result', ie=info_dict['extractor'])
2630 elif not info_dict.get('id'):
2631 raise ExtractorError('Extractor failed to obtain "id"', ie=info_dict['extractor'])
455a15e2 2632
c9969434
S
2633 def report_force_conversion(field, field_not, conversion):
2634 self.report_warning(
2635 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
2636 % (field, field_not, conversion))
2637
2638 def sanitize_string_field(info, string_field):
2639 field = info.get(string_field)
14f25df2 2640 if field is None or isinstance(field, str):
c9969434
S
2641 return
2642 report_force_conversion(string_field, 'a string', 'string')
14f25df2 2643 info[string_field] = str(field)
c9969434
S
2644
2645 def sanitize_numeric_fields(info):
2646 for numeric_field in self._NUMERIC_FIELDS:
2647 field = info.get(numeric_field)
f9934b96 2648 if field is None or isinstance(field, (int, float)):
c9969434
S
2649 continue
2650 report_force_conversion(numeric_field, 'numeric', 'int')
2651 info[numeric_field] = int_or_none(field)
2652
2653 sanitize_string_field(info_dict, 'id')
2654 sanitize_numeric_fields(info_dict)
3975b4d2 2655 if info_dict.get('section_end') and info_dict.get('section_start') is not None:
2656 info_dict['duration'] = round(info_dict['section_end'] - info_dict['section_start'], 3)
4c3f8c3f 2657 if (info_dict.get('duration') or 0) <= 0 and info_dict.pop('duration', None):
50e93e03 2658 self.report_warning('"duration" field is negative, there is an error in extractor')
be6217b2 2659
9eef7c4e 2660 chapters = info_dict.get('chapters') or []
a3976e07 2661 if chapters and chapters[0].get('start_time'):
2662 chapters.insert(0, {'start_time': 0})
2663
9eef7c4e 2664 dummy_chapter = {'end_time': 0, 'start_time': info_dict.get('duration')}
a3976e07 2665 for idx, (prev, current, next_) in enumerate(zip(
2666 (dummy_chapter, *chapters), chapters, (*chapters[1:], dummy_chapter)), 1):
9eef7c4e 2667 if current.get('start_time') is None:
2668 current['start_time'] = prev.get('end_time')
2669 if not current.get('end_time'):
2670 current['end_time'] = next_.get('start_time')
a3976e07 2671 if not current.get('title'):
2672 current['title'] = f'<Untitled Chapter {idx}>'
9eef7c4e 2673
dd82ffea
JMF
2674 if 'playlist' not in info_dict:
2675 # It isn't part of a playlist
2676 info_dict['playlist'] = None
2677 info_dict['playlist_index'] = None
2678
bc516a3f 2679 self._sanitize_thumbnails(info_dict)
d5519808 2680
536a55da 2681 thumbnail = info_dict.get('thumbnail')
bc516a3f 2682 thumbnails = info_dict.get('thumbnails')
536a55da
S
2683 if thumbnail:
2684 info_dict['thumbnail'] = sanitize_url(thumbnail)
2685 elif thumbnails:
d5519808
PH
2686 info_dict['thumbnail'] = thumbnails[-1]['url']
2687
ae30b840 2688 if info_dict.get('display_id') is None and 'id' in info_dict:
0afef30b
PH
2689 info_dict['display_id'] = info_dict['id']
2690
03f83004 2691 self._fill_common_fields(info_dict)
33d2fc2f 2692
05108a49
S
2693 for cc_kind in ('subtitles', 'automatic_captions'):
2694 cc = info_dict.get(cc_kind)
2695 if cc:
2696 for _, subtitle in cc.items():
2697 for subtitle_format in subtitle:
2698 if subtitle_format.get('url'):
2699 subtitle_format['url'] = sanitize_url(subtitle_format['url'])
2700 if subtitle_format.get('ext') is None:
2701 subtitle_format['ext'] = determine_ext(subtitle_format['url']).lower()
2702
2703 automatic_captions = info_dict.get('automatic_captions')
4bba3716 2704 subtitles = info_dict.get('subtitles')
4bba3716 2705
360e1ca5 2706 info_dict['requested_subtitles'] = self.process_subtitles(
05108a49 2707 info_dict['id'], subtitles, automatic_captions)
a504ced0 2708
aebb4f4b 2709 formats = self._get_formats(info_dict)
dd82ffea 2710
c154302c 2711 # Backward compatibility with InfoExtractor._sort_formats
9ebac355 2712 field_preference = (formats or [{}])[0].pop('__sort_fields', None)
c154302c 2713 if field_preference:
2714 info_dict['_format_sort_fields'] = field_preference
2715
bc344cd4 2716 info_dict['_has_drm'] = any( # or None ensures --clean-infojson removes it
2717 f.get('has_drm') and f['has_drm'] != 'maybe' for f in formats) or None
88acdbc2 2718 if not self.params.get('allow_unplayable_formats'):
bc344cd4 2719 formats = [f for f in formats if not f.get('has_drm') or f['has_drm'] == 'maybe']
17ffed18 2720
2721 if formats and all(f.get('acodec') == f.get('vcodec') == 'none' for f in formats):
2722 self.report_warning(
2723 f'{"This video is DRM protected and " if info_dict["_has_drm"] else ""}'
2724 'only images are available for download. Use --list-formats to see them'.capitalize())
88acdbc2 2725
319b6059 2726 get_from_start = not info_dict.get('is_live') or bool(self.params.get('live_from_start'))
2727 if not get_from_start:
2728 info_dict['title'] += ' ' + datetime.datetime.now().strftime('%Y-%m-%d %H:%M')
2729 if info_dict.get('is_live') and formats:
adbc4ec4 2730 formats = [f for f in formats if bool(f.get('is_from_start')) == get_from_start]
319b6059 2731 if get_from_start and not formats:
a44ca5a4 2732 self.raise_no_formats(info_dict, msg=(
2733 '--live-from-start is passed, but there are no formats that can be downloaded from the start. '
2734 'If you want to download from the current time, use --no-live-from-start'))
adbc4ec4 2735
73af5cc8
S
2736 def is_wellformed(f):
2737 url = f.get('url')
a5ac0c47 2738 if not url:
73af5cc8
S
2739 self.report_warning(
2740 '"url" field is missing or empty - skipping format, '
2741 'there is an error in extractor')
a5ac0c47
S
2742 return False
2743 if isinstance(url, bytes):
2744 sanitize_string_field(f, 'url')
2745 return True
73af5cc8
S
2746
2747 # Filter out malformed formats for better extraction robustness
1ac7f461 2748 formats = list(filter(is_wellformed, formats or []))
2749
2750 if not formats:
2751 self.raise_no_formats(info_dict)
73af5cc8 2752
39f32f17 2753 for format in formats:
c9969434
S
2754 sanitize_string_field(format, 'format_id')
2755 sanitize_numeric_fields(format)
dcf77cf1 2756 format['url'] = sanitize_url(format['url'])
39f32f17 2757 if format.get('ext') is None:
2758 format['ext'] = determine_ext(format['url']).lower()
2759 if format.get('protocol') is None:
2760 format['protocol'] = determine_protocol(format)
2761 if format.get('resolution') is None:
2762 format['resolution'] = self.format_resolution(format, default=None)
2763 if format.get('dynamic_range') is None and format.get('vcodec') != 'none':
2764 format['dynamic_range'] = 'SDR'
2765 if format.get('aspect_ratio') is None:
2766 format['aspect_ratio'] = try_call(lambda: round(format['width'] / format['height'], 2))
10025b71 2767 # For fragmented formats, "tbr" is often max bitrate and not average
2768 if (('manifest-filesize-approx' in self.params['compat_opts'] or not format.get('manifest_url'))
0dff8e4d 2769 and info_dict.get('duration') and format.get('tbr')
39f32f17 2770 and not format.get('filesize') and not format.get('filesize_approx')):
2771 format['filesize_approx'] = int(info_dict['duration'] * format['tbr'] * (1024 / 8))
6c5211ce 2772 format['http_headers'] = self._calc_headers(collections.ChainMap(format, info_dict), load_cookies=True)
2773
2774 # Safeguard against old/insecure infojson when using --load-info-json
2775 if info_dict.get('http_headers'):
2776 info_dict['http_headers'] = HTTPHeaderDict(info_dict['http_headers'])
2777 info_dict['http_headers'].pop('Cookie', None)
39f32f17 2778
2779 # This is copied to http_headers by the above _calc_headers and can now be removed
2780 if '__x_forwarded_for_ip' in info_dict:
2781 del info_dict['__x_forwarded_for_ip']
2782
c154302c 2783 self.sort_formats({
2784 'formats': formats,
2785 '_format_sort_fields': info_dict.get('_format_sort_fields')
2786 })
39f32f17 2787
2788 # Sanitize and group by format_id
2789 formats_dict = {}
2790 for i, format in enumerate(formats):
e74e3b63 2791 if not format.get('format_id'):
14f25df2 2792 format['format_id'] = str(i)
e2effb08
S
2793 else:
2794 # Sanitize format_id from characters used in format selector expression
ec85ded8 2795 format['format_id'] = re.sub(r'[\s,/+\[\]()]', '_', format['format_id'])
39f32f17 2796 formats_dict.setdefault(format['format_id'], []).append(format)
181c7053
S
2797
2798 # Make sure all formats have unique format_id
03b4de72 2799 common_exts = set(itertools.chain(*self._format_selection_exts.values()))
181c7053 2800 for format_id, ambiguous_formats in formats_dict.items():
48ee10ee 2801 ambigious_id = len(ambiguous_formats) > 1
2802 for i, format in enumerate(ambiguous_formats):
2803 if ambigious_id:
181c7053 2804 format['format_id'] = '%s-%d' % (format_id, i)
48ee10ee 2805 # Ensure there is no conflict between id and ext in format selection
2806 # See https://github.com/yt-dlp/yt-dlp/issues/1282
2807 if format['format_id'] != format['ext'] and format['format_id'] in common_exts:
2808 format['format_id'] = 'f%s' % format['format_id']
181c7053 2809
39f32f17 2810 if format.get('format') is None:
2811 format['format'] = '{id} - {res}{note}'.format(
2812 id=format['format_id'],
2813 res=self.format_resolution(format),
