]> jfr.im git - yt-dlp.git/blame - yt_dlp/YoutubeDL.py
[extractor/MLBTV] Detect live streams
[yt-dlp.git] / yt_dlp / YoutubeDL.py
CommitLineData
26e63931 1import collections
31bd3925 2import contextlib
9d2ecdbc 3import datetime
c1c9a79c 4import errno
31bd3925 5import fileinput
b5ae35ee 6import functools
8222d8de 7import io
b82f815f 8import itertools
8694c600 9import json
62fec3b2 10import locale
083c9df9 11import operator
8222d8de 12import os
f8271158 13import random
8222d8de
JMF
14import re
15import shutil
dca08720 16import subprocess
8222d8de 17import sys
21cd8fae 18import tempfile
8222d8de 19import time
67134eab 20import tokenize
8222d8de 21import traceback
524e2e4f 22import unicodedata
f9934b96 23import urllib.request
961ea474
S
24from string import ascii_letters
25
f8271158 26from .cache import Cache
14f25df2 27from .compat import compat_os_name, compat_shlex_quote
982ee69a 28from .cookies import load_cookies
f8271158 29from .downloader import FFmpegFD, get_suitable_downloader, shorten_protocol_name
30from .downloader.rtmp import rtmpdump_version
f8271158 31from .extractor import gen_extractor_classes, get_info_extractor
fe7866d0 32from .extractor.common import UnsupportedURLIE
f8271158 33from .extractor.openload import PhantomJSwrapper
34from .minicurses import format_text
35from .postprocessor import _PLUGIN_CLASSES as plugin_postprocessors
36from .postprocessor import (
37 EmbedThumbnailPP,
38 FFmpegFixupDuplicateMoovPP,
39 FFmpegFixupDurationPP,
40 FFmpegFixupM3u8PP,
41 FFmpegFixupM4aPP,
42 FFmpegFixupStretchedPP,
43 FFmpegFixupTimestampPP,
44 FFmpegMergerPP,
45 FFmpegPostProcessor,
ca9def71 46 FFmpegVideoConvertorPP,
f8271158 47 MoveFilesAfterDownloadPP,
48 get_postprocessor,
49)
ca9def71 50from .postprocessor.ffmpeg import resolve_mapping as resolve_recode_mapping
b5e7a2e6 51from .update import REPOSITORY, current_git_head, detect_variant
8c25f81b 52from .utils import (
f8271158 53 DEFAULT_OUTTMPL,
7b2c3f47 54 IDENTITY,
f8271158 55 LINK_TEMPLATES,
8dc59305 56 MEDIA_EXTENSIONS,
f8271158 57 NO_DEFAULT,
1d485a1a 58 NUMBER_RE,
f8271158 59 OUTTMPL_TYPES,
60 POSTPROCESS_WHEN,
61 STR_FORMAT_RE_TMPL,
62 STR_FORMAT_TYPES,
63 ContentTooShortError,
64 DateRange,
65 DownloadCancelled,
66 DownloadError,
67 EntryNotInPlaylist,
68 ExistingVideoReached,
69 ExtractorError,
70 GeoRestrictedError,
71 HEADRequest,
f8271158 72 ISO3166Utils,
73 LazyList,
74 MaxDownloadsReached,
19a03940 75 Namespace,
f8271158 76 PagedList,
77 PerRequestProxyHandler,
7e88d7d7 78 PlaylistEntries,
f8271158 79 Popen,
80 PostProcessingError,
81 ReExtractInfo,
82 RejectedVideoReached,
83 SameFileError,
84 UnavailableVideoError,
693f0600 85 UserNotLive,
f8271158 86 YoutubeDLCookieProcessor,
87 YoutubeDLHandler,
88 YoutubeDLRedirectHandler,
eedb7ba5
S
89 age_restricted,
90 args_to_str,
cb794ee0 91 bug_reports_message,
ce02ed60 92 date_from_str,
da4db748 93 deprecation_warning,
ce02ed60 94 determine_ext,
b5559424 95 determine_protocol,
c0384f22 96 encode_compat_str,
ce02ed60 97 encodeFilename,
a06916d9 98 error_to_compat_str,
47cdc68e 99 escapeHTML,
590bc6f6 100 expand_path,
90137ca4 101 filter_dict,
e29663c6 102 float_or_none,
02dbf93f 103 format_bytes,
e0fd9573 104 format_decimal_suffix,
f8271158 105 format_field,
525ef922 106 formatSeconds,
fc61aff4 107 get_compatible_ext,
0bb322b9 108 get_domain,
c9969434 109 int_or_none,
732044af 110 iri_to_uri,
34921b43 111 join_nonempty,
ce02ed60 112 locked_file,
0647d925 113 make_archive_id,
0202b52a 114 make_dir,
dca08720 115 make_HTTPS_handler,
8b7539d2 116 merge_headers,
3158150c 117 network_exceptions,
ec11a9f4 118 number_of_digits,
cd6fc19e 119 orderedSet,
5314b521 120 orderedSet_from_options,
083c9df9 121 parse_filesize,
ce02ed60 122 preferredencoding,
eedb7ba5 123 prepend_extension,
51fb4995 124 register_socks_protocols,
3efb96a6 125 remove_terminal_sequences,
cfb56d1a 126 render_table,
eedb7ba5 127 replace_extension,
ce02ed60 128 sanitize_filename,
1bb5c511 129 sanitize_path,
dcf77cf1 130 sanitize_url,
67dda517 131 sanitized_Request,
e5660ee6 132 std_headers,
1211bb6d 133 str_or_none,
e29663c6 134 strftime_or_none,
ce02ed60 135 subtitles_filename,
819e0531 136 supports_terminal_sequences,
b1f94422 137 system_identifier,
f2ebc5c7 138 timetuple_from_msec,
732044af 139 to_high_limit_path,
324ad820 140 traverse_obj,
fc61aff4 141 try_call,
6033d980 142 try_get,
29eb5174 143 url_basename,
7d1eb38a 144 variadic,
58b1f00d 145 version_tuple,
53973b4d 146 windows_enable_vt_mode,
ce02ed60
PH
147 write_json_file,
148 write_string,
4f026faf 149)
70b23409 150from .version import RELEASE_GIT_HEAD, VARIANT, __version__
8222d8de 151
e9c0cdd3
YCH
152if compat_os_name == 'nt':
153 import ctypes
154
2459b6e1 155
86e5f3ed 156class YoutubeDL:
8222d8de
JMF
157 """YoutubeDL class.
158
159 YoutubeDL objects are the ones responsible of downloading the
160 actual video file and writing it to disk if the user has requested
161 it, among some other tasks. In most cases there should be one per
162 program. As, given a video URL, the downloader doesn't know how to
163 extract all the needed information, task that InfoExtractors do, it
164 has to pass the URL to one of them.
165
166 For this, YoutubeDL objects have a method that allows
167 InfoExtractors to be registered in a given order. When it is passed
168 a URL, the YoutubeDL object handles it to the first InfoExtractor it
169 finds that reports being able to handle it. The InfoExtractor extracts
170 all the information about the video or videos the URL refers to, and
171 YoutubeDL process the extracted information, possibly using a File
172 Downloader to download the video.
173
174 YoutubeDL objects accept a lot of parameters. In order not to saturate
175 the object constructor with arguments, it receives a dictionary of
176 options instead. These options are available through the params
177 attribute for the InfoExtractors to use. The YoutubeDL also
178 registers itself as the downloader in charge for the InfoExtractors
179 that are added to it, so this is a "mutual registration".
180
181 Available options:
182
183 username: Username for authentication purposes.
184 password: Password for authentication purposes.
180940e0 185 videopassword: Password for accessing a video.
1da50aa3
S
186 ap_mso: Adobe Pass multiple-system operator identifier.
187 ap_username: Multiple-system operator account username.
188 ap_password: Multiple-system operator account password.
8222d8de
JMF
189 usenetrc: Use netrc for authentication instead.
190 verbose: Print additional info to stdout.
191 quiet: Do not print messages to stdout.
ad8915b7 192 no_warnings: Do not print out anything for warnings.
bb66c247 193 forceprint: A dict with keys WHEN mapped to a list of templates to
194 print to stdout. The allowed keys are video or any of the
195 items in utils.POSTPROCESS_WHEN.
ca30f449 196 For compatibility, a single list is also accepted
bb66c247 197 print_to_file: A dict with keys WHEN (same as forceprint) mapped to
198 a list of tuples with (template, filename)
8694c600 199 forcejson: Force printing info_dict as JSON.
63e0be34
PH
200 dump_single_json: Force printing the info_dict of the whole playlist
201 (or video) as a single JSON line.
c25228e5 202 force_write_download_archive: Force writing download archive regardless
203 of 'skip_download' or 'simulate'.
b7b04c78 204 simulate: Do not download the video files. If unset (or None),
205 simulate only if listsubtitles, listformats or list_thumbnails is used
eb8a4433 206 format: Video format code. see "FORMAT SELECTION" for more details.
093a1710 207 You can also pass a function. The function takes 'ctx' as
208 argument and returns the formats to download.
209 See "build_format_selector" for an implementation
63ad4d43 210 allow_unplayable_formats: Allow unplayable formats to be extracted and downloaded.
b7da73eb 211 ignore_no_formats_error: Ignore "No video formats" error. Usefull for
212 extracting metadata even if the video is not actually
213 available for download (experimental)
0930b11f 214 format_sort: A list of fields by which to sort the video formats.
215 See "Sorting Formats" for more details.
c25228e5 216 format_sort_force: Force the given format_sort. see "Sorting Formats"
217 for more details.
08d30158 218 prefer_free_formats: Whether to prefer video formats with free containers
219 over non-free ones of same quality.
c25228e5 220 allow_multiple_video_streams: Allow multiple video streams to be merged
221 into a single file
222 allow_multiple_audio_streams: Allow multiple audio streams to be merged
223 into a single file
0ba692ac 224 check_formats Whether to test if the formats are downloadable.
9f1a1c36 225 Can be True (check all), False (check none),
226 'selected' (check selected formats),
0ba692ac 227 or None (check only if requested by extractor)
4524baf0 228 paths: Dictionary of output paths. The allowed keys are 'home'
229 'temp' and the keys of OUTTMPL_TYPES (in utils.py)
de6000d9 230 outtmpl: Dictionary of templates for output names. Allowed keys
4524baf0 231 are 'default' and the keys of OUTTMPL_TYPES (in utils.py).
34488702 232 For compatibility with youtube-dl, a single string can also be used
a820dc72
RA
233 outtmpl_na_placeholder: Placeholder for unavailable meta fields.
234 restrictfilenames: Do not allow "&" and spaces in file names
235 trim_file_name: Limit length of filename (extension excluded)
4524baf0 236 windowsfilenames: Force the filenames to be windows compatible
b1940459 237 ignoreerrors: Do not stop on download/postprocessing errors.
238 Can be 'only_download' to ignore only download errors.
239 Default is 'only_download' for CLI, but False for API
26e2805c 240 skip_playlist_after_errors: Number of allowed failures until the rest of
241 the playlist is skipped
fe7866d0 242 allowed_extractors: List of regexes to match against extractor names that are allowed
0c3d0f51 243 overwrites: Overwrite all video and metadata files if True,
244 overwrite only non-video files if None
245 and don't overwrite any file if False
34488702 246 For compatibility with youtube-dl,
247 "nooverwrites" may also be used instead
c14e88f0 248 playlist_items: Specific indices of playlist to download.
75822ca7 249 playlistrandom: Download playlist items in random order.
7e9a6125 250 lazy_playlist: Process playlist entries as they are received.
8222d8de
JMF
251 matchtitle: Download only matching titles.
252 rejecttitle: Reject downloads for matching titles.
8bf9319e 253 logger: Log messages to a logging.Logger instance.
17ffed18 254 logtostderr: Print everything to stderr instead of stdout.
255 consoletitle: Display progress in console window's titlebar.
8222d8de
JMF
256 writedescription: Write the video description to a .description file
257 writeinfojson: Write the video description to a .info.json file
75d43ca0 258 clean_infojson: Remove private fields from the infojson
34488702 259 getcomments: Extract video comments. This will not be written to disk
06167fbb 260 unless writeinfojson is also given
1fb07d10 261 writeannotations: Write the video annotations to a .annotations.xml file
8222d8de 262 writethumbnail: Write the thumbnail image to a file
c25228e5 263 allow_playlist_files: Whether to write playlists' description, infojson etc
264 also to disk when using the 'write*' options
ec82d85a 265 write_all_thumbnails: Write all thumbnail formats to files
732044af 266 writelink: Write an internet shortcut file, depending on the
267 current platform (.url/.webloc/.desktop)
268 writeurllink: Write a Windows internet shortcut file (.url)
269 writewebloclink: Write a macOS internet shortcut file (.webloc)
270 writedesktoplink: Write a Linux internet shortcut file (.desktop)
8222d8de 271 writesubtitles: Write the video subtitles to a file
741dd8ea 272 writeautomaticsub: Write the automatically generated subtitles to a file
8222d8de 273 listsubtitles: Lists all available subtitles for the video
a504ced0 274 subtitlesformat: The format code for subtitles
c32b0aab 275 subtitleslangs: List of languages of the subtitles to download (can be regex).
276 The list may contain "all" to refer to all the available
277 subtitles. The language can be prefixed with a "-" to
62b58c09 278 exclude it from the requested languages, e.g. ['all', '-live_chat']
8222d8de
JMF
279 keepvideo: Keep the video file after post-processing
280 daterange: A DateRange object, download only if the upload_date is in the range.
281 skip_download: Skip the actual download of the video file
c35f9e72 282 cachedir: Location of the cache files in the filesystem.
a0e07d31 283 False to disable filesystem cache.
47192f92 284 noplaylist: Download single video instead of a playlist if in doubt.
8dbe9899
PH
285 age_limit: An integer representing the user's age in years.
286 Unsuitable videos for the given age are skipped.
5fe18bdb
PH
287 min_views: An integer representing the minimum view count the video
288 must have in order to not be skipped.
289 Videos without view count information are always
290 downloaded. None for no limit.
291 max_views: An integer representing the maximum view count.
292 Videos that are more popular than that are not
293 downloaded.
294 Videos without view count information are always
295 downloaded. None for no limit.
296 download_archive: File name of a file where all downloads are recorded.
c1c9a79c
PH
297 Videos already present in the file are not downloaded
298 again.
8a51f564 299 break_on_existing: Stop the download process after attempting to download a
300 file that is in the archive.
301 break_on_reject: Stop the download process when encountering a video that
302 has been filtered out.
b222c271 303 break_per_url: Whether break_on_reject and break_on_existing
304 should act on each input URL as opposed to for the entire queue
d76fa1f3 305 cookiefile: File name or text stream from where cookies should be read and dumped to
f59f5ef8 306 cookiesfrombrowser: A tuple containing the name of the browser, the profile
9bd13fe5 307 name/path from where cookies are loaded, the name of the keyring,
308 and the container name, e.g. ('chrome', ) or
309 ('vivaldi', 'default', 'BASICTEXT') or ('firefox', 'default', None, 'Meta')
f81c62a6 310 legacyserverconnect: Explicitly allow HTTPS connection to servers that do not
311 support RFC 5746 secure renegotiation
f59f5ef8 312 nocheckcertificate: Do not verify SSL certificates
bb58c9ed 313 client_certificate: Path to client certificate file in PEM format. May include the private key
314 client_certificate_key: Path to private key file for client certificate
315 client_certificate_password: Password for client certificate private key, if encrypted.
316 If not provided and the key is encrypted, yt-dlp will ask interactively
7e8c0af0 317 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
c6e07cf1 318 (Only supported by some extractors)
8b7539d2 319 http_headers: A dictionary of custom headers to be used for all requests
a1ee09e8 320 proxy: URL of the proxy server to use
38cce791 321 geo_verification_proxy: URL of the proxy to use for IP address verification
504f20dd 322 on geo-restricted sites.
e344693b 323 socket_timeout: Time to wait for unresponsive hosts, in seconds
0783b09b
PH
324 bidi_workaround: Work around buggy terminals without bidirectional text
325 support, using fridibi
a0ddb8a2 326 debug_printtraffic:Print out sent and received HTTP traffic
04b4d394
PH
327 default_search: Prepend this string if an input url is not valid.
328 'auto' for elaborate guessing
62fec3b2 329 encoding: Use this encoding instead of the system-specified.
134c913c 330 extract_flat: Whether to resolve and process url_results further
331 * False: Always process (default)
332 * True: Never process
333 * 'in_playlist': Do not process inside playlist/multi_video
334 * 'discard': Always process, but don't return the result
335 from inside playlist/multi_video
336 * 'discard_in_playlist': Same as "discard", but only for
337 playlists (not multi_video)
f2ebc5c7 338 wait_for_video: If given, wait for scheduled streams to become available.
339 The value should be a tuple containing the range
340 (min_secs, max_secs) to wait between retries
4f026faf 341 postprocessors: A list of dictionaries, each with an entry
71b640cc 342 * key: The name of the postprocessor. See
7a5c1cfe 343 yt_dlp/postprocessor/__init__.py for a list.
bb66c247 344 * when: When to run the postprocessor. Allowed values are
345 the entries of utils.POSTPROCESS_WHEN
56d868db 346 Assumed to be 'post_process' if not given
71b640cc
PH
347 progress_hooks: A list of functions that get called on download
348 progress, with a dictionary with the entries
5cda4eda 349 * status: One of "downloading", "error", or "finished".
ee69b99a 350 Check this first and ignore unknown values.
3ba7740d 351 * info_dict: The extracted info_dict
71b640cc 352
5cda4eda 353 If status is one of "downloading", or "finished", the
ee69b99a
PH
354 following properties may also be present:
355 * filename: The final filename (always present)
5cda4eda 356 * tmpfilename: The filename we're currently writing to
71b640cc
PH
357 * downloaded_bytes: Bytes on disk
358 * total_bytes: Size of the whole file, None if unknown
5cda4eda
PH
359 * total_bytes_estimate: Guess of the eventual file size,
360 None if unavailable.
361 * elapsed: The number of seconds since download started.
71b640cc
PH
362 * eta: The estimated time in seconds, None if unknown
363 * speed: The download speed in bytes/second, None if
364 unknown
5cda4eda
PH
365 * fragment_index: The counter of the currently
366 downloaded video fragment.
367 * fragment_count: The number of fragments (= individual
368 files that will be merged)
71b640cc
PH
369
370 Progress hooks are guaranteed to be called at least once
371 (with status "finished") if the download is successful.
819e0531 372 postprocessor_hooks: A list of functions that get called on postprocessing
373 progress, with a dictionary with the entries
374 * status: One of "started", "processing", or "finished".
375 Check this first and ignore unknown values.
376 * postprocessor: Name of the postprocessor
377 * info_dict: The extracted info_dict
378
379 Progress hooks are guaranteed to be called at least twice
380 (with status "started" and "finished") if the processing is successful.
fc61aff4 381 merge_output_format: "/" separated list of extensions to use when merging formats.
6b591b29 382 final_ext: Expected final extension; used to detect when the file was
59a7a13e 383 already downloaded and converted
6271f1ca
PH
384 fixup: Automatically correct known faults of the file.
385 One of:
386 - "never": do nothing
387 - "warn": only emit a warning
388 - "detect_or_warn": check whether we can do anything
62cd676c 389 about it, warn otherwise (default)
504f20dd 390 source_address: Client-side IP address to bind to.
1cf376f5 391 sleep_interval_requests: Number of seconds to sleep between requests
392 during extraction
7aa589a5
S
393 sleep_interval: Number of seconds to sleep before each download when
394 used alone or a lower bound of a range for randomized
395 sleep before each download (minimum possible number
396 of seconds to sleep) when used along with
397 max_sleep_interval.
398 max_sleep_interval:Upper bound of a range for randomized sleep before each
399 download (maximum possible number of seconds to sleep).
400 Must only be used along with sleep_interval.
401 Actual sleep time will be a random float from range
402 [sleep_interval; max_sleep_interval].
1cf376f5 403 sleep_interval_subtitles: Number of seconds to sleep before each subtitle download
cfb56d1a
PH
404 listformats: Print an overview of available video formats and exit.
405 list_thumbnails: Print a table of all thumbnails and exit.
0a41f331 406 match_filter: A function that gets called for every video with the signature
407 (info_dict, *, incomplete: bool) -> Optional[str]
408 For backward compatibility with youtube-dl, the signature
409 (info_dict) -> Optional[str] is also allowed.
410 - If it returns a message, the video is ignored.
411 - If it returns None, the video is downloaded.
412 - If it returns utils.NO_DEFAULT, the user is interactively
413 asked whether to download the video.
347de493 414 match_filter_func in utils.py is one example for this.
7e5db8c9 415 no_color: Do not emit color codes in output.
0a840f58 416 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
504f20dd 417 HTTP header
0a840f58 418 geo_bypass_country:
773f291d
S
419 Two-letter ISO 3166-2 country code that will be used for
420 explicit geographic restriction bypassing via faking
504f20dd 421 X-Forwarded-For HTTP header
5f95927a
S
422 geo_bypass_ip_block:
423 IP range in CIDR notation that will be used similarly to
504f20dd 424 geo_bypass_country
52a8a1e1 425 external_downloader: A dictionary of protocol keys and the executable of the
426 external downloader to use for it. The allowed protocols
427 are default|http|ftp|m3u8|dash|rtsp|rtmp|mms.
428 Set the value to 'native' to use the native downloader
53ed7066 429 compat_opts: Compatibility options. See "Differences in default behavior".
3acf6d38 430 The following options do not work when used through the API:
b5ae35ee 431 filename, abort-on-error, multistreams, no-live-chat, format-sort
dac5df5a 432 no-clean-infojson, no-playlist-metafiles, no-keep-subs, no-attach-info-json.
e4f02757 433 Refer __init__.py for their implementation
819e0531 434 progress_template: Dictionary of templates for progress outputs.
435 Allowed keys are 'download', 'postprocess',
436 'download-title' (console title) and 'postprocess-title'.
437 The template is mapped on a dictionary with keys 'progress' and 'info'
23326151 438 retry_sleep_functions: Dictionary of functions that takes the number of attempts
439 as argument and returns the time to sleep in seconds.
440 Allowed keys are 'http', 'fragment', 'file_access'
0f446365
SW
441 download_ranges: A callback function that gets called for every video with
442 the signature (info_dict, ydl) -> Iterable[Section].
443 Only the returned sections will be downloaded.
444 Each Section is a dict with the following keys:
5ec1b6b7 445 * start_time: Start time of the section in seconds
446 * end_time: End time of the section in seconds
447 * title: Section title (Optional)
448 * index: Section number (Optional)
0f446365 449 force_keyframes_at_cuts: Re-encode the video when downloading ranges to get precise cuts
a7dc6a89 450 noprogress: Do not print the progress bar
a831c2ea 451 live_from_start: Whether to download livestreams videos from the start
fe7e0c98 452
8222d8de 453 The following parameters are not used by YoutubeDL itself, they are used by
7a5c1cfe 454 the downloader (see yt_dlp/downloader/common.py):
51d9739f 455 nopart, updatetime, buffersize, ratelimit, throttledratelimit, min_filesize,
205a0654 456 max_filesize, test, noresizebuffer, retries, file_access_retries, fragment_retries,
a7dc6a89 457 continuedl, xattr_set_filesize, hls_use_mpegts, http_chunk_size,
59a7a13e 458 external_downloader_args, concurrent_fragment_downloads.
76b1bd67
JMF
459
460 The following options are used by the post processors:
c0b7d117
S
461 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
462 to the binary or its containing directory.
43820c03 463 postprocessor_args: A dictionary of postprocessor/executable keys (in lower case)
34488702 464 and a list of additional command-line arguments for the
465 postprocessor/executable. The dict can also have "PP+EXE" keys
466 which are used when the given exe is used by the given PP.
467 Use 'default' as the name for arguments to passed to all PP
468 For compatibility with youtube-dl, a single list of args
469 can also be used
e409895f 470
471 The following options are used by the extractors:
62bff2c1 472 extractor_retries: Number of times to retry for known errors
473 dynamic_mpd: Whether to process dynamic DASH manifests (default: True)
e409895f 474 hls_split_discontinuity: Split HLS playlists to different formats at
62bff2c1 475 discontinuities such as ad breaks (default: False)
5d3a0e79 476 extractor_args: A dictionary of arguments to be passed to the extractors.
477 See "EXTRACTOR ARGUMENTS" for details.
62b58c09 478 E.g. {'youtube': {'skip': ['dash', 'hls']}}
88f23a18 479 mark_watched: Mark videos watched (even with --simulate). Only for YouTube
1890fc63 480
481 The following options are deprecated and may be removed in the future:
482
fe7866d0 483 force_generic_extractor: Force downloader to use the generic extractor
484 - Use allowed_extractors = ['generic', 'default']
7e9a6125 485 playliststart: - Use playlist_items
486 Playlist item to start at.
487 playlistend: - Use playlist_items
488 Playlist item to end at.
489 playlistreverse: - Use playlist_items
490 Download playlist items in reverse order.
1890fc63 491 forceurl: - Use forceprint
492 Force printing final URL.
493 forcetitle: - Use forceprint
494 Force printing title.
495 forceid: - Use forceprint
496 Force printing ID.
497 forcethumbnail: - Use forceprint
498 Force printing thumbnail URL.
499 forcedescription: - Use forceprint
500 Force printing description.
501 forcefilename: - Use forceprint
502 Force printing final filename.
503 forceduration: - Use forceprint
504 Force printing duration.
505 allsubtitles: - Use subtitleslangs = ['all']
506 Downloads all the subtitles of the video
507 (requires writesubtitles or writeautomaticsub)
508 include_ads: - Doesn't work
509 Download ads as well
510 call_home: - Not implemented
511 Boolean, true iff we are allowed to contact the
512 yt-dlp servers for debugging.
