]> jfr.im git - yt-dlp.git/blame - yt_dlp/YoutubeDL.py
Fix bug in ae1035646a6be09c2aed3e22eb8910f341ddacfe
[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,
941e881e 111 is_path_like,
34921b43 112 join_nonempty,
ce02ed60 113 locked_file,
0647d925 114 make_archive_id,
0202b52a 115 make_dir,
dca08720 116 make_HTTPS_handler,
8b7539d2 117 merge_headers,
3158150c 118 network_exceptions,
ec11a9f4 119 number_of_digits,
cd6fc19e 120 orderedSet,
5314b521 121 orderedSet_from_options,
083c9df9 122 parse_filesize,
ce02ed60 123 preferredencoding,
eedb7ba5 124 prepend_extension,
51fb4995 125 register_socks_protocols,
3efb96a6 126 remove_terminal_sequences,
cfb56d1a 127 render_table,
eedb7ba5 128 replace_extension,
ce02ed60 129 sanitize_filename,
1bb5c511 130 sanitize_path,
dcf77cf1 131 sanitize_url,
67dda517 132 sanitized_Request,
e5660ee6 133 std_headers,
1211bb6d 134 str_or_none,
e29663c6 135 strftime_or_none,
ce02ed60 136 subtitles_filename,
819e0531 137 supports_terminal_sequences,
b1f94422 138 system_identifier,
f2ebc5c7 139 timetuple_from_msec,
732044af 140 to_high_limit_path,
324ad820 141 traverse_obj,
fc61aff4 142 try_call,
6033d980 143 try_get,
29eb5174 144 url_basename,
7d1eb38a 145 variadic,
58b1f00d 146 version_tuple,
53973b4d 147 windows_enable_vt_mode,
ce02ed60
PH
148 write_json_file,
149 write_string,
4f026faf 150)
70b23409 151from .version import RELEASE_GIT_HEAD, VARIANT, __version__
8222d8de 152
e9c0cdd3
YCH
153if compat_os_name == 'nt':
154 import ctypes
155
2459b6e1 156
86e5f3ed 157class YoutubeDL:
8222d8de
JMF
158 """YoutubeDL class.
159
160 YoutubeDL objects are the ones responsible of downloading the
161 actual video file and writing it to disk if the user has requested
162 it, among some other tasks. In most cases there should be one per
163 program. As, given a video URL, the downloader doesn't know how to
164 extract all the needed information, task that InfoExtractors do, it
165 has to pass the URL to one of them.
166
167 For this, YoutubeDL objects have a method that allows
168 InfoExtractors to be registered in a given order. When it is passed
169 a URL, the YoutubeDL object handles it to the first InfoExtractor it
170 finds that reports being able to handle it. The InfoExtractor extracts
171 all the information about the video or videos the URL refers to, and
172 YoutubeDL process the extracted information, possibly using a File
173 Downloader to download the video.
174
175 YoutubeDL objects accept a lot of parameters. In order not to saturate
176 the object constructor with arguments, it receives a dictionary of
177 options instead. These options are available through the params
178 attribute for the InfoExtractors to use. The YoutubeDL also
179 registers itself as the downloader in charge for the InfoExtractors
180 that are added to it, so this is a "mutual registration".
181
182 Available options:
183
184 username: Username for authentication purposes.
185 password: Password for authentication purposes.
180940e0 186 videopassword: Password for accessing a video.
1da50aa3
S
187 ap_mso: Adobe Pass multiple-system operator identifier.
188 ap_username: Multiple-system operator account username.
189 ap_password: Multiple-system operator account password.
8222d8de
JMF
190 usenetrc: Use netrc for authentication instead.
191 verbose: Print additional info to stdout.
192 quiet: Do not print messages to stdout.
ad8915b7 193 no_warnings: Do not print out anything for warnings.
bb66c247 194 forceprint: A dict with keys WHEN mapped to a list of templates to
195 print to stdout. The allowed keys are video or any of the
196 items in utils.POSTPROCESS_WHEN.
ca30f449 197 For compatibility, a single list is also accepted
bb66c247 198 print_to_file: A dict with keys WHEN (same as forceprint) mapped to
199 a list of tuples with (template, filename)
8694c600 200 forcejson: Force printing info_dict as JSON.
63e0be34
PH
201 dump_single_json: Force printing the info_dict of the whole playlist
202 (or video) as a single JSON line.
c25228e5 203 force_write_download_archive: Force writing download archive regardless
204 of 'skip_download' or 'simulate'.
b7b04c78 205 simulate: Do not download the video files. If unset (or None),
206 simulate only if listsubtitles, listformats or list_thumbnails is used
eb8a4433 207 format: Video format code. see "FORMAT SELECTION" for more details.
093a1710 208 You can also pass a function. The function takes 'ctx' as
209 argument and returns the formats to download.
210 See "build_format_selector" for an implementation
63ad4d43 211 allow_unplayable_formats: Allow unplayable formats to be extracted and downloaded.
b7da73eb 212 ignore_no_formats_error: Ignore "No video formats" error. Usefull for
213 extracting metadata even if the video is not actually
214 available for download (experimental)
0930b11f 215 format_sort: A list of fields by which to sort the video formats.
216 See "Sorting Formats" for more details.
c25228e5 217 format_sort_force: Force the given format_sort. see "Sorting Formats"
218 for more details.
08d30158 219 prefer_free_formats: Whether to prefer video formats with free containers
220 over non-free ones of same quality.
c25228e5 221 allow_multiple_video_streams: Allow multiple video streams to be merged
222 into a single file
223 allow_multiple_audio_streams: Allow multiple audio streams to be merged
224 into a single file
0ba692ac 225 check_formats Whether to test if the formats are downloadable.
9f1a1c36 226 Can be True (check all), False (check none),
227 'selected' (check selected formats),
0ba692ac 228 or None (check only if requested by extractor)
4524baf0 229 paths: Dictionary of output paths. The allowed keys are 'home'
230 'temp' and the keys of OUTTMPL_TYPES (in utils.py)
de6000d9 231 outtmpl: Dictionary of templates for output names. Allowed keys
4524baf0 232 are 'default' and the keys of OUTTMPL_TYPES (in utils.py).
34488702 233 For compatibility with youtube-dl, a single string can also be used
a820dc72
RA
234 outtmpl_na_placeholder: Placeholder for unavailable meta fields.
235 restrictfilenames: Do not allow "&" and spaces in file names
236 trim_file_name: Limit length of filename (extension excluded)
4524baf0 237 windowsfilenames: Force the filenames to be windows compatible
b1940459 238 ignoreerrors: Do not stop on download/postprocessing errors.
239 Can be 'only_download' to ignore only download errors.
240 Default is 'only_download' for CLI, but False for API
26e2805c 241 skip_playlist_after_errors: Number of allowed failures until the rest of
242 the playlist is skipped
fe7866d0 243 allowed_extractors: List of regexes to match against extractor names that are allowed
0c3d0f51 244 overwrites: Overwrite all video and metadata files if True,
245 overwrite only non-video files if None
246 and don't overwrite any file if False
34488702 247 For compatibility with youtube-dl,
248 "nooverwrites" may also be used instead
c14e88f0 249 playlist_items: Specific indices of playlist to download.
75822ca7 250 playlistrandom: Download playlist items in random order.
7e9a6125 251 lazy_playlist: Process playlist entries as they are received.
8222d8de
JMF
252 matchtitle: Download only matching titles.
253 rejecttitle: Reject downloads for matching titles.
8bf9319e 254 logger: Log messages to a logging.Logger instance.
17ffed18 255 logtostderr: Print everything to stderr instead of stdout.
256 consoletitle: Display progress in console window's titlebar.
8222d8de
JMF
257 writedescription: Write the video description to a .description file
258 writeinfojson: Write the video description to a .info.json file
75d43ca0 259 clean_infojson: Remove private fields from the infojson
34488702 260 getcomments: Extract video comments. This will not be written to disk
06167fbb 261 unless writeinfojson is also given
1fb07d10 262 writeannotations: Write the video annotations to a .annotations.xml file
8222d8de 263 writethumbnail: Write the thumbnail image to a file
c25228e5 264 allow_playlist_files: Whether to write playlists' description, infojson etc
265 also to disk when using the 'write*' options
ec82d85a 266 write_all_thumbnails: Write all thumbnail formats to files
732044af 267 writelink: Write an internet shortcut file, depending on the
268 current platform (.url/.webloc/.desktop)
269 writeurllink: Write a Windows internet shortcut file (.url)
270 writewebloclink: Write a macOS internet shortcut file (.webloc)
271 writedesktoplink: Write a Linux internet shortcut file (.desktop)
8222d8de 272 writesubtitles: Write the video subtitles to a file
741dd8ea 273 writeautomaticsub: Write the automatically generated subtitles to a file
8222d8de 274 listsubtitles: Lists all available subtitles for the video
a504ced0 275 subtitlesformat: The format code for subtitles
c32b0aab 276 subtitleslangs: List of languages of the subtitles to download (can be regex).
277 The list may contain "all" to refer to all the available
278 subtitles. The language can be prefixed with a "-" to
62b58c09 279 exclude it from the requested languages, e.g. ['all', '-live_chat']
8222d8de
JMF
280 keepvideo: Keep the video file after post-processing
281 daterange: A DateRange object, download only if the upload_date is in the range.
282 skip_download: Skip the actual download of the video file
c35f9e72 283 cachedir: Location of the cache files in the filesystem.
a0e07d31 284 False to disable filesystem cache.
47192f92 285 noplaylist: Download single video instead of a playlist if in doubt.
8dbe9899
PH
286 age_limit: An integer representing the user's age in years.
287 Unsuitable videos for the given age are skipped.
5fe18bdb
PH
288 min_views: An integer representing the minimum view count the video
289 must have in order to not be skipped.
290 Videos without view count information are always
291 downloaded. None for no limit.
292 max_views: An integer representing the maximum view count.
293 Videos that are more popular than that are not
294 downloaded.
295 Videos without view count information are always
296 downloaded. None for no limit.
ae103564 297 download_archive: A set, or the name of a file where all downloads are recorded.
298 Videos already present in the file are not downloaded 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"""
ae103564 726 archive = set()
ed39cac5 727 if fn is None:
ae103564 728 return archive
941e881e 729 elif not is_path_like(fn):
ae103564 730 return fn
731
49a57e70 732 self.write_debug(f'Loading archive file {fn!r}')
ed39cac5 733 try:
734 with locked_file(fn, 'r', encoding='utf-8') as archive_file:
735 for line in archive_file:
ae103564 736 archive.add(line.strip())
86e5f3ed 737 except OSError as ioe:
ed39cac5 738 if ioe.errno != errno.ENOENT:
739 raise
ae103564 740 return archive
ed39cac5 741
ae103564 742 self.archive = preload_download_archive(self.params.get('download_archive'))
ed39cac5 743
7d4111ed
PH
744 def warn_if_short_id(self, argv):
745 # short YouTube ID starting with dash?
746 idxs = [
747 i for i, a in enumerate(argv)
748 if re.match(r'^-[0-9A-Za-z_-]{10}$', a)]
749 if idxs:
750 correct_argv = (
7a5c1cfe 751 ['yt-dlp']
3089bc74
S
752 + [a for i, a in enumerate(argv) if i not in idxs]
753 + ['--'] + [argv[i] for i in idxs]
7d4111ed
PH
754 )
755 self.report_warning(
756 'Long argument string detected. '
49a57e70 757 'Use -- to separate parameters and URLs, like this:\n%s' %
7d4111ed
PH
758 args_to_str(correct_argv))
759
8222d8de
JMF
760 def add_info_extractor(self, ie):
761 """Add an InfoExtractor object to the end of the list."""
8b7491c8 762 ie_key = ie.ie_key()
763 self._ies[ie_key] = ie
e52d7f85 764 if not isinstance(ie, type):
8b7491c8 765 self._ies_instances[ie_key] = ie
e52d7f85 766 ie.set_downloader(self)
8222d8de 767
56c73665
JMF
768 def get_info_extractor(self, ie_key):
769 """
770 Get an instance of an IE with name ie_key, it will try to get one from
771 the _ies list, if there's no instance it will create a new one and add
772 it to the extractor list.
773 """
774 ie = self._ies_instances.get(ie_key)
775 if ie is None:
776 ie = get_info_extractor(ie_key)()
777 self.add_info_extractor(ie)
778 return ie
779
023fa8c4
JMF
780 def add_default_info_extractors(self):
781 """
782 Add the InfoExtractors returned by gen_extractors to the end of the list
783 """
fe7866d0 784 all_ies = {ie.IE_NAME.lower(): ie for ie in gen_extractor_classes()}
785 all_ies['end'] = UnsupportedURLIE()
786 try:
787 ie_names = orderedSet_from_options(
788 self.params.get('allowed_extractors', ['default']), {
789 'all': list(all_ies),
790 'default': [name for name, ie in all_ies.items() if ie._ENABLED],
791 }, use_regex=True)
792 except re.error as e:
793 raise ValueError(f'Wrong regex for allowed_extractors: {e.pattern}')
794 for name in ie_names:
795 self.add_info_extractor(all_ies[name])
796 self.write_debug(f'Loaded {len(ie_names)} extractors')
023fa8c4 797
56d868db 798 def add_post_processor(self, pp, when='post_process'):
8222d8de 799 """Add a PostProcessor object to the end of the chain."""
8aa0e7cd 800 assert when in POSTPROCESS_WHEN, f'Invalid when={when}'
5bfa4862 801 self._pps[when].append(pp)
8222d8de
JMF
802 pp.set_downloader(self)
803
ab8e5e51
AM
804 def add_post_hook(self, ph):
805 """Add the post hook"""
806 self._post_hooks.append(ph)
807
933605d7 808 def add_progress_hook(self, ph):
819e0531 809 """Add the download progress hook"""
933605d7 810 self._progress_hooks.append(ph)
8ab470f1 811
819e0531 812 def add_postprocessor_hook(self, ph):
813 """Add the postprocessing progress hook"""
814 self._postprocessor_hooks.append(ph)
5bfc8bee 815 for pps in self._pps.values():
816 for pp in pps:
817 pp.add_progress_hook(ph)
819e0531 818
1c088fa8 819 def _bidi_workaround(self, message):
5d681e96 820 if not hasattr(self, '_output_channel'):
1c088fa8
PH
821 return message
822
5d681e96 823 assert hasattr(self, '_output_process')
14f25df2 824 assert isinstance(message, str)
6febd1c1 825 line_count = message.count('\n') + 1
0f06bcd7 826 self._output_process.stdin.write((message + '\n').encode())
5d681e96 827 self._output_process.stdin.flush()
0f06bcd7 828 res = ''.join(self._output_channel.readline().decode()
9e1a5b84 829 for _ in range(line_count))
6febd1c1 830 return res[:-len('\n')]
1c088fa8 831
b35496d8 832 def _write_string(self, message, out=None, only_once=False):
833 if only_once:
834 if message in self._printed_messages:
835 return
836 self._printed_messages.add(message)
837 write_string(message, out=out, encoding=self.params.get('encoding'))
734f90bb 838
cf4f42cb 839 def to_stdout(self, message, skip_eol=False, quiet=None):
0760b0a7 840 """Print message to stdout"""
cf4f42cb 841 if quiet is not None:
da4db748 842 self.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument quiet. '
843 'Use "YoutubeDL.to_screen" instead')
8a82af35 844 if skip_eol is not False:
da4db748 845 self.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument skip_eol. '
846 'Use "YoutubeDL.to_screen" instead')
0bf9dc1e 847 self._write_string(f'{self._bidi_workaround(message)}\n', self._out_files.out)
cf4f42cb 848
849 def to_screen(self, message, skip_eol=False, quiet=None):
850 """Print message to screen if not in quiet mode"""
8bf9319e 851 if self.params.get('logger'):
43afe285 852 self.params['logger'].debug(message)
cf4f42cb 853 return
854 if (self.params.get('quiet') if quiet is None else quiet) and not self.params.get('verbose'):
855 return
856 self._write_string(
857 '%s%s' % (self._bidi_workaround(message), ('' if skip_eol else '\n')),
591bb9d3 858 self._out_files.screen)
8222d8de 859
b35496d8 860 def to_stderr(self, message, only_once=False):
0760b0a7 861 """Print message to stderr"""
14f25df2 862 assert isinstance(message, str)
8bf9319e 863 if self.params.get('logger'):
43afe285
IB
864 self.params['logger'].error(message)
865 else:
5792c950 866 self._write_string(f'{self._bidi_workaround(message)}\n', self._out_files.error, only_once=only_once)
cf4f42cb 867
868 def _send_console_code(self, code):
591bb9d3 869 if compat_os_name == 'nt' or not self._out_files.console:
cf4f42cb 870 return
591bb9d3 871 self._write_string(code, self._out_files.console)
8222d8de 872
1e5b9a95
PH
873 def to_console_title(self, message):
874 if not self.params.get('consoletitle', False):
875 return
3efb96a6 876 message = remove_terminal_sequences(message)
4bede0d8
C
877 if compat_os_name == 'nt':
878 if ctypes.windll.kernel32.GetConsoleWindow():
879 # c_wchar_p() might not be necessary if `message` is
880 # already of type unicode()
881 ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
cf4f42cb 882 else:
883 self._send_console_code(f'\033]0;{message}\007')
1e5b9a95 884
bdde425c 885 def save_console_title(self):
cf4f42cb 886 if not self.params.get('consoletitle') or self.params.get('simulate'):
bdde425c 887 return
592b7485 888 self._send_console_code('\033[22;0t') # Save the title on stack
bdde425c
PH
889
890 def restore_console_title(self):
cf4f42cb 891 if not self.params.get('consoletitle') or self.params.get('simulate'):
bdde425c 892 return
592b7485 893 self._send_console_code('\033[23;0t') # Restore the title from stack
bdde425c
PH
894
895 def __enter__(self):
896 self.save_console_title()
897 return self
898
899 def __exit__(self, *args):
900 self.restore_console_title()
f89197d7 901
dca08720 902 if self.params.get('cookiefile') is not None:
1bab3437 903 self.cookiejar.save(ignore_discard=True, ignore_expires=True)
bdde425c 904
fa9f30b8 905 def trouble(self, message=None, tb=None, is_error=True):
8222d8de
JMF
906 """Determine action to take when a download problem appears.
907
908 Depending on if the downloader has been configured to ignore
909 download errors or not, this method may throw an exception or
910 not when errors are found, after printing the message.