2814 note=format_field(format, 'format_note', ' (%s)'),
2815 )
dd82ffea 2816
9f1a1c36 2817 if self.params.get('check_formats') is True:
282f5709 2818 formats = LazyList(self._check_formats(formats[::-1]), reverse=True)
9f1a1c36 2819
88acdbc2 2820 if not formats or formats[0] is not info_dict:
b3d9ef88
JMF
2821 # only set the 'formats' fields if the original info_dict list them
2822 # otherwise we end up with a circular reference, the first (and unique)
f89197d7 2823 # element in the 'formats' field in info_dict is info_dict itself,
dfb1b146 2824 # which can't be exported to json
b3d9ef88 2825 info_dict['formats'] = formats
4ec82a72 2826
2827 info_dict, _ = self.pre_process(info_dict)
2828
6db9c4d5 2829 if self._match_entry(info_dict, incomplete=self._format_fields) is not None:
09b49e1f 2830 return info_dict
2831
2832 self.post_extract(info_dict)
2833 info_dict, _ = self.pre_process(info_dict, 'after_filter')
2834
093a1710 2835 # The pre-processors may have modified the formats
aebb4f4b 2836 formats = self._get_formats(info_dict)
093a1710 2837
e4221b70 2838 list_only = self.params.get('simulate') == 'list_only'
fa9f30b8 2839 interactive_format_selection = not list_only and self.format_selector == '-'
b7b04c78 2840 if self.params.get('list_thumbnails'):
2841 self.list_thumbnails(info_dict)
b7b04c78 2842 if self.params.get('listsubtitles'):
2843 if 'automatic_captions' in info_dict:
2844 self.list_subtitles(
2845 info_dict['id'], automatic_captions, 'automatic captions')
2846 self.list_subtitles(info_dict['id'], subtitles, 'subtitles')
fa9f30b8 2847 if self.params.get('listformats') or interactive_format_selection:
b69fd25c 2848 self.list_formats(info_dict)
169dbde9 2849 if list_only:
b7b04c78 2850 # Without this printing, -F --print-json will not work
17060584 2851 self.__forced_printings(info_dict)
c487cf00 2852 return info_dict
bfaae0a7 2853
187986a8 2854 format_selector = self.format_selector
fa9f30b8 2855 while True:
2856 if interactive_format_selection:
372a0f3b
IS
2857 req_format = input(self._format_screen('\nEnter format selector ', self.Styles.EMPHASIS)
2858 + '(Press ENTER for default, or Ctrl+C to quit)'
2859 + self._format_screen(': ', self.Styles.EMPHASIS))
fa9f30b8 2860 try:
372a0f3b 2861 format_selector = self.build_format_selector(req_format) if req_format else None
fa9f30b8 2862 except SyntaxError as err:
2863 self.report_error(err, tb=False, is_error=False)
2864 continue
2865
372a0f3b
IS
2866 if format_selector is None:
2867 req_format = self._default_format_spec(info_dict, download=download)
2868 self.write_debug(f'Default format spec: {req_format}')
2869 format_selector = self.build_format_selector(req_format)
2870
85e801a9 2871 formats_to_download = list(format_selector({
fa9f30b8 2872 'formats': formats,
85e801a9 2873 'has_merged_format': any('none' not in (f.get('acodec'), f.get('vcodec')) for f in formats),
906c0bdc 2874 'incomplete_formats': (all(f.get('vcodec') == 'none' for f in formats) # No formats with video
2875 or all(f.get('acodec') == 'none' for f in formats)), # OR, No formats with audio
85e801a9 2876 }))
fa9f30b8 2877 if interactive_format_selection and not formats_to_download:
2878 self.report_error('Requested format is not available', tb=False, is_error=False)
2879 continue
2880 break
317f7ab6 2881
dd82ffea 2882 if not formats_to_download:
b7da73eb 2883 if not self.params.get('ignore_no_formats_error'):
c0b6e5c7 2884 raise ExtractorError(
2885 'Requested format is not available. Use --list-formats for a list of available formats',
2886 expected=True, video_id=info_dict['id'], ie=info_dict['extractor'])
b62fa6d7 2887 self.report_warning('Requested format is not available')
2888 # Process what we can, even without any available formats.
2889 formats_to_download = [{}]
a13e6848 2890
0500ee3d 2891 requested_ranges = tuple(self.params.get('download_ranges', lambda *_: [{}])(info_dict, self))
5ec1b6b7 2892 best_format, downloaded_formats = formats_to_download[-1], []
b62fa6d7 2893 if download:
0500ee3d 2894 if best_format and requested_ranges:
5ec1b6b7 2895 def to_screen(*msg):
2896 self.to_screen(f'[info] {info_dict["id"]}: {" ".join(", ".join(variadic(m)) for m in msg)}')
2897
2898 to_screen(f'Downloading {len(formats_to_download)} format(s):',
2899 (f['format_id'] for f in formats_to_download))
0500ee3d 2900 if requested_ranges != ({}, ):
5ec1b6b7 2901 to_screen(f'Downloading {len(requested_ranges)} time ranges:',
fc2ba496 2902 (f'{c["start_time"]:.1f}-{c["end_time"]:.1f}' for c in requested_ranges))
a13e6848 2903 max_downloads_reached = False
5ec1b6b7 2904
0500ee3d 2905 for fmt, chapter in itertools.product(formats_to_download, requested_ranges):
5ec1b6b7 2906 new_info = self._copy_infodict(info_dict)
b7da73eb 2907 new_info.update(fmt)
3975b4d2 2908 offset, duration = info_dict.get('section_start') or 0, info_dict.get('duration') or float('inf')
fc2ba496 2909 end_time = offset + min(chapter.get('end_time', duration), duration)
b4e0d758 2910 # duration may not be accurate. So allow deviations <1sec
2911 if end_time == float('inf') or end_time > offset + duration + 1:
2912 end_time = None
3975b4d2 2913 if chapter or offset:
5ec1b6b7 2914 new_info.update({
3975b4d2 2915 'section_start': offset + chapter.get('start_time', 0),
b4e0d758 2916 'section_end': end_time,
5ec1b6b7 2917 'section_title': chapter.get('title'),
2918 'section_number': chapter.get('index'),
2919 })
2920 downloaded_formats.append(new_info)
a13e6848 2921 try:
2922 self.process_info(new_info)
2923 except MaxDownloadsReached:
2924 max_downloads_reached = True
415f8d51 2925 self._raise_pending_errors(new_info)
f46e2f9d 2926 # Remove copied info
2927 for key, val in tuple(new_info.items()):
2928 if info_dict.get(key) == val:
2929 new_info.pop(key)
a13e6848 2930 if max_downloads_reached:
2931 break
ebed8b37 2932
5ec1b6b7 2933 write_archive = {f.get('__write_download_archive', False) for f in downloaded_formats}
a13e6848 2934 assert write_archive.issubset({True, False, 'ignore'})
2935 if True in write_archive and False not in write_archive:
2936 self.record_download_archive(info_dict)
be72c624 2937
5ec1b6b7 2938 info_dict['requested_downloads'] = downloaded_formats
ed5835b4 2939 info_dict = self.run_all_pps('after_video', info_dict)
a13e6848 2940 if max_downloads_reached:
2941 raise MaxDownloadsReached()
ebed8b37 2942
49a57e70 2943 # We update the info dict with the selected best quality format (backwards compatibility)
be72c624 2944 info_dict.update(best_format)
dd82ffea
JMF
2945 return info_dict
2946
98c70d6f 2947 def process_subtitles(self, video_id, normal_subtitles, automatic_captions):
a504ced0 2948 """Select the requested subtitles and their format"""
d8a58ddc 2949 available_subs, normal_sub_langs = {}, []
98c70d6f
JMF
2950 if normal_subtitles and self.params.get('writesubtitles'):
2951 available_subs.update(normal_subtitles)
d8a58ddc 2952 normal_sub_langs = tuple(normal_subtitles.keys())
98c70d6f
JMF
2953 if automatic_captions and self.params.get('writeautomaticsub'):
2954 for lang, cap_info in automatic_captions.items():
360e1ca5
JMF
2955 if lang not in available_subs:
2956 available_subs[lang] = cap_info
2957
d2c8aadf 2958 if not available_subs or (
2959 not self.params.get('writesubtitles')
2960 and not self.params.get('writeautomaticsub')):
4d171848 2961 return None
a504ced0 2962
d8a58ddc 2963 all_sub_langs = tuple(available_subs.keys())
a504ced0 2964 if self.params.get('allsubtitles', False):
c32b0aab 2965 requested_langs = all_sub_langs
2966 elif self.params.get('subtitleslangs', False):
5314b521 2967 try:
2968 requested_langs = orderedSet_from_options(
2969 self.params.get('subtitleslangs'), {'all': all_sub_langs}, use_regex=True)
2970 except re.error as e:
2971 raise ValueError(f'Wrong regex for subtitlelangs: {e.pattern}')
a504ced0 2972 else:
376aa24b
SS
2973 requested_langs = LazyList(itertools.chain(
2974 ['en'] if 'en' in normal_sub_langs else [],
2975 filter(lambda f: f.startswith('en'), normal_sub_langs),
2976 ['en'] if 'en' in all_sub_langs else [],
2977 filter(lambda f: f.startswith('en'), all_sub_langs),
2978 normal_sub_langs, all_sub_langs,
2979 ))[:1]
ad3dc496 2980 if requested_langs:
d2c8aadf 2981 self.to_screen(f'[info] {video_id}: Downloading subtitles: {", ".join(requested_langs)}')
a504ced0
JMF
2982
2983 formats_query = self.params.get('subtitlesformat', 'best')
2984 formats_preference = formats_query.split('/') if formats_query else []
2985 subs = {}
2986 for lang in requested_langs:
2987 formats = available_subs.get(lang)
2988 if formats is None:
86e5f3ed 2989 self.report_warning(f'{lang} subtitles not available for {video_id}')
a504ced0 2990 continue
a504ced0
JMF
2991 for ext in formats_preference:
2992 if ext == 'best':
2993 f = formats[-1]
2994 break
2995 matches = list(filter(lambda f: f['ext'] == ext, formats))
2996 if matches:
2997 f = matches[-1]
2998 break
2999 else:
3000 f = formats[-1]
3001 self.report_warning(
3002 'No subtitle format found matching "%s" for language %s, '
3003 'using %s' % (formats_query, lang, f['ext']))