513 post_hooks: - Register a custom postprocessor
514 A list of functions that get called as the final step
515 for each video file, after all postprocessors have been
516 called. The filename will be passed as the only argument.
517 hls_prefer_native: - Use external_downloader = {'m3u8': 'native'} or {'m3u8': 'ffmpeg'}.
518 Use the native HLS downloader instead of ffmpeg/avconv
519 if True, otherwise use ffmpeg/avconv if False, otherwise
520 use downloader suggested by extractor if None.
521 prefer_ffmpeg: - avconv support is deprecated
522 If False, use avconv instead of ffmpeg if both are available,
523 otherwise prefer ffmpeg.
524 youtube_include_dash_manifest: - Use extractor_args
5d3a0e79 525 If True (default), DASH manifests and related
62bff2c1 526 data will be downloaded and processed by extractor.
527 You can reduce network I/O by disabling it if you don't
528 care about DASH. (only for youtube)
1890fc63 529 youtube_include_hls_manifest: - Use extractor_args
5d3a0e79 530 If True (default), HLS manifests and related
62bff2c1 531 data will be downloaded and processed by extractor.
532 You can reduce network I/O by disabling it if you don't
533 care about HLS. (only for youtube)
8222d8de
JMF
534 """
535
86e5f3ed 536 _NUMERIC_FIELDS = {
b8ed0f15 537 'width', 'height', 'asr', 'audio_channels', 'fps',
538 'tbr', 'abr', 'vbr', 'filesize', 'filesize_approx',
e6f21b3d 539 'timestamp', 'release_timestamp',
c9969434
S
540 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
541 'average_rating', 'comment_count', 'age_limit',
542 'start_time', 'end_time',
543 'chapter_number', 'season_number', 'episode_number',
544 'track_number', 'disc_number', 'release_year',
86e5f3ed 545 }
c9969434 546
6db9c4d5 547 _format_fields = {
548 # NB: Keep in sync with the docstring of extractor/common.py
a44ca5a4 549 'url', 'manifest_url', 'manifest_stream_number', 'ext', 'format', 'format_id', 'format_note',
b8ed0f15 550 'width', 'height', 'resolution', 'dynamic_range', 'tbr', 'abr', 'acodec', 'asr', 'audio_channels',
6db9c4d5 551 'vbr', 'fps', 'vcodec', 'container', 'filesize', 'filesize_approx',
552 'player_url', 'protocol', 'fragment_base_url', 'fragments', 'is_from_start',
553 'preference', 'language', 'language_preference', 'quality', 'source_preference',
554 'http_headers', 'stretched_ratio', 'no_resume', 'has_drm', 'downloader_options',
555 'page_url', 'app', 'play_path', 'tc_url', 'flash_version', 'rtmp_live', 'rtmp_conn', 'rtmp_protocol', 'rtmp_real_time'
556 }
48ee10ee 557 _format_selection_exts = {
8dc59305 558 'audio': set(MEDIA_EXTENSIONS.common_audio),
559 'video': set(MEDIA_EXTENSIONS.common_video + ('3gp', )),
560 'storyboards': set(MEDIA_EXTENSIONS.storyboards),
48ee10ee 561 }
562
3511266b 563 def __init__(self, params=None, auto_init=True):
883d4b1e 564 """Create a FileDownloader object with the given options.
565 @param auto_init Whether to load the default extractors and print header (if verbose).
49a57e70 566 Set to 'no_verbose_header' to not print the header
883d4b1e 567 """
e9f9a10f
JMF
568 if params is None:
569 params = {}
592b7485 570 self.params = params
8b7491c8 571 self._ies = {}
56c73665 572 self._ies_instances = {}
1e43a6f7 573 self._pps = {k: [] for k in POSTPROCESS_WHEN}
b35496d8 574 self._printed_messages = set()
1cf376f5 575 self._first_webpage_request = True
ab8e5e51 576 self._post_hooks = []
933605d7 577 self._progress_hooks = []
819e0531 578 self._postprocessor_hooks = []
8222d8de
JMF
579 self._download_retcode = 0
580 self._num_downloads = 0
9c906919 581 self._num_videos = 0
592b7485 582 self._playlist_level = 0
583 self._playlist_urls = set()
a0e07d31 584 self.cache = Cache(self)
34308b30 585
819e0531 586 windows_enable_vt_mode()
591bb9d3 587 stdout = sys.stderr if self.params.get('logtostderr') else sys.stdout
588 self._out_files = Namespace(
589 out=stdout,
590 error=sys.stderr,
591 screen=sys.stderr if self.params.get('quiet') else stdout,
592 console=None if compat_os_name == 'nt' else next(
cf4f42cb 593 filter(supports_terminal_sequences, (sys.stderr, sys.stdout)), None)
591bb9d3 594 )
595 self._allow_colors = Namespace(**{
596 type_: not self.params.get('no_color') and supports_terminal_sequences(stream)
64fa820c 597 for type_, stream in self._out_files.items_ if type_ != 'console'
591bb9d3 598 })
819e0531 599
6929b41a 600 # The code is left like this to be reused for future deprecations
601 MIN_SUPPORTED, MIN_RECOMMENDED = (3, 7), (3, 7)
eff42759 602 current_version = sys.version_info[:2]
603 if current_version < MIN_RECOMMENDED:
9d339c41 604 msg = ('Support for Python version %d.%d has been deprecated. '
24093d52 605 'See https://github.com/yt-dlp/yt-dlp/issues/3764 for more details.'
c6e07cf1 606 '\n You will no longer receive updates on this version')
eff42759 607 if current_version < MIN_SUPPORTED:
608 msg = 'Python version %d.%d is no longer supported'
609 self.deprecation_warning(
610 f'{msg}! Please update to Python %d.%d or above' % (*current_version, *MIN_RECOMMENDED))
a61f4b28 611
88acdbc2 612 if self.params.get('allow_unplayable_formats'):
613 self.report_warning(
ec11a9f4 614 f'You have asked for {self._format_err("UNPLAYABLE", self.Styles.EMPHASIS)} formats to be listed/downloaded. '
819e0531 615 'This is a developer option intended for debugging. \n'
616 ' If you experience any issues while using this option, '
ec11a9f4 617 f'{self._format_err("DO NOT", self.Styles.ERROR)} open a bug report')
88acdbc2 618
be5df5ee
S
619 def check_deprecated(param, option, suggestion):
620 if self.params.get(param) is not None:
86e5f3ed 621 self.report_warning(f'{option} is deprecated. Use {suggestion} instead')
be5df5ee
S
622 return True
623 return False
624
625 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
38cce791
YCH
626 if self.params.get('geo_verification_proxy') is None:
627 self.params['geo_verification_proxy'] = self.params['cn_verification_proxy']
628
0d1bb027 629 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
630 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
53ed7066 631 check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"')
0d1bb027 632
49a57e70 633 for msg in self.params.get('_warnings', []):
0d1bb027 634 self.report_warning(msg)
ee8dd27a 635 for msg in self.params.get('_deprecation_warnings', []):
da4db748 636 self.deprecated_feature(msg)
0d1bb027 637
8a82af35 638 self.params['compat_opts'] = set(self.params.get('compat_opts', ()))
8a82af35 639 if 'list-formats' in self.params['compat_opts']:
ec11a9f4 640 self.params['listformats_table'] = False
641
b5ae35ee 642 if 'overwrites' not in self.params and self.params.get('nooverwrites') is not None:
b868936c 643 # nooverwrites was unnecessarily changed to overwrites
644 # in 0c3d0f51778b153f65c21906031c2e091fcfb641
645 # This ensures compatibility with both keys
646 self.params['overwrites'] = not self.params['nooverwrites']
b5ae35ee 647 elif self.params.get('overwrites') is None:
648 self.params.pop('overwrites', None)
b868936c 649 else:
650 self.params['nooverwrites'] = not self.params['overwrites']
b9d973be 651
455a15e2 652 self.params.setdefault('forceprint', {})
653 self.params.setdefault('print_to_file', {})
bb66c247 654
655 # Compatibility with older syntax
ca30f449 656 if not isinstance(params['forceprint'], dict):
455a15e2 657 self.params['forceprint'] = {'video': params['forceprint']}
ca30f449 658
455a15e2 659 if self.params.get('bidi_workaround', False):
1c088fa8
PH
660 try:
661 import pty
662 master, slave = pty.openpty()
ac668111 663 width = shutil.get_terminal_size().columns
591bb9d3 664 width_args = [] if width is None else ['-w', str(width)]
665 sp_kwargs = {'stdin': subprocess.PIPE, 'stdout': slave, 'stderr': self._out_files.error}
5d681e96 666 try:
d3c93ec2 667 self._output_process = Popen(['bidiv'] + width_args, **sp_kwargs)
5d681e96 668 except OSError:
d3c93ec2 669 self._output_process = Popen(['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
5d681e96 670 self._output_channel = os.fdopen(master, 'rb')
1c088fa8 671 except OSError as ose:
66e7ace1 672 if ose.errno == errno.ENOENT:
49a57e70 673 self.report_warning(
674 'Could not find fribidi executable, ignoring --bidi-workaround. '
675 'Make sure that fribidi is an executable file in one of the directories in your $PATH.')
1c088fa8
PH
676 else:
677 raise
0783b09b 678
97ec5bc5 679 if auto_init:
680 if auto_init != 'no_verbose_header':
681 self.print_debug_header()
682 self.add_default_info_extractors()
683
3089bc74
S
684 if (sys.platform != 'win32'
685 and sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
455a15e2 686 and not self.params.get('restrictfilenames', False)):
e9137224 687 # Unicode filesystem API will throw errors (#1474, #13027)
34308b30 688 self.report_warning(
6febd1c1 689 'Assuming --restrict-filenames since file system encoding '
1b725173 690 'cannot encode all characters. '
6febd1c1 691 'Set the LC_ALL environment variable to fix this.')
4a98cdbf 692 self.params['restrictfilenames'] = True
34308b30 693
bf1824b3 694 self._parse_outtmpl()
486dd09e 695
187986a8 696 # Creating format selector here allows us to catch syntax errors before the extraction
697 self.format_selector = (
fa9f30b8 698 self.params.get('format') if self.params.get('format') in (None, '-')
093a1710 699 else self.params['format'] if callable(self.params['format'])
187986a8 700 else self.build_format_selector(self.params['format']))
701
8b7539d2 702 # Set http_headers defaults according to std_headers
703 self.params['http_headers'] = merge_headers(std_headers, self.params.get('http_headers', {}))
704
013b50b7 705 hooks = {
706 'post_hooks': self.add_post_hook,
707 'progress_hooks': self.add_progress_hook,
708 'postprocessor_hooks': self.add_postprocessor_hook,
709 }
710 for opt, fn in hooks.items():
711 for ph in self.params.get(opt, []):
712 fn(ph)
71b640cc 713
5bfc8bee 714 for pp_def_raw in self.params.get('postprocessors', []):
715 pp_def = dict(pp_def_raw)
716 when = pp_def.pop('when', 'post_process')
717 self.add_post_processor(
f9934b96 718 get_postprocessor(pp_def.pop('key'))(self, **pp_def),
5bfc8bee 719 when=when)
720
97ec5bc5 721 self._setup_opener()
51fb4995
YCH
722 register_socks_protocols()
723
ed39cac5 724 def preload_download_archive(fn):
725 """Preload the archive, if any is specified"""
726 if fn is None:
727 return False
49a57e70 728 self.write_debug(f'Loading archive file {fn!r}')
ed39cac5 729 try:
730 with locked_file(fn, 'r', encoding='utf-8') as archive_file:
731 for line in archive_file:
732 self.archive.add(line.strip())
86e5f3ed 733 except OSError as ioe:
ed39cac5 734 if ioe.errno != errno.ENOENT:
735 raise
736 return False
737 return True
738
739 self.archive = set()
740 preload_download_archive(self.params.get('download_archive'))
741
7d4111ed
PH
742 def warn_if_short_id(self, argv):
743 # short YouTube ID starting with dash?
744 idxs = [
745 i for i, a in enumerate(argv)
746 if re.match(r'^-[0-9A-Za-z_-]{10}$', a)]
747 if idxs:
748 correct_argv = (
7a5c1cfe 749 ['yt-dlp']
3089bc74
S
750 + [a for i, a in enumerate(argv) if i not in idxs]
751 + ['--'] + [argv[i] for i in idxs]
7d4111ed
PH
752 )
753 self.report_warning(
754 'Long argument string detected. '
49a57e70 755 'Use -- to separate parameters and URLs, like this:\n%s' %
7d4111ed
PH
756 args_to_str(correct_argv))
757
8222d8de
JMF
758 def add_info_extractor(self, ie):
759 """Add an InfoExtractor object to the end of the list."""
8b7491c8 760 ie_key = ie.ie_key()
761 self._ies[ie_key] = ie
e52d7f85 762 if not isinstance(ie, type):
8b7491c8 763 self._ies_instances[ie_key] = ie
e52d7f85 764 ie.set_downloader(self)
8222d8de 765
56c73665
JMF
766 def get_info_extractor(self, ie_key):
767 """
768 Get an instance of an IE with name ie_key, it will try to get one from
769 the _ies list, if there's no instance it will create a new one and add
770 it to the extractor list.
771 """
772 ie = self._ies_instances.get(ie_key)
773 if ie is None:
774 ie = get_info_extractor(ie_key)()
775 self.add_info_extractor(ie)
776 return ie
777
023fa8c4
JMF
778 def add_default_info_extractors(self):
779 """
780 Add the InfoExtractors returned by gen_extractors to the end of the list
781 """
fe7866d0 782 all_ies = {ie.IE_NAME.lower(): ie for ie in gen_extractor_classes()}
783 all_ies['end'] = UnsupportedURLIE()
784 try:
785 ie_names = orderedSet_from_options(
786 self.params.get('allowed_extractors', ['default']), {
787 'all': list(all_ies),
788 'default': [name for name, ie in all_ies.items() if ie._ENABLED],
789 }, use_regex=True)
790 except re.error as e:
791 raise ValueError(f'Wrong regex for allowed_extractors: {e.pattern}')
792 for name in ie_names:
793 self.add_info_extractor(all_ies[name])
794 self.write_debug(f'Loaded {len(ie_names)} extractors')
023fa8c4 795
56d868db 796 def add_post_processor(self, pp, when='post_process'):
8222d8de 797 """Add a PostProcessor object to the end of the chain."""
8aa0e7cd 798 assert when in POSTPROCESS_WHEN, f'Invalid when={when}'
5bfa4862 799 self._pps[when].append(pp)
8222d8de
JMF
800 pp.set_downloader(self)
801
ab8e5e51
AM
802 def add_post_hook(self, ph):
803 """Add the post hook"""
804 self._post_hooks.append(ph)
805
933605d7 806 def add_progress_hook(self, ph):
819e0531 807 """Add the download progress hook"""
933605d7 808 self._progress_hooks.append(ph)
8ab470f1 809
819e0531 810 def add_postprocessor_hook(self, ph):
811 """Add the postprocessing progress hook"""
812 self._postprocessor_hooks.append(ph)
5bfc8bee 813 for pps in self._pps.values():
814 for pp in pps:
815 pp.add_progress_hook(ph)
819e0531 816
1c088fa8 817 def _bidi_workaround(self, message):
5d681e96 818 if not hasattr(self, '_output_channel'):
1c088fa8
PH
819 return message
820
5d681e96 821 assert hasattr(self, '_output_process')
14f25df2 822 assert isinstance(message, str)
6febd1c1 823 line_count = message.count('\n') + 1
0f06bcd7 824 self._output_process.stdin.write((message + '\n').encode())
5d681e96 825 self._output_process.stdin.flush()
0f06bcd7 826 res = ''.join(self._output_channel.readline().decode()
9e1a5b84 827 for _ in range(line_count))
6febd1c1 828 return res[:-len('\n')]
1c088fa8 829
b35496d8 830 def _write_string(self, message, out=None, only_once=False):
831 if only_once:
832 if message in self._printed_messages:
833 return
834 self._printed_messages.add(message)
835 write_string(message, out=out, encoding=self.params.get('encoding'))
734f90bb 836
cf4f42cb 837 def to_stdout(self, message, skip_eol=False, quiet=None):
0760b0a7 838 """Print message to stdout"""
cf4f42cb 839 if quiet is not None:
da4db748 840 self.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument quiet. '
841 'Use "YoutubeDL.to_screen" instead')
8a82af35 842 if skip_eol is not False:
da4db748 843 self.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument skip_eol. '
844 'Use "YoutubeDL.to_screen" instead')
0bf9dc1e 845 self._write_string(f'{self._bidi_workaround(message)}\n', self._out_files.out)
cf4f42cb 846
847 def to_screen(self, message, skip_eol=False, quiet=None):
848 """Print message to screen if not in quiet mode"""
8bf9319e 849 if self.params.get('logger'):
43afe285 850 self.params['logger'].debug(message)
cf4f42cb 851 return
852 if (self.params.get('quiet') if quiet is None else quiet) and not self.params.get('verbose'):
853 return
854 self._write_string(
855 '%s%s' % (self._bidi_workaround(message), ('' if skip_eol else '\n')),
591bb9d3 856 self._out_files.screen)
8222d8de 857
b35496d8 858 def to_stderr(self, message, only_once=False):
0760b0a7 859 """Print message to stderr"""
14f25df2 860 assert isinstance(message, str)
8bf9319e 861 if self.params.get('logger'):
43afe285
IB
862 self.params['logger'].error(message)
863 else:
5792c950 864 self._write_string(f'{self._bidi_workaround(message)}\n', self._out_files.error, only_once=only_once)
cf4f42cb 865
866 def _send_console_code(self, code):
591bb9d3 867 if compat_os_name == 'nt' or not self._out_files.console:
cf4f42cb 868 return
591bb9d3 869 self._write_string(code, self._out_files.console)
8222d8de 870
1e5b9a95
PH
871 def to_console_title(self, message):
872 if not self.params.get('consoletitle', False):
873 return
3efb96a6 874 message = remove_terminal_sequences(message)
4bede0d8
C
875 if compat_os_name == 'nt':
876 if ctypes.windll.kernel32.GetConsoleWindow():
877 # c_wchar_p() might not be necessary if `message` is
878 # already of type unicode()
879 ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
cf4f42cb 880 else:
881 self._send_console_code(f'\033]0;{message}\007')
1e5b9a95 882
bdde425c 883 def save_console_title(self):
cf4f42cb 884 if not self.params.get('consoletitle') or self.params.get('simulate'):
bdde425c 885 return
592b7485 886 self._send_console_code('\033[22;0t') # Save the title on stack
bdde425c
PH
887
888 def restore_console_title(self):
cf4f42cb 889 if not self.params.get('consoletitle') or self.params.get('simulate'):
bdde425c 890 return
592b7485 891 self._send_console_code('\033[23;0t') # Restore the title from stack
bdde425c
PH
892
893 def __enter__(self):
894 self.save_console_title()
895 return self
896
897 def __exit__(self, *args):
898 self.restore_console_title()
f89197d7 899
dca08720 900 if self.params.get('cookiefile') is not None:
1bab3437 901 self.cookiejar.save(ignore_discard=True, ignore_expires=True)
bdde425c 902
fa9f30b8 903 def trouble(self, message=None, tb=None, is_error=True):
8222d8de
JMF
904 """Determine action to take when a download problem appears.
905
906 Depending on if the downloader has been configured to ignore
907 download errors or not, this method may throw an exception or
908 not when errors are found, after printing the message.
909
fa9f30b8 910 @param tb If given, is additional traceback information
911 @param is_error Whether to raise error according to ignorerrors
8222d8de
JMF
912 """
913 if message is not None:
914 self.to_stderr(message)
915 if self.params.get('verbose'):
916 if tb is None:
917 if sys.exc_info()[0]: # if .trouble has been called from an except block
6febd1c1 918 tb = ''
8222d8de 919 if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
6febd1c1 920 tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
c0384f22 921 tb += encode_compat_str(traceback.format_exc())
8222d8de
JMF
922 else:
923 tb_data = traceback.format_list(traceback.extract_stack())
6febd1c1 924 tb = ''.join(tb_data)
c19bc311 925 if tb:
926 self.to_stderr(tb)
fa9f30b8 927 if not is_error:
928 return
b1940459 929 if not self.params.get('ignoreerrors'):
8222d8de
JMF
930 if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
931 exc_info = sys.exc_info()[1].exc_info
932 else:
933 exc_info = sys.exc_info()
934 raise DownloadError(message, exc_info)
935 self._download_retcode = 1
936
19a03940 937 Styles = Namespace(
938 HEADERS='yellow',
939 EMPHASIS='light blue',
492272fe 940 FILENAME='green',
19a03940 941 ID='green',
942 DELIM='blue',
943 ERROR='red',
944 WARNING='yellow',
945 SUPPRESS='light black',
946 )
ec11a9f4 947
7578d77d 948 def _format_text(self, handle, allow_colors, text, f, fallback=None, *, test_encoding=False):
e5a998f3 949 text = str(text)
ec11a9f4 950 if test_encoding:
951 original_text = text
5c104538 952 # handle.encoding can be None. See https://github.com/yt-dlp/yt-dlp/issues/2711
953 encoding = self.params.get('encoding') or getattr(handle, 'encoding', None) or 'ascii'
ec11a9f4 954 text = text.encode(encoding, 'ignore').decode(encoding)
955 if fallback is not None and text != original_text:
956 text = fallback
7578d77d 957 return format_text(text, f) if allow_colors else text if fallback is None else fallback
ec11a9f4 958
591bb9d3 959 def _format_out(self, *args, **kwargs):
960 return self._format_text(self._out_files.out, self._allow_colors.out, *args, **kwargs)
961
ec11a9f4 962 def _format_screen(self, *args, **kwargs):
591bb9d3 963 return self._format_text(self._out_files.screen, self._allow_colors.screen, *args, **kwargs)
ec11a9f4 964
965 def _format_err(self, *args, **kwargs):
591bb9d3 966 return self._format_text(self._out_files.error, self._allow_colors.error, *args, **kwargs)
819e0531 967
c84aeac6 968 def report_warning(self, message, only_once=False):
8222d8de
JMF
969 '''
970 Print the message to stderr, it will be prefixed with 'WARNING:'
971 If stderr is a tty file the 'WARNING:' will be colored
972 '''
6d07ce01
JMF
973 if self.params.get('logger') is not None:
974 self.params['logger'].warning(message)
8222d8de 975 else:
ad8915b7
PH
976 if self.params.get('no_warnings'):
977 return
ec11a9f4 978 self.to_stderr(f'{self._format_err("WARNING:", self.Styles.WARNING)} {message}', only_once)
8222d8de 979
da4db748 980 def deprecation_warning(self, message, *, stacklevel=0):
981 deprecation_warning(
982 message, stacklevel=stacklevel + 1, printer=self.report_error, is_error=False)
983
984 def deprecated_feature(self, message):
ee8dd27a 985 if self.params.get('logger') is not None:
da4db748 986 self.params['logger'].warning(f'Deprecated Feature: {message}')
987 self.to_stderr(f'{self._format_err("Deprecated Feature:", self.Styles.ERROR)} {message}', True)
ee8dd27a 988
fa9f30b8 989 def report_error(self, message, *args, **kwargs):
8222d8de
JMF
990 '''
991 Do the same as trouble, but prefixes the message with 'ERROR:', colored
992 in red if stderr is a tty file.
993 '''
fa9f30b8 994 self.trouble(f'{self._format_err("ERROR:", self.Styles.ERROR)} {message}', *args, **kwargs)
8222d8de 995
b35496d8 996 def write_debug(self, message, only_once=False):
0760b0a7 997 '''Log debug message or Print message to stderr'''
998 if not self.params.get('verbose', False):
999 return
8a82af35 1000 message = f'[debug] {message}'
0760b0a7 1001 if self.params.get('logger'):
1002 self.params['logger'].debug(message)
1003 else:
b35496d8 1004 self.to_stderr(message, only_once)
0760b0a7 1005
8222d8de
JMF
1006 def report_file_already_downloaded(self, file_name):
1007 """Report file has already been fully downloaded."""
1008 try:
6febd1c1 1009 self.to_screen('[download] %s has already been downloaded' % file_name)
ce02ed60 1010 except UnicodeEncodeError:
6febd1c1 1011 self.to_screen('[download] The file has already been downloaded')
8222d8de 1012
0c3d0f51 1013 def report_file_delete(self, file_name):
1014 """Report that existing file will be deleted."""