911
fa9f30b8 912 @param tb If given, is additional traceback information
913 @param is_error Whether to raise error according to ignorerrors
8222d8de
JMF
914 """
915 if message is not None:
916 self.to_stderr(message)
917 if self.params.get('verbose'):
918 if tb is None:
919 if sys.exc_info()[0]: # if .trouble has been called from an except block
6febd1c1 920 tb = ''
8222d8de 921 if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
6febd1c1 922 tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
c0384f22 923 tb += encode_compat_str(traceback.format_exc())
8222d8de
JMF
924 else:
925 tb_data = traceback.format_list(traceback.extract_stack())
6febd1c1 926 tb = ''.join(tb_data)
c19bc311 927 if tb:
928 self.to_stderr(tb)
fa9f30b8 929 if not is_error:
930 return
b1940459 931 if not self.params.get('ignoreerrors'):
8222d8de
JMF
932 if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
933 exc_info = sys.exc_info()[1].exc_info
934 else:
935 exc_info = sys.exc_info()
936 raise DownloadError(message, exc_info)
937 self._download_retcode = 1
938
19a03940 939 Styles = Namespace(
940 HEADERS='yellow',
941 EMPHASIS='light blue',
492272fe 942 FILENAME='green',
19a03940 943 ID='green',
944 DELIM='blue',
945 ERROR='red',
946 WARNING='yellow',
947 SUPPRESS='light black',
948 )
ec11a9f4 949
7578d77d 950 def _format_text(self, handle, allow_colors, text, f, fallback=None, *, test_encoding=False):
e5a998f3 951 text = str(text)
ec11a9f4 952 if test_encoding:
953 original_text = text
5c104538 954 # handle.encoding can be None. See https://github.com/yt-dlp/yt-dlp/issues/2711
955 encoding = self.params.get('encoding') or getattr(handle, 'encoding', None) or 'ascii'
ec11a9f4 956 text = text.encode(encoding, 'ignore').decode(encoding)
957 if fallback is not None and text != original_text:
958 text = fallback
7578d77d 959 return format_text(text, f) if allow_colors else text if fallback is None else fallback
ec11a9f4 960
591bb9d3 961 def _format_out(self, *args, **kwargs):
962 return self._format_text(self._out_files.out, self._allow_colors.out, *args, **kwargs)
963
ec11a9f4 964 def _format_screen(self, *args, **kwargs):
591bb9d3 965 return self._format_text(self._out_files.screen, self._allow_colors.screen, *args, **kwargs)
ec11a9f4 966
967 def _format_err(self, *args, **kwargs):
591bb9d3 968 return self._format_text(self._out_files.error, self._allow_colors.error, *args, **kwargs)
819e0531 969
c84aeac6 970 def report_warning(self, message, only_once=False):
8222d8de
JMF
971 '''
972 Print the message to stderr, it will be prefixed with 'WARNING:'
973 If stderr is a tty file the 'WARNING:' will be colored
974 '''
6d07ce01
JMF
975 if self.params.get('logger') is not None:
976 self.params['logger'].warning(message)
8222d8de 977 else:
ad8915b7
PH
978 if self.params.get('no_warnings'):
979 return
ec11a9f4 980 self.to_stderr(f'{self._format_err("WARNING:", self.Styles.WARNING)} {message}', only_once)
8222d8de 981
da4db748 982 def deprecation_warning(self, message, *, stacklevel=0):
983 deprecation_warning(
984 message, stacklevel=stacklevel + 1, printer=self.report_error, is_error=False)
985
986 def deprecated_feature(self, message):
ee8dd27a 987 if self.params.get('logger') is not None:
da4db748 988 self.params['logger'].warning(f'Deprecated Feature: {message}')
989 self.to_stderr(f'{self._format_err("Deprecated Feature:", self.Styles.ERROR)} {message}', True)
ee8dd27a 990
fa9f30b8 991 def report_error(self, message, *args, **kwargs):
8222d8de
JMF
992 '''
993 Do the same as trouble, but prefixes the message with 'ERROR:', colored
994 in red if stderr is a tty file.
995 '''
fa9f30b8 996 self.trouble(f'{self._format_err("ERROR:", self.Styles.ERROR)} {message}', *args, **kwargs)
8222d8de 997
b35496d8 998 def write_debug(self, message, only_once=False):
0760b0a7 999 '''Log debug message or Print message to stderr'''
1000 if not self.params.get('verbose', False):
1001 return
8a82af35 1002 message = f'[debug] {message}'
0760b0a7 1003 if self.params.get('logger'):
1004 self.params['logger'].debug(message)
1005 else:
b35496d8 1006 self.to_stderr(message, only_once)
0760b0a7 1007
8222d8de
JMF
1008 def report_file_already_downloaded(self, file_name):
1009 """Report file has already been fully downloaded."""
1010 try:
6febd1c1 1011 self.to_screen('[download] %s has already been downloaded' % file_name)
ce02ed60 1012 except UnicodeEncodeError:
6febd1c1 1013 self.to_screen('[download] The file has already been downloaded')
8222d8de 1014
0c3d0f51 1015 def report_file_delete(self, file_name):
1016 """Report that existing file will be deleted."""
1017 try:
c25228e5 1018 self.to_screen('Deleting existing file %s' % file_name)
0c3d0f51 1019 except UnicodeEncodeError:
c25228e5 1020 self.to_screen('Deleting existing file')
0c3d0f51 1021
319b6059 1022 def raise_no_formats(self, info, forced=False, *, msg=None):
0a5a191a 1023 has_drm = info.get('_has_drm')
319b6059 1024 ignored, expected = self.params.get('ignore_no_formats_error'), bool(msg)
1025 msg = msg or has_drm and 'This video is DRM protected' or 'No video formats found!'
1026 if forced or not ignored:
1151c407 1027 raise ExtractorError(msg, video_id=info['id'], ie=info['extractor'],
319b6059 1028 expected=has_drm or ignored or expected)
88acdbc2 1029 else:
1030 self.report_warning(msg)
1031
de6000d9 1032 def parse_outtmpl(self):
bf1824b3 1033 self.deprecation_warning('"YoutubeDL.parse_outtmpl" is deprecated and may be removed in a future version')
1034 self._parse_outtmpl()
1035 return self.params['outtmpl']
1036
1037 def _parse_outtmpl(self):
7b2c3f47 1038 sanitize = IDENTITY
bf1824b3 1039 if self.params.get('restrictfilenames'): # Remove spaces in the default template
71ce444a 1040 sanitize = lambda x: x.replace(' - ', ' ').replace(' ', '-')
bf1824b3 1041
1042 outtmpl = self.params.setdefault('outtmpl', {})
1043 if not isinstance(outtmpl, dict):
1044 self.params['outtmpl'] = outtmpl = {'default': outtmpl}
1045 outtmpl.update({k: sanitize(v) for k, v in DEFAULT_OUTTMPL.items() if outtmpl.get(k) is None})
de6000d9 1046
21cd8fae 1047 def get_output_path(self, dir_type='', filename=None):
1048 paths = self.params.get('paths', {})
d2c8aadf 1049 assert isinstance(paths, dict), '"paths" parameter must be a dictionary'
21cd8fae 1050 path = os.path.join(
1051 expand_path(paths.get('home', '').strip()),
1052 expand_path(paths.get(dir_type, '').strip()) if dir_type else '',
1053 filename or '')
21cd8fae 1054 return sanitize_path(path, force=self.params.get('windowsfilenames'))
1055
76a264ac 1056 @staticmethod
901130bb 1057 def _outtmpl_expandpath(outtmpl):
1058 # expand_path translates '%%' into '%' and '$$' into '$'
1059 # correspondingly that is not what we want since we need to keep
1060 # '%%' intact for template dict substitution step. Working around
1061 # with boundary-alike separator hack.
1062 sep = ''.join([random.choice(ascii_letters) for _ in range(32)])
86e5f3ed 1063 outtmpl = outtmpl.replace('%%', f'%{sep}%').replace('$$', f'${sep}$')
901130bb 1064
1065 # outtmpl should be expand_path'ed before template dict substitution
1066 # because meta fields may contain env variables we don't want to
62b58c09 1067 # be expanded. E.g. for outtmpl "%(title)s.%(ext)s" and
901130bb 1068 # title "Hello $PATH", we don't want `$PATH` to be expanded.
1069 return expand_path(outtmpl).replace(sep, '')
1070
1071 @staticmethod
1072 def escape_outtmpl(outtmpl):
1073 ''' Escape any remaining strings like %s, %abc% etc. '''
1074 return re.sub(
1075 STR_FORMAT_RE_TMPL.format('', '(?![%(\0])'),
1076 lambda mobj: ('' if mobj.group('has_key') else '%') + mobj.group(0),
1077 outtmpl)
1078
1079 @classmethod
1080 def validate_outtmpl(cls, outtmpl):
76a264ac 1081 ''' @return None or Exception object '''
7d1eb38a 1082 outtmpl = re.sub(
47cdc68e 1083 STR_FORMAT_RE_TMPL.format('[^)]*', '[ljhqBUDS]'),
7d1eb38a 1084 lambda mobj: f'{mobj.group(0)[:-1]}s',
1085 cls._outtmpl_expandpath(outtmpl))
76a264ac 1086 try:
7d1eb38a 1087 cls.escape_outtmpl(outtmpl) % collections.defaultdict(int)
76a264ac 1088 return None
1089 except ValueError as err:
1090 return err
1091
03b4de72 1092 @staticmethod
1093 def _copy_infodict(info_dict):
1094 info_dict = dict(info_dict)
09b49e1f 1095 info_dict.pop('__postprocessors', None)
415f8d51 1096 info_dict.pop('__pending_error', None)
03b4de72 1097 return info_dict
1098
e0fd9573 1099 def prepare_outtmpl(self, outtmpl, info_dict, sanitize=False):
1100 """ Make the outtmpl and info_dict suitable for substitution: ydl.escape_outtmpl(outtmpl) % info_dict
1101 @param sanitize Whether to sanitize the output as a filename.
1102 For backward compatibility, a function can also be passed
1103 """
1104
6e84b215 1105 info_dict.setdefault('epoch', int(time.time())) # keep epoch consistent once set
143db31d 1106
03b4de72 1107 info_dict = self._copy_infodict(info_dict)
752cda38 1108 info_dict['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
53c18592 1109 formatSeconds(info_dict['duration'], '-' if sanitize else ':')
143db31d 1110 if info_dict.get('duration', None) is not None
1111 else None)
1d485a1a 1112 info_dict['autonumber'] = int(self.params.get('autonumber_start', 1) - 1 + self._num_downloads)
9c906919 1113 info_dict['video_autonumber'] = self._num_videos
752cda38 1114 if info_dict.get('resolution') is None:
1115 info_dict['resolution'] = self.format_resolution(info_dict, default=None)
143db31d 1116
e6f21b3d 1117 # For fields playlist_index, playlist_autonumber and autonumber convert all occurrences
143db31d 1118 # of %(field)s to %(field)0Nd for backward compatibility
1119 field_size_compat_map = {
0a5a191a 1120 'playlist_index': number_of_digits(info_dict.get('__last_playlist_index') or 0),
ec11a9f4 1121 'playlist_autonumber': number_of_digits(info_dict.get('n_entries') or 0),
752cda38 1122 'autonumber': self.params.get('autonumber_size') or 5,
143db31d 1123 }
752cda38 1124
385a27fa 1125 TMPL_DICT = {}
47cdc68e 1126 EXTERNAL_FORMAT_RE = re.compile(STR_FORMAT_RE_TMPL.format('[^)]*', f'[{STR_FORMAT_TYPES}ljhqBUDS]'))
385a27fa 1127 MATH_FUNCTIONS = {
1128 '+': float.__add__,
1129 '-': float.__sub__,
1130 }
e625be0d 1131 # Field is of the form key1.key2...
07a1250e 1132 # where keys (except first) can be string, int, slice or "{field, ...}"
1133 FIELD_INNER_RE = r'(?:\w+|%(num)s|%(num)s?(?::%(num)s?){1,2})' % {'num': r'(?:-?\d+)'}
1134 FIELD_RE = r'\w*(?:\.(?:%(inner)s|{%(field)s(?:,%(field)s)*}))*' % {
1135 'inner': FIELD_INNER_RE,
1136 'field': rf'\w*(?:\.{FIELD_INNER_RE})*'
1137 }
1d485a1a 1138 MATH_FIELD_RE = rf'(?:{FIELD_RE}|-?{NUMBER_RE})'
385a27fa 1139 MATH_OPERATORS_RE = r'(?:%s)' % '|'.join(map(re.escape, MATH_FUNCTIONS.keys()))
1d485a1a 1140 INTERNAL_FORMAT_RE = re.compile(rf'''(?x)
e625be0d 1141 (?P<negate>-)?
1d485a1a 1142 (?P<fields>{FIELD_RE})
1143 (?P<maths>(?:{MATH_OPERATORS_RE}{MATH_FIELD_RE})*)
e625be0d 1144 (?:>(?P<strf_format>.+?))?
34baa9fd 1145 (?P<remaining>
1146 (?P<alternate>(?<!\\),[^|&)]+)?
1147 (?:&(?P<replacement>.*?))?
1148 (?:\|(?P<default>.*?))?
1d485a1a 1149 )$''')
752cda38 1150
07a1250e 1151 def _traverse_infodict(fields):
1152 fields = [f for x in re.split(r'\.({.+?})\.?', fields)
1153 for f in ([x] if x.startswith('{') else x.split('.'))]
1154 for i in (0, -1):
1155 if fields and not fields[i]:
1156 fields.pop(i)
1157
1158 for i, f in enumerate(fields):
1159 if not f.startswith('{'):
1160 continue
1161 assert f.endswith('}'), f'No closing brace for {f} in {fields}'
1162 fields[i] = {k: k.split('.') for k in f[1:-1].split(',')}
1163
1164 return traverse_obj(info_dict, fields, is_user_input=True, traverse_string=True)
76a264ac 1165
752cda38 1166 def get_value(mdict):
1167 # Object traversal
2b8a2973 1168 value = _traverse_infodict(mdict['fields'])
752cda38 1169 # Negative
1170 if mdict['negate']:
1171 value = float_or_none(value)
1172 if value is not None:
1173 value *= -1
1174 # Do maths
385a27fa 1175 offset_key = mdict['maths']
1176 if offset_key:
752cda38 1177 value = float_or_none(value)
1178 operator = None
385a27fa 1179 while offset_key:
1180 item = re.match(
1181 MATH_FIELD_RE if operator else MATH_OPERATORS_RE,
1182 offset_key).group(0)
1183 offset_key = offset_key[len(item):]
1184 if operator is None:
752cda38 1185 operator = MATH_FUNCTIONS[item]
385a27fa 1186 continue
1187 item, multiplier = (item[1:], -1) if item[0] == '-' else (item, 1)
1188 offset = float_or_none(item)
1189 if offset is None:
2b8a2973 1190 offset = float_or_none(_traverse_infodict(item))
385a27fa 1191 try:
1192 value = operator(value, multiplier * offset)
1193 except (TypeError, ZeroDivisionError):
1194 return None
1195 operator = None
752cda38 1196 # Datetime formatting
1197 if mdict['strf_format']:
7c37ff97 1198 value = strftime_or_none(value, mdict['strf_format'].replace('\\,', ','))
752cda38 1199
a6bcaf71 1200 # XXX: Workaround for https://github.com/yt-dlp/yt-dlp/issues/4485
1201 if sanitize and value == '':
1202 value = None
752cda38 1203 return value
1204
b868936c 1205 na = self.params.get('outtmpl_na_placeholder', 'NA')
1206
e0fd9573 1207 def filename_sanitizer(key, value, restricted=self.params.get('restrictfilenames')):
5c3895ff 1208 return sanitize_filename(str(value), restricted=restricted, is_id=(
1209 bool(re.search(r'(^|[_.])id(\.|$)', key))
8a82af35 1210 if 'filename-sanitization' in self.params['compat_opts']
5c3895ff 1211 else NO_DEFAULT))
e0fd9573 1212
1213 sanitizer = sanitize if callable(sanitize) else filename_sanitizer
1214 sanitize = bool(sanitize)
1215
6e84b215 1216 def _dumpjson_default(obj):
1217 if isinstance(obj, (set, LazyList)):
1218 return list(obj)
adbc4ec4 1219 return repr(obj)
6e84b215 1220
752cda38 1221 def create_key(outer_mobj):
1222 if not outer_mobj.group('has_key'):
b836dc94 1223 return outer_mobj.group(0)
752cda38 1224 key = outer_mobj.group('key')
752cda38 1225 mobj = re.match(INTERNAL_FORMAT_RE, key)
e0fd9573 1226 initial_field = mobj.group('fields') if mobj else ''
e978789f 1227 value, replacement, default = None, None, na
7c37ff97 1228 while mobj:
e625be0d 1229 mobj = mobj.groupdict()
7c37ff97 1230 default = mobj['default'] if mobj['default'] is not None else default
752cda38 1231 value = get_value(mobj)
e978789f 1232 replacement = mobj['replacement']
7c37ff97 1233 if value is None and mobj['alternate']:
34baa9fd 1234 mobj = re.match(INTERNAL_FORMAT_RE, mobj['remaining'][1:])
7c37ff97 1235 else:
1236 break
752cda38 1237
b868936c 1238 fmt = outer_mobj.group('format')
752cda38 1239 if fmt == 's' and value is not None and key in field_size_compat_map.keys():
86e5f3ed 1240 fmt = f'0{field_size_compat_map[key]:d}d'
752cda38 1241
e978789f 1242 value = default if value is None else value if replacement is None else replacement
752cda38 1243
4476d2c7 1244 flags = outer_mobj.group('conversion') or ''
7d1eb38a 1245 str_fmt = f'{fmt[:-1]}s'
524e2e4f 1246 if fmt[-1] == 'l': # list
4476d2c7 1247 delim = '\n' if '#' in flags else ', '
9e907ebd 1248 value, fmt = delim.join(map(str, variadic(value, allowed_types=(str, bytes)))), str_fmt
524e2e4f 1249 elif fmt[-1] == 'j': # json
4476d2c7 1250 value, fmt = json.dumps(value, default=_dumpjson_default, indent=4 if '#' in flags else None), str_fmt
47cdc68e 1251 elif fmt[-1] == 'h': # html
1252 value, fmt = escapeHTML(value), str_fmt
524e2e4f 1253 elif fmt[-1] == 'q': # quoted
4476d2c7 1254 value = map(str, variadic(value) if '#' in flags else [value])
1255 value, fmt = ' '.join(map(compat_shlex_quote, value)), str_fmt
524e2e4f 1256 elif fmt[-1] == 'B': # bytes
0f06bcd7 1257 value = f'%{str_fmt}'.encode() % str(value).encode()
f5aa5cfb 1258 value, fmt = value.decode('utf-8', 'ignore'), 's'
524e2e4f 1259 elif fmt[-1] == 'U': # unicode normalized
524e2e4f 1260 value, fmt = unicodedata.normalize(
1261 # "+" = compatibility equivalence, "#" = NFD
4476d2c7 1262 'NF%s%s' % ('K' if '+' in flags else '', 'D' if '#' in flags else 'C'),
524e2e4f 1263 value), str_fmt
e0fd9573 1264 elif fmt[-1] == 'D': # decimal suffix
abbeeebc 1265 num_fmt, fmt = fmt[:-1].replace('#', ''), 's'
1266 value = format_decimal_suffix(value, f'%{num_fmt}f%s' if num_fmt else '%d%s',
1267 factor=1024 if '#' in flags else 1000)
37893bb0 1268 elif fmt[-1] == 'S': # filename sanitization
e0fd9573 1269 value, fmt = filename_sanitizer(initial_field, value, restricted='#' in flags), str_fmt
7d1eb38a 1270 elif fmt[-1] == 'c':
524e2e4f 1271 if value:
1272 value = str(value)[0]
76a264ac 1273 else:
524e2e4f 1274 fmt = str_fmt
76a264ac 1275 elif fmt[-1] not in 'rs': # numeric
a439a3a4 1276 value = float_or_none(value)
752cda38 1277 if value is None:
1278 value, fmt = default, 's'
901130bb 1279
752cda38 1280 if sanitize:
1281 if fmt[-1] == 'r':
1282 # If value is an object, sanitize might convert it to a string
1283 # So we convert it to repr first
7d1eb38a 1284 value, fmt = repr(value), str_fmt
639f1cea 1285 if fmt[-1] in 'csr':
e0fd9573 1286 value = sanitizer(initial_field, value)
901130bb 1287
b868936c 1288 key = '%s\0%s' % (key.replace('%', '%\0'), outer_mobj.group('format'))
385a27fa 1289 TMPL_DICT[key] = value
b868936c 1290 return '{prefix}%({key}){fmt}'.format(key=key, fmt=fmt, prefix=outer_mobj.group('prefix'))
752cda38 1291
385a27fa 1292 return EXTERNAL_FORMAT_RE.sub(create_key, outtmpl), TMPL_DICT
143db31d 1293
819e0531 1294 def evaluate_outtmpl(self, outtmpl, info_dict, *args, **kwargs):
1295 outtmpl, info_dict = self.prepare_outtmpl(outtmpl, info_dict, *args, **kwargs)
1296 return self.escape_outtmpl(outtmpl) % info_dict
1297
5127e92a 1298 def _prepare_filename(self, info_dict, *, outtmpl=None, tmpl_type=None):
1299 assert None in (outtmpl, tmpl_type), 'outtmpl and tmpl_type are mutually exclusive'
1300 if outtmpl is None:
bf1824b3 1301 outtmpl = self.params['outtmpl'].get(tmpl_type or 'default', self.params['outtmpl']['default'])
8222d8de 1302 try:
5127e92a 1303 outtmpl = self._outtmpl_expandpath(outtmpl)
e0fd9573 1304 filename = self.evaluate_outtmpl(outtmpl, info_dict, True)
6a0546e3 1305 if not filename:
1306 return None
15da37c7 1307
5127e92a 1308 if tmpl_type in ('', 'temp'):
6a0546e3 1309 final_ext, ext = self.params.get('final_ext'), info_dict.get('ext')
1310 if final_ext and ext and final_ext != ext and filename.endswith(f'.{final_ext}'):
1311 filename = replace_extension(filename, ext, final_ext)
5127e92a 1312 elif tmpl_type:
6a0546e3 1313 force_ext = OUTTMPL_TYPES[tmpl_type]
1314 if force_ext:
1315 filename = replace_extension(filename, force_ext, info_dict.get('ext'))
de6000d9 1316
bdc3fd2f
U
1317 # https://github.com/blackjack4494/youtube-dlc/issues/85
1318 trim_file_name = self.params.get('trim_file_name', False)
1319 if trim_file_name:
5c22c63d 1320 no_ext, *ext = filename.rsplit('.', 2)
1321 filename = join_nonempty(no_ext[:trim_file_name], *ext, delim='.')