3004 subs[lang] = f
3005 return subs
3006
bb66c247 3007 def _forceprint(self, key, info_dict):
3008 if info_dict is None:
3009 return
3010 info_copy = info_dict.copy()
17060584 3011 info_copy.setdefault('filename', self.prepare_filename(info_dict))
3012 if info_dict.get('requested_formats') is not None:
3013 # For RTMP URLs, also include the playpath
3014 info_copy['urls'] = '\n'.join(f['url'] + f.get('play_path', '') for f in info_dict['requested_formats'])
3015 elif info_dict.get('url'):
3016 info_copy['urls'] = info_dict['url'] + info_dict.get('play_path', '')
bb66c247 3017 info_copy['formats_table'] = self.render_formats_table(info_dict)
3018 info_copy['thumbnails_table'] = self.render_thumbnails_table(info_dict)
3019 info_copy['subtitles_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('subtitles'))
3020 info_copy['automatic_captions_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('automatic_captions'))
3021
3022 def format_tmpl(tmpl):
48c8424b 3023 mobj = re.fullmatch(r'([\w.:,]|-\d|(?P<dict>{([\w.:,]|-\d)+}))+=?', tmpl)
07a1250e 3024 if not mobj:
3025 return tmpl
48c8424b 3026
3027 fmt = '%({})s'
3028 if tmpl.startswith('{'):
6f2287cb 3029 tmpl, fmt = f'.{tmpl}', '%({})j'
48c8424b 3030 if tmpl.endswith('='):
3031 tmpl, fmt = tmpl[:-1], '{0} = %({0})#j'
3032 return '\n'.join(map(fmt.format, [tmpl] if mobj.group('dict') else tmpl.split(',')))
8130779d 3033
bb66c247 3034 for tmpl in self.params['forceprint'].get(key, []):
3035 self.to_stdout(self.evaluate_outtmpl(format_tmpl(tmpl), info_copy))
3036
3037 for tmpl, file_tmpl in self.params['print_to_file'].get(key, []):
5127e92a 3038 filename = self.prepare_filename(info_dict, outtmpl=file_tmpl)
bb66c247 3039 tmpl = format_tmpl(tmpl)
3040 self.to_screen(f'[info] Writing {tmpl!r} to: {filename}')
8d93e69d 3041 if self._ensure_dir_exists(filename):
9874e82b 3042 with open(filename, 'a', encoding='utf-8', newline='') as f:
3043 f.write(self.evaluate_outtmpl(tmpl, info_copy) + os.linesep)
ca30f449 3044
17060584 3045 return info_copy
3046
3047 def __forced_printings(self, info_dict, filename=None, incomplete=True):
bb66c247 3048 if (self.params.get('forcejson')
3049 or self.params['forceprint'].get('video')
3050 or self.params['print_to_file'].get('video')):
2b8a2973 3051 self.post_extract(info_dict)
17060584 3052 if filename:
3053 info_dict['filename'] = filename
b5f61b69 3054 info_copy = self._forceprint('video', info_dict)
3055
3056 def print_field(field, actual_field=None, optional=False):
3057 if actual_field is None:
3058 actual_field = field
3059 if self.params.get(f'force{field}') and (
3060 info_copy.get(field) is not None or (not optional and not incomplete)):
3061 self.to_stdout(info_copy[actual_field])
3062
3063 print_field('title')
3064 print_field('id')
3065 print_field('url', 'urls')
3066 print_field('thumbnail', optional=True)
3067 print_field('description', optional=True)
ad54c913 3068 print_field('filename')
b5f61b69 3069 if self.params.get('forceduration') and info_copy.get('duration') is not None:
3070 self.to_stdout(formatSeconds(info_copy['duration']))
3071 print_field('format')
53c18592 3072
2b8a2973 3073 if self.params.get('forcejson'):
6e84b215 3074 self.to_stdout(json.dumps(self.sanitize_info(info_dict)))
d06daf23 3075
e8e73840 3076 def dl(self, name, info, subtitle=False, test=False):
88acdbc2 3077 if not info.get('url'):
1151c407 3078 self.raise_no_formats(info, True)
e8e73840 3079
3080 if test:
3081 verbose = self.params.get('verbose')
3082 params = {
3083 'test': True,
a169858f 3084 'quiet': self.params.get('quiet') or not verbose,
e8e73840 3085 'verbose': verbose,
3086 'noprogress': not verbose,
3087 'nopart': True,
3088 'skip_unavailable_fragments': False,
3089 'keep_fragments': False,
3090 'overwrites': True,
3091 '_no_ytdl_file': True,
3092 }
3093 else:
3094 params = self.params
96fccc10 3095 fd = get_suitable_downloader(info, params, to_stdout=(name == '-'))(self, params)
e8e73840 3096 if not test:
3097 for ph in self._progress_hooks:
3098 fd.add_progress_hook(ph)
42676437
M
3099 urls = '", "'.join(
3100 (f['url'].split(',')[0] + ',<data>' if f['url'].startswith('data:') else f['url'])
3101 for f in info.get('requested_formats', []) or [info])
3a408f9d 3102 self.write_debug(f'Invoking {fd.FD_NAME} downloader on "{urls}"')
03b4de72 3103
adbc4ec4
THD
3104 # Note: Ideally info should be a deep-copied so that hooks cannot modify it.
3105 # But it may contain objects that are not deep-copyable
3106 new_info = self._copy_infodict(info)
e8e73840 3107 if new_info.get('http_headers') is None:
3108 new_info['http_headers'] = self._calc_headers(new_info)
3109 return fd.download(name, new_info, subtitle)
3110
e04938ab 3111 def existing_file(self, filepaths, *, default_overwrite=True):
3112 existing_files = list(filter(os.path.exists, orderedSet(filepaths)))
3113 if existing_files and not self.params.get('overwrites', default_overwrite):
3114 return existing_files[0]
3115
3116 for file in existing_files:
3117 self.report_file_delete(file)
3118 os.remove(file)
3119 return None
3120
8222d8de 3121 def process_info(self, info_dict):
09b49e1f 3122 """Process a single resolved IE result. (Modifies it in-place)"""
8222d8de
JMF
3123
3124 assert info_dict.get('_type', 'video') == 'video'
f46e2f9d 3125 original_infodict = info_dict
fd288278 3126
4513a41a 3127 if 'format' not in info_dict and 'ext' in info_dict:
8222d8de
JMF
3128 info_dict['format'] = info_dict['ext']
3129
c77495e3 3130 if self._match_entry(info_dict) is not None:
9e907ebd 3131 info_dict['__write_download_archive'] = 'ignore'
8222d8de
JMF
3132 return
3133
09b49e1f 3134 # Does nothing under normal operation - for backward compatibility of process_info
277d6ff5 3135 self.post_extract(info_dict)
119e40ef 3136
3137 def replace_info_dict(new_info):
3138 nonlocal info_dict
3139 if new_info == info_dict:
3140 return
3141 info_dict.clear()
3142 info_dict.update(new_info)
3143
3144 new_info, _ = self.pre_process(info_dict, 'video')
3145 replace_info_dict(new_info)
0c14d66a 3146 self._num_downloads += 1
8222d8de 3147
dcf64d43 3148 # info_dict['_filename'] needs to be set for backward compatibility
de6000d9 3149 info_dict['_filename'] = full_filename = self.prepare_filename(info_dict, warn=True)
3150 temp_filename = self.prepare_filename(info_dict, 'temp')
0202b52a 3151 files_to_move = {}
8222d8de
JMF
3152
3153 # Forced printings
4513a41a 3154 self.__forced_printings(info_dict, full_filename, incomplete=('format' not in info_dict))
8222d8de 3155
ca6d59d2 3156 def check_max_downloads():
3157 if self._num_downloads >= float(self.params.get('max_downloads') or 'inf'):
3158 raise MaxDownloadsReached()
3159
b7b04c78 3160 if self.params.get('simulate'):
9e907ebd 3161 info_dict['__write_download_archive'] = self.params.get('force_write_download_archive')
ca6d59d2 3162 check_max_downloads()
8222d8de
JMF
3163 return
3164
de6000d9 3165 if full_filename is None:
8222d8de 3166 return
e92caff5 3167 if not self._ensure_dir_exists(encodeFilename(full_filename)):
0202b52a 3168 return
e92caff5 3169 if not self._ensure_dir_exists(encodeFilename(temp_filename)):
8222d8de
JMF
3170 return
3171
80c03fa9 3172 if self._write_description('video', info_dict,
3173 self.prepare_filename(info_dict, 'description')) is None:
3174 return
3175
3176 sub_files = self._write_subtitles(info_dict, temp_filename)
3177 if sub_files is None:
3178 return
3179 files_to_move.update(dict(sub_files))
3180
3181 thumb_files = self._write_thumbnails(
3182 'video', info_dict, temp_filename, self.prepare_filename(info_dict, 'thumbnail'))
3183 if thumb_files is None:
3184 return
3185 files_to_move.update(dict(thumb_files))
8222d8de 3186
80c03fa9 3187 infofn = self.prepare_filename(info_dict, 'infojson')
3188 _infojson_written = self._write_info_json('video', info_dict, infofn)
3189 if _infojson_written:
dac5df5a 3190 info_dict['infojson_filename'] = infofn
e75bb0d6 3191 # For backward compatibility, even though it was a private field
80c03fa9 3192 info_dict['__infojson_filename'] = infofn
3193 elif _infojson_written is None:
3194 return
3195
3196 # Note: Annotations are deprecated
3197 annofn = None
1fb07d10 3198 if self.params.get('writeannotations', False):
de6000d9 3199 annofn = self.prepare_filename(info_dict, 'annotation')
80c03fa9 3200 if annofn:
e92caff5 3201 if not self._ensure_dir_exists(encodeFilename(annofn)):
0202b52a 3202 return
0c3d0f51 3203 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(annofn)):
6febd1c1 3204 self.to_screen('[info] Video annotations are already present')
ffddb112
RA
3205 elif not info_dict.get('annotations'):
3206 self.report_warning('There are no annotations to write.')
7b6fefc9
PH
3207 else:
3208 try:
6febd1c1 3209 self.to_screen('[info] Writing video annotations to: ' + annofn)
86e5f3ed 3210 with open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
7b6fefc9
PH
3211 annofile.write(info_dict['annotations'])
3212 except (KeyError, TypeError):
6febd1c1 3213 self.report_warning('There are no annotations to write.')