1015 try:
c25228e5 1016 self.to_screen('Deleting existing file %s' % file_name)
0c3d0f51 1017 except UnicodeEncodeError:
c25228e5 1018 self.to_screen('Deleting existing file')
0c3d0f51 1019
319b6059 1020 def raise_no_formats(self, info, forced=False, *, msg=None):
0a5a191a 1021 has_drm = info.get('_has_drm')
319b6059 1022 ignored, expected = self.params.get('ignore_no_formats_error'), bool(msg)
1023 msg = msg or has_drm and 'This video is DRM protected' or 'No video formats found!'
1024 if forced or not ignored:
1151c407 1025 raise ExtractorError(msg, video_id=info['id'], ie=info['extractor'],
319b6059 1026 expected=has_drm or ignored or expected)
88acdbc2 1027 else:
1028 self.report_warning(msg)
1029
de6000d9 1030 def parse_outtmpl(self):
bf1824b3 1031 self.deprecation_warning('"YoutubeDL.parse_outtmpl" is deprecated and may be removed in a future version')
1032 self._parse_outtmpl()
1033 return self.params['outtmpl']
1034
1035 def _parse_outtmpl(self):
7b2c3f47 1036 sanitize = IDENTITY
bf1824b3 1037 if self.params.get('restrictfilenames'): # Remove spaces in the default template
71ce444a 1038 sanitize = lambda x: x.replace(' - ', ' ').replace(' ', '-')
bf1824b3 1039
1040 outtmpl = self.params.setdefault('outtmpl', {})
1041 if not isinstance(outtmpl, dict):
1042 self.params['outtmpl'] = outtmpl = {'default': outtmpl}
1043 outtmpl.update({k: sanitize(v) for k, v in DEFAULT_OUTTMPL.items() if outtmpl.get(k) is None})
de6000d9 1044
21cd8fae 1045 def get_output_path(self, dir_type='', filename=None):
1046 paths = self.params.get('paths', {})
d2c8aadf 1047 assert isinstance(paths, dict), '"paths" parameter must be a dictionary'
21cd8fae 1048 path = os.path.join(
1049 expand_path(paths.get('home', '').strip()),
1050 expand_path(paths.get(dir_type, '').strip()) if dir_type else '',
1051 filename or '')
21cd8fae 1052 return sanitize_path(path, force=self.params.get('windowsfilenames'))
1053
76a264ac 1054 @staticmethod
901130bb 1055 def _outtmpl_expandpath(outtmpl):
1056 # expand_path translates '%%' into '%' and '$$' into '$'
1057 # correspondingly that is not what we want since we need to keep
1058 # '%%' intact for template dict substitution step. Working around
1059 # with boundary-alike separator hack.
1060 sep = ''.join([random.choice(ascii_letters) for _ in range(32)])
86e5f3ed 1061 outtmpl = outtmpl.replace('%%', f'%{sep}%').replace('$$', f'${sep}$')
901130bb 1062
1063 # outtmpl should be expand_path'ed before template dict substitution
1064 # because meta fields may contain env variables we don't want to
62b58c09 1065 # be expanded. E.g. for outtmpl "%(title)s.%(ext)s" and
901130bb 1066 # title "Hello $PATH", we don't want `$PATH` to be expanded.
1067 return expand_path(outtmpl).replace(sep, '')
1068
1069 @staticmethod
1070 def escape_outtmpl(outtmpl):
1071 ''' Escape any remaining strings like %s, %abc% etc. '''
1072 return re.sub(
1073 STR_FORMAT_RE_TMPL.format('', '(?![%(\0])'),
1074 lambda mobj: ('' if mobj.group('has_key') else '%') + mobj.group(0),
1075 outtmpl)
1076
1077 @classmethod
1078 def validate_outtmpl(cls, outtmpl):
76a264ac 1079 ''' @return None or Exception object '''
7d1eb38a 1080 outtmpl = re.sub(
47cdc68e 1081 STR_FORMAT_RE_TMPL.format('[^)]*', '[ljhqBUDS]'),
7d1eb38a 1082 lambda mobj: f'{mobj.group(0)[:-1]}s',
1083 cls._outtmpl_expandpath(outtmpl))
76a264ac 1084 try:
7d1eb38a 1085 cls.escape_outtmpl(outtmpl) % collections.defaultdict(int)
76a264ac 1086 return None
1087 except ValueError as err:
1088 return err
1089
03b4de72 1090 @staticmethod
1091 def _copy_infodict(info_dict):
1092 info_dict = dict(info_dict)
09b49e1f 1093 info_dict.pop('__postprocessors', None)
415f8d51 1094 info_dict.pop('__pending_error', None)
03b4de72 1095 return info_dict
1096
e0fd9573 1097 def prepare_outtmpl(self, outtmpl, info_dict, sanitize=False):
1098 """ Make the outtmpl and info_dict suitable for substitution: ydl.escape_outtmpl(outtmpl) % info_dict
1099 @param sanitize Whether to sanitize the output as a filename.
1100 For backward compatibility, a function can also be passed
1101 """
1102
6e84b215 1103 info_dict.setdefault('epoch', int(time.time())) # keep epoch consistent once set
143db31d 1104
03b4de72 1105 info_dict = self._copy_infodict(info_dict)
752cda38 1106 info_dict['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
53c18592 1107 formatSeconds(info_dict['duration'], '-' if sanitize else ':')
143db31d 1108 if info_dict.get('duration', None) is not None
1109 else None)
1d485a1a 1110 info_dict['autonumber'] = int(self.params.get('autonumber_start', 1) - 1 + self._num_downloads)
9c906919 1111 info_dict['video_autonumber'] = self._num_videos
752cda38 1112 if info_dict.get('resolution') is None:
1113 info_dict['resolution'] = self.format_resolution(info_dict, default=None)
143db31d 1114
e6f21b3d 1115 # For fields playlist_index, playlist_autonumber and autonumber convert all occurrences
143db31d 1116 # of %(field)s to %(field)0Nd for backward compatibility
1117 field_size_compat_map = {
0a5a191a 1118 'playlist_index': number_of_digits(info_dict.get('__last_playlist_index') or 0),
ec11a9f4 1119 'playlist_autonumber': number_of_digits(info_dict.get('n_entries') or 0),
752cda38 1120 'autonumber': self.params.get('autonumber_size') or 5,
143db31d 1121 }
752cda38 1122
385a27fa 1123 TMPL_DICT = {}
47cdc68e 1124 EXTERNAL_FORMAT_RE = re.compile(STR_FORMAT_RE_TMPL.format('[^)]*', f'[{STR_FORMAT_TYPES}ljhqBUDS]'))
385a27fa 1125 MATH_FUNCTIONS = {
1126 '+': float.__add__,
1127 '-': float.__sub__,
1128 }
e625be0d 1129 # Field is of the form key1.key2...
07a1250e 1130 # where keys (except first) can be string, int, slice or "{field, ...}"
1131 FIELD_INNER_RE = r'(?:\w+|%(num)s|%(num)s?(?::%(num)s?){1,2})' % {'num': r'(?:-?\d+)'}
1132 FIELD_RE = r'\w*(?:\.(?:%(inner)s|{%(field)s(?:,%(field)s)*}))*' % {
1133 'inner': FIELD_INNER_RE,
1134 'field': rf'\w*(?:\.{FIELD_INNER_RE})*'
1135 }
1d485a1a 1136 MATH_FIELD_RE = rf'(?:{FIELD_RE}|-?{NUMBER_RE})'
385a27fa 1137 MATH_OPERATORS_RE = r'(?:%s)' % '|'.join(map(re.escape, MATH_FUNCTIONS.keys()))
1d485a1a 1138 INTERNAL_FORMAT_RE = re.compile(rf'''(?x)
e625be0d 1139 (?P<negate>-)?
1d485a1a 1140 (?P<fields>{FIELD_RE})
1141 (?P<maths>(?:{MATH_OPERATORS_RE}{MATH_FIELD_RE})*)
e625be0d 1142 (?:>(?P<strf_format>.+?))?
34baa9fd 1143 (?P<remaining>
1144 (?P<alternate>(?<!\\),[^|&)]+)?
1145 (?:&(?P<replacement>.*?))?
1146 (?:\|(?P<default>.*?))?
1d485a1a 1147 )$''')
752cda38 1148
07a1250e 1149 def _traverse_infodict(fields):
1150 fields = [f for x in re.split(r'\.({.+?})\.?', fields)
1151 for f in ([x] if x.startswith('{') else x.split('.'))]
1152 for i in (0, -1):
1153 if fields and not fields[i]:
1154 fields.pop(i)
1155
1156 for i, f in enumerate(fields):
1157 if not f.startswith('{'):
1158 continue
1159 assert f.endswith('}'), f'No closing brace for {f} in {fields}'
1160 fields[i] = {k: k.split('.') for k in f[1:-1].split(',')}
1161
1162 return traverse_obj(info_dict, fields, is_user_input=True, traverse_string=True)
76a264ac 1163
752cda38 1164 def get_value(mdict):
1165 # Object traversal
2b8a2973 1166 value = _traverse_infodict(mdict['fields'])
752cda38 1167 # Negative
1168 if mdict['negate']:
1169 value = float_or_none(value)
1170 if value is not None:
1171 value *= -1
1172 # Do maths
385a27fa 1173 offset_key = mdict['maths']
1174 if offset_key:
752cda38 1175 value = float_or_none(value)
1176 operator = None
385a27fa 1177 while offset_key:
1178 item = re.match(
1179 MATH_FIELD_RE if operator else MATH_OPERATORS_RE,
1180 offset_key).group(0)
1181 offset_key = offset_key[len(item):]
1182 if operator is None:
752cda38 1183 operator = MATH_FUNCTIONS[item]
385a27fa 1184 continue
1185 item, multiplier = (item[1:], -1) if item[0] == '-' else (item, 1)
1186 offset = float_or_none(item)
1187 if offset is None:
2b8a2973 1188 offset = float_or_none(_traverse_infodict(item))
385a27fa 1189 try:
1190 value = operator(value, multiplier * offset)
1191 except (TypeError, ZeroDivisionError):
1192 return None
1193 operator = None
752cda38 1194 # Datetime formatting
1195 if mdict['strf_format']:
7c37ff97 1196 value = strftime_or_none(value, mdict['strf_format'].replace('\\,', ','))
752cda38 1197
a6bcaf71 1198 # XXX: Workaround for https://github.com/yt-dlp/yt-dlp/issues/4485
1199 if sanitize and value == '':
1200 value = None
752cda38 1201 return value
1202
b868936c 1203 na = self.params.get('outtmpl_na_placeholder', 'NA')
1204
e0fd9573 1205 def filename_sanitizer(key, value, restricted=self.params.get('restrictfilenames')):
5c3895ff 1206 return sanitize_filename(str(value), restricted=restricted, is_id=(
1207 bool(re.search(r'(^|[_.])id(\.|$)', key))
8a82af35 1208 if 'filename-sanitization' in self.params['compat_opts']
5c3895ff 1209 else NO_DEFAULT))
e0fd9573 1210
1211 sanitizer = sanitize if callable(sanitize) else filename_sanitizer
1212 sanitize = bool(sanitize)
1213
6e84b215 1214 def _dumpjson_default(obj):
1215 if isinstance(obj, (set, LazyList)):
1216 return list(obj)
adbc4ec4 1217 return repr(obj)
6e84b215 1218
752cda38 1219 def create_key(outer_mobj):
1220 if not outer_mobj.group('has_key'):
b836dc94 1221 return outer_mobj.group(0)
752cda38 1222 key = outer_mobj.group('key')
752cda38 1223 mobj = re.match(INTERNAL_FORMAT_RE, key)
e0fd9573 1224 initial_field = mobj.group('fields') if mobj else ''
e978789f 1225 value, replacement, default = None, None, na
7c37ff97 1226 while mobj:
e625be0d 1227 mobj = mobj.groupdict()
7c37ff97 1228 default = mobj['default'] if mobj['default'] is not None else default
752cda38 1229 value = get_value(mobj)
e978789f 1230 replacement = mobj['replacement']
7c37ff97 1231 if value is None and mobj['alternate']:
34baa9fd 1232 mobj = re.match(INTERNAL_FORMAT_RE, mobj['remaining'][1:])
7c37ff97 1233 else:
1234 break
752cda38 1235
b868936c 1236 fmt = outer_mobj.group('format')
752cda38 1237 if fmt == 's' and value is not None and key in field_size_compat_map.keys():
86e5f3ed 1238 fmt = f'0{field_size_compat_map[key]:d}d'
752cda38 1239
e978789f 1240 value = default if value is None else value if replacement is None else replacement
752cda38 1241
4476d2c7 1242 flags = outer_mobj.group('conversion') or ''
7d1eb38a 1243 str_fmt = f'{fmt[:-1]}s'
524e2e4f 1244 if fmt[-1] == 'l': # list
4476d2c7 1245 delim = '\n' if '#' in flags else ', '
9e907ebd 1246 value, fmt = delim.join(map(str, variadic(value, allowed_types=(str, bytes)))), str_fmt
524e2e4f 1247 elif fmt[-1] == 'j': # json
4476d2c7 1248 value, fmt = json.dumps(value, default=_dumpjson_default, indent=4 if '#' in flags else None), str_fmt
47cdc68e 1249 elif fmt[-1] == 'h': # html
1250 value, fmt = escapeHTML(value), str_fmt
524e2e4f 1251 elif fmt[-1] == 'q': # quoted
4476d2c7 1252 value = map(str, variadic(value) if '#' in flags else [value])
1253 value, fmt = ' '.join(map(compat_shlex_quote, value)), str_fmt
524e2e4f 1254 elif fmt[-1] == 'B': # bytes
0f06bcd7 1255 value = f'%{str_fmt}'.encode() % str(value).encode()
f5aa5cfb 1256 value, fmt = value.decode('utf-8', 'ignore'), 's'
524e2e4f 1257 elif fmt[-1] == 'U': # unicode normalized
524e2e4f 1258 value, fmt = unicodedata.normalize(
1259 # "+" = compatibility equivalence, "#" = NFD
4476d2c7 1260 'NF%s%s' % ('K' if '+' in flags else '', 'D' if '#' in flags else 'C'),
524e2e4f 1261 value), str_fmt
e0fd9573 1262 elif fmt[-1] == 'D': # decimal suffix
abbeeebc 1263 num_fmt, fmt = fmt[:-1].replace('#', ''), 's'
1264 value = format_decimal_suffix(value, f'%{num_fmt}f%s' if num_fmt else '%d%s',
1265 factor=1024 if '#' in flags else 1000)
37893bb0 1266 elif fmt[-1] == 'S': # filename sanitization
e0fd9573 1267 value, fmt = filename_sanitizer(initial_field, value, restricted='#' in flags), str_fmt
7d1eb38a 1268 elif fmt[-1] == 'c':
524e2e4f 1269 if value:
1270 value = str(value)[0]
76a264ac 1271 else:
524e2e4f 1272 fmt = str_fmt
76a264ac 1273 elif fmt[-1] not in 'rs': # numeric
a439a3a4 1274 value = float_or_none(value)
752cda38 1275 if value is None:
1276 value, fmt = default, 's'
901130bb 1277
752cda38 1278 if sanitize:
1279 if fmt[-1] == 'r':
1280 # If value is an object, sanitize might convert it to a string
1281 # So we convert it to repr first
7d1eb38a 1282 value, fmt = repr(value), str_fmt
639f1cea 1283 if fmt[-1] in 'csr':
e0fd9573 1284 value = sanitizer(initial_field, value)
901130bb 1285
b868936c 1286 key = '%s\0%s' % (key.replace('%', '%\0'), outer_mobj.group('format'))
385a27fa 1287 TMPL_DICT[key] = value
b868936c 1288 return '{prefix}%({key}){fmt}'.format(key=key, fmt=fmt, prefix=outer_mobj.group('prefix'))
752cda38 1289
385a27fa 1290 return EXTERNAL_FORMAT_RE.sub(create_key, outtmpl), TMPL_DICT
143db31d 1291
819e0531 1292 def evaluate_outtmpl(self, outtmpl, info_dict, *args, **kwargs):
1293 outtmpl, info_dict = self.prepare_outtmpl(outtmpl, info_dict, *args, **kwargs)
1294 return self.escape_outtmpl(outtmpl) % info_dict
1295
5127e92a 1296 def _prepare_filename(self, info_dict, *, outtmpl=None, tmpl_type=None):
1297 assert None in (outtmpl, tmpl_type), 'outtmpl and tmpl_type are mutually exclusive'
1298 if outtmpl is None:
bf1824b3 1299 outtmpl = self.params['outtmpl'].get(tmpl_type or 'default', self.params['outtmpl']['default'])
8222d8de 1300 try:
5127e92a 1301 outtmpl = self._outtmpl_expandpath(outtmpl)
e0fd9573 1302 filename = self.evaluate_outtmpl(outtmpl, info_dict, True)
6a0546e3 1303 if not filename:
1304 return None
15da37c7 1305
5127e92a 1306 if tmpl_type in ('', 'temp'):
6a0546e3 1307 final_ext, ext = self.params.get('final_ext'), info_dict.get('ext')
1308 if final_ext and ext and final_ext != ext and filename.endswith(f'.{final_ext}'):
1309 filename = replace_extension(filename, ext, final_ext)
5127e92a 1310 elif tmpl_type:
6a0546e3 1311 force_ext = OUTTMPL_TYPES[tmpl_type]
1312 if force_ext:
1313 filename = replace_extension(filename, force_ext, info_dict.get('ext'))
de6000d9 1314
bdc3fd2f
U
1315 # https://github.com/blackjack4494/youtube-dlc/issues/85
1316 trim_file_name = self.params.get('trim_file_name', False)
1317 if trim_file_name:
5c22c63d 1318 no_ext, *ext = filename.rsplit('.', 2)
1319 filename = join_nonempty(no_ext[:trim_file_name], *ext, delim='.')
bdc3fd2f 1320
0202b52a 1321 return filename
8222d8de 1322 except ValueError as err:
6febd1c1 1323 self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')')
8222d8de
JMF
1324 return None
1325
5127e92a 1326 def prepare_filename(self, info_dict, dir_type='', *, outtmpl=None, warn=False):
1327 """Generate the output filename"""
1328 if outtmpl:
1329 assert not dir_type, 'outtmpl and dir_type are mutually exclusive'
1330 dir_type = None
1331 filename = self._prepare_filename(info_dict, tmpl_type=dir_type, outtmpl=outtmpl)
80c03fa9 1332 if not filename and dir_type not in ('', 'temp'):
1333 return ''
de6000d9 1334
c84aeac6 1335 if warn:
21cd8fae 1336 if not self.params.get('paths'):
de6000d9 1337 pass
1338 elif filename == '-':
c84aeac6 1339 self.report_warning('--paths is ignored when an outputting to stdout', only_once=True)
de6000d9 1340 elif os.path.isabs(filename):
c84aeac6 1341 self.report_warning('--paths is ignored since an absolute path is given in output template', only_once=True)
de6000d9 1342 if filename == '-' or not filename:
1343 return filename
1344
21cd8fae 1345 return self.get_output_path(dir_type, filename)
0202b52a 1346
120fe513 1347 def _match_entry(self, info_dict, incomplete=False, silent=False):
ecdec191 1348 """ Returns None if the file should be downloaded """
8222d8de 1349
3bec830a 1350 video_title = info_dict.get('title', info_dict.get('id', 'entry'))
c77495e3 1351
8b0d7497 1352 def check_filter():
8b0d7497 1353 if 'title' in info_dict:
1354 # This can happen when we're just evaluating the playlist
1355 title = info_dict['title']
1356 matchtitle = self.params.get('matchtitle', False)
1357 if matchtitle:
1358 if not re.search(matchtitle, title, re.IGNORECASE):
1359 return '"' + title + '" title did not match pattern "' + matchtitle + '"'
1360 rejecttitle = self.params.get('rejecttitle', False)
1361 if rejecttitle:
1362 if re.search(rejecttitle, title, re.IGNORECASE):
1363 return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
1364 date = info_dict.get('upload_date')
1365 if date is not None:
1366 dateRange = self.params.get('daterange', DateRange())
1367 if date not in dateRange:
86e5f3ed 1368 return f'{date_from_str(date).isoformat()} upload date is not in range {dateRange}'
8b0d7497 1369 view_count = info_dict.get('view_count')
1370 if view_count is not None:
1371 min_views = self.params.get('min_views')
1372 if min_views is not None and view_count < min_views:
1373 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
1374 max_views = self.params.get('max_views')
1375 if max_views is not None and view_count > max_views:
1376 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
1377 if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')):
1378 return 'Skipping "%s" because it is age restricted' % video_title
8b0d7497 1379
8f18aca8 1380 match_filter = self.params.get('match_filter')
1381 if match_filter is not None:
1382 try:
1383 ret = match_filter(info_dict, incomplete=incomplete)
1384 except TypeError:
1385 # For backward compatibility
1386 ret = None if incomplete else match_filter(info_dict)
492272fe 1387 if ret is NO_DEFAULT:
1388 while True:
1389 filename = self._format_screen(self.prepare_filename(info_dict), self.Styles.FILENAME)
1390 reply = input(self._format_screen(
1391 f'Download "{filename}"? (Y/n): ', self.Styles.EMPHASIS)).lower().strip()
1392 if reply in {'y', ''}:
1393 return None
1394 elif reply == 'n':
1395 return f'Skipping {video_title}'
492272fe 1396 elif ret is not None:
8f18aca8 1397 return ret
8b0d7497 1398 return None
1399
c77495e3 1400 if self.in_download_archive(info_dict):
1401 reason = '%s has already been recorded in the archive' % video_title
1402 break_opt, break_err = 'break_on_existing', ExistingVideoReached
1403 else:
1404 reason = check_filter()
1405 break_opt, break_err = 'break_on_reject', RejectedVideoReached
8b0d7497 1406 if reason is not None:
120fe513 1407 if not silent:
1408 self.to_screen('[download] ' + reason)
c77495e3 1409 if self.params.get(break_opt, False):
1410 raise break_err()
8b0d7497 1411 return reason
fe7e0c98 1412
b6c45014
JMF
1413 @staticmethod
1414 def add_extra_info(info_dict, extra_info):
1415 '''Set the keys from extra_info in info dict if they are missing'''
1416 for key, value in extra_info.items():
1417 info_dict.setdefault(key, value)
1418
409e1828 1419 def extract_info(self, url, download=True, ie_key=None, extra_info=None,
61aa5ba3 1420 process=True, force_generic_extractor=False):
41d1cca3 1421 """
17ffed18 1422 Extract and return the information dictionary of the URL
41d1cca3 1423
1424 Arguments:
17ffed18 1425 @param url URL to extract
41d1cca3 1426
1427 Keyword arguments:
17ffed18 1428 @param download Whether to download videos
1429 @param process Whether to resolve all unresolved references (URLs, playlist items).
1430 Must be True for download to work
1431 @param ie_key Use only the extractor with this key
1432
1433 @param extra_info Dictionary containing the extra values to add to the info (For internal use only)
1434 @force_generic_extractor Force using the generic extractor (Deprecated; use ie_key='Generic')
41d1cca3 1435 """
fe7e0c98 1436
409e1828 1437 if extra_info is None:
1438 extra_info = {}
1439
61aa5ba3 1440 if not ie_key and force_generic_extractor:
d22dec74
S
1441 ie_key = 'Generic'
1442
8222d8de 1443 if ie_key:
fe7866d0 1444 ies = {ie_key: self._ies[ie_key]} if ie_key in self._ies else {}
8222d8de
JMF
1445 else:
1446 ies = self._ies
1447
fe7866d0 1448 for key, ie in ies.items():
8222d8de
JMF
1449 if not ie.suitable(url):
1450 continue
1451
1452 if not ie.working():
6febd1c1
PH
1453 self.report_warning('The program functionality for this site has been marked as broken, '
1454 'and will probably not work.')