bdc3fd2f 1322
0202b52a 1323 return filename
8222d8de 1324 except ValueError as err:
6febd1c1 1325 self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')')
8222d8de
JMF
1326 return None
1327
5127e92a 1328 def prepare_filename(self, info_dict, dir_type='', *, outtmpl=None, warn=False):
1329 """Generate the output filename"""
1330 if outtmpl:
1331 assert not dir_type, 'outtmpl and dir_type are mutually exclusive'
1332 dir_type = None
1333 filename = self._prepare_filename(info_dict, tmpl_type=dir_type, outtmpl=outtmpl)
80c03fa9 1334 if not filename and dir_type not in ('', 'temp'):
1335 return ''
de6000d9 1336
c84aeac6 1337 if warn:
21cd8fae 1338 if not self.params.get('paths'):
de6000d9 1339 pass
1340 elif filename == '-':
c84aeac6 1341 self.report_warning('--paths is ignored when an outputting to stdout', only_once=True)
de6000d9 1342 elif os.path.isabs(filename):
c84aeac6 1343 self.report_warning('--paths is ignored since an absolute path is given in output template', only_once=True)
de6000d9 1344 if filename == '-' or not filename:
1345 return filename
1346
21cd8fae 1347 return self.get_output_path(dir_type, filename)
0202b52a 1348
120fe513 1349 def _match_entry(self, info_dict, incomplete=False, silent=False):
ecdec191 1350 """ Returns None if the file should be downloaded """
8222d8de 1351
3bec830a 1352 video_title = info_dict.get('title', info_dict.get('id', 'entry'))
c77495e3 1353
8b0d7497 1354 def check_filter():
8b0d7497 1355 if 'title' in info_dict:
1356 # This can happen when we're just evaluating the playlist
1357 title = info_dict['title']
1358 matchtitle = self.params.get('matchtitle', False)
1359 if matchtitle:
1360 if not re.search(matchtitle, title, re.IGNORECASE):
1361 return '"' + title + '" title did not match pattern "' + matchtitle + '"'
1362 rejecttitle = self.params.get('rejecttitle', False)
1363 if rejecttitle:
1364 if re.search(rejecttitle, title, re.IGNORECASE):
1365 return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
1366 date = info_dict.get('upload_date')
1367 if date is not None:
1368 dateRange = self.params.get('daterange', DateRange())
1369 if date not in dateRange:
86e5f3ed 1370 return f'{date_from_str(date).isoformat()} upload date is not in range {dateRange}'
8b0d7497 1371 view_count = info_dict.get('view_count')
1372 if view_count is not None:
1373 min_views = self.params.get('min_views')
1374 if min_views is not None and view_count < min_views:
1375 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
1376 max_views = self.params.get('max_views')
1377 if max_views is not None and view_count > max_views:
1378 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
1379 if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')):
1380 return 'Skipping "%s" because it is age restricted' % video_title
8b0d7497 1381
8f18aca8 1382 match_filter = self.params.get('match_filter')
1383 if match_filter is not None:
1384 try:
1385 ret = match_filter(info_dict, incomplete=incomplete)
1386 except TypeError:
1387 # For backward compatibility
1388 ret = None if incomplete else match_filter(info_dict)
492272fe 1389 if ret is NO_DEFAULT:
1390 while True:
1391 filename = self._format_screen(self.prepare_filename(info_dict), self.Styles.FILENAME)
1392 reply = input(self._format_screen(
1393 f'Download "{filename}"? (Y/n): ', self.Styles.EMPHASIS)).lower().strip()
1394 if reply in {'y', ''}:
1395 return None
1396 elif reply == 'n':
1397 return f'Skipping {video_title}'
492272fe 1398 elif ret is not None:
8f18aca8 1399 return ret
8b0d7497 1400 return None
1401
c77495e3 1402 if self.in_download_archive(info_dict):
1403 reason = '%s has already been recorded in the archive' % video_title
1404 break_opt, break_err = 'break_on_existing', ExistingVideoReached
1405 else:
1406 reason = check_filter()
1407 break_opt, break_err = 'break_on_reject', RejectedVideoReached
8b0d7497 1408 if reason is not None:
120fe513 1409 if not silent:
1410 self.to_screen('[download] ' + reason)
c77495e3 1411 if self.params.get(break_opt, False):
1412 raise break_err()
8b0d7497 1413 return reason
fe7e0c98 1414
b6c45014
JMF
1415 @staticmethod
1416 def add_extra_info(info_dict, extra_info):
1417 '''Set the keys from extra_info in info dict if they are missing'''
1418 for key, value in extra_info.items():
1419 info_dict.setdefault(key, value)
1420
409e1828 1421 def extract_info(self, url, download=True, ie_key=None, extra_info=None,
61aa5ba3 1422 process=True, force_generic_extractor=False):
41d1cca3 1423 """
17ffed18 1424 Extract and return the information dictionary of the URL
41d1cca3 1425
1426 Arguments:
17ffed18 1427 @param url URL to extract
41d1cca3 1428
1429 Keyword arguments:
17ffed18 1430 @param download Whether to download videos
1431 @param process Whether to resolve all unresolved references (URLs, playlist items).
1432 Must be True for download to work
1433 @param ie_key Use only the extractor with this key
1434
1435 @param extra_info Dictionary containing the extra values to add to the info (For internal use only)
1436 @force_generic_extractor Force using the generic extractor (Deprecated; use ie_key='Generic')
41d1cca3 1437 """
fe7e0c98 1438
409e1828 1439 if extra_info is None:
1440 extra_info = {}
1441
61aa5ba3 1442 if not ie_key and force_generic_extractor:
d22dec74
S
1443 ie_key = 'Generic'
1444
8222d8de 1445 if ie_key:
fe7866d0 1446 ies = {ie_key: self._ies[ie_key]} if ie_key in self._ies else {}
8222d8de
JMF
1447 else:
1448 ies = self._ies
1449
fe7866d0 1450 for key, ie in ies.items():
8222d8de
JMF
1451 if not ie.suitable(url):
1452 continue
1453
1454 if not ie.working():
6febd1c1
PH
1455 self.report_warning('The program functionality for this site has been marked as broken, '
1456 'and will probably not work.')
8222d8de 1457
1151c407 1458 temp_id = ie.get_temp_id(url)
fe7866d0 1459 if temp_id is not None and self.in_download_archive({'id': temp_id, 'ie_key': key}):
1460 self.to_screen(f'[{key}] {temp_id}: has already been recorded in the archive')
5e5be0c0 1461 if self.params.get('break_on_existing', False):
1462 raise ExistingVideoReached()
a0566bbf 1463 break
fe7866d0 1464 return self.__extract_info(url, self.get_info_extractor(key), download, extra_info, process)
a0566bbf 1465 else:
fe7866d0 1466 extractors_restricted = self.params.get('allowed_extractors') not in (None, ['default'])
1467 self.report_error(f'No suitable extractor{format_field(ie_key, None, " (%s)")} found for URL {url}',
1468 tb=False if extractors_restricted else None)
a0566bbf 1469
7e88d7d7 1470 def _handle_extraction_exceptions(func):
b5ae35ee 1471 @functools.wraps(func)
a0566bbf 1472 def wrapper(self, *args, **kwargs):
6da22e7d 1473 while True:
1474 try:
1475 return func(self, *args, **kwargs)
1476 except (DownloadCancelled, LazyList.IndexError, PagedList.IndexError):
8222d8de 1477 raise
6da22e7d 1478 except ReExtractInfo as e:
1479 if e.expected:
1480 self.to_screen(f'{e}; Re-extracting data')
1481 else:
1482 self.to_stderr('\r')
1483 self.report_warning(f'{e}; Re-extracting data')
1484 continue
1485 except GeoRestrictedError as e:
1486 msg = e.msg
1487 if e.countries:
1488 msg += '\nThis video is available in %s.' % ', '.join(
1489 map(ISO3166Utils.short2full, e.countries))
1490 msg += '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1491 self.report_error(msg)
1492 except ExtractorError as e: # An error we somewhat expected
1493 self.report_error(str(e), e.format_traceback())
1494 except Exception as e:
1495 if self.params.get('ignoreerrors'):
1496 self.report_error(str(e), tb=encode_compat_str(traceback.format_exc()))
1497 else:
1498 raise
1499 break
a0566bbf 1500 return wrapper
1501
693f0600 1502 def _wait_for_video(self, ie_result={}):
f2ebc5c7 1503 if (not self.params.get('wait_for_video')
1504 or ie_result.get('_type', 'video') != 'video'
1505 or ie_result.get('formats') or ie_result.get('url')):
1506 return
1507
1508 format_dur = lambda dur: '%02d:%02d:%02d' % timetuple_from_msec(dur * 1000)[:-1]
1509 last_msg = ''
1510
1511 def progress(msg):
1512 nonlocal last_msg
a7dc6a89 1513 full_msg = f'{msg}\n'
1514 if not self.params.get('noprogress'):
1515 full_msg = msg + ' ' * (len(last_msg) - len(msg)) + '\r'
1516 elif last_msg:
1517 return
1518 self.to_screen(full_msg, skip_eol=True)
f2ebc5c7 1519 last_msg = msg
1520
1521 min_wait, max_wait = self.params.get('wait_for_video')
1522 diff = try_get(ie_result, lambda x: x['release_timestamp'] - time.time())
1523 if diff is None and ie_result.get('live_status') == 'is_upcoming':
16c620bc 1524 diff = round(random.uniform(min_wait, max_wait) if (max_wait and min_wait) else (max_wait or min_wait), 0)
f2ebc5c7 1525 self.report_warning('Release time of video is not known')
693f0600 1526 elif ie_result and (diff or 0) <= 0:
f2ebc5c7 1527 self.report_warning('Video should already be available according to extracted info')
38d79fd1 1528 diff = min(max(diff or 0, min_wait or 0), max_wait or float('inf'))
f2ebc5c7 1529 self.to_screen(f'[wait] Waiting for {format_dur(diff)} - Press Ctrl+C to try now')
1530
1531 wait_till = time.time() + diff
1532 try:
1533 while True:
1534 diff = wait_till - time.time()
1535 if diff <= 0:
1536 progress('')
1537 raise ReExtractInfo('[wait] Wait period ended', expected=True)
1538 progress(f'[wait] Remaining time until next attempt: {self._format_screen(format_dur(diff), self.Styles.EMPHASIS)}')
1539 time.sleep(1)
1540 except KeyboardInterrupt:
1541 progress('')
1542 raise ReExtractInfo('[wait] Interrupted by user', expected=True)
1543 except BaseException as e:
1544 if not isinstance(e, ReExtractInfo):
1545 self.to_screen('')
1546 raise
1547
7e88d7d7 1548 @_handle_extraction_exceptions
58f197b7 1549 def __extract_info(self, url, ie, download, extra_info, process):
693f0600 1550 try:
1551 ie_result = ie.extract(url)
1552 except UserNotLive as e:
1553 if process:
1554 if self.params.get('wait_for_video'):
1555 self.report_warning(e)
1556 self._wait_for_video()
1557 raise
a0566bbf 1558 if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
cb794ee0 1559 self.report_warning(f'Extractor {ie.IE_NAME} returned nothing{bug_reports_message()}')
a0566bbf 1560 return
1561 if isinstance(ie_result, list):
1562 # Backwards compatibility: old IE result format
1563 ie_result = {
1564 '_type': 'compat_list',
1565 'entries': ie_result,
1566 }
e37d0efb 1567 if extra_info.get('original_url'):
1568 ie_result.setdefault('original_url', extra_info['original_url'])
a0566bbf 1569 self.add_default_extra_info(ie_result, ie, url)
1570 if process:
f2ebc5c7 1571 self._wait_for_video(ie_result)
a0566bbf 1572 return self.process_ie_result(ie_result, download, extra_info)
8222d8de 1573 else:
a0566bbf 1574 return ie_result
fe7e0c98 1575
ea38e55f 1576 def add_default_extra_info(self, ie_result, ie, url):
6033d980 1577 if url is not None:
1578 self.add_extra_info(ie_result, {
1579 'webpage_url': url,
1580 'original_url': url,
57ebfca3 1581 })
1582 webpage_url = ie_result.get('webpage_url')
1583 if webpage_url:
1584 self.add_extra_info(ie_result, {
1585 'webpage_url_basename': url_basename(webpage_url),
1586 'webpage_url_domain': get_domain(webpage_url),
6033d980 1587 })
1588 if ie is not None:
1589 self.add_extra_info(ie_result, {
1590 'extractor': ie.IE_NAME,
1591 'extractor_key': ie.ie_key(),
1592 })
ea38e55f 1593
58adec46 1594 def process_ie_result(self, ie_result, download=True, extra_info=None):
8222d8de
JMF
1595 """
1596 Take the result of the ie(may be modified) and resolve all unresolved
1597 references (URLs, playlist items).
1598
1599 It will also download the videos if 'download'.
1600 Returns the resolved ie_result.
1601 """
58adec46 1602 if extra_info is None:
1603 extra_info = {}
e8ee972c
PH
1604 result_type = ie_result.get('_type', 'video')
1605
057a5206 1606 if result_type in ('url', 'url_transparent'):
8f97a15d 1607 ie_result['url'] = sanitize_url(
1608 ie_result['url'], scheme='http' if self.params.get('prefer_insecure') else 'https')
e37d0efb 1609 if ie_result.get('original_url'):
1610 extra_info.setdefault('original_url', ie_result['original_url'])
1611
057a5206 1612 extract_flat = self.params.get('extract_flat', False)
3089bc74
S
1613 if ((extract_flat == 'in_playlist' and 'playlist' in extra_info)
1614 or extract_flat is True):
ecb54191 1615 info_copy = ie_result.copy()
6033d980 1616 ie = try_get(ie_result.get('ie_key'), self.get_info_extractor)
360167b9 1617 if ie and not ie_result.get('id'):
4614bc22 1618 info_copy['id'] = ie.get_temp_id(ie_result['url'])
6033d980 1619 self.add_default_extra_info(info_copy, ie, ie_result['url'])
4614bc22 1620 self.add_extra_info(info_copy, extra_info)
b5475f11 1621 info_copy, _ = self.pre_process(info_copy)
ecb54191 1622 self.__forced_printings(info_copy, self.prepare_filename(info_copy), incomplete=True)
415f8d51 1623 self._raise_pending_errors(info_copy)
4614bc22 1624 if self.params.get('force_write_download_archive', False):
1625 self.record_download_archive(info_copy)
e8ee972c
PH
1626 return ie_result
1627
8222d8de 1628 if result_type == 'video':
b6c45014 1629 self.add_extra_info(ie_result, extra_info)
9c2b75b5 1630 ie_result = self.process_video_result(ie_result, download=download)
415f8d51 1631 self._raise_pending_errors(ie_result)
28b0eb0f 1632 additional_urls = (ie_result or {}).get('additional_urls')
9c2b75b5 1633 if additional_urls:
e9f4ccd1 1634 # TODO: Improve MetadataParserPP to allow setting a list
14f25df2 1635 if isinstance(additional_urls, str):
9c2b75b5 1636 additional_urls = [additional_urls]
1637 self.to_screen(
1638 '[info] %s: %d additional URL(s) requested' % (ie_result['id'], len(additional_urls)))
1639 self.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls))
1640 ie_result['additional_entries'] = [
1641 self.extract_info(
b69fd25c 1642 url, download, extra_info=extra_info,
9c2b75b5 1643 force_generic_extractor=self.params.get('force_generic_extractor'))
1644 for url in additional_urls
1645 ]
1646 return ie_result
8222d8de
JMF
1647 elif result_type == 'url':
1648 # We have to add extra_info to the results because it may be
1649 # contained in a playlist
07cce701 1650 return self.extract_info(
1651 ie_result['url'], download,
1652 ie_key=ie_result.get('ie_key'),
1653 extra_info=extra_info)
7fc3fa05
PH
1654 elif result_type == 'url_transparent':
1655 # Use the information from the embedding page
1656 info = self.extract_info(
1657 ie_result['url'], ie_key=ie_result.get('ie_key'),
1658 extra_info=extra_info, download=False, process=False)
1659
1640eb09
S
1660 # extract_info may return None when ignoreerrors is enabled and
1661 # extraction failed with an error, don't crash and return early
1662 # in this case
1663 if not info:
1664 return info
1665
3975b4d2 1666 exempted_fields = {'_type', 'url', 'ie_key'}
1667 if not ie_result.get('section_end') and ie_result.get('section_start') is None:
1668 # For video clips, the id etc of the clip extractor should be used
1669 exempted_fields |= {'id', 'extractor', 'extractor_key'}
1670
412c617d 1671 new_result = info.copy()
3975b4d2 1672 new_result.update(filter_dict(ie_result, lambda k, v: v is not None and k not in exempted_fields))
7fc3fa05 1673
0563f7ac
S
1674 # Extracted info may not be a video result (i.e.