86e5f3ed 3214 except OSError:
6febd1c1 3215 self.report_error('Cannot write annotations file: ' + annofn)
7b6fefc9 3216 return
1fb07d10 3217
732044af 3218 # Write internet shortcut files
08438d2c 3219 def _write_link_file(link_type):
60f3e995 3220 url = try_get(info_dict['webpage_url'], iri_to_uri)
3221 if not url:
3222 self.report_warning(
3223 f'Cannot write internet shortcut file because the actual URL of "{info_dict["webpage_url"]}" is unknown')
3224 return True
08438d2c 3225 linkfn = replace_extension(self.prepare_filename(info_dict, 'link'), link_type, info_dict.get('ext'))
0e6b018a
Z
3226 if not self._ensure_dir_exists(encodeFilename(linkfn)):
3227 return False
10e3742e 3228 if self.params.get('overwrites', True) and os.path.exists(encodeFilename(linkfn)):
08438d2c 3229 self.to_screen(f'[info] Internet shortcut (.{link_type}) is already present')
3230 return True
3231 try:
3232 self.to_screen(f'[info] Writing internet shortcut (.{link_type}) to: {linkfn}')
86e5f3ed 3233 with open(encodeFilename(to_high_limit_path(linkfn)), 'w', encoding='utf-8',
3234 newline='\r\n' if link_type == 'url' else '\n') as linkfile:
60f3e995 3235 template_vars = {'url': url}
08438d2c 3236 if link_type == 'desktop':
3237 template_vars['filename'] = linkfn[:-(len(link_type) + 1)]
3238 linkfile.write(LINK_TEMPLATES[link_type] % template_vars)
86e5f3ed 3239 except OSError:
08438d2c 3240 self.report_error(f'Cannot write internet shortcut {linkfn}')
3241 return False
732044af 3242 return True
3243
08438d2c 3244 write_links = {
3245 'url': self.params.get('writeurllink'),
3246 'webloc': self.params.get('writewebloclink'),
3247 'desktop': self.params.get('writedesktoplink'),
3248 }
3249 if self.params.get('writelink'):
3250 link_type = ('webloc' if sys.platform == 'darwin'
3251 else 'desktop' if sys.platform.startswith('linux')
3252 else 'url')
3253 write_links[link_type] = True
3254
3255 if any(should_write and not _write_link_file(link_type)
3256 for link_type, should_write in write_links.items()):
3257 return
732044af 3258
415f8d51 3259 new_info, files_to_move = self.pre_process(info_dict, 'before_dl', files_to_move)
3260 replace_info_dict(new_info)
56d868db 3261
a13e6848 3262 if self.params.get('skip_download'):
56d868db 3263 info_dict['filepath'] = temp_filename
3264 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
3265 info_dict['__files_to_move'] = files_to_move
f46e2f9d 3266 replace_info_dict(self.run_pp(MoveFilesAfterDownloadPP(self, False), info_dict))
9e907ebd 3267 info_dict['__write_download_archive'] = self.params.get('force_write_download_archive')
56d868db 3268 else:
3269 # Download
b868936c 3270 info_dict.setdefault('__postprocessors', [])
4340deca 3271 try:
0202b52a 3272
e04938ab 3273 def existing_video_file(*filepaths):
6b591b29 3274 ext = info_dict.get('ext')
e04938ab 3275 converted = lambda file: replace_extension(file, self.params.get('final_ext') or ext, ext)
3276 file = self.existing_file(itertools.chain(*zip(map(converted, filepaths), filepaths)),
3277 default_overwrite=False)
3278 if file:
3279 info_dict['ext'] = os.path.splitext(file)[1][1:]
3280 return file
0202b52a 3281
7b2c3f47 3282 fd, success = None, True
fccf90e7 3283 if info_dict.get('protocol') or info_dict.get('url'):
56ba69e4 3284 fd = get_suitable_downloader(info_dict, self.params, to_stdout=temp_filename == '-')
62b5c94c 3285 if fd != FFmpegFD and 'no-direct-merge' not in self.params['compat_opts'] and (
56ba69e4 3286 info_dict.get('section_start') or info_dict.get('section_end')):
7b2c3f47 3287 msg = ('This format cannot be partially downloaded' if FFmpegFD.available()
56ba69e4 3288 else 'You have requested downloading the video partially, but ffmpeg is not installed')
3289 self.report_error(f'{msg}. Aborting')
5ec1b6b7 3290 return
5ec1b6b7 3291
4340deca 3292 if info_dict.get('requested_formats') is not None:
0202b52a 3293 old_ext = info_dict['ext']
4e3b637d 3294 if self.params.get('merge_output_format') is None:
4e3b637d 3295 if (info_dict['ext'] == 'webm'
3296 and info_dict.get('thumbnails')
3297 # check with type instead of pp_key, __name__, or isinstance
3298 # since we dont want any custom PPs to trigger this
c487cf00 3299 and any(type(pp) == EmbedThumbnailPP for pp in self._pps['post_process'])): # noqa: E721
4e3b637d 3300 info_dict['ext'] = 'mkv'
3301 self.report_warning(
3302 'webm doesn\'t support embedding a thumbnail, mkv will be used')
124bc071 3303 new_ext = info_dict['ext']
0202b52a 3304
124bc071 3305 def correct_ext(filename, ext=new_ext):
96fccc10 3306 if filename == '-':
3307 return filename
0202b52a 3308 filename_real_ext = os.path.splitext(filename)[1][1:]
3309 filename_wo_ext = (
3310 os.path.splitext(filename)[0]
124bc071 3311 if filename_real_ext in (old_ext, new_ext)
0202b52a 3312 else filename)
86e5f3ed 3313 return f'{filename_wo_ext}.{ext}'
0202b52a 3314
38c6902b 3315 # Ensure filename always has a correct extension for successful merge
0202b52a 3316 full_filename = correct_ext(full_filename)
3317 temp_filename = correct_ext(temp_filename)
e04938ab 3318 dl_filename = existing_video_file(full_filename, temp_filename)
ad54c913 3319
1ea24129 3320 info_dict['__real_download'] = False
84078a8b 3321 # NOTE: Copy so that original format dicts are not modified
3322 info_dict['requested_formats'] = list(map(dict, info_dict['requested_formats']))
18e674b4 3323
7b2c3f47 3324 merger = FFmpegMergerPP(self)
adbc4ec4 3325 downloaded = []
dbf5416a 3326 if dl_filename is not None:
6c7274ec 3327 self.report_file_already_downloaded(dl_filename)
adbc4ec4 3328 elif fd:
ad54c913 3329 for f in info_dict['requested_formats'] if fd != FFmpegFD else []:
adbc4ec4
THD
3330 f['filepath'] = fname = prepend_extension(
3331 correct_ext(temp_filename, info_dict['ext']),
3332 'f%s' % f['format_id'], info_dict['ext'])
3333 downloaded.append(fname)
ad54c913 3334 info_dict['url'] = '\n'.join(f['url'] for f in info_dict['requested_formats'])
dbf5416a 3335 success, real_download = self.dl(temp_filename, info_dict)
3336 info_dict['__real_download'] = real_download
18e674b4 3337 else:
18e674b4 3338 if self.params.get('allow_unplayable_formats'):
3339 self.report_warning(
3340 'You have requested merging of multiple formats '
3341 'while also allowing unplayable formats to be downloaded. '
3342 'The formats won\'t be merged to prevent data corruption.')
3343 elif not merger.available:
e8969bda 3344 msg = 'You have requested merging of multiple formats but ffmpeg is not installed'
3345 if not self.params.get('ignoreerrors'):
3346 self.report_error(f'{msg}. Aborting due to --abort-on-error')
3347 return
3348 self.report_warning(f'{msg}. The formats won\'t be merged')
18e674b4 3349
96fccc10 3350 if temp_filename == '-':
adbc4ec4 3351 reason = ('using a downloader other than ffmpeg' if FFmpegFD.can_merge_formats(info_dict, self.params)
96fccc10 3352 else 'but the formats are incompatible for simultaneous download' if merger.available
3353 else 'but ffmpeg is not installed')
3354 self.report_warning(
3355 f'You have requested downloading multiple formats to stdout {reason}. '
3356 'The formats will be streamed one after the other')
3357 fname = temp_filename
ad54c913 3358 for f in info_dict['requested_formats']:
dbf5416a 3359 new_info = dict(info_dict)
3360 del new_info['requested_formats']
3361 new_info.update(f)
96fccc10 3362 if temp_filename != '-':
124bc071 3363 fname = prepend_extension(
3364 correct_ext(temp_filename, new_info['ext']),
3365 'f%s' % f['format_id'], new_info['ext'])
96fccc10 3366 if not self._ensure_dir_exists(fname):
3367 return
a21e0ab1 3368 f['filepath'] = fname
96fccc10 3369 downloaded.append(fname)
dbf5416a 3370 partial_success, real_download = self.dl(fname, new_info)
3371 info_dict['__real_download'] = info_dict['__real_download'] or real_download
3372 success = success and partial_success
adbc4ec4
THD
3373
3374 if downloaded and merger.available and not self.params.get('allow_unplayable_formats'):
3375 info_dict['__postprocessors'].append(merger)
3376 info_dict['__files_to_merge'] = downloaded
3377 # Even if there were no downloads, it is being merged only now
3378 info_dict['__real_download'] = True
3379 else:
3380 for file in downloaded:
3381 files_to_move[file] = None
4340deca
P
3382 else:
3383 # Just a single file
e04938ab 3384 dl_filename = existing_video_file(full_filename, temp_filename)
6c7274ec 3385 if dl_filename is None or dl_filename == temp_filename:
3386 # dl_filename == temp_filename could mean that the file was partially downloaded with --no-part.
3387 # So we should try to resume the download
e8e73840 3388 success, real_download = self.dl(temp_filename, info_dict)
0202b52a 3389 info_dict['__real_download'] = real_download
6c7274ec 3390 else:
3391 self.report_file_already_downloaded(dl_filename)
0202b52a 3392
0202b52a 3393 dl_filename = dl_filename or temp_filename
c571435f 3394 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
0202b52a 3395
3158150c 3396 except network_exceptions as err:
7960b056 3397 self.report_error('unable to download video data: %s' % error_to_compat_str(err))
4340deca 3398 return
86e5f3ed 3399 except OSError as err:
4340deca
P
3400 raise UnavailableVideoError(err)
3401 except (ContentTooShortError, ) as err:
86e5f3ed 3402 self.report_error(f'content too short (expected {err.expected} bytes and served {err.downloaded})')
4340deca 3403 return
8222d8de 3404
415f8d51 3405 self._raise_pending_errors(info_dict)
de6000d9 3406 if success and full_filename != '-':
f17f8651 3407
fd7cfb64 3408 def fixup():
3409 do_fixup = True
3410 fixup_policy = self.params.get('fixup')
3411 vid = info_dict['id']
3412
3413 if fixup_policy in ('ignore', 'never'):
3414 return
3415 elif fixup_policy == 'warn':
3fe75fdc 3416 do_fixup = 'warn'
f89b3e2d 3417 elif fixup_policy != 'force':
3418 assert fixup_policy in ('detect_or_warn', None)
3419 if not info_dict.get('__real_download'):
3420 do_fixup = False
fd7cfb64 3421
3422 def ffmpeg_fixup(cndn, msg, cls):
3fe75fdc 3423 if not (do_fixup and cndn):
fd7cfb64 3424 return
3fe75fdc 3425 elif do_fixup == 'warn':
fd7cfb64 3426 self.report_warning(f'{vid}: {msg}')
3427 return
3428 pp = cls(self)
3429 if pp.available:
3430 info_dict['__postprocessors'].append(pp)
3431 else:
3432 self.report_warning(f'{vid}: {msg}. Install ffmpeg to fix this automatically')
3433
3434 stretched_ratio = info_dict.get('stretched_ratio')
ca9def71
LNO
3435 ffmpeg_fixup(stretched_ratio not in (1, None),
3436 f'Non-uniform pixel ratio {stretched_ratio}',
3437 FFmpegFixupStretchedPP)
fd7cfb64 3438
993191c0 3439 downloader = get_suitable_downloader(info_dict, self.params) if 'protocol' in info_dict else None
24146491 3440 downloader = downloader.FD_NAME if downloader else None
adbc4ec4 3441
ca9def71
LNO
3442 ext = info_dict.get('ext')
3443 postprocessed_by_ffmpeg = info_dict.get('requested_formats') or any((
3444 isinstance(pp, FFmpegVideoConvertorPP)
3445 and resolve_recode_mapping(ext, pp.mapping)[0] not in (ext, None)
f73c1180 3446 ) for pp in self._pps['post_process'])
ca9def71
LNO
3447
3448 if not postprocessed_by_ffmpeg:
f73c1180 3449 ffmpeg_fixup(fd != FFmpegFD and ext == 'm4a'
3450 and info_dict.get('container') == 'm4a_dash',
f2df4071 3451 'writing DASH m4a. Only some players support this container',
3452 FFmpegFixupM4aPP)
24146491 3453 ffmpeg_fixup(downloader == 'hlsnative' and not self.params.get('hls_use_mpegts')
494f5230 3454 or info_dict.get('is_live') and self.params.get('hls_use_mpegts') is None,
adbc4ec4
THD
3455 'Possible MPEG-TS in MP4 container or malformed AAC timestamps',
3456 FFmpegFixupM3u8PP)
26010b5c 3457 ffmpeg_fixup(info_dict.get('is_live') and downloader == 'dashsegments',
adbc4ec4
THD
3458 'Possible duplicate MOOV atoms', FFmpegFixupDuplicateMoovPP)
3459
24146491 3460 ffmpeg_fixup(downloader == 'web_socket_fragment', 'Malformed timestamps detected', FFmpegFixupTimestampPP)
3461 ffmpeg_fixup(downloader == 'web_socket_fragment', 'Malformed duration detected', FFmpegFixupDurationPP)
fd7cfb64 3462
3463 fixup()
8222d8de 3464 try:
f46e2f9d 3465 replace_info_dict(self.post_process(dl_filename, info_dict, files_to_move))
af819c21 3466 except PostProcessingError as err:
3467 self.report_error('Postprocessing: %s' % str(err))
8222d8de 3468 return
ab8e5e51
AM
3469 try:
3470 for ph in self._post_hooks:
23c1a667 3471 ph(info_dict['filepath'])
ab8e5e51
AM
3472 except Exception as err:
3473 self.report_error('post hooks: %s' % str(err))
3474 return
9e907ebd 3475 info_dict['__write_download_archive'] = True
2d30509f 3476
c487cf00 3477 assert info_dict is original_infodict # Make sure the info_dict was modified in-place
a13e6848 3478 if self.params.get('force_write_download_archive'):
9e907ebd 3479 info_dict['__write_download_archive'] = True
ca6d59d2 3480 check_max_downloads()
8222d8de 3481
aa9369a2 3482 def __download_wrapper(self, func):
3483 @functools.wraps(func)
3484 def wrapper(*args, **kwargs):
3485 try:
3486 res = func(*args, **kwargs)
3487 except UnavailableVideoError as e:
3488 self.report_error(e)
b222c271 3489 except DownloadCancelled as e:
3490 self.to_screen(f'[info] {e}')
3491 if not self.params.get('break_per_url'):
3492 raise
fd404bec 3493 self._num_downloads = 0
aa9369a2 3494 else:
3495 if self.params.get('dump_single_json', False):
3496 self.post_extract(res)
3497 self.to_stdout(json.dumps(self.sanitize_info(res)))
3498 return wrapper
3499
8222d8de
JMF
3500 def download(self, url_list):
3501 """Download a given list of URLs."""