8222d8de 1455
1151c407 1456 temp_id = ie.get_temp_id(url)
fe7866d0 1457 if temp_id is not None and self.in_download_archive({'id': temp_id, 'ie_key': key}):
1458 self.to_screen(f'[{key}] {temp_id}: has already been recorded in the archive')
5e5be0c0 1459 if self.params.get('break_on_existing', False):
1460 raise ExistingVideoReached()
a0566bbf 1461 break
fe7866d0 1462 return self.__extract_info(url, self.get_info_extractor(key), download, extra_info, process)
a0566bbf 1463 else:
fe7866d0 1464 extractors_restricted = self.params.get('allowed_extractors') not in (None, ['default'])
1465 self.report_error(f'No suitable extractor{format_field(ie_key, None, " (%s)")} found for URL {url}',
1466 tb=False if extractors_restricted else None)
a0566bbf 1467
7e88d7d7 1468 def _handle_extraction_exceptions(func):
b5ae35ee 1469 @functools.wraps(func)
a0566bbf 1470 def wrapper(self, *args, **kwargs):
6da22e7d 1471 while True:
1472 try:
1473 return func(self, *args, **kwargs)
1474 except (DownloadCancelled, LazyList.IndexError, PagedList.IndexError):
8222d8de 1475 raise
6da22e7d 1476 except ReExtractInfo as e:
1477 if e.expected:
1478 self.to_screen(f'{e}; Re-extracting data')
1479 else:
1480 self.to_stderr('\r')
1481 self.report_warning(f'{e}; Re-extracting data')
1482 continue
1483 except GeoRestrictedError as e:
1484 msg = e.msg
1485 if e.countries:
1486 msg += '\nThis video is available in %s.' % ', '.join(
1487 map(ISO3166Utils.short2full, e.countries))
1488 msg += '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1489 self.report_error(msg)
1490 except ExtractorError as e: # An error we somewhat expected
1491 self.report_error(str(e), e.format_traceback())
1492 except Exception as e:
1493 if self.params.get('ignoreerrors'):
1494 self.report_error(str(e), tb=encode_compat_str(traceback.format_exc()))
1495 else:
1496 raise
1497 break
a0566bbf 1498 return wrapper
1499
693f0600 1500 def _wait_for_video(self, ie_result={}):
f2ebc5c7 1501 if (not self.params.get('wait_for_video')
1502 or ie_result.get('_type', 'video') != 'video'
1503 or ie_result.get('formats') or ie_result.get('url')):
1504 return
1505
1506 format_dur = lambda dur: '%02d:%02d:%02d' % timetuple_from_msec(dur * 1000)[:-1]
1507 last_msg = ''
1508
1509 def progress(msg):
1510 nonlocal last_msg
a7dc6a89 1511 full_msg = f'{msg}\n'
1512 if not self.params.get('noprogress'):
1513 full_msg = msg + ' ' * (len(last_msg) - len(msg)) + '\r'
1514 elif last_msg:
1515 return
1516 self.to_screen(full_msg, skip_eol=True)
f2ebc5c7 1517 last_msg = msg
1518
1519 min_wait, max_wait = self.params.get('wait_for_video')
1520 diff = try_get(ie_result, lambda x: x['release_timestamp'] - time.time())
1521 if diff is None and ie_result.get('live_status') == 'is_upcoming':
16c620bc 1522 diff = round(random.uniform(min_wait, max_wait) if (max_wait and min_wait) else (max_wait or min_wait), 0)
f2ebc5c7 1523 self.report_warning('Release time of video is not known')
693f0600 1524 elif ie_result and (diff or 0) <= 0:
f2ebc5c7 1525 self.report_warning('Video should already be available according to extracted info')
38d79fd1 1526 diff = min(max(diff or 0, min_wait or 0), max_wait or float('inf'))
f2ebc5c7 1527 self.to_screen(f'[wait] Waiting for {format_dur(diff)} - Press Ctrl+C to try now')
1528
1529 wait_till = time.time() + diff
1530 try:
1531 while True:
1532 diff = wait_till - time.time()
1533 if diff <= 0:
1534 progress('')
1535 raise ReExtractInfo('[wait] Wait period ended', expected=True)
1536 progress(f'[wait] Remaining time until next attempt: {self._format_screen(format_dur(diff), self.Styles.EMPHASIS)}')
1537 time.sleep(1)
1538 except KeyboardInterrupt:
1539 progress('')
1540 raise ReExtractInfo('[wait] Interrupted by user', expected=True)
1541 except BaseException as e:
1542 if not isinstance(e, ReExtractInfo):
1543 self.to_screen('')
1544 raise
1545
7e88d7d7 1546 @_handle_extraction_exceptions
58f197b7 1547 def __extract_info(self, url, ie, download, extra_info, process):
693f0600 1548 try:
1549 ie_result = ie.extract(url)
1550 except UserNotLive as e:
1551 if process:
1552 if self.params.get('wait_for_video'):
1553 self.report_warning(e)
1554 self._wait_for_video()
1555 raise
a0566bbf 1556 if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
cb794ee0 1557 self.report_warning(f'Extractor {ie.IE_NAME} returned nothing{bug_reports_message()}')
a0566bbf 1558 return
1559 if isinstance(ie_result, list):
1560 # Backwards compatibility: old IE result format
1561 ie_result = {
1562 '_type': 'compat_list',
1563 'entries': ie_result,
1564 }
e37d0efb 1565 if extra_info.get('original_url'):
1566 ie_result.setdefault('original_url', extra_info['original_url'])
a0566bbf 1567 self.add_default_extra_info(ie_result, ie, url)
1568 if process:
f2ebc5c7 1569 self._wait_for_video(ie_result)
a0566bbf 1570 return self.process_ie_result(ie_result, download, extra_info)
8222d8de 1571 else:
a0566bbf 1572 return ie_result
fe7e0c98 1573
ea38e55f 1574 def add_default_extra_info(self, ie_result, ie, url):
6033d980 1575 if url is not None:
1576 self.add_extra_info(ie_result, {
1577 'webpage_url': url,
1578 'original_url': url,
57ebfca3 1579 })
1580 webpage_url = ie_result.get('webpage_url')
1581 if webpage_url:
1582 self.add_extra_info(ie_result, {
1583 'webpage_url_basename': url_basename(webpage_url),
1584 'webpage_url_domain': get_domain(webpage_url),
6033d980 1585 })
1586 if ie is not None:
1587 self.add_extra_info(ie_result, {
1588 'extractor': ie.IE_NAME,
1589 'extractor_key': ie.ie_key(),
1590 })
ea38e55f 1591
58adec46 1592 def process_ie_result(self, ie_result, download=True, extra_info=None):
8222d8de
JMF
1593 """
1594 Take the result of the ie(may be modified) and resolve all unresolved
1595 references (URLs, playlist items).
1596
1597 It will also download the videos if 'download'.
1598 Returns the resolved ie_result.
1599 """
58adec46 1600 if extra_info is None:
1601 extra_info = {}
e8ee972c
PH
1602 result_type = ie_result.get('_type', 'video')
1603
057a5206 1604 if result_type in ('url', 'url_transparent'):
8f97a15d 1605 ie_result['url'] = sanitize_url(
1606 ie_result['url'], scheme='http' if self.params.get('prefer_insecure') else 'https')
e37d0efb 1607 if ie_result.get('original_url'):
1608 extra_info.setdefault('original_url', ie_result['original_url'])
1609
057a5206 1610 extract_flat = self.params.get('extract_flat', False)
3089bc74
S
1611 if ((extract_flat == 'in_playlist' and 'playlist' in extra_info)
1612 or extract_flat is True):
ecb54191 1613 info_copy = ie_result.copy()
6033d980 1614 ie = try_get(ie_result.get('ie_key'), self.get_info_extractor)
360167b9 1615 if ie and not ie_result.get('id'):
4614bc22 1616 info_copy['id'] = ie.get_temp_id(ie_result['url'])
6033d980 1617 self.add_default_extra_info(info_copy, ie, ie_result['url'])
4614bc22 1618 self.add_extra_info(info_copy, extra_info)
b5475f11 1619 info_copy, _ = self.pre_process(info_copy)
ecb54191 1620 self.__forced_printings(info_copy, self.prepare_filename(info_copy), incomplete=True)
415f8d51 1621 self._raise_pending_errors(info_copy)
4614bc22 1622 if self.params.get('force_write_download_archive', False):
1623 self.record_download_archive(info_copy)
e8ee972c
PH
1624 return ie_result
1625
8222d8de 1626 if result_type == 'video':
b6c45014 1627 self.add_extra_info(ie_result, extra_info)
9c2b75b5 1628 ie_result = self.process_video_result(ie_result, download=download)
415f8d51 1629 self._raise_pending_errors(ie_result)
28b0eb0f 1630 additional_urls = (ie_result or {}).get('additional_urls')
9c2b75b5 1631 if additional_urls:
e9f4ccd1 1632 # TODO: Improve MetadataParserPP to allow setting a list
14f25df2 1633 if isinstance(additional_urls, str):
9c2b75b5 1634 additional_urls = [additional_urls]
1635 self.to_screen(
1636 '[info] %s: %d additional URL(s) requested' % (ie_result['id'], len(additional_urls)))
1637 self.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls))
1638 ie_result['additional_entries'] = [
1639 self.extract_info(
b69fd25c 1640 url, download, extra_info=extra_info,
9c2b75b5 1641 force_generic_extractor=self.params.get('force_generic_extractor'))
1642 for url in additional_urls
1643 ]
1644 return ie_result
8222d8de
JMF
1645 elif result_type == 'url':
1646 # We have to add extra_info to the results because it may be
1647 # contained in a playlist
07cce701 1648 return self.extract_info(
1649 ie_result['url'], download,
1650 ie_key=ie_result.get('ie_key'),
1651 extra_info=extra_info)
7fc3fa05
PH
1652 elif result_type == 'url_transparent':
1653 # Use the information from the embedding page
1654 info = self.extract_info(
1655 ie_result['url'], ie_key=ie_result.get('ie_key'),
1656 extra_info=extra_info, download=False, process=False)
1657
1640eb09
S
1658 # extract_info may return None when ignoreerrors is enabled and
1659 # extraction failed with an error, don't crash and return early
1660 # in this case
1661 if not info:
1662 return info
1663
3975b4d2 1664 exempted_fields = {'_type', 'url', 'ie_key'}
1665 if not ie_result.get('section_end') and ie_result.get('section_start') is None:
1666 # For video clips, the id etc of the clip extractor should be used
1667 exempted_fields |= {'id', 'extractor', 'extractor_key'}
1668
412c617d 1669 new_result = info.copy()
3975b4d2 1670 new_result.update(filter_dict(ie_result, lambda k, v: v is not None and k not in exempted_fields))
7fc3fa05 1671
0563f7ac
S
1672 # Extracted info may not be a video result (i.e.
1673 # info.get('_type', 'video') != video) but rather an url or
1674 # url_transparent. In such cases outer metadata (from ie_result)
1675 # should be propagated to inner one (info). For this to happen
1676 # _type of info should be overridden with url_transparent. This
067aa17e 1677 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
0563f7ac
S
1678 if new_result.get('_type') == 'url':
1679 new_result['_type'] = 'url_transparent'
7fc3fa05
PH
1680
1681 return self.process_ie_result(
1682 new_result, download=download, extra_info=extra_info)
40fcba5e 1683 elif result_type in ('playlist', 'multi_video'):
30a074c2 1684 # Protect from infinite recursion due to recursively nested playlists
1685 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
1686 webpage_url = ie_result['webpage_url']
1687 if webpage_url in self._playlist_urls:
7e85e872 1688 self.to_screen(
30a074c2 1689 '[download] Skipping already downloaded playlist: %s'
1690 % ie_result.get('title') or ie_result.get('id'))
1691 return
7e85e872 1692
30a074c2 1693 self._playlist_level += 1
1694 self._playlist_urls.add(webpage_url)
03f83004 1695 self._fill_common_fields(ie_result, False)
bc516a3f 1696 self._sanitize_thumbnails(ie_result)
30a074c2 1697 try:
1698 return self.__process_playlist(ie_result, download)
1699 finally:
1700 self._playlist_level -= 1
1701 if not self._playlist_level:
1702 self._playlist_urls.clear()
8222d8de 1703 elif result_type == 'compat_list':
c9bf4114
PH
1704 self.report_warning(
1705 'Extractor %s returned a compat_list result. '
1706 'It needs to be updated.' % ie_result.get('extractor'))
5f6a1245 1707
8222d8de 1708 def _fixup(r):
b868936c 1709 self.add_extra_info(r, {
1710 'extractor': ie_result['extractor'],
1711 'webpage_url': ie_result['webpage_url'],
1712 'webpage_url_basename': url_basename(ie_result['webpage_url']),
0bb322b9 1713 'webpage_url_domain': get_domain(ie_result['webpage_url']),
b868936c 1714 'extractor_key': ie_result['extractor_key'],
1715 })
8222d8de
JMF
1716 return r
1717 ie_result['entries'] = [
b6c45014 1718 self.process_ie_result(_fixup(r), download, extra_info)
8222d8de
JMF
1719 for r in ie_result['entries']
1720 ]
1721 return ie_result
1722 else:
1723 raise Exception('Invalid result type: %s' % result_type)
1724
e92caff5 1725 def _ensure_dir_exists(self, path):
1726 return make_dir(path, self.report_error)
1727
3b603dbd 1728 @staticmethod
3bec830a 1729 def _playlist_infodict(ie_result, strict=False, **kwargs):
1730 info = {
1731 'playlist_count': ie_result.get('playlist_count'),
3b603dbd 1732 'playlist': ie_result.get('title') or ie_result.get('id'),
1733 'playlist_id': ie_result.get('id'),
1734 'playlist_title': ie_result.get('title'),
1735 'playlist_uploader': ie_result.get('uploader'),
1736 'playlist_uploader_id': ie_result.get('uploader_id'),
3b603dbd 1737 **kwargs,
1738 }
3bec830a 1739 if strict:
1740 return info
1741 return {
1742 **info,
1743 'playlist_index': 0,
1744 '__last_playlist_index': max(ie_result['requested_entries'] or (0, 0)),
1745 'extractor': ie_result['extractor'],
1746 'webpage_url': ie_result['webpage_url'],
1747 'webpage_url_basename': url_basename(ie_result['webpage_url']),
1748 'webpage_url_domain': get_domain(ie_result['webpage_url']),
1749 'extractor_key': ie_result['extractor_key'],
1750 }
3b603dbd 1751
30a074c2 1752 def __process_playlist(self, ie_result, download):
7e88d7d7 1753 """Process each entry in the playlist"""
f5ea4748 1754 assert ie_result['_type'] in ('playlist', 'multi_video')
1755
3bec830a 1756 common_info = self._playlist_infodict(ie_result, strict=True)
3955b207 1757 title = common_info.get('playlist') or '<Untitled>'
3bec830a 1758 if self._match_entry(common_info, incomplete=True) is not None:
1759 return
c6e07cf1 1760 self.to_screen(f'[download] Downloading {ie_result["_type"]}: {title}')
f0d785d3 1761
7e88d7d7 1762 all_entries = PlaylistEntries(self, ie_result)
7e9a6125 1763 entries = orderedSet(all_entries.get_requested_items(), lazy=True)
1764
1765 lazy = self.params.get('lazy_playlist')
1766 if lazy:
1767 resolved_entries, n_entries = [], 'N/A'
1768 ie_result['requested_entries'], ie_result['entries'] = None, None
1769 else:
1770 entries = resolved_entries = list(entries)
1771 n_entries = len(resolved_entries)
1772 ie_result['requested_entries'], ie_result['entries'] = tuple(zip(*resolved_entries)) or ([], [])
1773 if not ie_result.get('playlist_count'):
1774 # Better to do this after potentially exhausting entries
1775 ie_result['playlist_count'] = all_entries.get_full_count()
498f5606 1776
0647d925 1777 extra = self._playlist_infodict(ie_result, n_entries=int_or_none(n_entries))
1778 ie_copy = collections.ChainMap(ie_result, extra)
3bec830a 1779
e08a85d8 1780 _infojson_written = False
0bfc53d0 1781 write_playlist_files = self.params.get('allow_playlist_files', True)
1782 if write_playlist_files and self.params.get('list_thumbnails'):
1783 self.list_thumbnails(ie_result)
1784 if write_playlist_files and not self.params.get('simulate'):
e08a85d8 1785 _infojson_written = self._write_info_json(
1786 'playlist', ie_result, self.prepare_filename(ie_copy, 'pl_infojson'))
1787 if _infojson_written is None:
80c03fa9 1788 return
1789 if self._write_description('playlist', ie_result,
1790 self.prepare_filename(ie_copy, 'pl_description')) is None:
1791 return
681de68e 1792 # TODO: This should be passed to ThumbnailsConvertor if necessary
3bec830a 1793 self._write_thumbnails('playlist', ie_result, self.prepare_filename(ie_copy, 'pl_thumbnail'))
30a074c2 1794
7e9a6125 1795 if lazy:
1796 if self.params.get('playlistreverse') or self.params.get('playlistrandom'):
1797 self.report_warning('playlistreverse and playlistrandom are not supported with lazy_playlist', only_once=True)
1798 elif self.params.get('playlistreverse'):
1799 entries.reverse()
1800 elif self.params.get('playlistrandom'):
30a074c2 1801 random.shuffle(entries)
1802
7e88d7d7 1803 self.to_screen(f'[{ie_result["extractor"]}] Playlist {title}: Downloading {n_entries} videos'
1804 f'{format_field(ie_result, "playlist_count", " of %s")}')
30a074c2 1805
134c913c 1806 keep_resolved_entries = self.params.get('extract_flat') != 'discard'
1807 if self.params.get('extract_flat') == 'discard_in_playlist':
1808 keep_resolved_entries = ie_result['_type'] != 'playlist'
1809 if keep_resolved_entries:
1810 self.write_debug('The information of all playlist entries will be held in memory')
1811
26e2805c 1812 failures = 0
1813 max_failures = self.params.get('skip_playlist_after_errors') or float('inf')
7e9a6125 1814 for i, (playlist_index, entry) in enumerate(entries):
1815 if lazy:
1816 resolved_entries.append((playlist_index, entry))
3bec830a 1817 if not entry:
7e88d7d7 1818 continue
1819
7e88d7d7 1820 entry['__x_forwarded_for_ip'] = ie_result.get('__x_forwarded_for_ip')
7e9a6125 1821 if not lazy and 'playlist-index' in self.params.get('compat_opts', []):
1822 playlist_index = ie_result['requested_entries'][i]
1823
0647d925 1824 entry_copy = collections.ChainMap(entry, {
3bec830a 1825 **common_info,
3955b207 1826 'n_entries': int_or_none(n_entries),
71729754 1827 'playlist_index': playlist_index,
7e9a6125 1828 'playlist_autonumber': i + 1,
0647d925 1829 })
3bec830a 1830
0647d925 1831 if self._match_entry(entry_copy, incomplete=True) is not None:
f0ad6f8c 1832 # For compatabilty with youtube-dl. See https://github.com/yt-dlp/yt-dlp/issues/4369
1833 resolved_entries[i] = (playlist_index, NO_DEFAULT)
3bec830a 1834 continue
1835
1836 self.to_screen('[download] Downloading video %s of %s' % (
1837 self._format_screen(i + 1, self.Styles.ID), self._format_screen(n_entries, self.Styles.EMPHASIS)))
1838
a6ca61d4 1839 extra.update({
1840 'playlist_index': playlist_index,
1841 'playlist_autonumber': i + 1,
1842 })
3bec830a 1843 entry_result = self.__process_iterable_entry(entry, download, extra)
26e2805c 1844 if not entry_result:
1845 failures += 1
1846 if failures >= max_failures:
1847 self.report_error(
7e88d7d7 1848 f'Skipping the remaining entries in playlist "{title}" since {failures} items failed extraction')
26e2805c 1849 break
134c913c 1850 if keep_resolved_entries:
1851 resolved_entries[i] = (playlist_index, entry_result)
7e88d7d7 1852
1853 # Update with processed data
f0ad6f8c 1854 ie_result['requested_entries'] = [i for i, e in resolved_entries if e is not NO_DEFAULT]
1855 ie_result['entries'] = [e for _, e in resolved_entries if e is not NO_DEFAULT]
e08a85d8 1856
1857 # Write the updated info to json
cb96c5be 1858 if _infojson_written is True and self._write_info_json(
e08a85d8 1859 'updated playlist', ie_result,
1860 self.prepare_filename(ie_copy, 'pl_infojson'), overwrite=True) is None:
1861 return
ca30f449 1862
ed5835b4 1863 ie_result = self.run_all_pps('playlist', ie_result)
7e88d7d7 1864 self.to_screen(f'[download] Finished downloading playlist: {title}')
30a074c2 1865 return ie_result
1866
7e88d7d7 1867 @_handle_extraction_exceptions
a0566bbf 1868 def __process_iterable_entry(self, entry, download, extra_info):
1869 return self.process_ie_result(
1870 entry, download=download, extra_info=extra_info)
1871
67134eab
JMF
1872 def _build_format_filter(self, filter_spec):
1873 " Returns a function to filter the formats according to the filter_spec "
083c9df9
PH
1874
1875 OPERATORS = {
1876 '<': operator.lt,
1877 '<=': operator.le,
1878 '>': operator.gt,
1879 '>=': operator.ge,
1880 '=': operator.eq,
1881 '!=': operator.ne,
1882 }
67134eab 1883 operator_rex = re.compile(r'''(?x)\s*
187986a8 1884 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)\s*
1885 (?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1886 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)\s*
083c9df9 1887 ''' % '|'.join(map(re.escape, OPERATORS.keys())))
187986a8 1888 m = operator_rex.fullmatch(filter_spec)
9ddb6925
S
1889 if m:
1890 try:
1891 comparison_value = int(m.group('value'))
1892 except ValueError:
1893 comparison_value = parse_filesize(m.group('value'))
1894 if comparison_value is None:
1895 comparison_value = parse_filesize(m.group('value') + 'B')
1896 if comparison_value is None:
1897 raise ValueError(
1898 'Invalid value %r in format specification %r' % (
67134eab 1899 m.group('value'), filter_spec))
9ddb6925
S
1900 op = OPERATORS[m.group('op')]
1901
083c9df9 1902 if not m:
9ddb6925
S
1903 STR_OPERATORS = {
1904 '=': operator.eq,
10d33b34
YCH
1905 '^=': lambda attr, value: attr.startswith(value),
1906 '$=': lambda attr, value: attr.endswith(value),
1907 '*=': lambda attr, value: value in attr,
1ce9a3cb 1908 '~=': lambda attr, value: value.search(attr) is not None
9ddb6925 1909 }
187986a8 1910 str_operator_rex = re.compile(r'''(?x)\s*
1911 (?P<key>[a-zA-Z0-9._-]+)\s*
1ce9a3cb
LF
1912 (?P<negation>!\s*)?(?P<op>%s)\s*(?P<none_inclusive>\?\s*)?
1913 (?P<quote>["'])?
1914 (?P<value>(?(quote)(?:(?!(?P=quote))[^\\]|\\.)+|[\w.-]+))
1915 (?(quote)(?P=quote))\s*
9ddb6925 1916 ''' % '|'.join(map(re.escape, STR_OPERATORS.keys())))
187986a8 1917 m = str_operator_rex.fullmatch(filter_spec)
9ddb6925 1918 if m:
1ce9a3cb
LF
1919 if m.group('op') == '~=':
1920 comparison_value = re.compile(m.group('value'))
1921 else:
1922 comparison_value = re.sub(r'''\\([\\"'])''', r'\1', m.group('value'))
2cc779f4
S
1923 str_op = STR_OPERATORS[m.group('op')]
1924 if m.group('negation'):
e118a879 1925 op = lambda attr, value: not str_op(attr, value)
2cc779f4
S
1926 else:
1927 op = str_op
083c9df9 1928
9ddb6925 1929 if not m:
187986a8 1930 raise SyntaxError('Invalid filter specification %r' % filter_spec)
083c9df9
PH
1931
1932 def _filter(f):
1933 actual_value = f.get(m.group('key'))
1934 if actual_value is None:
1935 return m.group('none_inclusive')
1936 return op(actual_value, comparison_value)
67134eab
JMF
1937 return _filter
1938
9f1a1c36 1939 def _check_formats(self, formats):
1940 for f in formats:
1941 self.to_screen('[info] Testing format %s' % f['format_id'])
75689fe5 1942 path = self.get_output_path('temp')
1943 if not self._ensure_dir_exists(f'{path}/'):
1944 continue
1945 temp_file = tempfile.NamedTemporaryFile(suffix='.tmp', delete=False, dir=path or None)
9f1a1c36 1946 temp_file.close()
1947 try:
1948 success, _ = self.dl(temp_file.name, f, test=True)
8a82af35 1949 except (DownloadError, OSError, ValueError) + network_exceptions:
9f1a1c36 1950 success = False
1951 finally:
1952 if os.path.exists(temp_file.name):
1953 try:
1954 os.remove(temp_file.name)
1955 except OSError:
1956 self.report_warning('Unable to delete temporary file "%s"' % temp_file.name)
1957 if success:
1958 yield f
1959 else:
1960 self.to_screen('[info] Unable to download format %s. Skipping...' % f['format_id'])
1961
0017d9ad 1962 def _default_format_spec(self, info_dict, download=True):
0017d9ad 1963
af0f7428
S
1964 def can_merge():
1965 merger = FFmpegMergerPP(self)
1966 return merger.available and merger.can_merge()
1967
91ebc640 1968 prefer_best = (
b7b04c78 1969 not self.params.get('simulate')
91ebc640 1970 and download
1971 and (
1972 not can_merge()
21633673 1973 or info_dict.get('is_live') and not self.params.get('live_from_start')
bf1824b3 1974 or self.params['outtmpl']['default'] == '-'))
53ed7066 1975 compat = (
1976 prefer_best
1977 or self.params.get('allow_multiple_audio_streams', False)
8a82af35 1978 or 'format-spec' in self.params['compat_opts'])
91ebc640 1979
1980 return (
53ed7066 1981 'best/bestvideo+bestaudio' if prefer_best
1982 else 'bestvideo*+bestaudio/best' if not compat
91ebc640 1983 else 'bestvideo+bestaudio/best')
0017d9ad 1984
67134eab
JMF
1985 def build_format_selector(self, format_spec):
1986 def syntax_error(note, start):
1987 message = (
1988 'Invalid format specification: '
86e5f3ed 1989 '{}\n\t{}\n\t{}^'.format(note, format_spec, ' ' * start[1]))
67134eab
JMF
1990 return SyntaxError(message)
1991
1992 PICKFIRST = 'PICKFIRST'
1993 MERGE = 'MERGE'
1994 SINGLE = 'SINGLE'
0130afb7 1995 GROUP = 'GROUP'
67134eab
JMF
1996 FormatSelector = collections.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
1997
91ebc640 1998 allow_multiple_streams = {'audio': self.params.get('allow_multiple_audio_streams', False),
1999 'video': self.params.get('allow_multiple_video_streams', False)}
909d24dd 2000
9f1a1c36 2001 check_formats = self.params.get('check_formats') == 'selected'
e8e73840 2002
67134eab
JMF
2003 def _parse_filter(tokens):
2004 filter_parts = []
2005 for type, string, start, _, _ in tokens:
2006 if type == tokenize.OP and string == ']':
2007 return ''.join(filter_parts)
2008 else:
2009 filter_parts.append(string)
2010
232541df 2011 def _remove_unused_ops(tokens):
62b58c09
L
2012 # Remove operators that we don't use and join them with the surrounding strings.