1675 # info.get('_type', 'video') != video) but rather an url or
1676 # url_transparent. In such cases outer metadata (from ie_result)
1677 # should be propagated to inner one (info). For this to happen
1678 # _type of info should be overridden with url_transparent. This
067aa17e 1679 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
0563f7ac
S
1680 if new_result.get('_type') == 'url':
1681 new_result['_type'] = 'url_transparent'
7fc3fa05
PH
1682
1683 return self.process_ie_result(
1684 new_result, download=download, extra_info=extra_info)
40fcba5e 1685 elif result_type in ('playlist', 'multi_video'):
30a074c2 1686 # Protect from infinite recursion due to recursively nested playlists
1687 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
1688 webpage_url = ie_result['webpage_url']
1689 if webpage_url in self._playlist_urls:
7e85e872 1690 self.to_screen(
30a074c2 1691 '[download] Skipping already downloaded playlist: %s'
1692 % ie_result.get('title') or ie_result.get('id'))
1693 return
7e85e872 1694
30a074c2 1695 self._playlist_level += 1
1696 self._playlist_urls.add(webpage_url)
03f83004 1697 self._fill_common_fields(ie_result, False)
bc516a3f 1698 self._sanitize_thumbnails(ie_result)
30a074c2 1699 try:
1700 return self.__process_playlist(ie_result, download)
1701 finally:
1702 self._playlist_level -= 1
1703 if not self._playlist_level:
1704 self._playlist_urls.clear()
8222d8de 1705 elif result_type == 'compat_list':
c9bf4114
PH
1706 self.report_warning(
1707 'Extractor %s returned a compat_list result. '
1708 'It needs to be updated.' % ie_result.get('extractor'))
5f6a1245 1709
8222d8de 1710 def _fixup(r):
b868936c 1711 self.add_extra_info(r, {
1712 'extractor': ie_result['extractor'],
1713 'webpage_url': ie_result['webpage_url'],
1714 'webpage_url_basename': url_basename(ie_result['webpage_url']),
0bb322b9 1715 'webpage_url_domain': get_domain(ie_result['webpage_url']),
b868936c 1716 'extractor_key': ie_result['extractor_key'],
1717 })
8222d8de
JMF
1718 return r
1719 ie_result['entries'] = [
b6c45014 1720 self.process_ie_result(_fixup(r), download, extra_info)
8222d8de
JMF
1721 for r in ie_result['entries']
1722 ]
1723 return ie_result
1724 else:
1725 raise Exception('Invalid result type: %s' % result_type)
1726
e92caff5 1727 def _ensure_dir_exists(self, path):
1728 return make_dir(path, self.report_error)
1729
3b603dbd 1730 @staticmethod
3bec830a 1731 def _playlist_infodict(ie_result, strict=False, **kwargs):
1732 info = {
1733 'playlist_count': ie_result.get('playlist_count'),
3b603dbd 1734 'playlist': ie_result.get('title') or ie_result.get('id'),
1735 'playlist_id': ie_result.get('id'),
1736 'playlist_title': ie_result.get('title'),
1737 'playlist_uploader': ie_result.get('uploader'),
1738 'playlist_uploader_id': ie_result.get('uploader_id'),
3b603dbd 1739 **kwargs,
1740 }
3bec830a 1741 if strict:
1742 return info
1743 return {
1744 **info,
1745 'playlist_index': 0,
1746 '__last_playlist_index': max(ie_result['requested_entries'] or (0, 0)),
1747 'extractor': ie_result['extractor'],
1748 'webpage_url': ie_result['webpage_url'],
1749 'webpage_url_basename': url_basename(ie_result['webpage_url']),
1750 'webpage_url_domain': get_domain(ie_result['webpage_url']),
1751 'extractor_key': ie_result['extractor_key'],
1752 }
3b603dbd 1753
30a074c2 1754 def __process_playlist(self, ie_result, download):
7e88d7d7 1755 """Process each entry in the playlist"""
f5ea4748 1756 assert ie_result['_type'] in ('playlist', 'multi_video')
1757
3bec830a 1758 common_info = self._playlist_infodict(ie_result, strict=True)
3955b207 1759 title = common_info.get('playlist') or '<Untitled>'
3bec830a 1760 if self._match_entry(common_info, incomplete=True) is not None:
1761 return
c6e07cf1 1762 self.to_screen(f'[download] Downloading {ie_result["_type"]}: {title}')
f0d785d3 1763
7e88d7d7 1764 all_entries = PlaylistEntries(self, ie_result)
7e9a6125 1765 entries = orderedSet(all_entries.get_requested_items(), lazy=True)
1766
1767 lazy = self.params.get('lazy_playlist')
1768 if lazy:
1769 resolved_entries, n_entries = [], 'N/A'
1770 ie_result['requested_entries'], ie_result['entries'] = None, None
1771 else:
1772 entries = resolved_entries = list(entries)
1773 n_entries = len(resolved_entries)
1774 ie_result['requested_entries'], ie_result['entries'] = tuple(zip(*resolved_entries)) or ([], [])
1775 if not ie_result.get('playlist_count'):
1776 # Better to do this after potentially exhausting entries
1777 ie_result['playlist_count'] = all_entries.get_full_count()
498f5606 1778
0647d925 1779 extra = self._playlist_infodict(ie_result, n_entries=int_or_none(n_entries))
1780 ie_copy = collections.ChainMap(ie_result, extra)
3bec830a 1781
e08a85d8 1782 _infojson_written = False
0bfc53d0 1783 write_playlist_files = self.params.get('allow_playlist_files', True)
1784 if write_playlist_files and self.params.get('list_thumbnails'):
1785 self.list_thumbnails(ie_result)
1786 if write_playlist_files and not self.params.get('simulate'):
e08a85d8 1787 _infojson_written = self._write_info_json(
1788 'playlist', ie_result, self.prepare_filename(ie_copy, 'pl_infojson'))
1789 if _infojson_written is None:
80c03fa9 1790 return
1791 if self._write_description('playlist', ie_result,
1792 self.prepare_filename(ie_copy, 'pl_description')) is None:
1793 return
681de68e 1794 # TODO: This should be passed to ThumbnailsConvertor if necessary
3bec830a 1795 self._write_thumbnails('playlist', ie_result, self.prepare_filename(ie_copy, 'pl_thumbnail'))
30a074c2 1796
7e9a6125 1797 if lazy:
1798 if self.params.get('playlistreverse') or self.params.get('playlistrandom'):
1799 self.report_warning('playlistreverse and playlistrandom are not supported with lazy_playlist', only_once=True)
1800 elif self.params.get('playlistreverse'):
1801 entries.reverse()
1802 elif self.params.get('playlistrandom'):
30a074c2 1803 random.shuffle(entries)
1804
7e88d7d7 1805 self.to_screen(f'[{ie_result["extractor"]}] Playlist {title}: Downloading {n_entries} videos'
1806 f'{format_field(ie_result, "playlist_count", " of %s")}')
30a074c2 1807
134c913c 1808 keep_resolved_entries = self.params.get('extract_flat') != 'discard'
1809 if self.params.get('extract_flat') == 'discard_in_playlist':
1810 keep_resolved_entries = ie_result['_type'] != 'playlist'
1811 if keep_resolved_entries:
1812 self.write_debug('The information of all playlist entries will be held in memory')
1813
26e2805c 1814 failures = 0
1815 max_failures = self.params.get('skip_playlist_after_errors') or float('inf')
7e9a6125 1816 for i, (playlist_index, entry) in enumerate(entries):
1817 if lazy:
1818 resolved_entries.append((playlist_index, entry))
3bec830a 1819 if not entry:
7e88d7d7 1820 continue
1821
7e88d7d7 1822 entry['__x_forwarded_for_ip'] = ie_result.get('__x_forwarded_for_ip')
7e9a6125 1823 if not lazy and 'playlist-index' in self.params.get('compat_opts', []):
1824 playlist_index = ie_result['requested_entries'][i]
1825
0647d925 1826 entry_copy = collections.ChainMap(entry, {
3bec830a 1827 **common_info,
3955b207 1828 'n_entries': int_or_none(n_entries),
71729754 1829 'playlist_index': playlist_index,
7e9a6125 1830 'playlist_autonumber': i + 1,
0647d925 1831 })
3bec830a 1832
0647d925 1833 if self._match_entry(entry_copy, incomplete=True) is not None:
f0ad6f8c 1834 # For compatabilty with youtube-dl. See https://github.com/yt-dlp/yt-dlp/issues/4369
1835 resolved_entries[i] = (playlist_index, NO_DEFAULT)
3bec830a 1836 continue
1837
1838 self.to_screen('[download] Downloading video %s of %s' % (
1839 self._format_screen(i + 1, self.Styles.ID), self._format_screen(n_entries, self.Styles.EMPHASIS)))
1840
a6ca61d4 1841 extra.update({
1842 'playlist_index': playlist_index,
1843 'playlist_autonumber': i + 1,
1844 })
3bec830a 1845 entry_result = self.__process_iterable_entry(entry, download, extra)
26e2805c 1846 if not entry_result:
1847 failures += 1
1848 if failures >= max_failures:
1849 self.report_error(
7e88d7d7 1850 f'Skipping the remaining entries in playlist "{title}" since {failures} items failed extraction')
26e2805c 1851 break
134c913c 1852 if keep_resolved_entries:
1853 resolved_entries[i] = (playlist_index, entry_result)
7e88d7d7 1854
1855 # Update with processed data
f0ad6f8c 1856 ie_result['requested_entries'] = [i for i, e in resolved_entries if e is not NO_DEFAULT]
1857 ie_result['entries'] = [e for _, e in resolved_entries if e is not NO_DEFAULT]
e08a85d8 1858
1859 # Write the updated info to json
cb96c5be 1860 if _infojson_written is True and self._write_info_json(
e08a85d8 1861 'updated playlist', ie_result,
1862 self.prepare_filename(ie_copy, 'pl_infojson'), overwrite=True) is None:
1863 return
ca30f449 1864
ed5835b4 1865 ie_result = self.run_all_pps('playlist', ie_result)
7e88d7d7 1866 self.to_screen(f'[download] Finished downloading playlist: {title}')
30a074c2 1867 return ie_result
1868
7e88d7d7 1869 @_handle_extraction_exceptions
a0566bbf 1870 def __process_iterable_entry(self, entry, download, extra_info):
1871 return self.process_ie_result(
1872 entry, download=download, extra_info=extra_info)
1873
67134eab
JMF
1874 def _build_format_filter(self, filter_spec):
1875 " Returns a function to filter the formats according to the filter_spec "
083c9df9
PH
1876
1877 OPERATORS = {
1878 '<': operator.lt,
1879 '<=': operator.le,
1880 '>': operator.gt,
1881 '>=': operator.ge,
1882 '=': operator.eq,
1883 '!=': operator.ne,
1884 }
67134eab 1885 operator_rex = re.compile(r'''(?x)\s*
187986a8 1886 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)\s*
1887 (?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1888 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)\s*
083c9df9 1889 ''' % '|'.join(map(re.escape, OPERATORS.keys())))
187986a8 1890 m = operator_rex.fullmatch(filter_spec)
9ddb6925
S
1891 if m:
1892 try:
1893 comparison_value = int(m.group('value'))
1894 except ValueError:
1895 comparison_value = parse_filesize(m.group('value'))
1896 if comparison_value is None:
1897 comparison_value = parse_filesize(m.group('value') + 'B')
1898 if comparison_value is None:
1899 raise ValueError(
1900 'Invalid value %r in format specification %r' % (
67134eab 1901 m.group('value'), filter_spec))
9ddb6925
S
1902 op = OPERATORS[m.group('op')]
1903
083c9df9 1904 if not m:
9ddb6925
S
1905 STR_OPERATORS = {
1906 '=': operator.eq,
10d33b34
YCH
1907 '^=': lambda attr, value: attr.startswith(value),
1908 '$=': lambda attr, value: attr.endswith(value),
1909 '*=': lambda attr, value: value in attr,
1ce9a3cb 1910 '~=': lambda attr, value: value.search(attr) is not None
9ddb6925 1911 }
187986a8 1912 str_operator_rex = re.compile(r'''(?x)\s*
1913 (?P<key>[a-zA-Z0-9._-]+)\s*
1ce9a3cb
LF
1914 (?P<negation>!\s*)?(?P<op>%s)\s*(?P<none_inclusive>\?\s*)?
1915 (?P<quote>["'])?
1916 (?P<value>(?(quote)(?:(?!(?P=quote))[^\\]|\\.)+|[\w.-]+))
1917 (?(quote)(?P=quote))\s*
9ddb6925 1918 ''' % '|'.join(map(re.escape, STR_OPERATORS.keys())))
187986a8 1919 m = str_operator_rex.fullmatch(filter_spec)
9ddb6925 1920 if m:
1ce9a3cb
LF
1921 if m.group('op') == '~=':
1922 comparison_value = re.compile(m.group('value'))
1923 else:
1924 comparison_value = re.sub(r'''\\([\\"'])''', r'\1', m.group('value'))
2cc779f4
S
1925 str_op = STR_OPERATORS[m.group('op')]
1926 if m.group('negation'):
e118a879 1927 op = lambda attr, value: not str_op(attr, value)
2cc779f4
S
1928 else:
1929 op = str_op
083c9df9 1930
9ddb6925 1931 if not m:
187986a8 1932 raise SyntaxError('Invalid filter specification %r' % filter_spec)
083c9df9
PH
1933
1934 def _filter(f):
1935 actual_value = f.get(m.group('key'))
1936 if actual_value is None:
1937 return m.group('none_inclusive')
1938 return op(actual_value, comparison_value)
67134eab
JMF
1939 return _filter
1940
9f1a1c36 1941 def _check_formats(self, formats):
1942 for f in formats:
1943 self.to_screen('[info] Testing format %s' % f['format_id'])
75689fe5 1944 path = self.get_output_path('temp')
1945 if not self._ensure_dir_exists(f'{path}/'):
1946 continue
1947 temp_file = tempfile.NamedTemporaryFile(suffix='.tmp', delete=False, dir=path or None)
9f1a1c36 1948 temp_file.close()
1949 try:
1950 success, _ = self.dl(temp_file.name, f, test=True)
8a82af35 1951 except (DownloadError, OSError, ValueError) + network_exceptions:
9f1a1c36 1952 success = False
1953 finally:
1954 if os.path.exists(temp_file.name):
1955 try:
1956 os.remove(temp_file.name)
1957 except OSError:
1958 self.report_warning('Unable to delete temporary file "%s"' % temp_file.name)
1959 if success:
1960 yield f
1961 else:
1962 self.to_screen('[info] Unable to download format %s. Skipping...' % f['format_id'])
1963
0017d9ad 1964 def _default_format_spec(self, info_dict, download=True):
0017d9ad 1965
af0f7428
S
1966 def can_merge():
1967 merger = FFmpegMergerPP(self)
1968 return merger.available and merger.can_merge()
1969
91ebc640 1970 prefer_best = (
b7b04c78 1971 not self.params.get('simulate')
91ebc640 1972 and download
1973 and (
1974 not can_merge()
21633673 1975 or info_dict.get('is_live') and not self.params.get('live_from_start')
bf1824b3 1976 or self.params['outtmpl']['default'] == '-'))
53ed7066 1977 compat = (
1978 prefer_best
1979 or self.params.get('allow_multiple_audio_streams', False)
8a82af35 1980 or 'format-spec' in self.params['compat_opts'])
91ebc640 1981
1982 return (
53ed7066 1983 'best/bestvideo+bestaudio' if prefer_best
1984 else 'bestvideo*+bestaudio/best' if not compat
91ebc640 1985 else 'bestvideo+bestaudio/best')
0017d9ad 1986
67134eab
JMF
1987 def build_format_selector(self, format_spec):
1988 def syntax_error(note, start):
1989 message = (
1990 'Invalid format specification: '
86e5f3ed 1991 '{}\n\t{}\n\t{}^'.format(note, format_spec, ' ' * start[1]))
67134eab
JMF
1992 return SyntaxError(message)
1993
1994 PICKFIRST = 'PICKFIRST'
1995 MERGE = 'MERGE'
1996 SINGLE = 'SINGLE'
0130afb7 1997 GROUP = 'GROUP'
67134eab
JMF
1998 FormatSelector = collections.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
1999
91ebc640 2000 allow_multiple_streams = {'audio': self.params.get('allow_multiple_audio_streams', False),
2001 'video': self.params.get('allow_multiple_video_streams', False)}
909d24dd 2002
9f1a1c36 2003 check_formats = self.params.get('check_formats') == 'selected'
e8e73840 2004
67134eab
JMF
2005 def _parse_filter(tokens):
2006 filter_parts = []
2007 for type, string, start, _, _ in tokens:
2008 if type == tokenize.OP and string == ']':
2009 return ''.join(filter_parts)
2010 else:
2011 filter_parts.append(string)
2012
232541df 2013 def _remove_unused_ops(tokens):
62b58c09
L
2014 # Remove operators that we don't use and join them with the surrounding strings.