aa9369a2 3502 url_list = variadic(url_list) # Passing a single URL is a common mistake
bf1824b3 3503 outtmpl = self.params['outtmpl']['default']
3089bc74
S
3504 if (len(url_list) > 1
3505 and outtmpl != '-'
3506 and '%' not in outtmpl
3507 and self.params.get('max_downloads') != 1):
acd69589 3508 raise SameFileError(outtmpl)
8222d8de
JMF
3509
3510 for url in url_list:
aa9369a2 3511 self.__download_wrapper(self.extract_info)(
3512 url, force_generic_extractor=self.params.get('force_generic_extractor', False))
8222d8de
JMF
3513
3514 return self._download_retcode
3515
1dcc4c0c 3516 def download_with_info_file(self, info_filename):
31bd3925
JMF
3517 with contextlib.closing(fileinput.FileInput(
3518 [info_filename], mode='r',
3519 openhook=fileinput.hook_encoded('utf-8'))) as f:
3520 # FileInput doesn't have a read method, we can't call json.load
ab1de9cb 3521 infos = [self.sanitize_info(info, self.params.get('clean_infojson', True))
3522 for info in variadic(json.loads('\n'.join(f)))]
3523 for info in infos:
3524 try:
3525 self.__download_wrapper(self.process_ie_result)(info, download=True)
3526 except (DownloadError, EntryNotInPlaylist, ReExtractInfo) as e:
3527 if not isinstance(e, EntryNotInPlaylist):
3528 self.to_stderr('\r')
3529 webpage_url = info.get('webpage_url')
3530 if webpage_url is None:
3531 raise
aa9369a2 3532 self.report_warning(f'The info failed to download: {e}; trying with URL {webpage_url}')
ab1de9cb 3533 self.download([webpage_url])
d4943898 3534 return self._download_retcode
1dcc4c0c 3535
cb202fd2 3536 @staticmethod
8012d892 3537 def sanitize_info(info_dict, remove_private_keys=False):
3538 ''' Sanitize the infodict for converting to json '''
3ad56b42 3539 if info_dict is None:
3540 return info_dict
6e84b215 3541 info_dict.setdefault('epoch', int(time.time()))
6a5a30f9 3542 info_dict.setdefault('_type', 'video')
b5e7a2e6 3543 info_dict.setdefault('_version', {
3544 'version': __version__,
3545 'current_git_head': current_git_head(),
3546 'release_git_head': RELEASE_GIT_HEAD,
20314dd4 3547 'repository': ORIGIN,
b5e7a2e6 3548 })
09b49e1f 3549
8012d892 3550 if remove_private_keys:
0a5a191a 3551 reject = lambda k, v: v is None or k.startswith('__') or k in {
f46e2f9d 3552 'requested_downloads', 'requested_formats', 'requested_subtitles', 'requested_entries',
6f2287cb 3553 'entries', 'filepath', '_filename', 'filename', 'infojson_filename', 'original_url',
595ea4a9 3554 'playlist_autonumber',
6e84b215 3555 }
ae8f99e6 3556 else:
09b49e1f 3557 reject = lambda k, v: False
adbc4ec4
THD
3558
3559 def filter_fn(obj):
3560 if isinstance(obj, dict):
3561 return {k: filter_fn(v) for k, v in obj.items() if not reject(k, v)}
3562 elif isinstance(obj, (list, tuple, set, LazyList)):
3563 return list(map(filter_fn, obj))
3564 elif obj is None or isinstance(obj, (str, int, float, bool)):
3565 return obj
3566 else:
3567 return repr(obj)
3568
5226731e 3569 return filter_fn(info_dict)
cb202fd2 3570
8012d892 3571 @staticmethod
3572 def filter_requested_info(info_dict, actually_filter=True):
3573 ''' Alias of sanitize_info for backward compatibility '''
3574 return YoutubeDL.sanitize_info(info_dict, actually_filter)
3575
43d7f5a5 3576 def _delete_downloaded_files(self, *files_to_delete, info={}, msg=None):
3577 for filename in set(filter(None, files_to_delete)):
3578 if msg:
3579 self.to_screen(msg % filename)
3580 try:
3581 os.remove(filename)
3582 except OSError:
3583 self.report_warning(f'Unable to delete file {filename}')
3584 if filename in info.get('__files_to_move', []): # NB: Delete even if None
3585 del info['__files_to_move'][filename]
3586
ed5835b4 3587 @staticmethod
3588 def post_extract(info_dict):
3589 def actual_post_extract(info_dict):
3590 if info_dict.get('_type') in ('playlist', 'multi_video'):
3591 for video_dict in info_dict.get('entries', {}):
3592 actual_post_extract(video_dict or {})
3593 return
3594
09b49e1f 3595 post_extractor = info_dict.pop('__post_extractor', None) or (lambda: {})
3596 info_dict.update(post_extractor())
ed5835b4 3597
3598 actual_post_extract(info_dict or {})
3599
dcf64d43 3600 def run_pp(self, pp, infodict):
5bfa4862 3601 files_to_delete = []
dcf64d43 3602 if '__files_to_move' not in infodict:
3603 infodict['__files_to_move'] = {}
b1940459 3604 try:
3605 files_to_delete, infodict = pp.run(infodict)
3606 except PostProcessingError as e:
3607 # Must be True and not 'only_download'
3608 if self.params.get('ignoreerrors') is True:
3609 self.report_error(e)
3610 return infodict
3611 raise
3612
5bfa4862 3613 if not files_to_delete:
dcf64d43 3614 return infodict
5bfa4862 3615 if self.params.get('keepvideo', False):
3616 for f in files_to_delete:
dcf64d43 3617 infodict['__files_to_move'].setdefault(f, '')
5bfa4862 3618 else:
43d7f5a5 3619 self._delete_downloaded_files(
3620 *files_to_delete, info=infodict, msg='Deleting original file %s (pass -k to keep)')
dcf64d43 3621 return infodict
5bfa4862 3622
6f2287cb 3623 def run_all_pps(self, key, info, *, additional_pps=None):
17ba4343 3624 if key != 'video':
3625 self._forceprint(key, info)
3626 for pp in (additional_pps or []) + self._pps[key]:
3627 info = self.run_pp(pp, info)
ed5835b4 3628 return info
277d6ff5 3629
56d868db 3630 def pre_process(self, ie_info, key='pre_process', files_to_move=None):
5bfa4862 3631 info = dict(ie_info)
56d868db 3632 info['__files_to_move'] = files_to_move or {}
415f8d51 3633 try:
3634 info = self.run_all_pps(key, info)
3635 except PostProcessingError as err:
3636 msg = f'Preprocessing: {err}'
3637 info.setdefault('__pending_error', msg)
3638 self.report_error(msg, is_error=False)
56d868db 3639 return info, info.pop('__files_to_move', None)
5bfa4862 3640
f46e2f9d 3641 def post_process(self, filename, info, files_to_move=None):
8222d8de 3642 """Run all the postprocessors on the given file."""