2013 # E.g. 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
232541df
JMF
2014 ALLOWED_OPS = ('/', '+', ',', '(', ')')
2015 last_string, last_start, last_end, last_line = None, None, None, None
2016 for type, string, start, end, line in tokens:
2017 if type == tokenize.OP and string == '[':
2018 if last_string:
2019 yield tokenize.NAME, last_string, last_start, last_end, last_line
2020 last_string = None
2021 yield type, string, start, end, line
2022 # everything inside brackets will be handled by _parse_filter
2023 for type, string, start, end, line in tokens:
2024 yield type, string, start, end, line
2025 if type == tokenize.OP and string == ']':
2026 break
2027 elif type == tokenize.OP and string in ALLOWED_OPS:
2028 if last_string:
2029 yield tokenize.NAME, last_string, last_start, last_end, last_line
2030 last_string = None
2031 yield type, string, start, end, line
2032 elif type in [tokenize.NAME, tokenize.NUMBER, tokenize.OP]:
2033 if not last_string:
2034 last_string = string
2035 last_start = start
2036 last_end = end
2037 else:
2038 last_string += string
2039 if last_string:
2040 yield tokenize.NAME, last_string, last_start, last_end, last_line
2041
cf2ac6df 2042 def _parse_format_selection(tokens, inside_merge=False, inside_choice=False, inside_group=False):
67134eab
JMF
2043 selectors = []
2044 current_selector = None
2045 for type, string, start, _, _ in tokens:
2046 # ENCODING is only defined in python 3.x
2047 if type == getattr(tokenize, 'ENCODING', None):
2048 continue
2049 elif type in [tokenize.NAME, tokenize.NUMBER]:
2050 current_selector = FormatSelector(SINGLE, string, [])
2051 elif type == tokenize.OP:
cf2ac6df
JMF
2052 if string == ')':
2053 if not inside_group:
2054 # ')' will be handled by the parentheses group
2055 tokens.restore_last_token()
67134eab 2056 break
cf2ac6df 2057 elif inside_merge and string in ['/', ',']:
0130afb7
JMF
2058 tokens.restore_last_token()
2059 break
cf2ac6df
JMF
2060 elif inside_choice and string == ',':
2061 tokens.restore_last_token()
2062 break
2063 elif string == ',':
0a31a350
JMF
2064 if not current_selector:
2065 raise syntax_error('"," must follow a format selector', start)
67134eab
JMF
2066 selectors.append(current_selector)
2067 current_selector = None
2068 elif string == '/':
d96d604e
JMF
2069 if not current_selector:
2070 raise syntax_error('"/" must follow a format selector', start)
67134eab 2071 first_choice = current_selector
cf2ac6df 2072 second_choice = _parse_format_selection(tokens, inside_choice=True)
f5f4a27a 2073 current_selector = FormatSelector(PICKFIRST, (first_choice, second_choice), [])
67134eab
JMF
2074 elif string == '[':
2075 if not current_selector:
2076 current_selector = FormatSelector(SINGLE, 'best', [])
2077 format_filter = _parse_filter(tokens)
2078 current_selector.filters.append(format_filter)
0130afb7
JMF
2079 elif string == '(':
2080 if current_selector:
2081 raise syntax_error('Unexpected "("', start)
cf2ac6df
JMF
2082 group = _parse_format_selection(tokens, inside_group=True)
2083 current_selector = FormatSelector(GROUP, group, [])
67134eab 2084 elif string == '+':
d03cfdce 2085 if not current_selector:
2086 raise syntax_error('Unexpected "+"', start)
2087 selector_1 = current_selector
2088 selector_2 = _parse_format_selection(tokens, inside_merge=True)
2089 if not selector_2:
2090 raise syntax_error('Expected a selector', start)
2091 current_selector = FormatSelector(MERGE, (selector_1, selector_2), [])
67134eab 2092 else:
86e5f3ed 2093 raise syntax_error(f'Operator not recognized: "{string}"', start)
67134eab
JMF
2094 elif type == tokenize.ENDMARKER:
2095 break
2096 if current_selector:
2097 selectors.append(current_selector)
2098 return selectors
2099
f8d4ad9a 2100 def _merge(formats_pair):
2101 format_1, format_2 = formats_pair
2102
2103 formats_info = []
2104 formats_info.extend(format_1.get('requested_formats', (format_1,)))
2105 formats_info.extend(format_2.get('requested_formats', (format_2,)))
2106
2107 if not allow_multiple_streams['video'] or not allow_multiple_streams['audio']:
551f9388 2108 get_no_more = {'video': False, 'audio': False}
f8d4ad9a 2109 for (i, fmt_info) in enumerate(formats_info):
551f9388 2110 if fmt_info.get('acodec') == fmt_info.get('vcodec') == 'none':
2111 formats_info.pop(i)
2112 continue
2113 for aud_vid in ['audio', 'video']:
f8d4ad9a 2114 if not allow_multiple_streams[aud_vid] and fmt_info.get(aud_vid[0] + 'codec') != 'none':
2115 if get_no_more[aud_vid]:
2116 formats_info.pop(i)
f5510afe 2117 break
f8d4ad9a 2118 get_no_more[aud_vid] = True
2119
2120 if len(formats_info) == 1:
2121 return formats_info[0]
2122
2123 video_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('vcodec') != 'none']
2124 audio_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('acodec') != 'none']
2125
2126 the_only_video = video_fmts[0] if len(video_fmts) == 1 else None
2127 the_only_audio = audio_fmts[0] if len(audio_fmts) == 1 else None
2128
fc61aff4
LL
2129 output_ext = get_compatible_ext(
2130 vcodecs=[f.get('vcodec') for f in video_fmts],
2131 acodecs=[f.get('acodec') for f in audio_fmts],
2132 vexts=[f['ext'] for f in video_fmts],
2133 aexts=[f['ext'] for f in audio_fmts],
2134 preferences=(try_call(lambda: self.params['merge_output_format'].split('/'))
2135 or self.params.get('prefer_free_formats') and ('webm', 'mkv')))
f8d4ad9a 2136
975a0d0d 2137 filtered = lambda *keys: filter(None, (traverse_obj(fmt, *keys) for fmt in formats_info))
2138
f8d4ad9a 2139 new_dict = {
2140 'requested_formats': formats_info,
975a0d0d 2141 'format': '+'.join(filtered('format')),
2142 'format_id': '+'.join(filtered('format_id')),
f8d4ad9a 2143 'ext': output_ext,
975a0d0d 2144 'protocol': '+'.join(map(determine_protocol, formats_info)),
093a1710 2145 'language': '+'.join(orderedSet(filtered('language'))) or None,
2146 'format_note': '+'.join(orderedSet(filtered('format_note'))) or None,
2147 'filesize_approx': sum(filtered('filesize', 'filesize_approx')) or None,
975a0d0d 2148 'tbr': sum(filtered('tbr', 'vbr', 'abr')),
f8d4ad9a 2149 }
2150
2151 if the_only_video:
2152 new_dict.update({
2153 'width': the_only_video.get('width'),
2154 'height': the_only_video.get('height'),
2155 'resolution': the_only_video.get('resolution') or self.format_resolution(the_only_video),
2156 'fps': the_only_video.get('fps'),
49a57e70 2157 'dynamic_range': the_only_video.get('dynamic_range'),
f8d4ad9a 2158 'vcodec': the_only_video.get('vcodec'),
2159 'vbr': the_only_video.get('vbr'),
2160 'stretched_ratio': the_only_video.get('stretched_ratio'),
2161 })
2162
2163 if the_only_audio:
2164 new_dict.update({
2165 'acodec': the_only_audio.get('acodec'),
2166 'abr': the_only_audio.get('abr'),
975a0d0d 2167 'asr': the_only_audio.get('asr'),
b8ed0f15 2168 'audio_channels': the_only_audio.get('audio_channels')
f8d4ad9a 2169 })
2170
2171 return new_dict
2172
e8e73840 2173 def _check_formats(formats):
981052c9 2174 if not check_formats:
2175 yield from formats
b5ac45b1 2176 return
9f1a1c36 2177 yield from self._check_formats(formats)
e8e73840 2178
67134eab 2179 def _build_selector_function(selector):
909d24dd 2180 if isinstance(selector, list): # ,
67134eab
JMF
2181 fs = [_build_selector_function(s) for s in selector]
2182
317f7ab6 2183 def selector_function(ctx):
67134eab 2184 for f in fs:
981052c9 2185 yield from f(ctx)
67134eab 2186 return selector_function
909d24dd 2187
2188 elif selector.type == GROUP: # ()
0130afb7 2189 selector_function = _build_selector_function(selector.selector)
909d24dd 2190
2191 elif selector.type == PICKFIRST: # /
67134eab
JMF
2192 fs = [_build_selector_function(s) for s in selector.selector]
2193
317f7ab6 2194 def selector_function(ctx):
67134eab 2195 for f in fs:
317f7ab6 2196 picked_formats = list(f(ctx))
67134eab
JMF
2197 if picked_formats:
2198 return picked_formats
2199 return []
67134eab 2200
981052c9 2201 elif selector.type == MERGE: # +
2202 selector_1, selector_2 = map(_build_selector_function, selector.selector)
2203
2204 def selector_function(ctx):
adbc4ec4 2205 for pair in itertools.product(selector_1(ctx), selector_2(ctx)):
981052c9 2206 yield _merge(pair)
2207
909d24dd 2208 elif selector.type == SINGLE: # atom
598d185d 2209 format_spec = selector.selector or 'best'
909d24dd 2210
f8d4ad9a 2211 # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
909d24dd 2212 if format_spec == 'all':
2213 def selector_function(ctx):
9222c381 2214 yield from _check_formats(ctx['formats'][::-1])
f8d4ad9a 2215 elif format_spec == 'mergeall':
2216 def selector_function(ctx):
316f2650 2217 formats = list(_check_formats(
2218 f for f in ctx['formats'] if f.get('vcodec') != 'none' or f.get('acodec') != 'none'))
e01d6aa4 2219 if not formats:
2220 return
921b76ca 2221 merged_format = formats[-1]
2222 for f in formats[-2::-1]:
f8d4ad9a 2223 merged_format = _merge((merged_format, f))
2224 yield merged_format
909d24dd 2225
2226 else:
85e801a9 2227 format_fallback, seperate_fallback, format_reverse, format_idx = False, None, True, 1
eff63539 2228 mobj = re.match(
2229 r'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
2230 format_spec)
2231 if mobj is not None:
2232 format_idx = int_or_none(mobj.group('n'), default=1)
e8e73840 2233 format_reverse = mobj.group('bw')[0] == 'b'
eff63539 2234 format_type = (mobj.group('type') or [None])[0]
2235 not_format_type = {'v': 'a', 'a': 'v'}.get(format_type)
2236 format_modified = mobj.group('mod') is not None
909d24dd 2237
2238 format_fallback = not format_type and not format_modified # for b, w
8326b00a 2239 _filter_f = (
eff63539 2240 (lambda f: f.get('%scodec' % format_type) != 'none')
2241 if format_type and format_modified # bv*, ba*, wv*, wa*
2242 else (lambda f: f.get('%scodec' % not_format_type) == 'none')
2243 if format_type # bv, ba, wv, wa
2244 else (lambda f: f.get('vcodec') != 'none' and f.get('acodec') != 'none')
2245 if not format_modified # b, w
8326b00a 2246 else lambda f: True) # b*, w*
2247 filter_f = lambda f: _filter_f(f) and (
2248 f.get('vcodec') != 'none' or f.get('acodec') != 'none')
67134eab 2249 else:
48ee10ee 2250 if format_spec in self._format_selection_exts['audio']:
b11c04a8 2251 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none'
48ee10ee 2252 elif format_spec in self._format_selection_exts['video']:
b11c04a8 2253 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none' and f.get('vcodec') != 'none'
85e801a9 2254 seperate_fallback = lambda f: f.get('ext') == format_spec and f.get('vcodec') != 'none'
48ee10ee 2255 elif format_spec in self._format_selection_exts['storyboards']:
b11c04a8 2256 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') == 'none' and f.get('vcodec') == 'none'
2257 else:
b5ae35ee 2258 filter_f = lambda f: f.get('format_id') == format_spec # id
909d24dd 2259
2260 def selector_function(ctx):
2261 formats = list(ctx['formats'])
909d24dd 2262 matches = list(filter(filter_f, formats)) if filter_f is not None else formats
85e801a9 2263 if not matches:
2264 if format_fallback and ctx['incomplete_formats']:
2265 # for extractors with incomplete formats (audio only (soundcloud)
2266 # or video only (imgur)) best/worst will fallback to
2267 # best/worst {video,audio}-only format
2268 matches = formats
2269 elif seperate_fallback and not ctx['has_merged_format']:
2270 # for compatibility with youtube-dl when there is no pre-merged format
2271 matches = list(filter(seperate_fallback, formats))
981052c9 2272 matches = LazyList(_check_formats(matches[::-1 if format_reverse else 1]))
2273 try:
e8e73840 2274 yield matches[format_idx - 1]
4abea8ca 2275 except LazyList.IndexError:
981052c9 2276 return
083c9df9 2277
67134eab 2278 filters = [self._build_format_filter(f) for f in selector.filters]
083c9df9 2279
317f7ab6 2280 def final_selector(ctx):
adbc4ec4 2281 ctx_copy = dict(ctx)
67134eab 2282 for _filter in filters:
317f7ab6
S
2283 ctx_copy['formats'] = list(filter(_filter, ctx_copy['formats']))
2284 return selector_function(ctx_copy)
67134eab 2285 return final_selector
083c9df9 2286
0f06bcd7 2287 stream = io.BytesIO(format_spec.encode())
0130afb7 2288 try:
f9934b96 2289 tokens = list(_remove_unused_ops(tokenize.tokenize(stream.readline)))
0130afb7
JMF
2290 except tokenize.TokenError:
2291 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec)))
2292
86e5f3ed 2293 class TokenIterator:
0130afb7
JMF
2294 def __init__(self, tokens):
2295 self.tokens = tokens
2296 self.counter = 0
2297
2298 def __iter__(self):
2299 return self
2300
2301 def __next__(self):
2302 if self.counter >= len(self.tokens):
2303 raise StopIteration()
2304 value = self.tokens[self.counter]
2305 self.counter += 1
2306 return value
2307
2308 next = __next__
2309
2310 def restore_last_token(self):
2311 self.counter -= 1
2312
2313 parsed_selector = _parse_format_selection(iter(TokenIterator(tokens)))
67134eab 2314 return _build_selector_function(parsed_selector)
a9c58ad9 2315
e5660ee6 2316 def _calc_headers(self, info_dict):
8b7539d2 2317 res = merge_headers(self.params['http_headers'], info_dict.get('http_headers') or {})
e5660ee6 2318
c487cf00 2319 cookies = self._calc_cookies(info_dict['url'])
e5660ee6
JMF
2320 if cookies:
2321 res['Cookie'] = cookies
2322
0016b84e
S
2323 if 'X-Forwarded-For' not in res:
2324 x_forwarded_for_ip = info_dict.get('__x_forwarded_for_ip')
2325 if x_forwarded_for_ip:
2326 res['X-Forwarded-For'] = x_forwarded_for_ip
2327
e5660ee6
JMF
2328 return res
2329
c487cf00 2330 def _calc_cookies(self, url):
2331 pr = sanitized_Request(url)
e5660ee6 2332 self.cookiejar.add_cookie_header(pr)
662435f7 2333 return pr.get_header('Cookie')
e5660ee6 2334
9f1a1c36 2335 def _sort_thumbnails(self, thumbnails):
2336 thumbnails.sort(key=lambda t: (
2337 t.get('preference') if t.get('preference') is not None else -1,
2338 t.get('width') if t.get('width') is not None else -1,
2339 t.get('height') if t.get('height') is not None else -1,
2340 t.get('id') if t.get('id') is not None else '',
2341 t.get('url')))
2342
b0249bca 2343 def _sanitize_thumbnails(self, info_dict):
bc516a3f 2344 thumbnails = info_dict.get('thumbnails')
2345 if thumbnails is None:
2346 thumbnail = info_dict.get('thumbnail')
2347 if thumbnail:
2348 info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}]
9f1a1c36 2349 if not thumbnails:
2350 return
2351
2352 def check_thumbnails(thumbnails):
2353 for t in thumbnails:
2354 self.to_screen(f'[info] Testing thumbnail {t["id"]}')
2355 try:
2356 self.urlopen(HEADRequest(t['url']))
2357 except network_exceptions as err:
2358 self.to_screen(f'[info] Unable to connect to thumbnail {t["id"]} URL {t["url"]!r} - {err}. Skipping...')
2359 continue
2360 yield t
2361
2362 self._sort_thumbnails(thumbnails)
2363 for i, t in enumerate(thumbnails):
2364 if t.get('id') is None:
2365 t['id'] = '%d' % i
2366 if t.get('width') and t.get('height'):
2367 t['resolution'] = '%dx%d' % (t['width'], t['height'])
2368 t['url'] = sanitize_url(t['url'])
2369
2370 if self.params.get('check_formats') is True:
282f5709 2371 info_dict['thumbnails'] = LazyList(check_thumbnails(thumbnails[::-1]), reverse=True)
9f1a1c36 2372 else:
2373 info_dict['thumbnails'] = thumbnails
bc516a3f 2374
03f83004
LNO
2375 def _fill_common_fields(self, info_dict, is_video=True):
2376 # TODO: move sanitization here
2377 if is_video:
2378 # playlists are allowed to lack "title"
d4736fdb 2379 title = info_dict.get('title', NO_DEFAULT)
2380 if title is NO_DEFAULT:
03f83004
LNO
2381 raise ExtractorError('Missing "title" field in extractor result',
2382 video_id=info_dict['id'], ie=info_dict['extractor'])
d4736fdb 2383 info_dict['fulltitle'] = title
2384 if not title:
2385 if title == '':
2386 self.write_debug('Extractor gave empty title. Creating a generic title')
2387 else:
2388 self.report_warning('Extractor failed to obtain "title". Creating a generic title instead')
1d485a1a 2389 info_dict['title'] = f'{info_dict["extractor"].replace(":", "-")} video #{info_dict["id"]}'
03f83004
LNO
2390
2391 if info_dict.get('duration') is not None:
2392 info_dict['duration_string'] = formatSeconds(info_dict['duration'])
2393
2394 for ts_key, date_key in (
2395 ('timestamp', 'upload_date'),
2396 ('release_timestamp', 'release_date'),
2397 ('modified_timestamp', 'modified_date'),
2398 ):
2399 if info_dict.get(date_key) is None and info_dict.get(ts_key) is not None:
2400 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
2401 # see http://bugs.python.org/issue1646728)
19a03940 2402 with contextlib.suppress(ValueError, OverflowError, OSError):
03f83004
LNO
2403 upload_date = datetime.datetime.utcfromtimestamp(info_dict[ts_key])
2404 info_dict[date_key] = upload_date.strftime('%Y%m%d')
03f83004
LNO
2405
2406 live_keys = ('is_live', 'was_live')
2407 live_status = info_dict.get('live_status')
2408 if live_status is None:
2409 for key in live_keys:
2410 if info_dict.get(key) is False:
2411 continue
2412 if info_dict.get(key):
2413 live_status = key
2414 break
2415 if all(info_dict.get(key) is False for key in live_keys):
2416 live_status = 'not_live'
2417 if live_status:
2418 info_dict['live_status'] = live_status
2419 for key in live_keys:
2420 if info_dict.get(key) is None:
2421 info_dict[key] = (live_status == key)
2422
2423 # Auto generate title fields corresponding to the *_number fields when missing
2424 # in order to always have clean titles. This is very common for TV series.