2015 # E.g. 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
232541df
JMF
2016 ALLOWED_OPS = ('/', '+', ',', '(', ')')
2017 last_string, last_start, last_end, last_line = None, None, None, None
2018 for type, string, start, end, line in tokens:
2019 if type == tokenize.OP and string == '[':
2020 if last_string:
2021 yield tokenize.NAME, last_string, last_start, last_end, last_line
2022 last_string = None
2023 yield type, string, start, end, line
2024 # everything inside brackets will be handled by _parse_filter
2025 for type, string, start, end, line in tokens:
2026 yield type, string, start, end, line
2027 if type == tokenize.OP and string == ']':
2028 break
2029 elif type == tokenize.OP and string in ALLOWED_OPS:
2030 if last_string:
2031 yield tokenize.NAME, last_string, last_start, last_end, last_line
2032 last_string = None
2033 yield type, string, start, end, line
2034 elif type in [tokenize.NAME, tokenize.NUMBER, tokenize.OP]:
2035 if not last_string:
2036 last_string = string
2037 last_start = start
2038 last_end = end
2039 else:
2040 last_string += string
2041 if last_string:
2042 yield tokenize.NAME, last_string, last_start, last_end, last_line
2043
cf2ac6df 2044 def _parse_format_selection(tokens, inside_merge=False, inside_choice=False, inside_group=False):
67134eab
JMF
2045 selectors = []
2046 current_selector = None
2047 for type, string, start, _, _ in tokens:
2048 # ENCODING is only defined in python 3.x
2049 if type == getattr(tokenize, 'ENCODING', None):
2050 continue
2051 elif type in [tokenize.NAME, tokenize.NUMBER]:
2052 current_selector = FormatSelector(SINGLE, string, [])
2053 elif type == tokenize.OP:
cf2ac6df
JMF
2054 if string == ')':
2055 if not inside_group:
2056 # ')' will be handled by the parentheses group
2057 tokens.restore_last_token()
67134eab 2058 break
cf2ac6df 2059 elif inside_merge and string in ['/', ',']:
0130afb7
JMF
2060 tokens.restore_last_token()
2061 break
cf2ac6df
JMF
2062 elif inside_choice and string == ',':
2063 tokens.restore_last_token()
2064 break
2065 elif string == ',':
0a31a350
JMF
2066 if not current_selector:
2067 raise syntax_error('"," must follow a format selector', start)
67134eab
JMF
2068 selectors.append(current_selector)
2069 current_selector = None
2070 elif string == '/':
d96d604e
JMF
2071 if not current_selector:
2072 raise syntax_error('"/" must follow a format selector', start)
67134eab 2073 first_choice = current_selector
cf2ac6df 2074 second_choice = _parse_format_selection(tokens, inside_choice=True)
f5f4a27a 2075 current_selector = FormatSelector(PICKFIRST, (first_choice, second_choice), [])
67134eab
JMF
2076 elif string == '[':
2077 if not current_selector:
2078 current_selector = FormatSelector(SINGLE, 'best', [])
2079 format_filter = _parse_filter(tokens)
2080 current_selector.filters.append(format_filter)
0130afb7
JMF
2081 elif string == '(':
2082 if current_selector:
2083 raise syntax_error('Unexpected "("', start)
cf2ac6df
JMF
2084 group = _parse_format_selection(tokens, inside_group=True)
2085 current_selector = FormatSelector(GROUP, group, [])
67134eab 2086 elif string == '+':
d03cfdce 2087 if not current_selector:
2088 raise syntax_error('Unexpected "+"', start)
2089 selector_1 = current_selector
2090 selector_2 = _parse_format_selection(tokens, inside_merge=True)
2091 if not selector_2:
2092 raise syntax_error('Expected a selector', start)
2093 current_selector = FormatSelector(MERGE, (selector_1, selector_2), [])
67134eab 2094 else:
86e5f3ed 2095 raise syntax_error(f'Operator not recognized: "{string}"', start)
67134eab
JMF
2096 elif type == tokenize.ENDMARKER:
2097 break
2098 if current_selector:
2099 selectors.append(current_selector)
2100 return selectors
2101
f8d4ad9a 2102 def _merge(formats_pair):
2103 format_1, format_2 = formats_pair
2104
2105 formats_info = []
2106 formats_info.extend(format_1.get('requested_formats', (format_1,)))
2107 formats_info.extend(format_2.get('requested_formats', (format_2,)))
2108
2109 if not allow_multiple_streams['video'] or not allow_multiple_streams['audio']:
551f9388 2110 get_no_more = {'video': False, 'audio': False}
f8d4ad9a 2111 for (i, fmt_info) in enumerate(formats_info):
551f9388 2112 if fmt_info.get('acodec') == fmt_info.get('vcodec') == 'none':
2113 formats_info.pop(i)
2114 continue
2115 for aud_vid in ['audio', 'video']:
f8d4ad9a 2116 if not allow_multiple_streams[aud_vid] and fmt_info.get(aud_vid[0] + 'codec') != 'none':
2117 if get_no_more[aud_vid]:
2118 formats_info.pop(i)
f5510afe 2119 break
f8d4ad9a 2120 get_no_more[aud_vid] = True
2121
2122 if len(formats_info) == 1:
2123 return formats_info[0]
2124
2125 video_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('vcodec') != 'none']
2126 audio_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('acodec') != 'none']
2127
2128 the_only_video = video_fmts[0] if len(video_fmts) == 1 else None
2129 the_only_audio = audio_fmts[0] if len(audio_fmts) == 1 else None
2130
fc61aff4
LL
2131 output_ext = get_compatible_ext(
2132 vcodecs=[f.get('vcodec') for f in video_fmts],
2133 acodecs=[f.get('acodec') for f in audio_fmts],
2134 vexts=[f['ext'] for f in video_fmts],
2135 aexts=[f['ext'] for f in audio_fmts],
2136 preferences=(try_call(lambda: self.params['merge_output_format'].split('/'))
2137 or self.params.get('prefer_free_formats') and ('webm', 'mkv')))
f8d4ad9a 2138
975a0d0d 2139 filtered = lambda *keys: filter(None, (traverse_obj(fmt, *keys) for fmt in formats_info))
2140
f8d4ad9a 2141 new_dict = {
2142 'requested_formats': formats_info,
975a0d0d 2143 'format': '+'.join(filtered('format')),
2144 'format_id': '+'.join(filtered('format_id')),
f8d4ad9a 2145 'ext': output_ext,
975a0d0d 2146 'protocol': '+'.join(map(determine_protocol, formats_info)),
093a1710 2147 'language': '+'.join(orderedSet(filtered('language'))) or None,
2148 'format_note': '+'.join(orderedSet(filtered('format_note'))) or None,
2149 'filesize_approx': sum(filtered('filesize', 'filesize_approx')) or None,
975a0d0d 2150 'tbr': sum(filtered('tbr', 'vbr', 'abr')),
f8d4ad9a 2151 }
2152
2153 if the_only_video:
2154 new_dict.update({
2155 'width': the_only_video.get('width'),
2156 'height': the_only_video.get('height'),
2157 'resolution': the_only_video.get('resolution') or self.format_resolution(the_only_video),
2158 'fps': the_only_video.get('fps'),
49a57e70 2159 'dynamic_range': the_only_video.get('dynamic_range'),
f8d4ad9a 2160 'vcodec': the_only_video.get('vcodec'),
2161 'vbr': the_only_video.get('vbr'),
2162 'stretched_ratio': the_only_video.get('stretched_ratio'),
2163 })
2164
2165 if the_only_audio:
2166 new_dict.update({
2167 'acodec': the_only_audio.get('acodec'),
2168 'abr': the_only_audio.get('abr'),
975a0d0d 2169 'asr': the_only_audio.get('asr'),
b8ed0f15 2170 'audio_channels': the_only_audio.get('audio_channels')
f8d4ad9a 2171 })
2172
2173 return new_dict
2174
e8e73840 2175 def _check_formats(formats):
981052c9 2176 if not check_formats:
2177 yield from formats
b5ac45b1 2178 return
9f1a1c36 2179 yield from self._check_formats(formats)
e8e73840 2180
67134eab 2181 def _build_selector_function(selector):
909d24dd 2182 if isinstance(selector, list): # ,
67134eab
JMF
2183 fs = [_build_selector_function(s) for s in selector]
2184
317f7ab6 2185 def selector_function(ctx):
67134eab 2186 for f in fs:
981052c9 2187 yield from f(ctx)
67134eab 2188 return selector_function
909d24dd 2189
2190 elif selector.type == GROUP: # ()
0130afb7 2191 selector_function = _build_selector_function(selector.selector)
909d24dd 2192
2193 elif selector.type == PICKFIRST: # /
67134eab
JMF
2194 fs = [_build_selector_function(s) for s in selector.selector]
2195
317f7ab6 2196 def selector_function(ctx):
67134eab 2197 for f in fs:
317f7ab6 2198 picked_formats = list(f(ctx))
67134eab
JMF
2199 if picked_formats:
2200 return picked_formats
2201 return []
67134eab 2202
981052c9 2203 elif selector.type == MERGE: # +
2204 selector_1, selector_2 = map(_build_selector_function, selector.selector)
2205
2206 def selector_function(ctx):
adbc4ec4 2207 for pair in itertools.product(selector_1(ctx), selector_2(ctx)):
981052c9 2208 yield _merge(pair)
2209
909d24dd 2210 elif selector.type == SINGLE: # atom
598d185d 2211 format_spec = selector.selector or 'best'
909d24dd 2212
f8d4ad9a 2213 # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
909d24dd 2214 if format_spec == 'all':
2215 def selector_function(ctx):
9222c381 2216 yield from _check_formats(ctx['formats'][::-1])
f8d4ad9a 2217 elif format_spec == 'mergeall':
2218 def selector_function(ctx):
316f2650 2219 formats = list(_check_formats(
2220 f for f in ctx['formats'] if f.get('vcodec') != 'none' or f.get('acodec') != 'none'))
e01d6aa4 2221 if not formats:
2222 return
921b76ca 2223 merged_format = formats[-1]
2224 for f in formats[-2::-1]:
f8d4ad9a 2225 merged_format = _merge((merged_format, f))
2226 yield merged_format
909d24dd 2227
2228 else:
85e801a9 2229 format_fallback, seperate_fallback, format_reverse, format_idx = False, None, True, 1
eff63539 2230 mobj = re.match(
2231 r'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
2232 format_spec)
2233 if mobj is not None:
2234 format_idx = int_or_none(mobj.group('n'), default=1)
e8e73840 2235 format_reverse = mobj.group('bw')[0] == 'b'
eff63539 2236 format_type = (mobj.group('type') or [None])[0]
2237 not_format_type = {'v': 'a', 'a': 'v'}.get(format_type)
2238 format_modified = mobj.group('mod') is not None
909d24dd 2239
2240 format_fallback = not format_type and not format_modified # for b, w
8326b00a 2241 _filter_f = (
eff63539 2242 (lambda f: f.get('%scodec' % format_type) != 'none')
2243 if format_type and format_modified # bv*, ba*, wv*, wa*
2244 else (lambda f: f.get('%scodec' % not_format_type) == 'none')
2245 if format_type # bv, ba, wv, wa
2246 else (lambda f: f.get('vcodec') != 'none' and f.get('acodec') != 'none')
2247 if not format_modified # b, w
8326b00a 2248 else lambda f: True) # b*, w*
2249 filter_f = lambda f: _filter_f(f) and (
2250 f.get('vcodec') != 'none' or f.get('acodec') != 'none')
67134eab 2251 else:
48ee10ee 2252 if format_spec in self._format_selection_exts['audio']:
b11c04a8 2253 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none'
48ee10ee 2254 elif format_spec in self._format_selection_exts['video']:
b11c04a8 2255 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none' and f.get('vcodec') != 'none'
85e801a9 2256 seperate_fallback = lambda f: f.get('ext') == format_spec and f.get('vcodec') != 'none'
48ee10ee 2257 elif format_spec in self._format_selection_exts['storyboards']:
b11c04a8 2258 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') == 'none' and f.get('vcodec') == 'none'
2259 else:
b5ae35ee 2260 filter_f = lambda f: f.get('format_id') == format_spec # id
909d24dd 2261
2262 def selector_function(ctx):
2263 formats = list(ctx['formats'])
909d24dd 2264 matches = list(filter(filter_f, formats)) if filter_f is not None else formats
85e801a9 2265 if not matches:
2266 if format_fallback and ctx['incomplete_formats']:
2267 # for extractors with incomplete formats (audio only (soundcloud)
2268 # or video only (imgur)) best/worst will fallback to
2269 # best/worst {video,audio}-only format
2270 matches = formats
2271 elif seperate_fallback and not ctx['has_merged_format']:
2272 # for compatibility with youtube-dl when there is no pre-merged format
2273 matches = list(filter(seperate_fallback, formats))
981052c9 2274 matches = LazyList(_check_formats(matches[::-1 if format_reverse else 1]))
2275 try:
e8e73840 2276 yield matches[format_idx - 1]
4abea8ca 2277 except LazyList.IndexError:
981052c9 2278 return
083c9df9 2279
67134eab 2280 filters = [self._build_format_filter(f) for f in selector.filters]
083c9df9 2281
317f7ab6 2282 def final_selector(ctx):
adbc4ec4 2283 ctx_copy = dict(ctx)
67134eab 2284 for _filter in filters:
317f7ab6
S
2285 ctx_copy['formats'] = list(filter(_filter, ctx_copy['formats']))
2286 return selector_function(ctx_copy)
67134eab 2287 return final_selector
083c9df9 2288
0f06bcd7 2289 stream = io.BytesIO(format_spec.encode())
0130afb7 2290 try:
f9934b96 2291 tokens = list(_remove_unused_ops(tokenize.tokenize(stream.readline)))
0130afb7
JMF
2292 except tokenize.TokenError:
2293 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec)))
2294
86e5f3ed 2295 class TokenIterator:
0130afb7
JMF
2296 def __init__(self, tokens):
2297 self.tokens = tokens
2298 self.counter = 0
2299
2300 def __iter__(self):
2301 return self
2302
2303 def __next__(self):
2304 if self.counter >= len(self.tokens):
2305 raise StopIteration()
2306 value = self.tokens[self.counter]
2307 self.counter += 1
2308 return value
2309
2310 next = __next__
2311
2312 def restore_last_token(self):
2313 self.counter -= 1
2314
2315 parsed_selector = _parse_format_selection(iter(TokenIterator(tokens)))
67134eab 2316 return _build_selector_function(parsed_selector)
a9c58ad9 2317
e5660ee6 2318 def _calc_headers(self, info_dict):
8b7539d2 2319 res = merge_headers(self.params['http_headers'], info_dict.get('http_headers') or {})
e5660ee6 2320
c487cf00 2321 cookies = self._calc_cookies(info_dict['url'])
e5660ee6
JMF
2322 if cookies:
2323 res['Cookie'] = cookies
2324
0016b84e
S
2325 if 'X-Forwarded-For' not in res:
2326 x_forwarded_for_ip = info_dict.get('__x_forwarded_for_ip')
2327 if x_forwarded_for_ip:
2328 res['X-Forwarded-For'] = x_forwarded_for_ip
2329
e5660ee6
JMF
2330 return res
2331
c487cf00 2332 def _calc_cookies(self, url):
2333 pr = sanitized_Request(url)
e5660ee6 2334 self.cookiejar.add_cookie_header(pr)
662435f7 2335 return pr.get_header('Cookie')
e5660ee6 2336
9f1a1c36 2337 def _sort_thumbnails(self, thumbnails):
2338 thumbnails.sort(key=lambda t: (
2339 t.get('preference') if t.get('preference') is not None else -1,
2340 t.get('width') if t.get('width') is not None else -1,
2341 t.get('height') if t.get('height') is not None else -1,
2342 t.get('id') if t.get('id') is not None else '',
2343 t.get('url')))
2344
b0249bca 2345 def _sanitize_thumbnails(self, info_dict):
bc516a3f 2346 thumbnails = info_dict.get('thumbnails')
2347 if thumbnails is None:
2348 thumbnail = info_dict.get('thumbnail')
2349 if thumbnail:
2350 info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}]
9f1a1c36 2351 if not thumbnails:
2352 return
2353
2354 def check_thumbnails(thumbnails):
2355 for t in thumbnails:
2356 self.to_screen(f'[info] Testing thumbnail {t["id"]}')
2357 try:
2358 self.urlopen(HEADRequest(t['url']))
2359 except network_exceptions as err:
2360 self.to_screen(f'[info] Unable to connect to thumbnail {t["id"]} URL {t["url"]!r} - {err}. Skipping...')
2361 continue
2362 yield t
2363
2364 self._sort_thumbnails(thumbnails)
2365 for i, t in enumerate(thumbnails):
2366 if t.get('id') is None:
2367 t['id'] = '%d' % i
2368 if t.get('width') and t.get('height'):
2369 t['resolution'] = '%dx%d' % (t['width'], t['height'])
2370 t['url'] = sanitize_url(t['url'])
2371
2372 if self.params.get('check_formats') is True:
282f5709 2373 info_dict['thumbnails'] = LazyList(check_thumbnails(thumbnails[::-1]), reverse=True)
9f1a1c36 2374 else:
2375 info_dict['thumbnails'] = thumbnails
bc516a3f 2376
03f83004
LNO
2377 def _fill_common_fields(self, info_dict, is_video=True):
2378 # TODO: move sanitization here
2379 if is_video:
2380 # playlists are allowed to lack "title"
d4736fdb 2381 title = info_dict.get('title', NO_DEFAULT)
2382 if title is NO_DEFAULT:
03f83004
LNO
2383 raise ExtractorError('Missing "title" field in extractor result',
2384 video_id=info_dict['id'], ie=info_dict['extractor'])
d4736fdb 2385 info_dict['fulltitle'] = title
2386 if not title:
2387 if title == '':
2388 self.write_debug('Extractor gave empty title. Creating a generic title')
2389 else:
2390 self.report_warning('Extractor failed to obtain "title". Creating a generic title instead')
1d485a1a 2391 info_dict['title'] = f'{info_dict["extractor"].replace(":", "-")} video #{info_dict["id"]}'
03f83004
LNO
2392
2393 if info_dict.get('duration') is not None:
2394 info_dict['duration_string'] = formatSeconds(info_dict['duration'])
2395
2396 for ts_key, date_key in (
2397 ('timestamp', 'upload_date'),
2398 ('release_timestamp', 'release_date'),
2399 ('modified_timestamp', 'modified_date'),
2400 ):
2401 if info_dict.get(date_key) is None and info_dict.get(ts_key) is not None:
2402 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
2403 # see http://bugs.python.org/issue1646728)
19a03940 2404 with contextlib.suppress(ValueError, OverflowError, OSError):
03f83004
LNO
2405 upload_date = datetime.datetime.utcfromtimestamp(info_dict[ts_key])
2406 info_dict[date_key] = upload_date.strftime('%Y%m%d')
03f83004
LNO
2407
2408 live_keys = ('is_live', 'was_live')
2409 live_status = info_dict.get('live_status')
2410 if live_status is None:
2411 for key in live_keys:
2412 if info_dict.get(key) is False:
2413 continue
2414 if info_dict.get(key):
2415 live_status = key
2416 break
2417 if all(info_dict.get(key) is False for key in live_keys):
2418 live_status = 'not_live'
2419 if live_status:
2420 info_dict['live_status'] = live_status
2421 for key in live_keys:
2422 if info_dict.get(key) is None:
2423 info_dict[key] = (live_status == key)
2424
2425 # Auto generate title fields corresponding to the *_number fields when missing
2426 # in order to always have clean titles. This is very common for TV series.