8222d8de 3643 info['filepath'] = filename
dcf64d43 3644 info['__files_to_move'] = files_to_move or {}
ed5835b4 3645 info = self.run_all_pps('post_process', info, additional_pps=info.get('__postprocessors'))
dcf64d43 3646 info = self.run_pp(MoveFilesAfterDownloadPP(self), info)
3647 del info['__files_to_move']
ed5835b4 3648 return self.run_all_pps('after_move', info)
c1c9a79c 3649
5db07df6 3650 def _make_archive_id(self, info_dict):
e9fef7ee
S
3651 video_id = info_dict.get('id')
3652 if not video_id:
3653 return
5db07df6
PH
3654 # Future-proof against any change in case
3655 # and backwards compatibility with prior versions
e9fef7ee 3656 extractor = info_dict.get('extractor_key') or info_dict.get('ie_key') # key in a playlist
7012b23c 3657 if extractor is None:
1211bb6d
S
3658 url = str_or_none(info_dict.get('url'))
3659 if not url:
3660 return
e9fef7ee 3661 # Try to find matching extractor for the URL and take its ie_key
8b7491c8 3662 for ie_key, ie in self._ies.items():
1211bb6d 3663 if ie.suitable(url):
8b7491c8 3664 extractor = ie_key
e9fef7ee
S
3665 break
3666 else:
3667 return
0647d925 3668 return make_archive_id(extractor, video_id)
5db07df6
PH
3669
3670 def in_download_archive(self, info_dict):
ae103564 3671 if not self.archive:
5db07df6
PH
3672 return False
3673
1e8fe57e 3674 vid_ids = [self._make_archive_id(info_dict)]
c200096c 3675 vid_ids.extend(info_dict.get('_old_archive_ids') or [])
1e8fe57e 3676 return any(id_ in self.archive for id_ in vid_ids)
c1c9a79c
PH
3677
3678 def record_download_archive(self, info_dict):
3679 fn = self.params.get('download_archive')
3680 if fn is None:
3681 return
5db07df6
PH
3682 vid_id = self._make_archive_id(info_dict)
3683 assert vid_id
ae103564 3684
a13e6848 3685 self.write_debug(f'Adding to archive: {vid_id}')
9c935fbc 3686 if is_path_like(fn):
ae103564 3687 with locked_file(fn, 'a', encoding='utf-8') as archive_file:
3688 archive_file.write(vid_id + '\n')
a45e8619 3689 self.archive.add(vid_id)
dd82ffea 3690
8c51aa65 3691 @staticmethod
8abeeb94 3692 def format_resolution(format, default='unknown'):
9359f3d4 3693 if format.get('vcodec') == 'none' and format.get('acodec') != 'none':
fb04e403 3694 return 'audio only'
f49d89ee
PH
3695 if format.get('resolution') is not None:
3696 return format['resolution']
35615307 3697 if format.get('width') and format.get('height'):
ff51ed58 3698 return '%dx%d' % (format['width'], format['height'])
35615307 3699 elif format.get('height'):
ff51ed58 3700 return '%sp' % format['height']
35615307 3701 elif format.get('width'):
ff51ed58 3702 return '%dx?' % format['width']
3703 return default
8c51aa65 3704
8130779d 3705 def _list_format_headers(self, *headers):
3706 if self.params.get('listformats_table', True) is not False:
591bb9d3 3707 return [self._format_out(header, self.Styles.HEADERS) for header in headers]
8130779d 3708 return headers
3709
c57f7757
PH
3710 def _format_note(self, fdict):
3711 res = ''
3712 if fdict.get('ext') in ['f4f', 'f4m']:
f304da8a 3713 res += '(unsupported)'
32f90364
PH
3714 if fdict.get('language'):
3715 if res:
3716 res += ' '
f304da8a 3717 res += '[%s]' % fdict['language']
c57f7757 3718 if fdict.get('format_note') is not None:
f304da8a 3719 if res:
3720 res += ' '
3721 res += fdict['format_note']
c57f7757 3722 if fdict.get('tbr') is not None:
f304da8a 3723 if res:
3724 res += ', '
3725 res += '%4dk' % fdict['tbr']
c57f7757
PH
3726 if fdict.get('container') is not None:
3727 if res:
3728 res += ', '
3729 res += '%s container' % fdict['container']
3089bc74
S
3730 if (fdict.get('vcodec') is not None
3731 and fdict.get('vcodec') != 'none'):
c57f7757
PH
3732 if res:
3733 res += ', '
3734 res += fdict['vcodec']
91c7271a 3735 if fdict.get('vbr') is not None:
c57f7757
PH
3736 res += '@'
3737 elif fdict.get('vbr') is not None and fdict.get('abr') is not None:
3738 res += 'video@'
3739 if fdict.get('vbr') is not None:
3740 res += '%4dk' % fdict['vbr']
fbb21cf5 3741 if fdict.get('fps') is not None:
5d583bdf
S
3742 if res:
3743 res += ', '
3744 res += '%sfps' % fdict['fps']
c57f7757
PH
3745 if fdict.get('acodec') is not None:
3746 if res:
3747 res += ', '
3748 if fdict['acodec'] == 'none':
3749 res += 'video only'
3750 else:
3751 res += '%-5s' % fdict['acodec']
3752 elif fdict.get('abr') is not None:
3753 if res:
3754 res += ', '
3755 res += 'audio'
3756 if fdict.get('abr') is not None:
3757 res += '@%3dk' % fdict['abr']
3758 if fdict.get('asr') is not None:
3759 res += ' (%5dHz)' % fdict['asr']
3760 if fdict.get('filesize') is not None:
3761 if res:
3762 res += ', '
3763 res += format_bytes(fdict['filesize'])
9732d77e
PH
3764 elif fdict.get('filesize_approx') is not None:
3765 if res:
3766 res += ', '
3767 res += '~' + format_bytes(fdict['filesize_approx'])
c57f7757 3768 return res
91c7271a 3769
aebb4f4b 3770 def _get_formats(self, info_dict):
3771 if info_dict.get('formats') is None:
3772 if info_dict.get('url') and info_dict.get('_type', 'video') == 'video':
3773 return [info_dict]
3774 return []
3775 return info_dict['formats']
b69fd25c 3776
aebb4f4b 3777 def render_formats_table(self, info_dict):
3778 formats = self._get_formats(info_dict)
3779 if not formats:
3780 return
8130779d 3781 if not self.params.get('listformats_table', True) is not False:
76d321f6 3782 table = [
3783 [
3784 format_field(f, 'format_id'),
3785 format_field(f, 'ext'),
3786 self.format_resolution(f),
8130779d 3787 self._format_note(f)
d5d1df8a 3788 ] for f in formats if (f.get('preference') or 0) >= -1000]
8130779d 3789 return render_table(['format code', 'extension', 'resolution', 'note'], table, extra_gap=1)
3790
d816f61f 3791 def simplified_codec(f, field):
3792 assert field in ('acodec', 'vcodec')
337734d4 3793 codec = f.get(field)
f5ea4748 3794 if not codec:
3795 return 'unknown'
3796 elif codec != 'none':
d816f61f 3797 return '.'.join(codec.split('.')[:4])
3798
3799 if field == 'vcodec' and f.get('acodec') == 'none':
3800 return 'images'
3801 elif field == 'acodec' and f.get('vcodec') == 'none':
3802 return ''
3803 return self._format_out('audio only' if field == 'vcodec' else 'video only',
3804 self.Styles.SUPPRESS)
3805
591bb9d3 3806 delim = self._format_out('\u2502', self.Styles.DELIM, '|', test_encoding=True)
8130779d 3807 table = [
3808 [
591bb9d3 3809 self._format_out(format_field(f, 'format_id'), self.Styles.ID),
8130779d 3810 format_field(f, 'ext'),
3811 format_field(f, func=self.format_resolution, ignore=('audio only', 'images')),
563e0bf8 3812 format_field(f, 'fps', '\t%d', func=round),
8130779d 3813 format_field(f, 'dynamic_range', '%s', ignore=(None, 'SDR')).replace('HDR', ''),
b8ed0f15 3814 format_field(f, 'audio_channels', '\t%s'),
0dff8e4d 3815 delim, (
3816 format_field(f, 'filesize', ' \t%s', func=format_bytes)
3817 or format_field(f, 'filesize_approx', '≈\t%s', func=format_bytes)
3818 or format_field(try_call(lambda: format_bytes(int(info_dict['duration'] * f['tbr'] * (1024 / 8)))),
3819 None, self._format_out('~\t%s', self.Styles.SUPPRESS))),
563e0bf8 3820 format_field(f, 'tbr', '\t%dk', func=round),
8130779d 3821 shorten_protocol_name(f.get('protocol', '')),
3822 delim,
d816f61f 3823 simplified_codec(f, 'vcodec'),
563e0bf8 3824 format_field(f, 'vbr', '\t%dk', func=round),
d816f61f 3825 simplified_codec(f, 'acodec'),
563e0bf8 3826 format_field(f, 'abr', '\t%dk', func=round),
ae61d108 3827 format_field(f, 'asr', '\t%s', func=format_decimal_suffix),
bc344cd4 3828 join_nonempty(format_field(f, 'language', '[%s]'), join_nonempty(
3829 self._format_out('UNSUPPORTED', self.Styles.BAD_FORMAT) if f.get('ext') in ('f4f', 'f4m') else None,
3830 (self._format_out('Maybe DRM', self.Styles.WARNING) if f.get('has_drm') == 'maybe'
3831 else self._format_out('DRM', self.Styles.BAD_FORMAT) if f.get('has_drm') else None),
3832 format_field(f, 'format_note'),
3833 format_field(f, 'container', ignore=(None, f.get('ext'))),
3834 delim=', '), delim=' '),
8130779d 3835 ] for f in formats if f.get('preference') is None or f['preference'] >= -1000]
3836 header_line = self._list_format_headers(
b8ed0f15 3837 'ID', 'EXT', 'RESOLUTION', '\tFPS', 'HDR', 'CH', delim, '\tFILESIZE', '\tTBR', 'PROTO',
8130779d 3838 delim, 'VCODEC', '\tVBR', 'ACODEC', '\tABR', '\tASR', 'MORE INFO')
3839
3840 return render_table(
3841 header_line, table, hide_empty=True,
591bb9d3 3842 delim=self._format_out('\u2500', self.Styles.DELIM, '-', test_encoding=True))
8130779d 3843
3844 def render_thumbnails_table(self, info_dict):
88f23a18 3845 thumbnails = list(info_dict.get('thumbnails') or [])
cfb56d1a 3846 if not thumbnails:
8130779d 3847 return None