2425 for field in ('chapter', 'season', 'episode'):
2426 if info_dict.get('%s_number' % field) is not None and not info_dict.get(field):
2427 info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field])
2428
415f8d51 2429 def _raise_pending_errors(self, info):
2430 err = info.pop('__pending_error', None)
2431 if err:
2432 self.report_error(err, tb=False)
2433
dd82ffea
JMF
2434 def process_video_result(self, info_dict, download=True):
2435 assert info_dict.get('_type', 'video') == 'video'
9c906919 2436 self._num_videos += 1
dd82ffea 2437
bec1fad2 2438 if 'id' not in info_dict:
fc08bdd6 2439 raise ExtractorError('Missing "id" field in extractor result', ie=info_dict['extractor'])
2440 elif not info_dict.get('id'):
2441 raise ExtractorError('Extractor failed to obtain "id"', ie=info_dict['extractor'])
455a15e2 2442
c9969434
S
2443 def report_force_conversion(field, field_not, conversion):
2444 self.report_warning(
2445 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
2446 % (field, field_not, conversion))
2447
2448 def sanitize_string_field(info, string_field):
2449 field = info.get(string_field)
14f25df2 2450 if field is None or isinstance(field, str):
c9969434
S
2451 return
2452 report_force_conversion(string_field, 'a string', 'string')
14f25df2 2453 info[string_field] = str(field)
c9969434
S
2454
2455 def sanitize_numeric_fields(info):
2456 for numeric_field in self._NUMERIC_FIELDS:
2457 field = info.get(numeric_field)
f9934b96 2458 if field is None or isinstance(field, (int, float)):
c9969434
S
2459 continue
2460 report_force_conversion(numeric_field, 'numeric', 'int')
2461 info[numeric_field] = int_or_none(field)
2462
2463 sanitize_string_field(info_dict, 'id')
2464 sanitize_numeric_fields(info_dict)
3975b4d2 2465 if info_dict.get('section_end') and info_dict.get('section_start') is not None:
2466 info_dict['duration'] = round(info_dict['section_end'] - info_dict['section_start'], 3)
4c3f8c3f 2467 if (info_dict.get('duration') or 0) <= 0 and info_dict.pop('duration', None):
50e93e03 2468 self.report_warning('"duration" field is negative, there is an error in extractor')
be6217b2 2469
9eef7c4e 2470 chapters = info_dict.get('chapters') or []
a3976e07 2471 if chapters and chapters[0].get('start_time'):
2472 chapters.insert(0, {'start_time': 0})
2473
9eef7c4e 2474 dummy_chapter = {'end_time': 0, 'start_time': info_dict.get('duration')}
a3976e07 2475 for idx, (prev, current, next_) in enumerate(zip(
2476 (dummy_chapter, *chapters), chapters, (*chapters[1:], dummy_chapter)), 1):
9eef7c4e 2477 if current.get('start_time') is None:
2478 current['start_time'] = prev.get('end_time')
2479 if not current.get('end_time'):
2480 current['end_time'] = next_.get('start_time')
a3976e07 2481 if not current.get('title'):
2482 current['title'] = f'<Untitled Chapter {idx}>'
9eef7c4e 2483
dd82ffea
JMF
2484 if 'playlist' not in info_dict:
2485 # It isn't part of a playlist
2486 info_dict['playlist'] = None
2487 info_dict['playlist_index'] = None
2488
bc516a3f 2489 self._sanitize_thumbnails(info_dict)
d5519808 2490
536a55da 2491 thumbnail = info_dict.get('thumbnail')
bc516a3f 2492 thumbnails = info_dict.get('thumbnails')
536a55da
S
2493 if thumbnail:
2494 info_dict['thumbnail'] = sanitize_url(thumbnail)
2495 elif thumbnails:
d5519808
PH
2496 info_dict['thumbnail'] = thumbnails[-1]['url']
2497
ae30b840 2498 if info_dict.get('display_id') is None and 'id' in info_dict:
0afef30b
PH
2499 info_dict['display_id'] = info_dict['id']
2500
03f83004 2501 self._fill_common_fields(info_dict)
33d2fc2f 2502
05108a49
S
2503 for cc_kind in ('subtitles', 'automatic_captions'):
2504 cc = info_dict.get(cc_kind)
2505 if cc:
2506 for _, subtitle in cc.items():
2507 for subtitle_format in subtitle:
2508 if subtitle_format.get('url'):
2509 subtitle_format['url'] = sanitize_url(subtitle_format['url'])
2510 if subtitle_format.get('ext') is None:
2511 subtitle_format['ext'] = determine_ext(subtitle_format['url']).lower()
2512
2513 automatic_captions = info_dict.get('automatic_captions')
4bba3716 2514 subtitles = info_dict.get('subtitles')
4bba3716 2515
360e1ca5 2516 info_dict['requested_subtitles'] = self.process_subtitles(
05108a49 2517 info_dict['id'], subtitles, automatic_captions)
a504ced0 2518
dd82ffea
JMF
2519 if info_dict.get('formats') is None:
2520 # There's only one format available
2521 formats = [info_dict]
2522 else:
2523 formats = info_dict['formats']
2524
0a5a191a 2525 # or None ensures --clean-infojson removes it
2526 info_dict['_has_drm'] = any(f.get('has_drm') for f in formats) or None
88acdbc2 2527 if not self.params.get('allow_unplayable_formats'):
2528 formats = [f for f in formats if not f.get('has_drm')]
17ffed18 2529
2530 if formats and all(f.get('acodec') == f.get('vcodec') == 'none' for f in formats):
2531 self.report_warning(
2532 f'{"This video is DRM protected and " if info_dict["_has_drm"] else ""}'
2533 'only images are available for download. Use --list-formats to see them'.capitalize())
88acdbc2 2534
319b6059 2535 get_from_start = not info_dict.get('is_live') or bool(self.params.get('live_from_start'))
2536 if not get_from_start:
2537 info_dict['title'] += ' ' + datetime.datetime.now().strftime('%Y-%m-%d %H:%M')
2538 if info_dict.get('is_live') and formats:
adbc4ec4 2539 formats = [f for f in formats if bool(f.get('is_from_start')) == get_from_start]
319b6059 2540 if get_from_start and not formats:
a44ca5a4 2541 self.raise_no_formats(info_dict, msg=(
2542 '--live-from-start is passed, but there are no formats that can be downloaded from the start. '
2543 'If you want to download from the current time, use --no-live-from-start'))
adbc4ec4 2544
73af5cc8
S
2545 def is_wellformed(f):
2546 url = f.get('url')
a5ac0c47 2547 if not url:
73af5cc8
S
2548 self.report_warning(
2549 '"url" field is missing or empty - skipping format, '
2550 'there is an error in extractor')
a5ac0c47
S
2551 return False
2552 if isinstance(url, bytes):
2553 sanitize_string_field(f, 'url')
2554 return True
73af5cc8
S
2555
2556 # Filter out malformed formats for better extraction robustness
1ac7f461 2557 formats = list(filter(is_wellformed, formats or []))
2558
2559 if not formats:
2560 self.raise_no_formats(info_dict)
73af5cc8 2561
181c7053
S
2562 formats_dict = {}
2563
dd82ffea 2564 # We check that all the formats have the format and format_id fields
db95dc13 2565 for i, format in enumerate(formats):
c9969434
S
2566 sanitize_string_field(format, 'format_id')
2567 sanitize_numeric_fields(format)
dcf77cf1 2568 format['url'] = sanitize_url(format['url'])
e74e3b63 2569 if not format.get('format_id'):
14f25df2 2570 format['format_id'] = str(i)
e2effb08
S
2571 else:
2572 # Sanitize format_id from characters used in format selector expression
ec85ded8 2573 format['format_id'] = re.sub(r'[\s,/+\[\]()]', '_', format['format_id'])
181c7053
S
2574 format_id = format['format_id']
2575 if format_id not in formats_dict:
2576 formats_dict[format_id] = []
2577 formats_dict[format_id].append(format)
2578
2579 # Make sure all formats have unique format_id
03b4de72 2580 common_exts = set(itertools.chain(*self._format_selection_exts.values()))
181c7053 2581 for format_id, ambiguous_formats in formats_dict.items():
48ee10ee 2582 ambigious_id = len(ambiguous_formats) > 1
2583 for i, format in enumerate(ambiguous_formats):
2584 if ambigious_id:
181c7053 2585 format['format_id'] = '%s-%d' % (format_id, i)
48ee10ee 2586 if format.get('ext') is None:
2587 format['ext'] = determine_ext(format['url']).lower()
2588 # Ensure there is no conflict between id and ext in format selection
2589 # See https://github.com/yt-dlp/yt-dlp/issues/1282
2590 if format['format_id'] != format['ext'] and format['format_id'] in common_exts:
2591 format['format_id'] = 'f%s' % format['format_id']
181c7053
S
2592
2593 for i, format in enumerate(formats):
8c51aa65 2594 if format.get('format') is None:
6febd1c1 2595 format['format'] = '{id} - {res}{note}'.format(
8c51aa65
JMF
2596 id=format['format_id'],
2597 res=self.format_resolution(format),
b868936c 2598 note=format_field(format, 'format_note', ' (%s)'),
8c51aa65 2599 )
6f0be937 2600 if format.get('protocol') is None:
b5559424 2601 format['protocol'] = determine_protocol(format)
239df021 2602 if format.get('resolution') is None:
2603 format['resolution'] = self.format_resolution(format, default=None)
176f1866 2604 if format.get('dynamic_range') is None and format.get('vcodec') != 'none':
2605 format['dynamic_range'] = 'SDR'
f2fe69c7 2606 if (info_dict.get('duration') and format.get('tbr')
2607 and not format.get('filesize') and not format.get('filesize_approx')):
56ba69e4 2608 format['filesize_approx'] = int(info_dict['duration'] * format['tbr'] * (1024 / 8))
f2fe69c7 2609
e5660ee6
JMF
2610 # Add HTTP headers, so that external programs can use them from the
2611 # json output
2612 full_format_info = info_dict.copy()
2613 full_format_info.update(format)
2614 format['http_headers'] = self._calc_headers(full_format_info)
0016b84e
S
2615 # Remove private housekeeping stuff
2616 if '__x_forwarded_for_ip' in info_dict:
2617 del info_dict['__x_forwarded_for_ip']
dd82ffea 2618
9f1a1c36 2619 if self.params.get('check_formats') is True:
282f5709 2620 formats = LazyList(self._check_formats(formats[::-1]), reverse=True)
9f1a1c36 2621
88acdbc2 2622 if not formats or formats[0] is not info_dict:
b3d9ef88
JMF
2623 # only set the 'formats' fields if the original info_dict list them
2624 # otherwise we end up with a circular reference, the first (and unique)
f89197d7 2625 # element in the 'formats' field in info_dict is info_dict itself,
dfb1b146 2626 # which can't be exported to json
b3d9ef88 2627 info_dict['formats'] = formats
4ec82a72 2628
2629 info_dict, _ = self.pre_process(info_dict)
2630
6db9c4d5 2631 if self._match_entry(info_dict, incomplete=self._format_fields) is not None:
09b49e1f 2632 return info_dict
2633
2634 self.post_extract(info_dict)
2635 info_dict, _ = self.pre_process(info_dict, 'after_filter')
2636
093a1710 2637 # The pre-processors may have modified the formats
2638 formats = info_dict.get('formats', [info_dict])
2639
fa9f30b8 2640 list_only = self.params.get('simulate') is None and (
2641 self.params.get('list_thumbnails') or self.params.get('listformats') or self.params.get('listsubtitles'))
2642 interactive_format_selection = not list_only and self.format_selector == '-'
b7b04c78 2643 if self.params.get('list_thumbnails'):
2644 self.list_thumbnails(info_dict)
b7b04c78 2645 if self.params.get('listsubtitles'):
2646 if 'automatic_captions' in info_dict:
2647 self.list_subtitles(
2648 info_dict['id'], automatic_captions, 'automatic captions')
2649 self.list_subtitles(info_dict['id'], subtitles, 'subtitles')
fa9f30b8 2650 if self.params.get('listformats') or interactive_format_selection:
b69fd25c 2651 self.list_formats(info_dict)
169dbde9 2652 if list_only:
b7b04c78 2653 # Without this printing, -F --print-json will not work
169dbde9 2654 self.__forced_printings(info_dict, self.prepare_filename(info_dict), incomplete=True)
c487cf00 2655 return info_dict
bfaae0a7 2656
187986a8 2657 format_selector = self.format_selector
2658 if format_selector is None:
0017d9ad 2659 req_format = self._default_format_spec(info_dict, download=download)
0760b0a7 2660 self.write_debug('Default format spec: %s' % req_format)
187986a8 2661 format_selector = self.build_format_selector(req_format)
317f7ab6 2662
fa9f30b8 2663 while True:
2664 if interactive_format_selection:
2665 req_format = input(
2666 self._format_screen('\nEnter format selector: ', self.Styles.EMPHASIS))
2667 try:
2668 format_selector = self.build_format_selector(req_format)
2669 except SyntaxError as err:
2670 self.report_error(err, tb=False, is_error=False)
2671 continue
2672
85e801a9 2673 formats_to_download = list(format_selector({
fa9f30b8 2674 'formats': formats,
85e801a9 2675 'has_merged_format': any('none' not in (f.get('acodec'), f.get('vcodec')) for f in formats),
2676 'incomplete_formats': (
2677 # All formats are video-only or
2678 all(f.get('vcodec') != 'none' and f.get('acodec') == 'none' for f in formats)
2679 # all formats are audio-only
2680 or all(f.get('vcodec') == 'none' and f.get('acodec') != 'none' for f in formats)),
2681 }))
fa9f30b8 2682 if interactive_format_selection and not formats_to_download:
2683 self.report_error('Requested format is not available', tb=False, is_error=False)
2684 continue
2685 break
317f7ab6 2686
dd82ffea 2687 if not formats_to_download:
b7da73eb 2688 if not self.params.get('ignore_no_formats_error'):
c0b6e5c7 2689 raise ExtractorError(
2690 'Requested format is not available. Use --list-formats for a list of available formats',
2691 expected=True, video_id=info_dict['id'], ie=info_dict['extractor'])
b62fa6d7 2692 self.report_warning('Requested format is not available')
2693 # Process what we can, even without any available formats.
2694 formats_to_download = [{}]
a13e6848 2695
5ec1b6b7 2696 requested_ranges = self.params.get('download_ranges')
2697 if requested_ranges:
2698 requested_ranges = tuple(requested_ranges(info_dict, self))
2699
2700 best_format, downloaded_formats = formats_to_download[-1], []
b62fa6d7 2701 if download:
2702 if best_format:
5ec1b6b7 2703 def to_screen(*msg):
2704 self.to_screen(f'[info] {info_dict["id"]}: {" ".join(", ".join(variadic(m)) for m in msg)}')
2705
2706 to_screen(f'Downloading {len(formats_to_download)} format(s):',
2707 (f['format_id'] for f in formats_to_download))
2708 if requested_ranges:
2709 to_screen(f'Downloading {len(requested_ranges)} time ranges:',
2710 (f'{int(c["start_time"])}-{int(c["end_time"])}' for c in requested_ranges))
a13e6848 2711 max_downloads_reached = False
5ec1b6b7 2712
2713 for fmt, chapter in itertools.product(formats_to_download, requested_ranges or [{}]):
2714 new_info = self._copy_infodict(info_dict)
b7da73eb 2715 new_info.update(fmt)
3975b4d2 2716 offset, duration = info_dict.get('section_start') or 0, info_dict.get('duration') or float('inf')
2717 if chapter or offset:
5ec1b6b7 2718 new_info.update({
3975b4d2 2719 'section_start': offset + chapter.get('start_time', 0),
bc401608 2720 'section_end': offset + min(chapter.get('end_time', duration), duration),
5ec1b6b7 2721 'section_title': chapter.get('title'),
2722 'section_number': chapter.get('index'),
2723 })
2724 downloaded_formats.append(new_info)
a13e6848 2725 try:
2726 self.process_info(new_info)
2727 except MaxDownloadsReached:
2728 max_downloads_reached = True
415f8d51 2729 self._raise_pending_errors(new_info)
f46e2f9d 2730 # Remove copied info
2731 for key, val in tuple(new_info.items()):
2732 if info_dict.get(key) == val:
2733 new_info.pop(key)
a13e6848 2734 if max_downloads_reached:
2735 break
ebed8b37 2736
5ec1b6b7 2737 write_archive = {f.get('__write_download_archive', False) for f in downloaded_formats}
a13e6848 2738 assert write_archive.issubset({True, False, 'ignore'})
2739 if True in write_archive and False not in write_archive:
2740 self.record_download_archive(info_dict)
be72c624 2741
5ec1b6b7 2742 info_dict['requested_downloads'] = downloaded_formats
ed5835b4 2743 info_dict = self.run_all_pps('after_video', info_dict)
a13e6848 2744 if max_downloads_reached:
2745 raise MaxDownloadsReached()
ebed8b37 2746
49a57e70 2747 # We update the info dict with the selected best quality format (backwards compatibility)
be72c624 2748 info_dict.update(best_format)
dd82ffea
JMF
2749 return info_dict
2750
98c70d6f 2751 def process_subtitles(self, video_id, normal_subtitles, automatic_captions):
a504ced0 2752 """Select the requested subtitles and their format"""
d8a58ddc 2753 available_subs, normal_sub_langs = {}, []
98c70d6f
JMF
2754 if normal_subtitles and self.params.get('writesubtitles'):
2755 available_subs.update(normal_subtitles)
d8a58ddc 2756 normal_sub_langs = tuple(normal_subtitles.keys())
98c70d6f
JMF
2757 if automatic_captions and self.params.get('writeautomaticsub'):
2758 for lang, cap_info in automatic_captions.items():
360e1ca5
JMF
2759 if lang not in available_subs:
2760 available_subs[lang] = cap_info
2761
d2c8aadf 2762 if not available_subs or (
2763 not self.params.get('writesubtitles')
2764 and not self.params.get('writeautomaticsub')):
4d171848 2765 return None
a504ced0 2766
d8a58ddc 2767 all_sub_langs = tuple(available_subs.keys())
a504ced0 2768 if self.params.get('allsubtitles', False):
c32b0aab 2769 requested_langs = all_sub_langs
2770 elif self.params.get('subtitleslangs', False):
5314b521 2771 try:
2772 requested_langs = orderedSet_from_options(
2773 self.params.get('subtitleslangs'), {'all': all_sub_langs}, use_regex=True)
2774 except re.error as e:
2775 raise ValueError(f'Wrong regex for subtitlelangs: {e.pattern}')
d8a58ddc 2776 elif normal_sub_langs:
2777 requested_langs = ['en'] if 'en' in normal_sub_langs else normal_sub_langs[:1]
a504ced0 2778 else:
d8a58ddc 2779 requested_langs = ['en'] if 'en' in all_sub_langs else all_sub_langs[:1]
ad3dc496 2780 if requested_langs:
d2c8aadf 2781 self.to_screen(f'[info] {video_id}: Downloading subtitles: {", ".join(requested_langs)}')
a504ced0
JMF
2782
2783 formats_query = self.params.get('subtitlesformat', 'best')
2784 formats_preference = formats_query.split('/') if formats_query else []
2785 subs = {}
2786 for lang in requested_langs:
2787 formats = available_subs.get(lang)
2788 if formats is None:
86e5f3ed 2789 self.report_warning(f'{lang} subtitles not available for {video_id}')
a504ced0 2790 continue
a504ced0
JMF
2791 for ext in formats_preference:
2792 if ext == 'best':
2793 f = formats[-1]
2794 break
2795 matches = list(filter(lambda f: f['ext'] == ext, formats))
2796 if matches:
2797 f = matches[-1]
2798 break
2799 else:
2800 f = formats[-1]
2801 self.report_warning(
2802 'No subtitle format found matching "%s" for language %s, '
2803 'using %s' % (formats_query, lang, f['ext']))
2804 subs[lang] = f
2805 return subs
2806
bb66c247 2807 def _forceprint(self, key, info_dict):
2808 if info_dict is None:
2809 return
2810 info_copy = info_dict.copy()
2811 info_copy['formats_table'] = self.render_formats_table(info_dict)
2812 info_copy['thumbnails_table'] = self.render_thumbnails_table(info_dict)
2813 info_copy['subtitles_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('subtitles'))
2814 info_copy['automatic_captions_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('automatic_captions'))
2815
2816 def format_tmpl(tmpl):
48c8424b 2817 mobj = re.fullmatch(r'([\w.:,]|-\d|(?P<dict>{([\w.:,]|-\d)+}))+=?', tmpl)
07a1250e 2818 if not mobj:
2819 return tmpl
48c8424b 2820
2821 fmt = '%({})s'
2822 if tmpl.startswith('{'):
2823 tmpl = f'.{tmpl}'
2824 if tmpl.endswith('='):
2825 tmpl, fmt = tmpl[:-1], '{0} = %({0})#j'
2826 return '\n'.join(map(fmt.format, [tmpl] if mobj.group('dict') else tmpl.split(',')))
8130779d 2827
bb66c247 2828 for tmpl in self.params['forceprint'].get(key, []):
2829 self.to_stdout(self.evaluate_outtmpl(format_tmpl(tmpl), info_copy))
2830
2831 for tmpl, file_tmpl in self.params['print_to_file'].get(key, []):
5127e92a 2832 filename = self.prepare_filename(info_dict, outtmpl=file_tmpl)
bb66c247 2833 tmpl = format_tmpl(tmpl)
2834 self.to_screen(f'[info] Writing {tmpl!r} to: {filename}')
8d93e69d 2835 if self._ensure_dir_exists(filename):
86e5f3ed 2836 with open(filename, 'a', encoding='utf-8') as f:
8d93e69d 2837 f.write(self.evaluate_outtmpl(tmpl, info_copy) + '\n')
ca30f449 2838
d06daf23 2839 def __forced_printings(self, info_dict, filename, incomplete):
53c18592 2840 def print_mandatory(field, actual_field=None):
2841 if actual_field is None:
2842 actual_field = field
d06daf23 2843 if (self.params.get('force%s' % field, False)
53c18592 2844 and (not incomplete or info_dict.get(actual_field) is not None)):
2845 self.to_stdout(info_dict[actual_field])
d06daf23
S
2846
2847 def print_optional(field):
2848 if (self.params.get('force%s' % field, False)
2849 and info_dict.get(field) is not None):
2850 self.to_stdout(info_dict[field])
2851
53c18592 2852 info_dict = info_dict.copy()
2853 if filename is not None:
2854 info_dict['filename'] = filename
2855 if info_dict.get('requested_formats') is not None:
2856 # For RTMP URLs, also include the playpath
2857 info_dict['urls'] = '\n'.join(f['url'] + f.get('play_path', '') for f in info_dict['requested_formats'])
10331a26 2858 elif info_dict.get('url'):
53c18592 2859 info_dict['urls'] = info_dict['url'] + info_dict.get('play_path', '')
2860
bb66c247 2861 if (self.params.get('forcejson')
2862 or self.params['forceprint'].get('video')
2863 or self.params['print_to_file'].get('video')):
2b8a2973 2864 self.post_extract(info_dict)
bb66c247 2865 self._forceprint('video', info_dict)
53c18592 2866
d06daf23
S
2867 print_mandatory('title')
2868 print_mandatory('id')
53c18592 2869 print_mandatory('url', 'urls')
d06daf23
S
2870 print_optional('thumbnail')
2871 print_optional('description')
53c18592 2872 print_optional('filename')
b868936c 2873 if self.params.get('forceduration') and info_dict.get('duration') is not None:
d06daf23
S
2874 self.to_stdout(formatSeconds(info_dict['duration']))
2875 print_mandatory('format')
53c18592 2876
2b8a2973 2877 if self.params.get('forcejson'):
6e84b215 2878 self.to_stdout(json.dumps(self.sanitize_info(info_dict)))
d06daf23 2879
e8e73840 2880 def dl(self, name, info, subtitle=False, test=False):
88acdbc2 2881 if not info.get('url'):
1151c407 2882 self.raise_no_formats(info, True)
e8e73840 2883
2884 if test:
2885 verbose = self.params.get('verbose')
2886 params = {
2887 'test': True,
a169858f 2888 'quiet': self.params.get('quiet') or not verbose,
e8e73840 2889 'verbose': verbose,
2890 'noprogress': not verbose,
2891 'nopart': True,
2892 'skip_unavailable_fragments': False,
2893 'keep_fragments': False,
2894 'overwrites': True,
2895 '_no_ytdl_file': True,
2896 }
2897 else:
2898 params = self.params
96fccc10 2899 fd = get_suitable_downloader(info, params, to_stdout=(name == '-'))(self, params)
e8e73840 2900 if not test:
2901 for ph in self._progress_hooks:
2902 fd.add_progress_hook(ph)
42676437
M
2903 urls = '", "'.join(
2904 (f['url'].split(',')[0] + ',<data>' if f['url'].startswith('data:') else f['url'])
2905 for f in info.get('requested_formats', []) or [info])
3a408f9d 2906 self.write_debug(f'Invoking {fd.FD_NAME} downloader on "{urls}"')
03b4de72 2907
adbc4ec4
THD
2908 # Note: Ideally info should be a deep-copied so that hooks cannot modify it.
2909 # But it may contain objects that are not deep-copyable
2910 new_info = self._copy_infodict(info)
e8e73840 2911 if new_info.get('http_headers') is None:
2912 new_info['http_headers'] = self._calc_headers(new_info)
2913 return fd.download(name, new_info, subtitle)
2914
e04938ab 2915 def existing_file(self, filepaths, *, default_overwrite=True):
2916 existing_files = list(filter(os.path.exists, orderedSet(filepaths)))
2917 if existing_files and not self.params.get('overwrites', default_overwrite):
2918 return existing_files[0]
2919
2920 for file in existing_files:
2921 self.report_file_delete(file)
2922 os.remove(file)
2923 return None
2924
8222d8de 2925 def process_info(self, info_dict):
09b49e1f 2926 """Process a single resolved IE result. (Modifies it in-place)"""
8222d8de
JMF
2927
2928 assert info_dict.get('_type', 'video') == 'video'
f46e2f9d 2929 original_infodict = info_dict
fd288278 2930
4513a41a 2931 if 'format' not in info_dict and 'ext' in info_dict:
8222d8de
JMF
2932 info_dict['format'] = info_dict['ext']
2933
09b49e1f 2934 # This is mostly just for backward compatibility of process_info
2935 # As a side-effect, this allows for format-specific filters
c77495e3 2936 if self._match_entry(info_dict) is not None:
9e907ebd 2937 info_dict['__write_download_archive'] = 'ignore'
8222d8de
JMF
2938 return
2939
09b49e1f 2940 # Does nothing under normal operation - for backward compatibility of process_info
277d6ff5 2941 self.post_extract(info_dict)
0c14d66a 2942 self._num_downloads += 1
8222d8de 2943
dcf64d43 2944 # info_dict['_filename'] needs to be set for backward compatibility
de6000d9 2945 info_dict['_filename'] = full_filename = self.prepare_filename(info_dict, warn=True)
2946 temp_filename = self.prepare_filename(info_dict, 'temp')
0202b52a 2947 files_to_move = {}
8222d8de
JMF
2948
2949 # Forced printings
4513a41a 2950 self.__forced_printings(info_dict, full_filename, incomplete=('format' not in info_dict))
8222d8de 2951
ca6d59d2 2952 def check_max_downloads():
2953 if self._num_downloads >= float(self.params.get('max_downloads') or 'inf'):
2954 raise MaxDownloadsReached()
2955
b7b04c78 2956 if self.params.get('simulate'):
9e907ebd 2957 info_dict['__write_download_archive'] = self.params.get('force_write_download_archive')
ca6d59d2 2958 check_max_downloads()
8222d8de
JMF
2959 return
2960
de6000d9 2961 if full_filename is None:
8222d8de 2962 return
e92caff5 2963 if not self._ensure_dir_exists(encodeFilename(full_filename)):
0202b52a 2964 return
e92caff5 2965 if not self._ensure_dir_exists(encodeFilename(temp_filename)):
8222d8de
JMF
2966 return
2967
80c03fa9 2968 if self._write_description('video', info_dict,
2969 self.prepare_filename(info_dict, 'description')) is None:
2970 return
2971
2972 sub_files = self._write_subtitles(info_dict, temp_filename)
2973 if sub_files is None:
2974 return
2975 files_to_move.update(dict(sub_files))
2976
2977 thumb_files = self._write_thumbnails(
2978 'video', info_dict, temp_filename, self.prepare_filename(info_dict, 'thumbnail'))
2979 if thumb_files is None:
2980 return
2981 files_to_move.update(dict(thumb_files))
8222d8de 2982
80c03fa9 2983 infofn = self.prepare_filename(info_dict, 'infojson')
2984 _infojson_written = self._write_info_json('video', info_dict, infofn)
2985 if _infojson_written:
dac5df5a 2986 info_dict['infojson_filename'] = infofn
e75bb0d6 2987 # For backward compatibility, even though it was a private field
80c03fa9 2988 info_dict['__infojson_filename'] = infofn
2989 elif _infojson_written is None:
2990 return
2991
2992 # Note: Annotations are deprecated
2993 annofn = None
1fb07d10 2994 if self.params.get('writeannotations', False):
de6000d9 2995 annofn = self.prepare_filename(info_dict, 'annotation')
80c03fa9 2996 if annofn:
e92caff5 2997 if not self._ensure_dir_exists(encodeFilename(annofn)):
0202b52a 2998 return
0c3d0f51 2999 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(annofn)):
6febd1c1 3000 self.to_screen('[info] Video annotations are already present')
ffddb112
RA
3001 elif not info_dict.get('annotations'):
3002 self.report_warning('There are no annotations to write.')