2427 for field in ('chapter', 'season', 'episode'):
2428 if info_dict.get('%s_number' % field) is not None and not info_dict.get(field):
2429 info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field])
2430
415f8d51 2431 def _raise_pending_errors(self, info):
2432 err = info.pop('__pending_error', None)
2433 if err:
2434 self.report_error(err, tb=False)
2435
dd82ffea
JMF
2436 def process_video_result(self, info_dict, download=True):
2437 assert info_dict.get('_type', 'video') == 'video'
9c906919 2438 self._num_videos += 1
dd82ffea 2439
bec1fad2 2440 if 'id' not in info_dict:
fc08bdd6 2441 raise ExtractorError('Missing "id" field in extractor result', ie=info_dict['extractor'])
2442 elif not info_dict.get('id'):
2443 raise ExtractorError('Extractor failed to obtain "id"', ie=info_dict['extractor'])
455a15e2 2444
c9969434
S
2445 def report_force_conversion(field, field_not, conversion):
2446 self.report_warning(
2447 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
2448 % (field, field_not, conversion))
2449
2450 def sanitize_string_field(info, string_field):
2451 field = info.get(string_field)
14f25df2 2452 if field is None or isinstance(field, str):
c9969434
S
2453 return
2454 report_force_conversion(string_field, 'a string', 'string')
14f25df2 2455 info[string_field] = str(field)
c9969434
S
2456
2457 def sanitize_numeric_fields(info):
2458 for numeric_field in self._NUMERIC_FIELDS:
2459 field = info.get(numeric_field)
f9934b96 2460 if field is None or isinstance(field, (int, float)):
c9969434
S
2461 continue
2462 report_force_conversion(numeric_field, 'numeric', 'int')
2463 info[numeric_field] = int_or_none(field)
2464
2465 sanitize_string_field(info_dict, 'id')
2466 sanitize_numeric_fields(info_dict)
3975b4d2 2467 if info_dict.get('section_end') and info_dict.get('section_start') is not None:
2468 info_dict['duration'] = round(info_dict['section_end'] - info_dict['section_start'], 3)
4c3f8c3f 2469 if (info_dict.get('duration') or 0) <= 0 and info_dict.pop('duration', None):
50e93e03 2470 self.report_warning('"duration" field is negative, there is an error in extractor')
be6217b2 2471
9eef7c4e 2472 chapters = info_dict.get('chapters') or []
a3976e07 2473 if chapters and chapters[0].get('start_time'):
2474 chapters.insert(0, {'start_time': 0})
2475
9eef7c4e 2476 dummy_chapter = {'end_time': 0, 'start_time': info_dict.get('duration')}
a3976e07 2477 for idx, (prev, current, next_) in enumerate(zip(
2478 (dummy_chapter, *chapters), chapters, (*chapters[1:], dummy_chapter)), 1):
9eef7c4e 2479 if current.get('start_time') is None:
2480 current['start_time'] = prev.get('end_time')
2481 if not current.get('end_time'):
2482 current['end_time'] = next_.get('start_time')
a3976e07 2483 if not current.get('title'):
2484 current['title'] = f'<Untitled Chapter {idx}>'
9eef7c4e 2485
dd82ffea
JMF
2486 if 'playlist' not in info_dict:
2487 # It isn't part of a playlist
2488 info_dict['playlist'] = None
2489 info_dict['playlist_index'] = None
2490
bc516a3f 2491 self._sanitize_thumbnails(info_dict)
d5519808 2492
536a55da 2493 thumbnail = info_dict.get('thumbnail')
bc516a3f 2494 thumbnails = info_dict.get('thumbnails')
536a55da
S
2495 if thumbnail:
2496 info_dict['thumbnail'] = sanitize_url(thumbnail)
2497 elif thumbnails:
d5519808
PH
2498 info_dict['thumbnail'] = thumbnails[-1]['url']
2499
ae30b840 2500 if info_dict.get('display_id') is None and 'id' in info_dict:
0afef30b
PH
2501 info_dict['display_id'] = info_dict['id']
2502
03f83004 2503 self._fill_common_fields(info_dict)
33d2fc2f 2504
05108a49
S
2505 for cc_kind in ('subtitles', 'automatic_captions'):
2506 cc = info_dict.get(cc_kind)
2507 if cc:
2508 for _, subtitle in cc.items():
2509 for subtitle_format in subtitle:
2510 if subtitle_format.get('url'):
2511 subtitle_format['url'] = sanitize_url(subtitle_format['url'])
2512 if subtitle_format.get('ext') is None:
2513 subtitle_format['ext'] = determine_ext(subtitle_format['url']).lower()
2514
2515 automatic_captions = info_dict.get('automatic_captions')
4bba3716 2516 subtitles = info_dict.get('subtitles')
4bba3716 2517
360e1ca5 2518 info_dict['requested_subtitles'] = self.process_subtitles(
05108a49 2519 info_dict['id'], subtitles, automatic_captions)
a504ced0 2520
dd82ffea
JMF
2521 if info_dict.get('formats') is None:
2522 # There's only one format available
2523 formats = [info_dict]
2524 else:
2525 formats = info_dict['formats']
2526
0a5a191a 2527 # or None ensures --clean-infojson removes it
2528 info_dict['_has_drm'] = any(f.get('has_drm') for f in formats) or None
88acdbc2 2529 if not self.params.get('allow_unplayable_formats'):
2530 formats = [f for f in formats if not f.get('has_drm')]
17ffed18 2531
2532 if formats and all(f.get('acodec') == f.get('vcodec') == 'none' for f in formats):
2533 self.report_warning(
2534 f'{"This video is DRM protected and " if info_dict["_has_drm"] else ""}'
2535 'only images are available for download. Use --list-formats to see them'.capitalize())
88acdbc2 2536
319b6059 2537 get_from_start = not info_dict.get('is_live') or bool(self.params.get('live_from_start'))
2538 if not get_from_start:
2539 info_dict['title'] += ' ' + datetime.datetime.now().strftime('%Y-%m-%d %H:%M')
2540 if info_dict.get('is_live') and formats:
adbc4ec4 2541 formats = [f for f in formats if bool(f.get('is_from_start')) == get_from_start]
319b6059 2542 if get_from_start and not formats:
a44ca5a4 2543 self.raise_no_formats(info_dict, msg=(
2544 '--live-from-start is passed, but there are no formats that can be downloaded from the start. '
2545 'If you want to download from the current time, use --no-live-from-start'))
adbc4ec4 2546
73af5cc8
S
2547 def is_wellformed(f):
2548 url = f.get('url')
a5ac0c47 2549 if not url:
73af5cc8
S
2550 self.report_warning(
2551 '"url" field is missing or empty - skipping format, '
2552 'there is an error in extractor')
a5ac0c47
S
2553 return False
2554 if isinstance(url, bytes):
2555 sanitize_string_field(f, 'url')
2556 return True
73af5cc8
S
2557
2558 # Filter out malformed formats for better extraction robustness
1ac7f461 2559 formats = list(filter(is_wellformed, formats or []))
2560
2561 if not formats:
2562 self.raise_no_formats(info_dict)
73af5cc8 2563
181c7053
S
2564 formats_dict = {}
2565
dd82ffea 2566 # We check that all the formats have the format and format_id fields
db95dc13 2567 for i, format in enumerate(formats):
c9969434
S
2568 sanitize_string_field(format, 'format_id')
2569 sanitize_numeric_fields(format)
dcf77cf1 2570 format['url'] = sanitize_url(format['url'])
e74e3b63 2571 if not format.get('format_id'):
14f25df2 2572 format['format_id'] = str(i)
e2effb08
S
2573 else:
2574 # Sanitize format_id from characters used in format selector expression
ec85ded8 2575 format['format_id'] = re.sub(r'[\s,/+\[\]()]', '_', format['format_id'])
181c7053
S
2576 format_id = format['format_id']
2577 if format_id not in formats_dict:
2578 formats_dict[format_id] = []
2579 formats_dict[format_id].append(format)
2580
2581 # Make sure all formats have unique format_id
03b4de72 2582 common_exts = set(itertools.chain(*self._format_selection_exts.values()))
181c7053 2583 for format_id, ambiguous_formats in formats_dict.items():
48ee10ee 2584 ambigious_id = len(ambiguous_formats) > 1
2585 for i, format in enumerate(ambiguous_formats):
2586 if ambigious_id:
181c7053 2587 format['format_id'] = '%s-%d' % (format_id, i)
48ee10ee 2588 if format.get('ext') is None:
2589 format['ext'] = determine_ext(format['url']).lower()
2590 # Ensure there is no conflict between id and ext in format selection
2591 # See https://github.com/yt-dlp/yt-dlp/issues/1282
2592 if format['format_id'] != format['ext'] and format['format_id'] in common_exts:
2593 format['format_id'] = 'f%s' % format['format_id']
181c7053
S
2594
2595 for i, format in enumerate(formats):
8c51aa65 2596 if format.get('format') is None:
6febd1c1 2597 format['format'] = '{id} - {res}{note}'.format(
8c51aa65
JMF
2598 id=format['format_id'],
2599 res=self.format_resolution(format),
b868936c 2600 note=format_field(format, 'format_note', ' (%s)'),
8c51aa65 2601 )
6f0be937 2602 if format.get('protocol') is None:
b5559424 2603 format['protocol'] = determine_protocol(format)
239df021 2604 if format.get('resolution') is None:
2605 format['resolution'] = self.format_resolution(format, default=None)
176f1866 2606 if format.get('dynamic_range') is None and format.get('vcodec') != 'none':
2607 format['dynamic_range'] = 'SDR'
f2fe69c7 2608 if (info_dict.get('duration') and format.get('tbr')
2609 and not format.get('filesize') and not format.get('filesize_approx')):
56ba69e4 2610 format['filesize_approx'] = int(info_dict['duration'] * format['tbr'] * (1024 / 8))
f2fe69c7 2611
e5660ee6
JMF
2612 # Add HTTP headers, so that external programs can use them from the
2613 # json output
2614 full_format_info = info_dict.copy()
2615 full_format_info.update(format)
2616 format['http_headers'] = self._calc_headers(full_format_info)
0016b84e
S
2617 # Remove private housekeeping stuff
2618 if '__x_forwarded_for_ip' in info_dict:
2619 del info_dict['__x_forwarded_for_ip']
dd82ffea 2620
9f1a1c36 2621 if self.params.get('check_formats') is True:
282f5709 2622 formats = LazyList(self._check_formats(formats[::-1]), reverse=True)
9f1a1c36 2623
88acdbc2 2624 if not formats or formats[0] is not info_dict:
b3d9ef88
JMF
2625 # only set the 'formats' fields if the original info_dict list them
2626 # otherwise we end up with a circular reference, the first (and unique)
f89197d7 2627 # element in the 'formats' field in info_dict is info_dict itself,
dfb1b146 2628 # which can't be exported to json
b3d9ef88 2629 info_dict['formats'] = formats
4ec82a72 2630
2631 info_dict, _ = self.pre_process(info_dict)
2632
6db9c4d5 2633 if self._match_entry(info_dict, incomplete=self._format_fields) is not None:
09b49e1f 2634 return info_dict
2635
2636 self.post_extract(info_dict)
2637 info_dict, _ = self.pre_process(info_dict, 'after_filter')
2638
093a1710 2639 # The pre-processors may have modified the formats
2640 formats = info_dict.get('formats', [info_dict])
2641
fa9f30b8 2642 list_only = self.params.get('simulate') is None and (
2643 self.params.get('list_thumbnails') or self.params.get('listformats') or self.params.get('listsubtitles'))
2644 interactive_format_selection = not list_only and self.format_selector == '-'
b7b04c78 2645 if self.params.get('list_thumbnails'):
2646 self.list_thumbnails(info_dict)
b7b04c78 2647 if self.params.get('listsubtitles'):
2648 if 'automatic_captions' in info_dict:
2649 self.list_subtitles(
2650 info_dict['id'], automatic_captions, 'automatic captions')
2651 self.list_subtitles(info_dict['id'], subtitles, 'subtitles')
fa9f30b8 2652 if self.params.get('listformats') or interactive_format_selection:
b69fd25c 2653 self.list_formats(info_dict)
169dbde9 2654 if list_only:
b7b04c78 2655 # Without this printing, -F --print-json will not work
169dbde9 2656 self.__forced_printings(info_dict, self.prepare_filename(info_dict), incomplete=True)
c487cf00 2657 return info_dict
bfaae0a7 2658
187986a8 2659 format_selector = self.format_selector
2660 if format_selector is None:
0017d9ad 2661 req_format = self._default_format_spec(info_dict, download=download)
0760b0a7 2662 self.write_debug('Default format spec: %s' % req_format)
187986a8 2663 format_selector = self.build_format_selector(req_format)
317f7ab6 2664
fa9f30b8 2665 while True:
2666 if interactive_format_selection:
2667 req_format = input(
2668 self._format_screen('\nEnter format selector: ', self.Styles.EMPHASIS))
2669 try:
2670 format_selector = self.build_format_selector(req_format)
2671 except SyntaxError as err:
2672 self.report_error(err, tb=False, is_error=False)
2673 continue
2674
85e801a9 2675 formats_to_download = list(format_selector({
fa9f30b8 2676 'formats': formats,
85e801a9 2677 'has_merged_format': any('none' not in (f.get('acodec'), f.get('vcodec')) for f in formats),
2678 'incomplete_formats': (
2679 # All formats are video-only or
2680 all(f.get('vcodec') != 'none' and f.get('acodec') == 'none' for f in formats)
2681 # all formats are audio-only
2682 or all(f.get('vcodec') == 'none' and f.get('acodec') != 'none' for f in formats)),
2683 }))
fa9f30b8 2684 if interactive_format_selection and not formats_to_download:
2685 self.report_error('Requested format is not available', tb=False, is_error=False)
2686 continue
2687 break
317f7ab6 2688
dd82ffea 2689 if not formats_to_download:
b7da73eb 2690 if not self.params.get('ignore_no_formats_error'):
c0b6e5c7 2691 raise ExtractorError(
2692 'Requested format is not available. Use --list-formats for a list of available formats',
2693 expected=True, video_id=info_dict['id'], ie=info_dict['extractor'])
b62fa6d7 2694 self.report_warning('Requested format is not available')
2695 # Process what we can, even without any available formats.
2696 formats_to_download = [{}]
a13e6848 2697
5ec1b6b7 2698 requested_ranges = self.params.get('download_ranges')
2699 if requested_ranges:
2700 requested_ranges = tuple(requested_ranges(info_dict, self))
2701
2702 best_format, downloaded_formats = formats_to_download[-1], []
b62fa6d7 2703 if download:
2704 if best_format:
5ec1b6b7 2705 def to_screen(*msg):
2706 self.to_screen(f'[info] {info_dict["id"]}: {" ".join(", ".join(variadic(m)) for m in msg)}')
2707
2708 to_screen(f'Downloading {len(formats_to_download)} format(s):',
2709 (f['format_id'] for f in formats_to_download))
2710 if requested_ranges:
2711 to_screen(f'Downloading {len(requested_ranges)} time ranges:',
2712 (f'{int(c["start_time"])}-{int(c["end_time"])}' for c in requested_ranges))
a13e6848 2713 max_downloads_reached = False
5ec1b6b7 2714
2715 for fmt, chapter in itertools.product(formats_to_download, requested_ranges or [{}]):
2716 new_info = self._copy_infodict(info_dict)
b7da73eb 2717 new_info.update(fmt)
3975b4d2 2718 offset, duration = info_dict.get('section_start') or 0, info_dict.get('duration') or float('inf')
2719 if chapter or offset:
5ec1b6b7 2720 new_info.update({
3975b4d2 2721 'section_start': offset + chapter.get('start_time', 0),
bc401608 2722 'section_end': offset + min(chapter.get('end_time', duration), duration),
5ec1b6b7 2723 'section_title': chapter.get('title'),
2724 'section_number': chapter.get('index'),
2725 })
2726 downloaded_formats.append(new_info)
a13e6848 2727 try:
2728 self.process_info(new_info)
2729 except MaxDownloadsReached:
2730 max_downloads_reached = True
415f8d51 2731 self._raise_pending_errors(new_info)
f46e2f9d 2732 # Remove copied info
2733 for key, val in tuple(new_info.items()):
2734 if info_dict.get(key) == val:
2735 new_info.pop(key)
a13e6848 2736 if max_downloads_reached:
2737 break
ebed8b37 2738
5ec1b6b7 2739 write_archive = {f.get('__write_download_archive', False) for f in downloaded_formats}
a13e6848 2740 assert write_archive.issubset({True, False, 'ignore'})
2741 if True in write_archive and False not in write_archive:
2742 self.record_download_archive(info_dict)
be72c624 2743
5ec1b6b7 2744 info_dict['requested_downloads'] = downloaded_formats
ed5835b4 2745 info_dict = self.run_all_pps('after_video', info_dict)
a13e6848 2746 if max_downloads_reached:
2747 raise MaxDownloadsReached()
ebed8b37 2748
49a57e70 2749 # We update the info dict with the selected best quality format (backwards compatibility)
be72c624 2750 info_dict.update(best_format)
dd82ffea
JMF
2751 return info_dict
2752
98c70d6f 2753 def process_subtitles(self, video_id, normal_subtitles, automatic_captions):
a504ced0 2754 """Select the requested subtitles and their format"""
d8a58ddc 2755 available_subs, normal_sub_langs = {}, []
98c70d6f
JMF
2756 if normal_subtitles and self.params.get('writesubtitles'):
2757 available_subs.update(normal_subtitles)
d8a58ddc 2758 normal_sub_langs = tuple(normal_subtitles.keys())
98c70d6f
JMF
2759 if automatic_captions and self.params.get('writeautomaticsub'):
2760 for lang, cap_info in automatic_captions.items():
360e1ca5
JMF
2761 if lang not in available_subs:
2762 available_subs[lang] = cap_info
2763
d2c8aadf 2764 if not available_subs or (
2765 not self.params.get('writesubtitles')
2766 and not self.params.get('writeautomaticsub')):
4d171848 2767 return None
a504ced0 2768
d8a58ddc 2769 all_sub_langs = tuple(available_subs.keys())
a504ced0 2770 if self.params.get('allsubtitles', False):
c32b0aab 2771 requested_langs = all_sub_langs
2772 elif self.params.get('subtitleslangs', False):
5314b521 2773 try:
2774 requested_langs = orderedSet_from_options(
2775 self.params.get('subtitleslangs'), {'all': all_sub_langs}, use_regex=True)
2776 except re.error as e:
2777 raise ValueError(f'Wrong regex for subtitlelangs: {e.pattern}')
d8a58ddc 2778 elif normal_sub_langs:
2779 requested_langs = ['en'] if 'en' in normal_sub_langs else normal_sub_langs[:1]
a504ced0 2780 else:
d8a58ddc 2781 requested_langs = ['en'] if 'en' in all_sub_langs else all_sub_langs[:1]
ad3dc496 2782 if requested_langs:
d2c8aadf 2783 self.to_screen(f'[info] {video_id}: Downloading subtitles: {", ".join(requested_langs)}')
a504ced0
JMF
2784
2785 formats_query = self.params.get('subtitlesformat', 'best')
2786 formats_preference = formats_query.split('/') if formats_query else []
2787 subs = {}
2788 for lang in requested_langs:
2789 formats = available_subs.get(lang)
2790 if formats is None:
86e5f3ed 2791 self.report_warning(f'{lang} subtitles not available for {video_id}')
a504ced0 2792 continue
a504ced0
JMF
2793 for ext in formats_preference:
2794 if ext == 'best':
2795 f = formats[-1]
2796 break
2797 matches = list(filter(lambda f: f['ext'] == ext, formats))
2798 if matches:
2799 f = matches[-1]
2800 break
2801 else:
2802 f = formats[-1]
2803 self.report_warning(
2804 'No subtitle format found matching "%s" for language %s, '
2805 'using %s' % (formats_query, lang, f['ext']))
2806 subs[lang] = f
2807 return subs
2808
bb66c247 2809 def _forceprint(self, key, info_dict):
2810 if info_dict is None:
2811 return
2812 info_copy = info_dict.copy()
2813 info_copy['formats_table'] = self.render_formats_table(info_dict)
2814 info_copy['thumbnails_table'] = self.render_thumbnails_table(info_dict)
2815 info_copy['subtitles_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('subtitles'))
2816 info_copy['automatic_captions_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('automatic_captions'))
2817
2818 def format_tmpl(tmpl):
48c8424b 2819 mobj = re.fullmatch(r'([\w.:,]|-\d|(?P<dict>{([\w.:,]|-\d)+}))+=?', tmpl)
07a1250e 2820 if not mobj:
2821 return tmpl
48c8424b 2822
2823 fmt = '%({})s'
2824 if tmpl.startswith('{'):
2825 tmpl = f'.{tmpl}'
2826 if tmpl.endswith('='):
2827 tmpl, fmt = tmpl[:-1], '{0} = %({0})#j'
2828 return '\n'.join(map(fmt.format, [tmpl] if mobj.group('dict') else tmpl.split(',')))
8130779d 2829
bb66c247 2830 for tmpl in self.params['forceprint'].get(key, []):
2831 self.to_stdout(self.evaluate_outtmpl(format_tmpl(tmpl), info_copy))
2832
2833 for tmpl, file_tmpl in self.params['print_to_file'].get(key, []):
5127e92a 2834 filename = self.prepare_filename(info_dict, outtmpl=file_tmpl)
bb66c247 2835 tmpl = format_tmpl(tmpl)
2836 self.to_screen(f'[info] Writing {tmpl!r} to: {filename}')
8d93e69d 2837 if self._ensure_dir_exists(filename):
86e5f3ed 2838 with open(filename, 'a', encoding='utf-8') as f:
8d93e69d 2839 f.write(self.evaluate_outtmpl(tmpl, info_copy) + '\n')
ca30f449 2840
d06daf23 2841 def __forced_printings(self, info_dict, filename, incomplete):
53c18592 2842 def print_mandatory(field, actual_field=None):
2843 if actual_field is None:
2844 actual_field = field
d06daf23 2845 if (self.params.get('force%s' % field, False)
53c18592 2846 and (not incomplete or info_dict.get(actual_field) is not None)):
2847 self.to_stdout(info_dict[actual_field])
d06daf23
S
2848
2849 def print_optional(field):
2850 if (self.params.get('force%s' % field, False)
2851 and info_dict.get(field) is not None):
2852 self.to_stdout(info_dict[field])
2853
53c18592 2854 info_dict = info_dict.copy()
2855 if filename is not None:
2856 info_dict['filename'] = filename
2857 if info_dict.get('requested_formats') is not None:
2858 # For RTMP URLs, also include the playpath
2859 info_dict['urls'] = '\n'.join(f['url'] + f.get('play_path', '') for f in info_dict['requested_formats'])
10331a26 2860 elif info_dict.get('url'):
53c18592 2861 info_dict['urls'] = info_dict['url'] + info_dict.get('play_path', '')
2862
bb66c247 2863 if (self.params.get('forcejson')
2864 or self.params['forceprint'].get('video')
2865 or self.params['print_to_file'].get('video')):
2b8a2973 2866 self.post_extract(info_dict)
bb66c247 2867 self._forceprint('video', info_dict)
53c18592 2868
d06daf23
S
2869 print_mandatory('title')
2870 print_mandatory('id')
53c18592 2871 print_mandatory('url', 'urls')
d06daf23
S
2872 print_optional('thumbnail')
2873 print_optional('description')
53c18592 2874 print_optional('filename')
b868936c 2875 if self.params.get('forceduration') and info_dict.get('duration') is not None:
d06daf23
S
2876 self.to_stdout(formatSeconds(info_dict['duration']))
2877 print_mandatory('format')
53c18592 2878
2b8a2973 2879 if self.params.get('forcejson'):
6e84b215 2880 self.to_stdout(json.dumps(self.sanitize_info(info_dict)))
d06daf23 2881
e8e73840 2882 def dl(self, name, info, subtitle=False, test=False):
88acdbc2 2883 if not info.get('url'):
1151c407 2884 self.raise_no_formats(info, True)
e8e73840 2885
2886 if test:
2887 verbose = self.params.get('verbose')
2888 params = {
2889 'test': True,
a169858f 2890 'quiet': self.params.get('quiet') or not verbose,
e8e73840 2891 'verbose': verbose,
2892 'noprogress': not verbose,
2893 'nopart': True,
2894 'skip_unavailable_fragments': False,
2895 'keep_fragments': False,
2896 'overwrites': True,
2897 '_no_ytdl_file': True,
2898 }
2899 else:
2900 params = self.params
96fccc10 2901 fd = get_suitable_downloader(info, params, to_stdout=(name == '-'))(self, params)
e8e73840 2902 if not test:
2903 for ph in self._progress_hooks:
2904 fd.add_progress_hook(ph)
42676437
M
2905 urls = '", "'.join(
2906 (f['url'].split(',')[0] + ',<data>' if f['url'].startswith('data:') else f['url'])
2907 for f in info.get('requested_formats', []) or [info])
3a408f9d 2908 self.write_debug(f'Invoking {fd.FD_NAME} downloader on "{urls}"')
03b4de72 2909
adbc4ec4
THD
2910 # Note: Ideally info should be a deep-copied so that hooks cannot modify it.