3848 return render_table(
ec11a9f4 3849 self._list_format_headers('ID', 'Width', 'Height', 'URL'),
177662e0 3850 [[t.get('id'), t.get('width') or 'unknown', t.get('height') or 'unknown', t['url']] for t in thumbnails])
2412044c 3851
8130779d 3852 def render_subtitles_table(self, video_id, subtitles):
2412044c 3853 def _row(lang, formats):
49c258e1 3854 exts, names = zip(*((f['ext'], f.get('name') or 'unknown') for f in reversed(formats)))
2412044c 3855 if len(set(names)) == 1:
7aee40c1 3856 names = [] if names[0] == 'unknown' else names[:1]
2412044c 3857 return [lang, ', '.join(names), ', '.join(exts)]
3858
8130779d 3859 if not subtitles:
3860 return None
3861 return render_table(
ec11a9f4 3862 self._list_format_headers('Language', 'Name', 'Formats'),
2412044c 3863 [_row(lang, formats) for lang, formats in subtitles.items()],
8130779d 3864 hide_empty=True)
3865
3866 def __list_table(self, video_id, name, func, *args):
3867 table = func(*args)
3868 if not table:
3869 self.to_screen(f'{video_id} has no {name}')
3870 return
3871 self.to_screen(f'[info] Available {name} for {video_id}:')
3872 self.to_stdout(table)
3873
3874 def list_formats(self, info_dict):
3875 self.__list_table(info_dict['id'], 'formats', self.render_formats_table, info_dict)
3876
3877 def list_thumbnails(self, info_dict):
3878 self.__list_table(info_dict['id'], 'thumbnails', self.render_thumbnails_table, info_dict)
3879
3880 def list_subtitles(self, video_id, subtitles, name='subtitles'):
3881 self.__list_table(video_id, name, self.render_subtitles_table, video_id, subtitles)
a504ced0 3882
dca08720
PH
3883 def print_debug_header(self):
3884 if not self.params.get('verbose'):
3885 return
49a57e70 3886
a057779d 3887 from . import _IN_CLI # Must be delayed import
3888
560738f3 3889 # These imports can be slow. So import them only as needed
3890 from .extractor.extractors import _LAZY_LOADER
e756f45b
M
3891 from .extractor.extractors import (
3892 _PLUGIN_CLASSES as plugin_ies,
3893 _PLUGIN_OVERRIDES as plugin_ie_overrides
3894 )
560738f3 3895
49a57e70 3896 def get_encoding(stream):
2a938746 3897 ret = str(getattr(stream, 'encoding', 'missing (%s)' % type(stream).__name__))
8417f26b
SS
3898 additional_info = []
3899 if os.environ.get('TERM', '').lower() == 'dumb':
3900 additional_info.append('dumb')
49a57e70 3901 if not supports_terminal_sequences(stream):
53973b4d 3902 from .utils import WINDOWS_VT_MODE # Must be imported locally
8417f26b
SS
3903 additional_info.append('No VT' if WINDOWS_VT_MODE is False else 'No ANSI')
3904 if additional_info:
3905 ret = f'{ret} ({",".join(additional_info)})'
49a57e70 3906 return ret
3907
591bb9d3 3908 encoding_str = 'Encodings: locale %s, fs %s, pref %s, %s' % (
49a57e70 3909 locale.getpreferredencoding(),
3910 sys.getfilesystemencoding(),
591bb9d3 3911 self.get_encoding(),
3912 ', '.join(
64fa820c 3913 f'{key} {get_encoding(stream)}' for key, stream in self._out_files.items_
591bb9d3 3914 if stream is not None and key != 'console')
3915 )
883d4b1e 3916
3917 logger = self.params.get('logger')
3918 if logger:
3919 write_debug = lambda msg: logger.debug(f'[debug] {msg}')
3920 write_debug(encoding_str)
3921 else:
96565c7e 3922 write_string(f'[debug] {encoding_str}\n', encoding=None)
49a57e70 3923 write_debug = lambda msg: self._write_string(f'[debug] {msg}\n')
734f90bb 3924
4c88ff87 3925 source = detect_variant()
70b23409 3926 if VARIANT not in (None, 'pip'):
3927 source += '*'
a5387729 3928 klass = type(self)
36eaf303 3929 write_debug(join_nonempty(
20314dd4 3930 f'{REPOSITORY.rpartition("/")[2]} version',
3931 f'{CHANNEL.rpartition("@")[2]}@{__version__}',
3932 not ORIGIN.startswith('yt-dlp/') and f'from {ORIGIN}',
29cb20bd 3933 f'[{RELEASE_GIT_HEAD[:9]}]' if RELEASE_GIT_HEAD else '',
36eaf303 3934 '' if source == 'unknown' else f'({source})',
a5387729 3935 '' if _IN_CLI else 'API' if klass == YoutubeDL else f'API:{self.__module__}.{klass.__qualname__}',
36eaf303 3936 delim=' '))
497074f0 3937
3938 if not _IN_CLI:
3939 write_debug(f'params: {self.params}')
3940
6e21fdd2 3941 if not _LAZY_LOADER:
3942 if os.environ.get('YTDLP_NO_LAZY_EXTRACTORS'):
49a57e70 3943 write_debug('Lazy loading extractors is forcibly disabled')
6e21fdd2 3944 else:
49a57e70 3945 write_debug('Lazy loading extractors is disabled')
8a82af35 3946 if self.params['compat_opts']:
3947 write_debug('Compatibility options: %s' % ', '.join(self.params['compat_opts']))
36eaf303 3948
b5e7a2e6 3949 if current_git_head():
3950 write_debug(f'Git HEAD: {current_git_head()}')
b1f94422 3951 write_debug(system_identifier())
d28b5171 3952
8913ef74 3953 exe_versions, ffmpeg_features = FFmpegPostProcessor.get_versions_and_features(self)
3954 ffmpeg_features = {key for key, val in ffmpeg_features.items() if val}
3955 if ffmpeg_features:
19a03940 3956 exe_versions['ffmpeg'] += ' (%s)' % ','.join(sorted(ffmpeg_features))
8913ef74 3957
4c83c967 3958 exe_versions['rtmpdump'] = rtmpdump_version()
feee8d32 3959 exe_versions['phantomjs'] = PhantomJSwrapper._version()
d28b5171 3960 exe_str = ', '.join(
2831b468 3961 f'{exe} {v}' for exe, v in sorted(exe_versions.items()) if v
3962 ) or 'none'
49a57e70 3963 write_debug('exe versions: %s' % exe_str)
dca08720 3964
1d485a1a 3965 from .compat.compat_utils import get_package_info
9b8ee23b 3966 from .dependencies import available_dependencies
3967
3968 write_debug('Optional libraries: %s' % (', '.join(sorted({
1d485a1a 3969 join_nonempty(*get_package_info(m)) for m in available_dependencies.values()
9b8ee23b 3970 })) or 'none'))
2831b468 3971
227bf1a3 3972 write_debug(f'Proxy map: {self.proxies}')
8a8b5452 3973 write_debug(f'Request Handlers: {", ".join(rh.RH_NAME for rh in self._request_director.handlers.values())}')
e756f45b
M
3974 for plugin_type, plugins in {'Extractor': plugin_ies, 'Post-Processor': plugin_pps}.items():
3975 display_list = ['%s%s' % (
8e40b9d1 3976 klass.__name__, '' if klass.__name__ == name else f' as {name}')
e756f45b
M
3977 for name, klass in plugins.items()]
3978 if plugin_type == 'Extractor':
3979 display_list.extend(f'{plugins[-1].IE_NAME.partition("+")[2]} ({parent.__name__})'
3980 for parent, plugins in plugin_ie_overrides.items())
3981 if not display_list:
3982 continue
3983 write_debug(f'{plugin_type} Plugins: {", ".join(sorted(display_list))}')
3984
8e40b9d1
M
3985 plugin_dirs = plugin_directories()
3986 if plugin_dirs:
3987 write_debug(f'Plugin directories: {plugin_dirs}')
3988
49a57e70 3989 # Not implemented
3990 if False and self.params.get('call_home'):
0f06bcd7 3991 ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode()
49a57e70 3992 write_debug('Public IP address: %s' % ipaddr)
58b1f00d 3993 latest_version = self.urlopen(
0f06bcd7 3994 'https://yt-dl.org/latest/version').read().decode()
58b1f00d
PH
3995 if version_tuple(latest_version) > version_tuple(__version__):
3996 self.report_warning(
3997 'You are using an outdated version (newest version: %s)! '
3998 'See https://yt-dl.org/update if you need help updating.' %
3999 latest_version)
4000
227bf1a3 4001 @functools.cached_property
4002 def proxies(self):
4003 """Global proxy configuration"""
dca08720 4004 opts_proxy = self.params.get('proxy')
dca08720
PH
4005 if opts_proxy is not None:
4006 if opts_proxy == '':
227bf1a3 4007 opts_proxy = '__noproxy__'
4008 proxies = {'all': opts_proxy}
dca08720 4009 else:
ac668111 4010 proxies = urllib.request.getproxies()
227bf1a3 4011 # compat. Set HTTPS_PROXY to __noproxy__ to revert
dca08720
PH
4012 if 'http' in proxies and 'https' not in proxies:
4013 proxies['https'] = proxies['http']
227bf1a3 4014
4015 return proxies
62fec3b2 4016
c365dba8 4017 @functools.cached_property
4018 def cookiejar(self):
4019 """Global cookiejar instance"""
4020 return load_cookies(
4021 self.params.get('cookiefile'), self.params.get('cookiesfrombrowser'), self)
4022
227bf1a3 4023 @property
4024 def _opener(self):
4025 """
4026 Get a urllib OpenerDirector from the Urllib handler (deprecated).
4027 """
62b5c94c 4028 self.deprecation_warning('YoutubeDL._opener is deprecated, use YoutubeDL.urlopen()')
227bf1a3 4029 handler = self._request_director.handlers['Urllib']
4030 return handler._get_instance(cookiejar=self.cookiejar, proxies=self.proxies)
4031
c365dba8 4032 def urlopen(self, req):
4033 """ Start an HTTP download """
4034 if isinstance(req, str):
227bf1a3 4035 req = Request(req)
4036 elif isinstance(req, urllib.request.Request):
3d2623a8 4037 self.deprecation_warning(
4038 'Passing a urllib.request.Request object to YoutubeDL.urlopen() is deprecated. '
4039 'Use yt_dlp.networking.common.Request instead.')