7b6fefc9
PH
3003 else:
3004 try:
6febd1c1 3005 self.to_screen('[info] Writing video annotations to: ' + annofn)
86e5f3ed 3006 with open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
7b6fefc9
PH
3007 annofile.write(info_dict['annotations'])
3008 except (KeyError, TypeError):
6febd1c1 3009 self.report_warning('There are no annotations to write.')
86e5f3ed 3010 except OSError:
6febd1c1 3011 self.report_error('Cannot write annotations file: ' + annofn)
7b6fefc9 3012 return
1fb07d10 3013
732044af 3014 # Write internet shortcut files
08438d2c 3015 def _write_link_file(link_type):
60f3e995 3016 url = try_get(info_dict['webpage_url'], iri_to_uri)
3017 if not url:
3018 self.report_warning(
3019 f'Cannot write internet shortcut file because the actual URL of "{info_dict["webpage_url"]}" is unknown')
3020 return True
08438d2c 3021 linkfn = replace_extension(self.prepare_filename(info_dict, 'link'), link_type, info_dict.get('ext'))
0e6b018a
Z
3022 if not self._ensure_dir_exists(encodeFilename(linkfn)):
3023 return False
10e3742e 3024 if self.params.get('overwrites', True) and os.path.exists(encodeFilename(linkfn)):
08438d2c 3025 self.to_screen(f'[info] Internet shortcut (.{link_type}) is already present')
3026 return True
3027 try:
3028 self.to_screen(f'[info] Writing internet shortcut (.{link_type}) to: {linkfn}')
86e5f3ed 3029 with open(encodeFilename(to_high_limit_path(linkfn)), 'w', encoding='utf-8',
3030 newline='\r\n' if link_type == 'url' else '\n') as linkfile:
60f3e995 3031 template_vars = {'url': url}
08438d2c 3032 if link_type == 'desktop':
3033 template_vars['filename'] = linkfn[:-(len(link_type) + 1)]
3034 linkfile.write(LINK_TEMPLATES[link_type] % template_vars)
86e5f3ed 3035 except OSError:
08438d2c 3036 self.report_error(f'Cannot write internet shortcut {linkfn}')
3037 return False
732044af 3038 return True
3039
08438d2c 3040 write_links = {
3041 'url': self.params.get('writeurllink'),
3042 'webloc': self.params.get('writewebloclink'),
3043 'desktop': self.params.get('writedesktoplink'),
3044 }
3045 if self.params.get('writelink'):
3046 link_type = ('webloc' if sys.platform == 'darwin'
3047 else 'desktop' if sys.platform.startswith('linux')
3048 else 'url')
3049 write_links[link_type] = True
3050
3051 if any(should_write and not _write_link_file(link_type)
3052 for link_type, should_write in write_links.items()):
3053 return
732044af 3054
f46e2f9d 3055 def replace_info_dict(new_info):
3056 nonlocal info_dict
3057 if new_info == info_dict:
3058 return
3059 info_dict.clear()
3060 info_dict.update(new_info)
3061
415f8d51 3062 new_info, files_to_move = self.pre_process(info_dict, 'before_dl', files_to_move)
3063 replace_info_dict(new_info)
56d868db 3064
a13e6848 3065 if self.params.get('skip_download'):
56d868db 3066 info_dict['filepath'] = temp_filename
3067 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
3068 info_dict['__files_to_move'] = files_to_move
f46e2f9d 3069 replace_info_dict(self.run_pp(MoveFilesAfterDownloadPP(self, False), info_dict))
9e907ebd 3070 info_dict['__write_download_archive'] = self.params.get('force_write_download_archive')
56d868db 3071 else:
3072 # Download
b868936c 3073 info_dict.setdefault('__postprocessors', [])
4340deca 3074 try:
0202b52a 3075
e04938ab 3076 def existing_video_file(*filepaths):
6b591b29 3077 ext = info_dict.get('ext')
e04938ab 3078 converted = lambda file: replace_extension(file, self.params.get('final_ext') or ext, ext)
3079 file = self.existing_file(itertools.chain(*zip(map(converted, filepaths), filepaths)),
3080 default_overwrite=False)
3081 if file:
3082 info_dict['ext'] = os.path.splitext(file)[1][1:]
3083 return file
0202b52a 3084
7b2c3f47 3085 fd, success = None, True
fccf90e7 3086 if info_dict.get('protocol') or info_dict.get('url'):
56ba69e4 3087 fd = get_suitable_downloader(info_dict, self.params, to_stdout=temp_filename == '-')
3088 if fd is not FFmpegFD and (
3089 info_dict.get('section_start') or info_dict.get('section_end')):
7b2c3f47 3090 msg = ('This format cannot be partially downloaded' if FFmpegFD.available()
56ba69e4 3091 else 'You have requested downloading the video partially, but ffmpeg is not installed')
3092 self.report_error(f'{msg}. Aborting')
5ec1b6b7 3093 return
5ec1b6b7 3094
4340deca 3095 if info_dict.get('requested_formats') is not None:
81cd954a 3096 requested_formats = info_dict['requested_formats']
0202b52a 3097 old_ext = info_dict['ext']
4e3b637d 3098 if self.params.get('merge_output_format') is None:
4e3b637d 3099 if (info_dict['ext'] == 'webm'
3100 and info_dict.get('thumbnails')
3101 # check with type instead of pp_key, __name__, or isinstance
3102 # since we dont want any custom PPs to trigger this
c487cf00 3103 and any(type(pp) == EmbedThumbnailPP for pp in self._pps['post_process'])): # noqa: E721
4e3b637d 3104 info_dict['ext'] = 'mkv'
3105 self.report_warning(
3106 'webm doesn\'t support embedding a thumbnail, mkv will be used')
124bc071 3107 new_ext = info_dict['ext']
0202b52a 3108
124bc071 3109 def correct_ext(filename, ext=new_ext):
96fccc10 3110 if filename == '-':
3111 return filename
0202b52a 3112 filename_real_ext = os.path.splitext(filename)[1][1:]
3113 filename_wo_ext = (
3114 os.path.splitext(filename)[0]
124bc071 3115 if filename_real_ext in (old_ext, new_ext)
0202b52a 3116 else filename)
86e5f3ed 3117 return f'{filename_wo_ext}.{ext}'
0202b52a 3118
38c6902b 3119 # Ensure filename always has a correct extension for successful merge
0202b52a 3120 full_filename = correct_ext(full_filename)
3121 temp_filename = correct_ext(temp_filename)
e04938ab 3122 dl_filename = existing_video_file(full_filename, temp_filename)
1ea24129 3123 info_dict['__real_download'] = False
18e674b4 3124
7b2c3f47 3125 merger = FFmpegMergerPP(self)
adbc4ec4 3126 downloaded = []
dbf5416a 3127 if dl_filename is not None:
6c7274ec 3128 self.report_file_already_downloaded(dl_filename)
adbc4ec4
THD
3129 elif fd:
3130 for f in requested_formats if fd != FFmpegFD else []:
3131 f['filepath'] = fname = prepend_extension(
3132 correct_ext(temp_filename, info_dict['ext']),
3133 'f%s' % f['format_id'], info_dict['ext'])
3134 downloaded.append(fname)
dbf5416a 3135 info_dict['url'] = '\n'.join(f['url'] for f in requested_formats)
3136 success, real_download = self.dl(temp_filename, info_dict)
3137 info_dict['__real_download'] = real_download
18e674b4 3138 else:
18e674b4 3139 if self.params.get('allow_unplayable_formats'):
3140 self.report_warning(
3141 'You have requested merging of multiple formats '
3142 'while also allowing unplayable formats to be downloaded. '
3143 'The formats won\'t be merged to prevent data corruption.')
3144 elif not merger.available:
e8969bda 3145 msg = 'You have requested merging of multiple formats but ffmpeg is not installed'
3146 if not self.params.get('ignoreerrors'):
3147 self.report_error(f'{msg}. Aborting due to --abort-on-error')
3148 return
3149 self.report_warning(f'{msg}. The formats won\'t be merged')
18e674b4 3150
96fccc10 3151 if temp_filename == '-':
adbc4ec4 3152 reason = ('using a downloader other than ffmpeg' if FFmpegFD.can_merge_formats(info_dict, self.params)
96fccc10 3153 else 'but the formats are incompatible for simultaneous download' if merger.available
3154 else 'but ffmpeg is not installed')
3155 self.report_warning(
3156 f'You have requested downloading multiple formats to stdout {reason}. '
3157 'The formats will be streamed one after the other')
3158 fname = temp_filename
dbf5416a 3159 for f in requested_formats:
3160 new_info = dict(info_dict)
3161 del new_info['requested_formats']
3162 new_info.update(f)
96fccc10 3163 if temp_filename != '-':
124bc071 3164 fname = prepend_extension(
3165 correct_ext(temp_filename, new_info['ext']),
3166 'f%s' % f['format_id'], new_info['ext'])
96fccc10 3167 if not self._ensure_dir_exists(fname):
3168 return
a21e0ab1 3169 f['filepath'] = fname
96fccc10 3170 downloaded.append(fname)
dbf5416a 3171 partial_success, real_download = self.dl(fname, new_info)
3172 info_dict['__real_download'] = info_dict['__real_download'] or real_download
3173 success = success and partial_success
adbc4ec4
THD
3174
3175 if downloaded and merger.available and not self.params.get('allow_unplayable_formats'):
3176 info_dict['__postprocessors'].append(merger)
3177 info_dict['__files_to_merge'] = downloaded
3178 # Even if there were no downloads, it is being merged only now
3179 info_dict['__real_download'] = True
3180 else:
3181 for file in downloaded:
3182 files_to_move[file] = None
4340deca
P
3183 else:
3184 # Just a single file
e04938ab 3185 dl_filename = existing_video_file(full_filename, temp_filename)
6c7274ec 3186 if dl_filename is None or dl_filename == temp_filename:
3187 # dl_filename == temp_filename could mean that the file was partially downloaded with --no-part.
3188 # So we should try to resume the download
e8e73840 3189 success, real_download = self.dl(temp_filename, info_dict)
0202b52a 3190 info_dict['__real_download'] = real_download
6c7274ec 3191 else:
3192 self.report_file_already_downloaded(dl_filename)
0202b52a 3193
0202b52a 3194 dl_filename = dl_filename or temp_filename
c571435f 3195 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
0202b52a 3196
3158150c 3197 except network_exceptions as err:
7960b056 3198 self.report_error('unable to download video data: %s' % error_to_compat_str(err))
4340deca 3199 return
86e5f3ed 3200 except OSError as err:
4340deca
P
3201 raise UnavailableVideoError(err)
3202 except (ContentTooShortError, ) as err:
86e5f3ed 3203 self.report_error(f'content too short (expected {err.expected} bytes and served {err.downloaded})')
4340deca 3204 return
8222d8de 3205
415f8d51 3206 self._raise_pending_errors(info_dict)
de6000d9 3207 if success and full_filename != '-':
f17f8651 3208
fd7cfb64 3209 def fixup():
3210 do_fixup = True
3211 fixup_policy = self.params.get('fixup')
3212 vid = info_dict['id']
3213
3214 if fixup_policy in ('ignore', 'never'):
3215 return
3216 elif fixup_policy == 'warn':
3fe75fdc 3217 do_fixup = 'warn'
f89b3e2d 3218 elif fixup_policy != 'force':
3219 assert fixup_policy in ('detect_or_warn', None)
3220 if not info_dict.get('__real_download'):
3221 do_fixup = False
fd7cfb64 3222
3223 def ffmpeg_fixup(cndn, msg, cls):
3fe75fdc 3224 if not (do_fixup and cndn):
fd7cfb64 3225 return
3fe75fdc 3226 elif do_fixup == 'warn':
fd7cfb64 3227 self.report_warning(f'{vid}: {msg}')
3228 return
3229 pp = cls(self)
3230 if pp.available:
3231 info_dict['__postprocessors'].append(pp)
3232 else:
3233 self.report_warning(f'{vid}: {msg}. Install ffmpeg to fix this automatically')
3234
3235 stretched_ratio = info_dict.get('stretched_ratio')
ca9def71
LNO
3236 ffmpeg_fixup(stretched_ratio not in (1, None),
3237 f'Non-uniform pixel ratio {stretched_ratio}',
3238 FFmpegFixupStretchedPP)
fd7cfb64 3239
993191c0 3240 downloader = get_suitable_downloader(info_dict, self.params) if 'protocol' in info_dict else None
24146491 3241 downloader = downloader.FD_NAME if downloader else None
adbc4ec4 3242
ca9def71
LNO
3243 ext = info_dict.get('ext')
3244 postprocessed_by_ffmpeg = info_dict.get('requested_formats') or any((
3245 isinstance(pp, FFmpegVideoConvertorPP)
3246 and resolve_recode_mapping(ext, pp.mapping)[0] not in (ext, None)
3247 ) for pp in self._pps['post_process'])
3248
3249 if not postprocessed_by_ffmpeg:
3250 ffmpeg_fixup(ext == 'm4a' and info_dict.get('container') == 'm4a_dash',
f2df4071 3251 'writing DASH m4a. Only some players support this container',
3252 FFmpegFixupM4aPP)
24146491 3253 ffmpeg_fixup(downloader == 'hlsnative' and not self.params.get('hls_use_mpegts')
494f5230 3254 or info_dict.get('is_live') and self.params.get('hls_use_mpegts') is None,
adbc4ec4
THD
3255 'Possible MPEG-TS in MP4 container or malformed AAC timestamps',
3256 FFmpegFixupM3u8PP)
3257 ffmpeg_fixup(info_dict.get('is_live') and downloader == 'DashSegmentsFD',
3258 'Possible duplicate MOOV atoms', FFmpegFixupDuplicateMoovPP)
3259
24146491 3260 ffmpeg_fixup(downloader == 'web_socket_fragment', 'Malformed timestamps detected', FFmpegFixupTimestampPP)
3261 ffmpeg_fixup(downloader == 'web_socket_fragment', 'Malformed duration detected', FFmpegFixupDurationPP)
fd7cfb64 3262
3263 fixup()
8222d8de 3264 try:
f46e2f9d 3265 replace_info_dict(self.post_process(dl_filename, info_dict, files_to_move))
af819c21 3266 except PostProcessingError as err:
3267 self.report_error('Postprocessing: %s' % str(err))
8222d8de 3268 return
ab8e5e51
AM
3269 try:
3270 for ph in self._post_hooks:
23c1a667 3271 ph(info_dict['filepath'])
ab8e5e51
AM
3272 except Exception as err:
3273 self.report_error('post hooks: %s' % str(err))
3274 return
9e907ebd 3275 info_dict['__write_download_archive'] = True
2d30509f 3276
c487cf00 3277 assert info_dict is original_infodict # Make sure the info_dict was modified in-place
a13e6848 3278 if self.params.get('force_write_download_archive'):
9e907ebd 3279 info_dict['__write_download_archive'] = True
ca6d59d2 3280 check_max_downloads()
8222d8de 3281
aa9369a2 3282 def __download_wrapper(self, func):
3283 @functools.wraps(func)
3284 def wrapper(*args, **kwargs):
3285 try:
3286 res = func(*args, **kwargs)
3287 except UnavailableVideoError as e:
3288 self.report_error(e)
b222c271 3289 except DownloadCancelled as e:
3290 self.to_screen(f'[info] {e}')
3291 if not self.params.get('break_per_url'):
3292 raise
fd404bec 3293 self._num_downloads = 0
aa9369a2 3294 else:
3295 if self.params.get('dump_single_json', False):
3296 self.post_extract(res)
3297 self.to_stdout(json.dumps(self.sanitize_info(res)))
3298 return wrapper
3299
8222d8de
JMF
3300 def download(self, url_list):
3301 """Download a given list of URLs."""
aa9369a2 3302 url_list = variadic(url_list) # Passing a single URL is a common mistake
bf1824b3 3303 outtmpl = self.params['outtmpl']['default']
3089bc74
S
3304 if (len(url_list) > 1
3305 and outtmpl != '-'
3306 and '%' not in outtmpl
3307 and self.params.get('max_downloads') != 1):
acd69589 3308 raise SameFileError(outtmpl)
8222d8de
JMF
3309
3310 for url in url_list:
aa9369a2 3311 self.__download_wrapper(self.extract_info)(
3312 url, force_generic_extractor=self.params.get('force_generic_extractor', False))
8222d8de
JMF
3313
3314 return self._download_retcode
3315
1dcc4c0c 3316 def download_with_info_file(self, info_filename):
31bd3925
JMF
3317 with contextlib.closing(fileinput.FileInput(
3318 [info_filename], mode='r',
3319 openhook=fileinput.hook_encoded('utf-8'))) as f:
3320 # FileInput doesn't have a read method, we can't call json.load
8012d892 3321 info = self.sanitize_info(json.loads('\n'.join(f)), self.params.get('clean_infojson', True))
d4943898 3322 try:
aa9369a2 3323 self.__download_wrapper(self.process_ie_result)(info, download=True)
f2ebc5c7 3324 except (DownloadError, EntryNotInPlaylist, ReExtractInfo) as e:
bf5f605e 3325 if not isinstance(e, EntryNotInPlaylist):
3326 self.to_stderr('\r')
d4943898
JMF
3327 webpage_url = info.get('webpage_url')
3328 if webpage_url is not None:
aa9369a2 3329 self.report_warning(f'The info failed to download: {e}; trying with URL {webpage_url}')
d4943898
JMF
3330 return self.download([webpage_url])
3331 else:
3332 raise
3333 return self._download_retcode
1dcc4c0c 3334
cb202fd2 3335 @staticmethod
8012d892 3336 def sanitize_info(info_dict, remove_private_keys=False):
3337 ''' Sanitize the infodict for converting to json '''
3ad56b42 3338 if info_dict is None:
3339 return info_dict
6e84b215 3340 info_dict.setdefault('epoch', int(time.time()))
6a5a30f9 3341 info_dict.setdefault('_type', 'video')
b5e7a2e6 3342 info_dict.setdefault('_version', {
3343 'version': __version__,
3344 'current_git_head': current_git_head(),
3345 'release_git_head': RELEASE_GIT_HEAD,
3346 'repository': REPOSITORY,
3347 })
09b49e1f 3348
8012d892 3349 if remove_private_keys:
0a5a191a 3350 reject = lambda k, v: v is None or k.startswith('__') or k in {
f46e2f9d 3351 'requested_downloads', 'requested_formats', 'requested_subtitles', 'requested_entries',
0a5a191a 3352 'entries', 'filepath', '_filename', 'infojson_filename', 'original_url', 'playlist_autonumber',
6e84b215 3353 }
ae8f99e6 3354 else:
09b49e1f 3355 reject = lambda k, v: False
adbc4ec4
THD
3356
3357 def filter_fn(obj):
3358 if isinstance(obj, dict):
3359 return {k: filter_fn(v) for k, v in obj.items() if not reject(k, v)}
3360 elif isinstance(obj, (list, tuple, set, LazyList)):
3361 return list(map(filter_fn, obj))
3362 elif obj is None or isinstance(obj, (str, int, float, bool)):
3363 return obj
3364 else:
3365 return repr(obj)
3366
5226731e 3367 return filter_fn(info_dict)
cb202fd2 3368
8012d892 3369 @staticmethod
3370 def filter_requested_info(info_dict, actually_filter=True):
3371 ''' Alias of sanitize_info for backward compatibility '''
3372 return YoutubeDL.sanitize_info(info_dict, actually_filter)
3373
43d7f5a5 3374 def _delete_downloaded_files(self, *files_to_delete, info={}, msg=None):
3375 for filename in set(filter(None, files_to_delete)):
3376 if msg:
3377 self.to_screen(msg % filename)
3378 try:
3379 os.remove(filename)
3380 except OSError:
3381 self.report_warning(f'Unable to delete file {filename}')
3382 if filename in info.get('__files_to_move', []): # NB: Delete even if None
3383 del info['__files_to_move'][filename]
3384
ed5835b4 3385 @staticmethod
3386 def post_extract(info_dict):
3387 def actual_post_extract(info_dict):
3388 if info_dict.get('_type') in ('playlist', 'multi_video'):
3389 for video_dict in info_dict.get('entries', {}):
3390 actual_post_extract(video_dict or {})
3391 return
3392
09b49e1f 3393 post_extractor = info_dict.pop('__post_extractor', None) or (lambda: {})
3394 info_dict.update(post_extractor())
ed5835b4 3395
3396 actual_post_extract(info_dict or {})
3397
dcf64d43 3398 def run_pp(self, pp, infodict):
5bfa4862 3399 files_to_delete = []
dcf64d43 3400 if '__files_to_move' not in infodict:
3401 infodict['__files_to_move'] = {}
b1940459 3402 try:
3403 files_to_delete, infodict = pp.run(infodict)
3404 except PostProcessingError as e:
3405 # Must be True and not 'only_download'
3406 if self.params.get('ignoreerrors') is True:
3407 self.report_error(e)
3408 return infodict
3409 raise
3410
5bfa4862 3411 if not files_to_delete:
dcf64d43 3412 return infodict
5bfa4862 3413 if self.params.get('keepvideo', False):
3414 for f in files_to_delete:
dcf64d43 3415 infodict['__files_to_move'].setdefault(f, '')
5bfa4862 3416 else:
43d7f5a5 3417 self._delete_downloaded_files(
3418 *files_to_delete, info=infodict, msg='Deleting original file %s (pass -k to keep)')
dcf64d43 3419 return infodict
5bfa4862 3420
ed5835b4 3421 def run_all_pps(self, key, info, *, additional_pps=None):
bb66c247 3422 self._forceprint(key, info)
ed5835b4 3423 for pp in (additional_pps or []) + self._pps[key]:
dc5f409c 3424 info = self.run_pp(pp, info)
ed5835b4 3425 return info
277d6ff5 3426
56d868db 3427 def pre_process(self, ie_info, key='pre_process', files_to_move=None):
5bfa4862 3428 info = dict(ie_info)
56d868db 3429 info['__files_to_move'] = files_to_move or {}
415f8d51 3430 try:
3431 info = self.run_all_pps(key, info)
3432 except PostProcessingError as err:
3433 msg = f'Preprocessing: {err}'
3434 info.setdefault('__pending_error', msg)
3435 self.report_error(msg, is_error=False)
56d868db 3436 return info, info.pop('__files_to_move', None)
5bfa4862 3437
f46e2f9d 3438 def post_process(self, filename, info, files_to_move=None):
8222d8de 3439 """Run all the postprocessors on the given file."""