2911 # But it may contain objects that are not deep-copyable
2912 new_info = self._copy_infodict(info)
e8e73840 2913 if new_info.get('http_headers') is None:
2914 new_info['http_headers'] = self._calc_headers(new_info)
2915 return fd.download(name, new_info, subtitle)
2916
e04938ab 2917 def existing_file(self, filepaths, *, default_overwrite=True):
2918 existing_files = list(filter(os.path.exists, orderedSet(filepaths)))
2919 if existing_files and not self.params.get('overwrites', default_overwrite):
2920 return existing_files[0]
2921
2922 for file in existing_files:
2923 self.report_file_delete(file)
2924 os.remove(file)
2925 return None
2926
8222d8de 2927 def process_info(self, info_dict):
09b49e1f 2928 """Process a single resolved IE result. (Modifies it in-place)"""
8222d8de
JMF
2929
2930 assert info_dict.get('_type', 'video') == 'video'
f46e2f9d 2931 original_infodict = info_dict
fd288278 2932
4513a41a 2933 if 'format' not in info_dict and 'ext' in info_dict:
8222d8de
JMF
2934 info_dict['format'] = info_dict['ext']
2935
09b49e1f 2936 # This is mostly just for backward compatibility of process_info
2937 # As a side-effect, this allows for format-specific filters
c77495e3 2938 if self._match_entry(info_dict) is not None:
9e907ebd 2939 info_dict['__write_download_archive'] = 'ignore'
8222d8de
JMF
2940 return
2941
09b49e1f 2942 # Does nothing under normal operation - for backward compatibility of process_info
277d6ff5 2943 self.post_extract(info_dict)
0c14d66a 2944 self._num_downloads += 1
8222d8de 2945
dcf64d43 2946 # info_dict['_filename'] needs to be set for backward compatibility
de6000d9 2947 info_dict['_filename'] = full_filename = self.prepare_filename(info_dict, warn=True)
2948 temp_filename = self.prepare_filename(info_dict, 'temp')
0202b52a 2949 files_to_move = {}
8222d8de
JMF
2950
2951 # Forced printings
4513a41a 2952 self.__forced_printings(info_dict, full_filename, incomplete=('format' not in info_dict))
8222d8de 2953
ca6d59d2 2954 def check_max_downloads():
2955 if self._num_downloads >= float(self.params.get('max_downloads') or 'inf'):
2956 raise MaxDownloadsReached()
2957
b7b04c78 2958 if self.params.get('simulate'):
9e907ebd 2959 info_dict['__write_download_archive'] = self.params.get('force_write_download_archive')
ca6d59d2 2960 check_max_downloads()
8222d8de
JMF
2961 return
2962
de6000d9 2963 if full_filename is None:
8222d8de 2964 return
e92caff5 2965 if not self._ensure_dir_exists(encodeFilename(full_filename)):
0202b52a 2966 return
e92caff5 2967 if not self._ensure_dir_exists(encodeFilename(temp_filename)):
8222d8de
JMF
2968 return
2969
80c03fa9 2970 if self._write_description('video', info_dict,
2971 self.prepare_filename(info_dict, 'description')) is None:
2972 return
2973
2974 sub_files = self._write_subtitles(info_dict, temp_filename)
2975 if sub_files is None:
2976 return
2977 files_to_move.update(dict(sub_files))
2978
2979 thumb_files = self._write_thumbnails(
2980 'video', info_dict, temp_filename, self.prepare_filename(info_dict, 'thumbnail'))
2981 if thumb_files is None:
2982 return
2983 files_to_move.update(dict(thumb_files))
8222d8de 2984
80c03fa9 2985 infofn = self.prepare_filename(info_dict, 'infojson')
2986 _infojson_written = self._write_info_json('video', info_dict, infofn)
2987 if _infojson_written:
dac5df5a 2988 info_dict['infojson_filename'] = infofn
e75bb0d6 2989 # For backward compatibility, even though it was a private field
80c03fa9 2990 info_dict['__infojson_filename'] = infofn
2991 elif _infojson_written is None:
2992 return
2993
2994 # Note: Annotations are deprecated
2995 annofn = None
1fb07d10 2996 if self.params.get('writeannotations', False):
de6000d9 2997 annofn = self.prepare_filename(info_dict, 'annotation')
80c03fa9 2998 if annofn:
e92caff5 2999 if not self._ensure_dir_exists(encodeFilename(annofn)):
0202b52a 3000 return
0c3d0f51 3001 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(annofn)):
6febd1c1 3002 self.to_screen('[info] Video annotations are already present')
ffddb112
RA
3003 elif not info_dict.get('annotations'):
3004 self.report_warning('There are no annotations to write.')
7b6fefc9
PH
3005 else:
3006 try:
6febd1c1 3007 self.to_screen('[info] Writing video annotations to: ' + annofn)
86e5f3ed 3008 with open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
7b6fefc9
PH
3009 annofile.write(info_dict['annotations'])
3010 except (KeyError, TypeError):
6febd1c1 3011 self.report_warning('There are no annotations to write.')
86e5f3ed 3012 except OSError:
6febd1c1 3013 self.report_error('Cannot write annotations file: ' + annofn)
7b6fefc9 3014 return
1fb07d10 3015
732044af 3016 # Write internet shortcut files
08438d2c 3017 def _write_link_file(link_type):
60f3e995 3018 url = try_get(info_dict['webpage_url'], iri_to_uri)
3019 if not url:
3020 self.report_warning(
3021 f'Cannot write internet shortcut file because the actual URL of "{info_dict["webpage_url"]}" is unknown')
3022 return True
08438d2c 3023 linkfn = replace_extension(self.prepare_filename(info_dict, 'link'), link_type, info_dict.get('ext'))
0e6b018a
Z
3024 if not self._ensure_dir_exists(encodeFilename(linkfn)):
3025 return False
10e3742e 3026 if self.params.get('overwrites', True) and os.path.exists(encodeFilename(linkfn)):
08438d2c 3027 self.to_screen(f'[info] Internet shortcut (.{link_type}) is already present')
3028 return True
3029 try:
3030 self.to_screen(f'[info] Writing internet shortcut (.{link_type}) to: {linkfn}')
86e5f3ed 3031 with open(encodeFilename(to_high_limit_path(linkfn)), 'w', encoding='utf-8',
3032 newline='\r\n' if link_type == 'url' else '\n') as linkfile:
60f3e995 3033 template_vars = {'url': url}
08438d2c 3034 if link_type == 'desktop':
3035 template_vars['filename'] = linkfn[:-(len(link_type) + 1)]
3036 linkfile.write(LINK_TEMPLATES[link_type] % template_vars)
86e5f3ed 3037 except OSError:
08438d2c 3038 self.report_error(f'Cannot write internet shortcut {linkfn}')
3039 return False
732044af 3040 return True
3041
08438d2c 3042 write_links = {
3043 'url': self.params.get('writeurllink'),
3044 'webloc': self.params.get('writewebloclink'),
3045 'desktop': self.params.get('writedesktoplink'),
3046 }
3047 if self.params.get('writelink'):
3048 link_type = ('webloc' if sys.platform == 'darwin'
3049 else 'desktop' if sys.platform.startswith('linux')
3050 else 'url')
3051 write_links[link_type] = True
3052
3053 if any(should_write and not _write_link_file(link_type)
3054 for link_type, should_write in write_links.items()):
3055 return
732044af 3056
f46e2f9d 3057 def replace_info_dict(new_info):
3058 nonlocal info_dict
3059 if new_info == info_dict:
3060 return
3061 info_dict.clear()
3062 info_dict.update(new_info)
3063
415f8d51 3064 new_info, files_to_move = self.pre_process(info_dict, 'before_dl', files_to_move)
3065 replace_info_dict(new_info)
56d868db 3066
a13e6848 3067 if self.params.get('skip_download'):
56d868db 3068 info_dict['filepath'] = temp_filename
3069 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
3070 info_dict['__files_to_move'] = files_to_move
f46e2f9d 3071 replace_info_dict(self.run_pp(MoveFilesAfterDownloadPP(self, False), info_dict))
9e907ebd 3072 info_dict['__write_download_archive'] = self.params.get('force_write_download_archive')
56d868db 3073 else:
3074 # Download
b868936c 3075 info_dict.setdefault('__postprocessors', [])
4340deca 3076 try:
0202b52a 3077
e04938ab 3078 def existing_video_file(*filepaths):
6b591b29 3079 ext = info_dict.get('ext')
e04938ab 3080 converted = lambda file: replace_extension(file, self.params.get('final_ext') or ext, ext)
3081 file = self.existing_file(itertools.chain(*zip(map(converted, filepaths), filepaths)),
3082 default_overwrite=False)
3083 if file:
3084 info_dict['ext'] = os.path.splitext(file)[1][1:]
3085 return file
0202b52a 3086
7b2c3f47 3087 fd, success = None, True
fccf90e7 3088 if info_dict.get('protocol') or info_dict.get('url'):
56ba69e4 3089 fd = get_suitable_downloader(info_dict, self.params, to_stdout=temp_filename == '-')
3090 if fd is not FFmpegFD and (
3091 info_dict.get('section_start') or info_dict.get('section_end')):
7b2c3f47 3092 msg = ('This format cannot be partially downloaded' if FFmpegFD.available()
56ba69e4 3093 else 'You have requested downloading the video partially, but ffmpeg is not installed')
3094 self.report_error(f'{msg}. Aborting')
5ec1b6b7 3095 return
5ec1b6b7 3096
4340deca 3097 if info_dict.get('requested_formats') is not None:
81cd954a 3098 requested_formats = info_dict['requested_formats']
0202b52a 3099 old_ext = info_dict['ext']
4e3b637d 3100 if self.params.get('merge_output_format') is None:
4e3b637d 3101 if (info_dict['ext'] == 'webm'
3102 and info_dict.get('thumbnails')
3103 # check with type instead of pp_key, __name__, or isinstance
3104 # since we dont want any custom PPs to trigger this
c487cf00 3105 and any(type(pp) == EmbedThumbnailPP for pp in self._pps['post_process'])): # noqa: E721
4e3b637d 3106 info_dict['ext'] = 'mkv'
3107 self.report_warning(
3108 'webm doesn\'t support embedding a thumbnail, mkv will be used')
124bc071 3109 new_ext = info_dict['ext']
0202b52a 3110
124bc071 3111 def correct_ext(filename, ext=new_ext):
96fccc10 3112 if filename == '-':
3113 return filename
0202b52a 3114 filename_real_ext = os.path.splitext(filename)[1][1:]
3115 filename_wo_ext = (
3116 os.path.splitext(filename)[0]
124bc071 3117 if filename_real_ext in (old_ext, new_ext)
0202b52a 3118 else filename)
86e5f3ed 3119 return f'{filename_wo_ext}.{ext}'
0202b52a 3120
38c6902b 3121 # Ensure filename always has a correct extension for successful merge
0202b52a 3122 full_filename = correct_ext(full_filename)
3123 temp_filename = correct_ext(temp_filename)
e04938ab 3124 dl_filename = existing_video_file(full_filename, temp_filename)
1ea24129 3125 info_dict['__real_download'] = False
18e674b4 3126
7b2c3f47 3127 merger = FFmpegMergerPP(self)
adbc4ec4 3128 downloaded = []
dbf5416a 3129 if dl_filename is not None:
6c7274ec 3130 self.report_file_already_downloaded(dl_filename)
adbc4ec4
THD
3131 elif fd:
3132 for f in requested_formats if fd != FFmpegFD else []:
3133 f['filepath'] = fname = prepend_extension(
3134 correct_ext(temp_filename, info_dict['ext']),
3135 'f%s' % f['format_id'], info_dict['ext'])
3136 downloaded.append(fname)
dbf5416a 3137 info_dict['url'] = '\n'.join(f['url'] for f in requested_formats)
3138 success, real_download = self.dl(temp_filename, info_dict)
3139 info_dict['__real_download'] = real_download
18e674b4 3140 else:
18e674b4 3141 if self.params.get('allow_unplayable_formats'):
3142 self.report_warning(
3143 'You have requested merging of multiple formats '
3144 'while also allowing unplayable formats to be downloaded. '
3145 'The formats won\'t be merged to prevent data corruption.')
3146 elif not merger.available:
e8969bda 3147 msg = 'You have requested merging of multiple formats but ffmpeg is not installed'
3148 if not self.params.get('ignoreerrors'):
3149 self.report_error(f'{msg}. Aborting due to --abort-on-error')
3150 return
3151 self.report_warning(f'{msg}. The formats won\'t be merged')
18e674b4 3152
96fccc10 3153 if temp_filename == '-':
adbc4ec4 3154 reason = ('using a downloader other than ffmpeg' if FFmpegFD.can_merge_formats(info_dict, self.params)
96fccc10 3155 else 'but the formats are incompatible for simultaneous download' if merger.available
3156 else 'but ffmpeg is not installed')
3157 self.report_warning(
3158 f'You have requested downloading multiple formats to stdout {reason}. '
3159 'The formats will be streamed one after the other')
3160 fname = temp_filename
dbf5416a 3161 for f in requested_formats:
3162 new_info = dict(info_dict)
3163 del new_info['requested_formats']
3164 new_info.update(f)
96fccc10 3165 if temp_filename != '-':
124bc071 3166 fname = prepend_extension(
3167 correct_ext(temp_filename, new_info['ext']),
3168 'f%s' % f['format_id'], new_info['ext'])
96fccc10 3169 if not self._ensure_dir_exists(fname):
3170 return
a21e0ab1 3171 f['filepath'] = fname
96fccc10 3172 downloaded.append(fname)
dbf5416a 3173 partial_success, real_download = self.dl(fname, new_info)
3174 info_dict['__real_download'] = info_dict['__real_download'] or real_download
3175 success = success and partial_success
adbc4ec4
THD
3176
3177 if downloaded and merger.available and not self.params.get('allow_unplayable_formats'):
3178 info_dict['__postprocessors'].append(merger)
3179 info_dict['__files_to_merge'] = downloaded
3180 # Even if there were no downloads, it is being merged only now
3181 info_dict['__real_download'] = True
3182 else:
3183 for file in downloaded:
3184 files_to_move[file] = None
4340deca
P
3185 else:
3186 # Just a single file
e04938ab 3187 dl_filename = existing_video_file(full_filename, temp_filename)
6c7274ec 3188 if dl_filename is None or dl_filename == temp_filename:
3189 # dl_filename == temp_filename could mean that the file was partially downloaded with --no-part.