227bf1a3 4040 req = urllib_req_to_req(req)
4041 assert isinstance(req, Request)
4042
4043 # compat: Assume user:pass url params are basic auth
4044 url, basic_auth_header = extract_basic_auth(req.url)
4045 if basic_auth_header:
4046 req.headers['Authorization'] = basic_auth_header
4047 req.url = sanitize_url(url)
4048
4049 clean_proxies(proxies=req.proxies, headers=req.headers)
4050 clean_headers(req.headers)
4051
4052 try:
4053 return self._request_director.send(req)
4054 except NoSupportingHandlers as e:
4055 for ue in e.unsupported_errors:
4056 if not (ue.handler and ue.msg):
4057 continue
4058 if ue.handler.RH_KEY == 'Urllib' and 'unsupported url scheme: "file"' in ue.msg.lower():
4059 raise RequestError(
4060 'file:// URLs are disabled by default in yt-dlp for security reasons. '
4061 'Use --enable-file-urls to enable at your own risk.', cause=ue) from ue
8a8b5452 4062 if 'unsupported proxy type: "https"' in ue.msg.lower():
4063 raise RequestError(
4064 'To use an HTTPS proxy for this request, one of the following dependencies needs to be installed: requests')
227bf1a3 4065 raise
4066 except SSLError as e:
4067 if 'UNSAFE_LEGACY_RENEGOTIATION_DISABLED' in str(e):
4068 raise RequestError('UNSAFE_LEGACY_RENEGOTIATION_DISABLED: Try using --legacy-server-connect', cause=e) from e
4069 elif 'SSLV3_ALERT_HANDSHAKE_FAILURE' in str(e):
4070 raise RequestError(
4071 'SSLV3_ALERT_HANDSHAKE_FAILURE: The server may not support the current cipher list. '
4072 'Try using --legacy-server-connect', cause=e) from e
4073 raise
4074 except HTTPError as e: # TODO: Remove in a future release
4075 raise _CompatHTTPError(e) from e
4076
db7b054a 4077 def build_request_director(self, handlers, preferences=None):
227bf1a3 4078 logger = _YDLLogger(self)
6148833f 4079 headers = self.params['http_headers'].copy()
227bf1a3 4080 proxies = self.proxies.copy()
4081 clean_headers(headers)
4082 clean_proxies(proxies, headers)
4083
4084 director = RequestDirector(logger=logger, verbose=self.params.get('debug_printtraffic'))
4085 for handler in handlers:
4086 director.add_handler(handler(
4087 logger=logger,
4088 headers=headers,
4089 cookiejar=self.cookiejar,
4090 proxies=proxies,
4091 prefer_system_certs='no-certifi' in self.params['compat_opts'],
4092 verify=not self.params.get('nocheckcertificate'),
4093 **traverse_obj(self.params, {
4094 'verbose': 'debug_printtraffic',
4095 'source_address': 'source_address',
4096 'timeout': 'socket_timeout',
75dc8e67 4097 'legacy_ssl_support': 'legacyserverconnect',
227bf1a3 4098 'enable_file_urls': 'enable_file_urls',
4099 'client_cert': {
4100 'client_certificate': 'client_certificate',
4101 'client_certificate_key': 'client_certificate_key',
4102 'client_certificate_password': 'client_certificate_password',
4103 },
4104 }),
4105 ))
db7b054a 4106 director.preferences.update(preferences or [])
8a8b5452 4107 if 'prefer-legacy-http-handler' in self.params['compat_opts']:
4108 director.preferences.add(lambda rh, _: 500 if rh.RH_KEY == 'Urllib' else 0)
227bf1a3 4109 return director
c365dba8 4110
62fec3b2
PH
4111 def encode(self, s):
4112 if isinstance(s, bytes):
4113 return s # Already encoded
4114
4115 try:
4116 return s.encode(self.get_encoding())
4117 except UnicodeEncodeError as err:
4118 err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.'
4119 raise
4120
4121 def get_encoding(self):
4122 encoding = self.params.get('encoding')
4123 if encoding is None:
4124 encoding = preferredencoding()
4125 return encoding
ec82d85a 4126
e08a85d8 4127 def _write_info_json(self, label, ie_result, infofn, overwrite=None):
cb96c5be 4128 ''' Write infojson and returns True = written, 'exists' = Already exists, False = skip, None = error '''
e08a85d8 4129 if overwrite is None:
4130 overwrite = self.params.get('overwrites', True)
80c03fa9 4131 if not self.params.get('writeinfojson'):
4132 return False
4133 elif not infofn:
4134 self.write_debug(f'Skipping writing {label} infojson')
4135 return False
4136 elif not self._ensure_dir_exists(infofn):
4137 return None
e08a85d8 4138 elif not overwrite and os.path.exists(infofn):
80c03fa9 4139 self.to_screen(f'[info] {label.title()} metadata is already present')
cb96c5be 4140 return 'exists'
4141
4142 self.to_screen(f'[info] Writing {label} metadata as JSON to: {infofn}')
4143 try:
4144 write_json_file(self.sanitize_info(ie_result, self.params.get('clean_infojson', True)), infofn)
4145 return True
86e5f3ed 4146 except OSError:
cb96c5be 4147 self.report_error(f'Cannot write {label} metadata to JSON file {infofn}')
4148 return None
80c03fa9 4149
4150 def _write_description(self, label, ie_result, descfn):
4151 ''' Write description and returns True = written, False = skip, None = error '''
4152 if not self.params.get('writedescription'):
4153 return False
4154 elif not descfn:
4155 self.write_debug(f'Skipping writing {label} description')
4156 return False
4157 elif not self._ensure_dir_exists(descfn):
4158 return None
4159 elif not self.params.get('overwrites', True) and os.path.exists(descfn):
4160 self.to_screen(f'[info] {label.title()} description is already present')
4161 elif ie_result.get('description') is None:
88fb9425 4162 self.to_screen(f'[info] There\'s no {label} description to write')
80c03fa9 4163 return False
4164 else:
4165 try:
4166 self.to_screen(f'[info] Writing {label} description to: {descfn}')
86e5f3ed 4167 with open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
80c03fa9 4168 descfile.write(ie_result['description'])
86e5f3ed 4169 except OSError:
80c03fa9 4170 self.report_error(f'Cannot write {label} description file {descfn}')
4171 return None
4172 return True
4173
4174 def _write_subtitles(self, info_dict, filename):
4175 ''' Write subtitles to file and return list of (sub_filename, final_sub_filename); or None if error'''
4176 ret = []
4177 subtitles = info_dict.get('requested_subtitles')
88fb9425 4178 if not (self.params.get('writesubtitles') or self.params.get('writeautomaticsub')):
80c03fa9 4179 # subtitles download errors are already managed as troubles in relevant IE
4180 # that way it will silently go on when used with unsupporting IE
4181 return ret
88fb9425 4182 elif not subtitles:
c8bc203f 4183 self.to_screen('[info] There are no subtitles for the requested languages')
88fb9425 4184 return ret
80c03fa9 4185 sub_filename_base = self.prepare_filename(info_dict, 'subtitle')
4186 if not sub_filename_base:
4187 self.to_screen('[info] Skipping writing video subtitles')
4188 return ret
88fb9425 4189
80c03fa9 4190 for sub_lang, sub_info in subtitles.items():
4191 sub_format = sub_info['ext']
4192 sub_filename = subtitles_filename(filename, sub_lang, sub_format, info_dict.get('ext'))
4193 sub_filename_final = subtitles_filename(sub_filename_base, sub_lang, sub_format, info_dict.get('ext'))
e04938ab 4194 existing_sub = self.existing_file((sub_filename_final, sub_filename))
4195 if existing_sub:
80c03fa9 4196 self.to_screen(f'[info] Video subtitle {sub_lang}.{sub_format} is already present')
e04938ab 4197 sub_info['filepath'] = existing_sub
4198 ret.append((existing_sub, sub_filename_final))
80c03fa9 4199 continue
4200
4201 self.to_screen(f'[info] Writing video subtitles to: {sub_filename}')
4202 if sub_info.get('data') is not None:
4203 try:
4204 # Use newline='' to prevent conversion of newline characters
4205 # See https://github.com/ytdl-org/youtube-dl/issues/10268
86e5f3ed 4206 with open(sub_filename, 'w', encoding='utf-8', newline='') as subfile:
80c03fa9 4207 subfile.write(sub_info['data'])
4208 sub_info['filepath'] = sub_filename
4209 ret.append((sub_filename, sub_filename_final))
4210 continue
86e5f3ed 4211 except OSError:
80c03fa9 4212 self.report_error(f'Cannot write video subtitles file {sub_filename}')
4213 return None
4214
4215 try:
4216 sub_copy = sub_info.copy()
4217 sub_copy.setdefault('http_headers', info_dict.get('http_headers'))
4218 self.dl(sub_filename, sub_copy, subtitle=True)
4219 sub_info['filepath'] = sub_filename
4220 ret.append((sub_filename, sub_filename_final))
6020e05d 4221 except (DownloadError, ExtractorError, IOError, OSError, ValueError) + network_exceptions as err:
c70c418d 4222 msg = f'Unable to download video subtitles for {sub_lang!r}: {err}'
6020e05d 4223 if self.params.get('ignoreerrors') is not True: # False or 'only_download'
c70c418d 4224 if not self.params.get('ignoreerrors'):
4225 self.report_error(msg)
4226 raise DownloadError(msg)
4227 self.report_warning(msg)
519804a9 4228 return ret
80c03fa9 4229
4230 def _write_thumbnails(self, label, info_dict, filename, thumb_filename_base=None):
2acd1d55 4231 ''' Write thumbnails to file and return list of (thumb_filename, final_thumb_filename); or None if error '''
6c4fd172 4232 write_all = self.params.get('write_all_thumbnails', False)
80c03fa9 4233 thumbnails, ret = [], []
6c4fd172 4234 if write_all or self.params.get('writethumbnail', False):
0202b52a 4235 thumbnails = info_dict.get('thumbnails') or []
88fb9425 4236 if not thumbnails:
c8bc203f 4237 self.to_screen(f'[info] There are no {label} thumbnails to download')
88fb9425 4238 return ret
6c4fd172 4239 multiple = write_all and len(thumbnails) > 1
ec82d85a 4240
80c03fa9 4241 if thumb_filename_base is None:
4242 thumb_filename_base = filename
4243 if thumbnails and not thumb_filename_base:
4244 self.write_debug(f'Skipping writing {label} thumbnail')
4245 return ret
4246
a40e0b37 4247 if thumbnails and not self._ensure_dir_exists(filename):
2acd1d55
R
4248 return None
4249
dd0228ce 4250 for idx, t in list(enumerate(thumbnails))[::-1]:
80c03fa9 4251 thumb_ext = (f'{t["id"]}.' if multiple else '') + determine_ext(t['url'], 'jpg')
aa9369a2 4252 thumb_display_id = f'{label} thumbnail {t["id"]}'
80c03fa9 4253 thumb_filename = replace_extension(filename, thumb_ext, info_dict.get('ext'))
4254 thumb_filename_final = replace_extension(thumb_filename_base, thumb_ext, info_dict.get('ext'))
ec82d85a 4255
e04938ab 4256 existing_thumb = self.existing_file((thumb_filename_final, thumb_filename))
4257 if existing_thumb:
aa9369a2 4258 self.to_screen('[info] %s is already present' % (
4259 thumb_display_id if multiple else f'{label} thumbnail').capitalize())
e04938ab 4260 t['filepath'] = existing_thumb
4261 ret.append((existing_thumb, thumb_filename_final))
ec82d85a 4262 else:
80c03fa9 4263 self.to_screen(f'[info] Downloading {thumb_display_id} ...')
ec82d85a 4264 try:
227bf1a3 4265 uf = self.urlopen(Request(t['url'], headers=t.get('http_headers', {})))
80c03fa9 4266 self.to_screen(f'[info] Writing {thumb_display_id} to: {thumb_filename}')
d3d89c32 4267 with open(encodeFilename(thumb_filename), 'wb') as thumbf:
ec82d85a 4268 shutil.copyfileobj(uf, thumbf)
80c03fa9 4269 ret.append((thumb_filename, thumb_filename_final))
885cc0b7 4270 t['filepath'] = thumb_filename
3158150c 4271 except network_exceptions as err:
3d2623a8 4272 if isinstance(err, HTTPError) and err.status == 404:
ad54c913 4273 self.to_screen(f'[info] {thumb_display_id.title()} does not exist')
4274 else:
4275 self.report_warning(f'Unable to download {thumb_display_id}: {err}')
dd0228ce 4276 thumbnails.pop(idx)
6c4fd172 4277 if ret and not write_all:
4278 break
0202b52a 4279 return ret