8222d8de 3440 info['filepath'] = filename
dcf64d43 3441 info['__files_to_move'] = files_to_move or {}
ed5835b4 3442 info = self.run_all_pps('post_process', info, additional_pps=info.get('__postprocessors'))
dcf64d43 3443 info = self.run_pp(MoveFilesAfterDownloadPP(self), info)
3444 del info['__files_to_move']
ed5835b4 3445 return self.run_all_pps('after_move', info)
c1c9a79c 3446
5db07df6 3447 def _make_archive_id(self, info_dict):
e9fef7ee
S
3448 video_id = info_dict.get('id')
3449 if not video_id:
3450 return
5db07df6
PH
3451 # Future-proof against any change in case
3452 # and backwards compatibility with prior versions
e9fef7ee 3453 extractor = info_dict.get('extractor_key') or info_dict.get('ie_key') # key in a playlist
7012b23c 3454 if extractor is None:
1211bb6d
S
3455 url = str_or_none(info_dict.get('url'))
3456 if not url:
3457 return
e9fef7ee 3458 # Try to find matching extractor for the URL and take its ie_key
8b7491c8 3459 for ie_key, ie in self._ies.items():
1211bb6d 3460 if ie.suitable(url):
8b7491c8 3461 extractor = ie_key
e9fef7ee
S
3462 break
3463 else:
3464 return
0647d925 3465 return make_archive_id(extractor, video_id)
5db07df6
PH
3466
3467 def in_download_archive(self, info_dict):
3468 fn = self.params.get('download_archive')
3469 if fn is None:
3470 return False
3471
1e8fe57e 3472 vid_ids = [self._make_archive_id(info_dict)]
c200096c 3473 vid_ids.extend(info_dict.get('_old_archive_ids') or [])
1e8fe57e 3474 return any(id_ in self.archive for id_ in vid_ids)
c1c9a79c
PH
3475
3476 def record_download_archive(self, info_dict):
3477 fn = self.params.get('download_archive')
3478 if fn is None:
3479 return
5db07df6
PH
3480 vid_id = self._make_archive_id(info_dict)
3481 assert vid_id
a13e6848 3482 self.write_debug(f'Adding to archive: {vid_id}')
c1c9a79c 3483 with locked_file(fn, 'a', encoding='utf-8') as archive_file:
6febd1c1 3484 archive_file.write(vid_id + '\n')
a45e8619 3485 self.archive.add(vid_id)
dd82ffea 3486
8c51aa65 3487 @staticmethod
8abeeb94 3488 def format_resolution(format, default='unknown'):
9359f3d4 3489 if format.get('vcodec') == 'none' and format.get('acodec') != 'none':
fb04e403 3490 return 'audio only'
f49d89ee
PH
3491 if format.get('resolution') is not None:
3492 return format['resolution']
35615307 3493 if format.get('width') and format.get('height'):
ff51ed58 3494 return '%dx%d' % (format['width'], format['height'])
35615307 3495 elif format.get('height'):
ff51ed58 3496 return '%sp' % format['height']
35615307 3497 elif format.get('width'):
ff51ed58 3498 return '%dx?' % format['width']
3499 return default
8c51aa65 3500
8130779d 3501 def _list_format_headers(self, *headers):
3502 if self.params.get('listformats_table', True) is not False:
591bb9d3 3503 return [self._format_out(header, self.Styles.HEADERS) for header in headers]
8130779d 3504 return headers
3505
c57f7757
PH
3506 def _format_note(self, fdict):
3507 res = ''
3508 if fdict.get('ext') in ['f4f', 'f4m']:
f304da8a 3509 res += '(unsupported)'
32f90364
PH
3510 if fdict.get('language'):
3511 if res:
3512 res += ' '
f304da8a 3513 res += '[%s]' % fdict['language']
c57f7757 3514 if fdict.get('format_note') is not None:
f304da8a 3515 if res:
3516 res += ' '
3517 res += fdict['format_note']
c57f7757 3518 if fdict.get('tbr') is not None:
f304da8a 3519 if res:
3520 res += ', '
3521 res += '%4dk' % fdict['tbr']
c57f7757
PH
3522 if fdict.get('container') is not None:
3523 if res:
3524 res += ', '
3525 res += '%s container' % fdict['container']
3089bc74
S
3526 if (fdict.get('vcodec') is not None
3527 and fdict.get('vcodec') != 'none'):
c57f7757
PH
3528 if res:
3529 res += ', '
3530 res += fdict['vcodec']
91c7271a 3531 if fdict.get('vbr') is not None:
c57f7757
PH
3532 res += '@'
3533 elif fdict.get('vbr') is not None and fdict.get('abr') is not None:
3534 res += 'video@'
3535 if fdict.get('vbr') is not None:
3536 res += '%4dk' % fdict['vbr']
fbb21cf5 3537 if fdict.get('fps') is not None:
5d583bdf
S
3538 if res:
3539 res += ', '
3540 res += '%sfps' % fdict['fps']
c57f7757
PH
3541 if fdict.get('acodec') is not None:
3542 if res:
3543 res += ', '
3544 if fdict['acodec'] == 'none':
3545 res += 'video only'
3546 else:
3547 res += '%-5s' % fdict['acodec']
3548 elif fdict.get('abr') is not None:
3549 if res:
3550 res += ', '
3551 res += 'audio'
3552 if fdict.get('abr') is not None:
3553 res += '@%3dk' % fdict['abr']
3554 if fdict.get('asr') is not None:
3555 res += ' (%5dHz)' % fdict['asr']
3556 if fdict.get('filesize') is not None:
3557 if res:
3558 res += ', '
3559 res += format_bytes(fdict['filesize'])
9732d77e
PH
3560 elif fdict.get('filesize_approx') is not None:
3561 if res:
3562 res += ', '
3563 res += '~' + format_bytes(fdict['filesize_approx'])
c57f7757 3564 return res
91c7271a 3565
8130779d 3566 def render_formats_table(self, info_dict):
b69fd25c 3567 if not info_dict.get('formats') and not info_dict.get('url'):
8130779d 3568 return None
b69fd25c 3569
94badb25 3570 formats = info_dict.get('formats', [info_dict])
8130779d 3571 if not self.params.get('listformats_table', True) is not False:
76d321f6 3572 table = [
3573 [
3574 format_field(f, 'format_id'),
3575 format_field(f, 'ext'),
3576 self.format_resolution(f),
8130779d 3577 self._format_note(f)
3578 ] for f in formats if f.get('preference') is None or f['preference'] >= -1000]
3579 return render_table(['format code', 'extension', 'resolution', 'note'], table, extra_gap=1)
3580
d816f61f 3581 def simplified_codec(f, field):
3582 assert field in ('acodec', 'vcodec')
3583 codec = f.get(field, 'unknown')
f5ea4748 3584 if not codec:
3585 return 'unknown'
3586 elif codec != 'none':
d816f61f 3587 return '.'.join(codec.split('.')[:4])
3588
3589 if field == 'vcodec' and f.get('acodec') == 'none':
3590 return 'images'
3591 elif field == 'acodec' and f.get('vcodec') == 'none':
3592 return ''
3593 return self._format_out('audio only' if field == 'vcodec' else 'video only',
3594 self.Styles.SUPPRESS)
3595
591bb9d3 3596 delim = self._format_out('\u2502', self.Styles.DELIM, '|', test_encoding=True)
8130779d 3597 table = [
3598 [
591bb9d3 3599 self._format_out(format_field(f, 'format_id'), self.Styles.ID),
8130779d 3600 format_field(f, 'ext'),
3601 format_field(f, func=self.format_resolution, ignore=('audio only', 'images')),
563e0bf8 3602 format_field(f, 'fps', '\t%d', func=round),
8130779d 3603 format_field(f, 'dynamic_range', '%s', ignore=(None, 'SDR')).replace('HDR', ''),
b8ed0f15 3604 format_field(f, 'audio_channels', '\t%s'),
8130779d 3605 delim,
3606 format_field(f, 'filesize', ' \t%s', func=format_bytes) + format_field(f, 'filesize_approx', '~\t%s', func=format_bytes),
563e0bf8 3607 format_field(f, 'tbr', '\t%dk', func=round),
8130779d 3608 shorten_protocol_name(f.get('protocol', '')),
3609 delim,
d816f61f 3610 simplified_codec(f, 'vcodec'),
563e0bf8 3611 format_field(f, 'vbr', '\t%dk', func=round),
d816f61f 3612 simplified_codec(f, 'acodec'),
563e0bf8 3613 format_field(f, 'abr', '\t%dk', func=round),
ae61d108 3614 format_field(f, 'asr', '\t%s', func=format_decimal_suffix),
8130779d 3615 join_nonempty(
591bb9d3 3616 self._format_out('UNSUPPORTED', 'light red') if f.get('ext') in ('f4f', 'f4m') else None,
8130779d 3617 format_field(f, 'language', '[%s]'),
3618 join_nonempty(format_field(f, 'format_note'),
3619 format_field(f, 'container', ignore=(None, f.get('ext'))),
3620 delim=', '),
3621 delim=' '),
3622 ] for f in formats if f.get('preference') is None or f['preference'] >= -1000]
3623 header_line = self._list_format_headers(
b8ed0f15 3624 'ID', 'EXT', 'RESOLUTION', '\tFPS', 'HDR', 'CH', delim, '\tFILESIZE', '\tTBR', 'PROTO',
8130779d 3625 delim, 'VCODEC', '\tVBR', 'ACODEC', '\tABR', '\tASR', 'MORE INFO')
3626
3627 return render_table(
3628 header_line, table, hide_empty=True,
591bb9d3 3629 delim=self._format_out('\u2500', self.Styles.DELIM, '-', test_encoding=True))
8130779d 3630
3631 def render_thumbnails_table(self, info_dict):
88f23a18 3632 thumbnails = list(info_dict.get('thumbnails') or [])
cfb56d1a 3633 if not thumbnails:
8130779d 3634 return None
3635 return render_table(
ec11a9f4 3636 self._list_format_headers('ID', 'Width', 'Height', 'URL'),
6970b600 3637 [[t.get('id'), t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails])
2412044c 3638
8130779d 3639 def render_subtitles_table(self, video_id, subtitles):
2412044c 3640 def _row(lang, formats):
49c258e1 3641 exts, names = zip(*((f['ext'], f.get('name') or 'unknown') for f in reversed(formats)))
2412044c 3642 if len(set(names)) == 1:
7aee40c1 3643 names = [] if names[0] == 'unknown' else names[:1]
2412044c 3644 return [lang, ', '.join(names), ', '.join(exts)]
3645
8130779d 3646 if not subtitles:
3647 return None
3648 return render_table(
ec11a9f4 3649 self._list_format_headers('Language', 'Name', 'Formats'),
2412044c 3650 [_row(lang, formats) for lang, formats in subtitles.items()],
8130779d 3651 hide_empty=True)
3652
3653 def __list_table(self, video_id, name, func, *args):
3654 table = func(*args)
3655 if not table:
3656 self.to_screen(f'{video_id} has no {name}')
3657 return
3658 self.to_screen(f'[info] Available {name} for {video_id}:')
3659 self.to_stdout(table)
3660
3661 def list_formats(self, info_dict):
3662 self.__list_table(info_dict['id'], 'formats', self.render_formats_table, info_dict)
3663
3664 def list_thumbnails(self, info_dict):
3665 self.__list_table(info_dict['id'], 'thumbnails', self.render_thumbnails_table, info_dict)
3666
3667 def list_subtitles(self, video_id, subtitles, name='subtitles'):
3668 self.__list_table(video_id, name, self.render_subtitles_table, video_id, subtitles)
a504ced0 3669
dca08720
PH
3670 def urlopen(self, req):
3671 """ Start an HTTP download """
f9934b96 3672 if isinstance(req, str):
67dda517 3673 req = sanitized_Request(req)
19a41fc6 3674 return self._opener.open(req, timeout=self._socket_timeout)
dca08720
PH
3675
3676 def print_debug_header(self):
3677 if not self.params.get('verbose'):
3678 return
49a57e70 3679
560738f3 3680 # These imports can be slow. So import them only as needed
3681 from .extractor.extractors import _LAZY_LOADER
3682 from .extractor.extractors import _PLUGIN_CLASSES as plugin_extractors
3683
49a57e70 3684 def get_encoding(stream):
2a938746 3685 ret = str(getattr(stream, 'encoding', 'missing (%s)' % type(stream).__name__))
49a57e70 3686 if not supports_terminal_sequences(stream):
53973b4d 3687 from .utils import WINDOWS_VT_MODE # Must be imported locally
e3c7d495 3688 ret += ' (No VT)' if WINDOWS_VT_MODE is False else ' (No ANSI)'
49a57e70 3689 return ret
3690
591bb9d3 3691 encoding_str = 'Encodings: locale %s, fs %s, pref %s, %s' % (
49a57e70 3692 locale.getpreferredencoding(),
3693 sys.getfilesystemencoding(),
591bb9d3 3694 self.get_encoding(),
3695 ', '.join(
64fa820c 3696 f'{key} {get_encoding(stream)}' for key, stream in self._out_files.items_
591bb9d3 3697 if stream is not None and key != 'console')
3698 )
883d4b1e 3699
3700 logger = self.params.get('logger')
3701 if logger:
3702 write_debug = lambda msg: logger.debug(f'[debug] {msg}')
3703 write_debug(encoding_str)
3704 else:
96565c7e 3705 write_string(f'[debug] {encoding_str}\n', encoding=None)
49a57e70 3706 write_debug = lambda msg: self._write_string(f'[debug] {msg}\n')
734f90bb 3707
4c88ff87 3708 source = detect_variant()
70b23409 3709 if VARIANT not in (None, 'pip'):
3710 source += '*'
36eaf303 3711 write_debug(join_nonempty(
b5e7a2e6 3712 f'{"yt-dlp" if REPOSITORY == "yt-dlp/yt-dlp" else REPOSITORY} version',
3713 __version__,
36eaf303 3714 f'[{RELEASE_GIT_HEAD}]' if RELEASE_GIT_HEAD else '',
3715 '' if source == 'unknown' else f'({source})',
3716 delim=' '))
6e21fdd2 3717 if not _LAZY_LOADER:
3718 if os.environ.get('YTDLP_NO_LAZY_EXTRACTORS'):
49a57e70 3719 write_debug('Lazy loading extractors is forcibly disabled')
6e21fdd2 3720 else:
49a57e70 3721 write_debug('Lazy loading extractors is disabled')
3ae5e797 3722 if plugin_extractors or plugin_postprocessors:
49a57e70 3723 write_debug('Plugins: %s' % [
3ae5e797 3724 '%s%s' % (klass.__name__, '' if klass.__name__ == name else f' as {name}')
3725 for name, klass in itertools.chain(plugin_extractors.items(), plugin_postprocessors.items())])
8a82af35 3726 if self.params['compat_opts']:
3727 write_debug('Compatibility options: %s' % ', '.join(self.params['compat_opts']))
36eaf303 3728
b5e7a2e6 3729 if current_git_head():
3730 write_debug(f'Git HEAD: {current_git_head()}')
b1f94422 3731 write_debug(system_identifier())
d28b5171 3732
8913ef74 3733 exe_versions, ffmpeg_features = FFmpegPostProcessor.get_versions_and_features(self)
3734 ffmpeg_features = {key for key, val in ffmpeg_features.items() if val}
3735 if ffmpeg_features:
19a03940 3736 exe_versions['ffmpeg'] += ' (%s)' % ','.join(sorted(ffmpeg_features))
8913ef74 3737
4c83c967 3738 exe_versions['rtmpdump'] = rtmpdump_version()
feee8d32 3739 exe_versions['phantomjs'] = PhantomJSwrapper._version()
d28b5171 3740 exe_str = ', '.join(
2831b468 3741 f'{exe} {v}' for exe, v in sorted(exe_versions.items()) if v
3742 ) or 'none'
49a57e70 3743 write_debug('exe versions: %s' % exe_str)
dca08720 3744
1d485a1a 3745 from .compat.compat_utils import get_package_info
9b8ee23b 3746 from .dependencies import available_dependencies
3747
3748 write_debug('Optional libraries: %s' % (', '.join(sorted({
1d485a1a 3749 join_nonempty(*get_package_info(m)) for m in available_dependencies.values()
9b8ee23b 3750 })) or 'none'))
2831b468 3751
97ec5bc5 3752 self._setup_opener()
dca08720
PH
3753 proxy_map = {}
3754 for handler in self._opener.handlers:
3755 if hasattr(handler, 'proxies'):
3756 proxy_map.update(handler.proxies)
49a57e70 3757 write_debug(f'Proxy map: {proxy_map}')
dca08720 3758
49a57e70 3759 # Not implemented
3760 if False and self.params.get('call_home'):
0f06bcd7 3761 ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode()
49a57e70 3762 write_debug('Public IP address: %s' % ipaddr)
58b1f00d 3763 latest_version = self.urlopen(
0f06bcd7 3764 'https://yt-dl.org/latest/version').read().decode()
58b1f00d
PH
3765 if version_tuple(latest_version) > version_tuple(__version__):
3766 self.report_warning(
3767 'You are using an outdated version (newest version: %s)! '
3768 'See https://yt-dl.org/update if you need help updating.' %
3769 latest_version)
3770
e344693b 3771 def _setup_opener(self):
97ec5bc5 3772 if hasattr(self, '_opener'):
3773 return
6ad14cab 3774 timeout_val = self.params.get('socket_timeout')
17bddf3e 3775 self._socket_timeout = 20 if timeout_val is None else float(timeout_val)
6ad14cab 3776
982ee69a 3777 opts_cookiesfrombrowser = self.params.get('cookiesfrombrowser')
dca08720
PH
3778 opts_cookiefile = self.params.get('cookiefile')
3779 opts_proxy = self.params.get('proxy')
3780
982ee69a 3781 self.cookiejar = load_cookies(opts_cookiefile, opts_cookiesfrombrowser, self)
dca08720 3782
6a3f4c3f 3783 cookie_processor = YoutubeDLCookieProcessor(self.cookiejar)
dca08720
PH
3784 if opts_proxy is not None:
3785 if opts_proxy == '':
3786 proxies = {}
3787 else:
3788 proxies = {'http': opts_proxy, 'https': opts_proxy}
3789 else:
ac668111 3790 proxies = urllib.request.getproxies()
067aa17e 3791 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
dca08720
PH
3792 if 'http' in proxies and 'https' not in proxies:
3793 proxies['https'] = proxies['http']
91410c9b 3794 proxy_handler = PerRequestProxyHandler(proxies)
a0ddb8a2
PH
3795
3796 debuglevel = 1 if self.params.get('debug_printtraffic') else 0
be4a824d
PH
3797 https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel)
3798 ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel)
fca6dba8 3799 redirect_handler = YoutubeDLRedirectHandler()
f9934b96 3800 data_handler = urllib.request.DataHandler()
6240b0a2
JMF
3801
3802 # When passing our own FileHandler instance, build_opener won't add the
3803 # default FileHandler and allows us to disable the file protocol, which
3804 # can be used for malicious purposes (see
067aa17e 3805 # https://github.com/ytdl-org/youtube-dl/issues/8227)
ac668111 3806 file_handler = urllib.request.FileHandler()
6240b0a2
JMF
3807
3808 def file_open(*args, **kwargs):
ac668111 3809 raise urllib.error.URLError('file:// scheme is explicitly disabled in yt-dlp for security reasons')
6240b0a2
JMF
3810 file_handler.file_open = file_open
3811
ac668111 3812 opener = urllib.request.build_opener(
fca6dba8 3813 proxy_handler, https_handler, cookie_processor, ydlh, redirect_handler, data_handler, file_handler)
2461f79d 3814
dca08720
PH
3815 # Delete the default user-agent header, which would otherwise apply in
3816 # cases where our custom HTTP handler doesn't come into play
067aa17e 3817 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
dca08720
PH
3818 opener.addheaders = []
3819 self._opener = opener
62fec3b2
PH
3820
3821 def encode(self, s):
3822 if isinstance(s, bytes):
3823 return s # Already encoded
3824
3825 try:
3826 return s.encode(self.get_encoding())
3827 except UnicodeEncodeError as err:
3828 err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.'
3829 raise
3830
3831 def get_encoding(self):
3832 encoding = self.params.get('encoding')
3833 if encoding is None:
3834 encoding = preferredencoding()
3835 return encoding
ec82d85a 3836
e08a85d8 3837 def _write_info_json(self, label, ie_result, infofn, overwrite=None):
cb96c5be 3838 ''' Write infojson and returns True = written, 'exists' = Already exists, False = skip, None = error '''
e08a85d8 3839 if overwrite is None:
3840 overwrite = self.params.get('overwrites', True)
80c03fa9 3841 if not self.params.get('writeinfojson'):
3842 return False
3843 elif not infofn:
3844 self.write_debug(f'Skipping writing {label} infojson')
3845 return False
3846 elif not self._ensure_dir_exists(infofn):
3847 return None
e08a85d8 3848 elif not overwrite and os.path.exists(infofn):
80c03fa9 3849 self.to_screen(f'[info] {label.title()} metadata is already present')
cb96c5be 3850 return 'exists'
3851
3852 self.to_screen(f'[info] Writing {label} metadata as JSON to: {infofn}')
3853 try:
3854 write_json_file(self.sanitize_info(ie_result, self.params.get('clean_infojson', True)), infofn)
3855 return True
86e5f3ed 3856 except OSError:
cb96c5be 3857 self.report_error(f'Cannot write {label} metadata to JSON file {infofn}')
3858 return None
80c03fa9 3859
3860 def _write_description(self, label, ie_result, descfn):
3861 ''' Write description and returns True = written, False = skip, None = error '''
3862 if not self.params.get('writedescription'):
3863 return False
3864 elif not descfn:
3865 self.write_debug(f'Skipping writing {label} description')
3866 return False
3867 elif not self._ensure_dir_exists(descfn):
3868 return None
3869 elif not self.params.get('overwrites', True) and os.path.exists(descfn):
3870 self.to_screen(f'[info] {label.title()} description is already present')
3871 elif ie_result.get('description') is None:
3872 self.report_warning(f'There\'s no {label} description to write')
3873 return False
3874 else:
3875 try:
3876 self.to_screen(f'[info] Writing {label} description to: {descfn}')
86e5f3ed 3877 with open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
80c03fa9 3878 descfile.write(ie_result['description'])
86e5f3ed 3879 except OSError:
80c03fa9 3880 self.report_error(f'Cannot write {label} description file {descfn}')
3881 return None
3882 return True
3883
3884 def _write_subtitles(self, info_dict, filename):
3885 ''' Write subtitles to file and return list of (sub_filename, final_sub_filename); or None if error'''
3886 ret = []
3887 subtitles = info_dict.get('requested_subtitles')
3888 if not subtitles or not (self.params.get('writesubtitles') or self.params.get('writeautomaticsub')):
3889 # subtitles download errors are already managed as troubles in relevant IE
3890 # that way it will silently go on when used with unsupporting IE
3891 return ret
3892
3893 sub_filename_base = self.prepare_filename(info_dict, 'subtitle')
3894 if not sub_filename_base:
3895 self.to_screen('[info] Skipping writing video subtitles')
3896 return ret
3897 for sub_lang, sub_info in subtitles.items():
3898 sub_format = sub_info['ext']
3899 sub_filename = subtitles_filename(filename, sub_lang, sub_format, info_dict.get('ext'))
3900 sub_filename_final = subtitles_filename(sub_filename_base, sub_lang, sub_format, info_dict.get('ext'))
e04938ab 3901 existing_sub = self.existing_file((sub_filename_final, sub_filename))
3902 if existing_sub:
80c03fa9 3903 self.to_screen(f'[info] Video subtitle {sub_lang}.{sub_format} is already present')
e04938ab 3904 sub_info['filepath'] = existing_sub
3905 ret.append((existing_sub, sub_filename_final))
80c03fa9 3906 continue
3907
3908 self.to_screen(f'[info] Writing video subtitles to: {sub_filename}')
3909 if sub_info.get('data') is not None:
3910 try:
3911 # Use newline='' to prevent conversion of newline characters
3912 # See https://github.com/ytdl-org/youtube-dl/issues/10268
86e5f3ed 3913 with open(sub_filename, 'w', encoding='utf-8', newline='') as subfile:
80c03fa9 3914 subfile.write(sub_info['data'])
3915 sub_info['filepath'] = sub_filename
3916 ret.append((sub_filename, sub_filename_final))
3917 continue
86e5f3ed 3918 except OSError:
80c03fa9 3919 self.report_error(f'Cannot write video subtitles file {sub_filename}')
3920 return None
3921
3922 try:
3923 sub_copy = sub_info.copy()
3924 sub_copy.setdefault('http_headers', info_dict.get('http_headers'))
3925 self.dl(sub_filename, sub_copy, subtitle=True)
3926 sub_info['filepath'] = sub_filename
3927 ret.append((sub_filename, sub_filename_final))
6020e05d 3928 except (DownloadError, ExtractorError, IOError, OSError, ValueError) + network_exceptions as err:
c70c418d 3929 msg = f'Unable to download video subtitles for {sub_lang!r}: {err}'
6020e05d 3930 if self.params.get('ignoreerrors') is not True: # False or 'only_download'
c70c418d 3931 if not self.params.get('ignoreerrors'):
3932 self.report_error(msg)
3933 raise DownloadError(msg)
3934 self.report_warning(msg)
519804a9 3935 return ret
80c03fa9 3936
3937 def _write_thumbnails(self, label, info_dict, filename, thumb_filename_base=None):
3938 ''' Write thumbnails to file and return list of (thumb_filename, final_thumb_filename) '''
6c4fd172 3939 write_all = self.params.get('write_all_thumbnails', False)
80c03fa9 3940 thumbnails, ret = [], []
6c4fd172 3941 if write_all or self.params.get('writethumbnail', False):
0202b52a 3942 thumbnails = info_dict.get('thumbnails') or []
6c4fd172 3943 multiple = write_all and len(thumbnails) > 1
ec82d85a 3944
80c03fa9 3945 if thumb_filename_base is None:
3946 thumb_filename_base = filename
3947 if thumbnails and not thumb_filename_base:
3948 self.write_debug(f'Skipping writing {label} thumbnail')
3949 return ret
3950
dd0228ce 3951 for idx, t in list(enumerate(thumbnails))[::-1]:
80c03fa9 3952 thumb_ext = (f'{t["id"]}.' if multiple else '') + determine_ext(t['url'], 'jpg')
aa9369a2 3953 thumb_display_id = f'{label} thumbnail {t["id"]}'
80c03fa9 3954 thumb_filename = replace_extension(filename, thumb_ext, info_dict.get('ext'))
3955 thumb_filename_final = replace_extension(thumb_filename_base, thumb_ext, info_dict.get('ext'))
ec82d85a 3956
e04938ab 3957 existing_thumb = self.existing_file((thumb_filename_final, thumb_filename))
3958 if existing_thumb:
aa9369a2 3959 self.to_screen('[info] %s is already present' % (
3960 thumb_display_id if multiple else f'{label} thumbnail').capitalize())
e04938ab 3961 t['filepath'] = existing_thumb
3962 ret.append((existing_thumb, thumb_filename_final))
ec82d85a 3963 else:
80c03fa9 3964 self.to_screen(f'[info] Downloading {thumb_display_id} ...')
ec82d85a 3965 try:
297e9952 3966 uf = self.urlopen(sanitized_Request(t['url'], headers=t.get('http_headers', {})))
80c03fa9 3967 self.to_screen(f'[info] Writing {thumb_display_id} to: {thumb_filename}')
d3d89c32 3968 with open(encodeFilename(thumb_filename), 'wb') as thumbf:
ec82d85a 3969 shutil.copyfileobj(uf, thumbf)
80c03fa9 3970 ret.append((thumb_filename, thumb_filename_final))
885cc0b7 3971 t['filepath'] = thumb_filename
3158150c 3972 except network_exceptions as err:
dd0228ce 3973 thumbnails.pop(idx)
80c03fa9 3974 self.report_warning(f'Unable to download {thumb_display_id}: {err}')
6c4fd172 3975 if ret and not write_all:
3976 break
0202b52a 3977 return ret