3190 # So we should try to resume the download
e8e73840 3191 success, real_download = self.dl(temp_filename, info_dict)
0202b52a 3192 info_dict['__real_download'] = real_download
6c7274ec 3193 else:
3194 self.report_file_already_downloaded(dl_filename)
0202b52a 3195
0202b52a 3196 dl_filename = dl_filename or temp_filename
c571435f 3197 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
0202b52a 3198
3158150c 3199 except network_exceptions as err:
7960b056 3200 self.report_error('unable to download video data: %s' % error_to_compat_str(err))
4340deca 3201 return
86e5f3ed 3202 except OSError as err:
4340deca
P
3203 raise UnavailableVideoError(err)
3204 except (ContentTooShortError, ) as err:
86e5f3ed 3205 self.report_error(f'content too short (expected {err.expected} bytes and served {err.downloaded})')
4340deca 3206 return
8222d8de 3207
415f8d51 3208 self._raise_pending_errors(info_dict)
de6000d9 3209 if success and full_filename != '-':
f17f8651 3210
fd7cfb64 3211 def fixup():
3212 do_fixup = True
3213 fixup_policy = self.params.get('fixup')
3214 vid = info_dict['id']
3215
3216 if fixup_policy in ('ignore', 'never'):
3217 return
3218 elif fixup_policy == 'warn':
3fe75fdc 3219 do_fixup = 'warn'
f89b3e2d 3220 elif fixup_policy != 'force':
3221 assert fixup_policy in ('detect_or_warn', None)
3222 if not info_dict.get('__real_download'):
3223 do_fixup = False
fd7cfb64 3224
3225 def ffmpeg_fixup(cndn, msg, cls):
3fe75fdc 3226 if not (do_fixup and cndn):
fd7cfb64 3227 return
3fe75fdc 3228 elif do_fixup == 'warn':
fd7cfb64 3229 self.report_warning(f'{vid}: {msg}')
3230 return
3231 pp = cls(self)
3232 if pp.available:
3233 info_dict['__postprocessors'].append(pp)
3234 else:
3235 self.report_warning(f'{vid}: {msg}. Install ffmpeg to fix this automatically')
3236
3237 stretched_ratio = info_dict.get('stretched_ratio')
ca9def71
LNO
3238 ffmpeg_fixup(stretched_ratio not in (1, None),
3239 f'Non-uniform pixel ratio {stretched_ratio}',
3240 FFmpegFixupStretchedPP)
fd7cfb64 3241
993191c0 3242 downloader = get_suitable_downloader(info_dict, self.params) if 'protocol' in info_dict else None
24146491 3243 downloader = downloader.FD_NAME if downloader else None
adbc4ec4 3244
ca9def71
LNO
3245 ext = info_dict.get('ext')
3246 postprocessed_by_ffmpeg = info_dict.get('requested_formats') or any((
3247 isinstance(pp, FFmpegVideoConvertorPP)
3248 and resolve_recode_mapping(ext, pp.mapping)[0] not in (ext, None)
3249 ) for pp in self._pps['post_process'])
3250
3251 if not postprocessed_by_ffmpeg:
3252 ffmpeg_fixup(ext == 'm4a' and info_dict.get('container') == 'm4a_dash',
f2df4071 3253 'writing DASH m4a. Only some players support this container',
3254 FFmpegFixupM4aPP)
24146491 3255 ffmpeg_fixup(downloader == 'hlsnative' and not self.params.get('hls_use_mpegts')
494f5230 3256 or info_dict.get('is_live') and self.params.get('hls_use_mpegts') is None,
adbc4ec4
THD
3257 'Possible MPEG-TS in MP4 container or malformed AAC timestamps',
3258 FFmpegFixupM3u8PP)
3259 ffmpeg_fixup(info_dict.get('is_live') and downloader == 'DashSegmentsFD',
3260 'Possible duplicate MOOV atoms', FFmpegFixupDuplicateMoovPP)
3261
24146491 3262 ffmpeg_fixup(downloader == 'web_socket_fragment', 'Malformed timestamps detected', FFmpegFixupTimestampPP)
3263 ffmpeg_fixup(downloader == 'web_socket_fragment', 'Malformed duration detected', FFmpegFixupDurationPP)
fd7cfb64 3264
3265 fixup()
8222d8de 3266 try:
f46e2f9d 3267 replace_info_dict(self.post_process(dl_filename, info_dict, files_to_move))
af819c21 3268 except PostProcessingError as err:
3269 self.report_error('Postprocessing: %s' % str(err))
8222d8de 3270 return
ab8e5e51
AM
3271 try:
3272 for ph in self._post_hooks:
23c1a667 3273 ph(info_dict['filepath'])
ab8e5e51
AM
3274 except Exception as err:
3275 self.report_error('post hooks: %s' % str(err))
3276 return
9e907ebd 3277 info_dict['__write_download_archive'] = True
2d30509f 3278
c487cf00 3279 assert info_dict is original_infodict # Make sure the info_dict was modified in-place
a13e6848 3280 if self.params.get('force_write_download_archive'):
9e907ebd 3281 info_dict['__write_download_archive'] = True
ca6d59d2 3282 check_max_downloads()
8222d8de 3283
aa9369a2 3284 def __download_wrapper(self, func):
3285 @functools.wraps(func)
3286 def wrapper(*args, **kwargs):
3287 try:
3288 res = func(*args, **kwargs)
3289 except UnavailableVideoError as e:
3290 self.report_error(e)
b222c271 3291 except DownloadCancelled as e:
3292 self.to_screen(f'[info] {e}')
3293 if not self.params.get('break_per_url'):
3294 raise
fd404bec 3295 self._num_downloads = 0
aa9369a2 3296 else:
3297 if self.params.get('dump_single_json', False):
3298 self.post_extract(res)
3299 self.to_stdout(json.dumps(self.sanitize_info(res)))
3300 return wrapper
3301
8222d8de
JMF
3302 def download(self, url_list):
3303 """Download a given list of URLs."""
aa9369a2 3304 url_list = variadic(url_list) # Passing a single URL is a common mistake
bf1824b3 3305 outtmpl = self.params['outtmpl']['default']
3089bc74
S
3306 if (len(url_list) > 1
3307 and outtmpl != '-'
3308 and '%' not in outtmpl
3309 and self.params.get('max_downloads') != 1):
acd69589 3310 raise SameFileError(outtmpl)
8222d8de
JMF
3311
3312 for url in url_list:
aa9369a2 3313 self.__download_wrapper(self.extract_info)(
3314 url, force_generic_extractor=self.params.get('force_generic_extractor', False))
8222d8de
JMF
3315
3316 return self._download_retcode
3317
1dcc4c0c 3318 def download_with_info_file(self, info_filename):
31bd3925
JMF
3319 with contextlib.closing(fileinput.FileInput(
3320 [info_filename], mode='r',
3321 openhook=fileinput.hook_encoded('utf-8'))) as f:
3322 # FileInput doesn't have a read method, we can't call json.load
8012d892 3323 info = self.sanitize_info(json.loads('\n'.join(f)), self.params.get('clean_infojson', True))
d4943898 3324 try:
aa9369a2 3325 self.__download_wrapper(self.process_ie_result)(info, download=True)
f2ebc5c7 3326 except (DownloadError, EntryNotInPlaylist, ReExtractInfo) as e:
bf5f605e 3327 if not isinstance(e, EntryNotInPlaylist):
3328 self.to_stderr('\r')
d4943898
JMF
3329 webpage_url = info.get('webpage_url')
3330 if webpage_url is not None:
aa9369a2 3331 self.report_warning(f'The info failed to download: {e}; trying with URL {webpage_url}')
d4943898
JMF
3332 return self.download([webpage_url])
3333 else:
3334 raise
3335 return self._download_retcode
1dcc4c0c 3336
cb202fd2 3337 @staticmethod
8012d892 3338 def sanitize_info(info_dict, remove_private_keys=False):
3339 ''' Sanitize the infodict for converting to json '''
3ad56b42 3340 if info_dict is None:
3341 return info_dict
6e84b215 3342 info_dict.setdefault('epoch', int(time.time()))
6a5a30f9 3343 info_dict.setdefault('_type', 'video')
b5e7a2e6 3344 info_dict.setdefault('_version', {
3345 'version': __version__,
3346 'current_git_head': current_git_head(),
3347 'release_git_head': RELEASE_GIT_HEAD,
3348 'repository': REPOSITORY,
3349 })
09b49e1f 3350
8012d892 3351 if remove_private_keys:
0a5a191a 3352 reject = lambda k, v: v is None or k.startswith('__') or k in {
f46e2f9d 3353 'requested_downloads', 'requested_formats', 'requested_subtitles', 'requested_entries',
0a5a191a 3354 'entries', 'filepath', '_filename', 'infojson_filename', 'original_url', 'playlist_autonumber',
6e84b215 3355 }
ae8f99e6 3356 else:
09b49e1f 3357 reject = lambda k, v: False
adbc4ec4
THD
3358
3359 def filter_fn(obj):
3360 if isinstance(obj, dict):
3361 return {k: filter_fn(v) for k, v in obj.items() if not reject(k, v)}
3362 elif isinstance(obj, (list, tuple, set, LazyList)):
3363 return list(map(filter_fn, obj))
3364 elif obj is None or isinstance(obj, (str, int, float, bool)):
3365 return obj
3366 else:
3367 return repr(obj)
3368
5226731e 3369 return filter_fn(info_dict)
cb202fd2 3370
8012d892 3371 @staticmethod
3372 def filter_requested_info(info_dict, actually_filter=True):
3373 ''' Alias of sanitize_info for backward compatibility '''
3374 return YoutubeDL.sanitize_info(info_dict, actually_filter)
3375
43d7f5a5 3376 def _delete_downloaded_files(self, *files_to_delete, info={}, msg=None):
3377 for filename in set(filter(None, files_to_delete)):
3378 if msg:
3379 self.to_screen(msg % filename)
3380 try:
3381 os.remove(filename)
3382 except OSError:
3383 self.report_warning(f'Unable to delete file {filename}')
3384 if filename in info.get('__files_to_move', []): # NB: Delete even if None
3385 del info['__files_to_move'][filename]
3386
ed5835b4 3387 @staticmethod
3388 def post_extract(info_dict):
3389 def actual_post_extract(info_dict):
3390 if info_dict.get('_type') in ('playlist', 'multi_video'):
3391 for video_dict in info_dict.get('entries', {}):
3392 actual_post_extract(video_dict or {})
3393 return
3394
09b49e1f 3395 post_extractor = info_dict.pop('__post_extractor', None) or (lambda: {})
3396 info_dict.update(post_extractor())
ed5835b4 3397
3398 actual_post_extract(info_dict or {})
3399
dcf64d43 3400 def run_pp(self, pp, infodict):
5bfa4862 3401 files_to_delete = []
dcf64d43 3402 if '__files_to_move' not in infodict:
3403 infodict['__files_to_move'] = {}
b1940459 3404 try:
3405 files_to_delete, infodict = pp.run(infodict)
3406 except PostProcessingError as e:
3407 # Must be True and not 'only_download'
3408 if self.params.get('ignoreerrors') is True:
3409 self.report_error(e)
3410 return infodict
3411 raise
3412
5bfa4862 3413 if not files_to_delete:
dcf64d43 3414 return infodict
5bfa4862 3415 if self.params.get('keepvideo', False):
3416 for f in files_to_delete:
dcf64d43 3417 infodict['__files_to_move'].setdefault(f, '')
5bfa4862 3418 else:
43d7f5a5 3419 self._delete_downloaded_files(
3420 *files_to_delete, info=infodict, msg='Deleting original file %s (pass -k to keep)')
dcf64d43 3421 return infodict
5bfa4862 3422
ed5835b4 3423 def run_all_pps(self, key, info, *, additional_pps=None):
bb66c247 3424 self._forceprint(key, info)
ed5835b4 3425 for pp in (additional_pps or []) + self._pps[key]:
dc5f409c 3426 info = self.run_pp(pp, info)
ed5835b4 3427 return info
277d6ff5 3428
56d868db 3429 def pre_process(self, ie_info, key='pre_process', files_to_move=None):
5bfa4862 3430 info = dict(ie_info)
56d868db 3431 info['__files_to_move'] = files_to_move or {}
415f8d51 3432 try:
3433 info = self.run_all_pps(key, info)
3434 except PostProcessingError as err:
3435 msg = f'Preprocessing: {err}'
3436 info.setdefault('__pending_error', msg)
3437 self.report_error(msg, is_error=False)
56d868db 3438 return info, info.pop('__files_to_move', None)
5bfa4862 3439
f46e2f9d 3440 def post_process(self, filename, info, files_to_move=None):
8222d8de 3441 """Run all the postprocessors on the given file."""
8222d8de 3442 info['filepath'] = filename
dcf64d43 3443 info['__files_to_move'] = files_to_move or {}
ed5835b4 3444 info = self.run_all_pps('post_process', info, additional_pps=info.get('__postprocessors'))
dcf64d43 3445 info = self.run_pp(MoveFilesAfterDownloadPP(self), info)
3446 del info['__files_to_move']
ed5835b4 3447 return self.run_all_pps('after_move', info)
c1c9a79c 3448
5db07df6 3449 def _make_archive_id(self, info_dict):
e9fef7ee
S
3450 video_id = info_dict.get('id')
3451 if not video_id:
3452 return
5db07df6
PH
3453 # Future-proof against any change in case
3454 # and backwards compatibility with prior versions
e9fef7ee 3455 extractor = info_dict.get('extractor_key') or info_dict.get('ie_key') # key in a playlist
7012b23c 3456 if extractor is None:
1211bb6d
S
3457 url = str_or_none(info_dict.get('url'))
3458 if not url:
3459 return
e9fef7ee 3460 # Try to find matching extractor for the URL and take its ie_key
8b7491c8 3461 for ie_key, ie in self._ies.items():
1211bb6d 3462 if ie.suitable(url):
8b7491c8 3463 extractor = ie_key
e9fef7ee
S
3464 break
3465 else:
3466 return
0647d925 3467 return make_archive_id(extractor, video_id)
5db07df6
PH
3468
3469 def in_download_archive(self, info_dict):
ae103564 3470 if not self.archive:
5db07df6
PH
3471 return False
3472
1e8fe57e 3473 vid_ids = [self._make_archive_id(info_dict)]
c200096c 3474 vid_ids.extend(info_dict.get('_old_archive_ids') or [])
1e8fe57e 3475 return any(id_ in self.archive for id_ in vid_ids)
c1c9a79c
PH
3476
3477 def record_download_archive(self, info_dict):
3478 fn = self.params.get('download_archive')
3479 if fn is None:
3480 return
5db07df6
PH
3481 vid_id = self._make_archive_id(info_dict)
3482 assert vid_id
ae103564 3483
a13e6848 3484 self.write_debug(f'Adding to archive: {vid_id}')
ae103564 3485 if isinstance(fn, os.PathLike):
3486 with locked_file(fn, 'a', encoding='utf-8') as archive_file:
3487 archive_file.write(vid_id + '\n')
a45e8619 3488 self.archive.add(vid_id)
dd82ffea 3489
8c51aa65 3490 @staticmethod
8abeeb94 3491 def format_resolution(format, default='unknown'):
9359f3d4 3492 if format.get('vcodec') == 'none' and format.get('acodec') != 'none':
fb04e403 3493 return 'audio only'
f49d89ee
PH
3494 if format.get('resolution') is not None:
3495 return format['resolution']
35615307 3496 if format.get('width') and format.get('height'):
ff51ed58 3497 return '%dx%d' % (format['width'], format['height'])
35615307 3498 elif format.get('height'):
ff51ed58 3499 return '%sp' % format['height']
35615307 3500 elif format.get('width'):
ff51ed58 3501 return '%dx?' % format['width']
3502 return default
8c51aa65 3503
8130779d 3504 def _list_format_headers(self, *headers):
3505 if self.params.get('listformats_table', True) is not False:
591bb9d3 3506 return [self._format_out(header, self.Styles.HEADERS) for header in headers]
8130779d 3507 return headers
3508
c57f7757
PH
3509 def _format_note(self, fdict):
3510 res = ''
3511 if fdict.get('ext') in ['f4f', 'f4m']:
f304da8a 3512 res += '(unsupported)'
32f90364
PH
3513 if fdict.get('language'):
3514 if res:
3515 res += ' '
f304da8a 3516 res += '[%s]' % fdict['language']
c57f7757 3517 if fdict.get('format_note') is not None:
f304da8a 3518 if res:
3519 res += ' '
3520 res += fdict['format_note']
c57f7757 3521 if fdict.get('tbr') is not None:
f304da8a 3522 if res:
3523 res += ', '
3524 res += '%4dk' % fdict['tbr']
c57f7757
PH
3525 if fdict.get('container') is not None:
3526 if res:
3527 res += ', '
3528 res += '%s container' % fdict['container']
3089bc74
S
3529 if (fdict.get('vcodec') is not None
3530 and fdict.get('vcodec') != 'none'):
c57f7757
PH
3531 if res:
3532 res += ', '
3533 res += fdict['vcodec']
91c7271a 3534 if fdict.get('vbr') is not None:
c57f7757
PH
3535 res += '@'
3536 elif fdict.get('vbr') is not None and fdict.get('abr') is not None:
3537 res += 'video@'
3538 if fdict.get('vbr') is not None:
3539 res += '%4dk' % fdict['vbr']
fbb21cf5 3540 if fdict.get('fps') is not None:
5d583bdf
S
3541 if res:
3542 res += ', '
3543 res += '%sfps' % fdict['fps']
c57f7757
PH
3544 if fdict.get('acodec') is not None:
3545 if res:
3546 res += ', '
3547 if fdict['acodec'] == 'none':
3548 res += 'video only'
3549 else:
3550 res += '%-5s' % fdict['acodec']
3551 elif fdict.get('abr') is not None:
3552 if res:
3553 res += ', '
3554 res += 'audio'
3555 if fdict.get('abr') is not None:
3556 res += '@%3dk' % fdict['abr']
3557 if fdict.get('asr') is not None:
3558 res += ' (%5dHz)' % fdict['asr']
3559 if fdict.get('filesize') is not None:
3560 if res:
3561 res += ', '
3562 res += format_bytes(fdict['filesize'])
9732d77e
PH
3563 elif fdict.get('filesize_approx') is not None:
3564 if res:
3565 res += ', '
3566 res += '~' + format_bytes(fdict['filesize_approx'])
c57f7757 3567 return res
91c7271a 3568
8130779d 3569 def render_formats_table(self, info_dict):
b69fd25c 3570 if not info_dict.get('formats') and not info_dict.get('url'):
8130779d 3571 return None
b69fd25c 3572
94badb25 3573 formats = info_dict.get('formats', [info_dict])
8130779d 3574 if not self.params.get('listformats_table', True) is not False:
76d321f6 3575 table = [
3576 [
3577 format_field(f, 'format_id'),
3578 format_field(f, 'ext'),
3579 self.format_resolution(f),
8130779d 3580 self._format_note(f)
3581 ] for f in formats if f.get('preference') is None or f['preference'] >= -1000]
3582 return render_table(['format code', 'extension', 'resolution', 'note'], table, extra_gap=1)
3583
d816f61f 3584 def simplified_codec(f, field):
3585 assert field in ('acodec', 'vcodec')
3586 codec = f.get(field, 'unknown')
f5ea4748 3587 if not codec:
3588 return 'unknown'
3589 elif codec != 'none':
d816f61f 3590 return '.'.join(codec.split('.')[:4])
3591
3592 if field == 'vcodec' and f.get('acodec') == 'none':
3593 return 'images'
3594 elif field == 'acodec' and f.get('vcodec') == 'none':
3595 return ''
3596 return self._format_out('audio only' if field == 'vcodec' else 'video only',
3597 self.Styles.SUPPRESS)
3598
591bb9d3 3599 delim = self._format_out('\u2502', self.Styles.DELIM, '|', test_encoding=True)
8130779d 3600 table = [
3601 [
591bb9d3 3602 self._format_out(format_field(f, 'format_id'), self.Styles.ID),
8130779d 3603 format_field(f, 'ext'),
3604 format_field(f, func=self.format_resolution, ignore=('audio only', 'images')),
563e0bf8 3605 format_field(f, 'fps', '\t%d', func=round),
8130779d 3606 format_field(f, 'dynamic_range', '%s', ignore=(None, 'SDR')).replace('HDR', ''),
b8ed0f15 3607 format_field(f, 'audio_channels', '\t%s'),
8130779d 3608 delim,
3609 format_field(f, 'filesize', ' \t%s', func=format_bytes) + format_field(f, 'filesize_approx', '~\t%s', func=format_bytes),
563e0bf8 3610 format_field(f, 'tbr', '\t%dk', func=round),
8130779d 3611 shorten_protocol_name(f.get('protocol', '')),
3612 delim,
d816f61f 3613 simplified_codec(f, 'vcodec'),
563e0bf8 3614 format_field(f, 'vbr', '\t%dk', func=round),
d816f61f 3615 simplified_codec(f, 'acodec'),
563e0bf8 3616 format_field(f, 'abr', '\t%dk', func=round),
ae61d108 3617 format_field(f, 'asr', '\t%s', func=format_decimal_suffix),
8130779d 3618 join_nonempty(
591bb9d3 3619 self._format_out('UNSUPPORTED', 'light red') if f.get('ext') in ('f4f', 'f4m') else None,
8130779d 3620 format_field(f, 'language', '[%s]'),
3621 join_nonempty(format_field(f, 'format_note'),
3622 format_field(f, 'container', ignore=(None, f.get('ext'))),
3623 delim=', '),
3624 delim=' '),
3625 ] for f in formats if f.get('preference') is None or f['preference'] >= -1000]
3626 header_line = self._list_format_headers(
b8ed0f15 3627 'ID', 'EXT', 'RESOLUTION', '\tFPS', 'HDR', 'CH', delim, '\tFILESIZE', '\tTBR', 'PROTO',
8130779d 3628 delim, 'VCODEC', '\tVBR', 'ACODEC', '\tABR', '\tASR', 'MORE INFO')
3629
3630 return render_table(
3631 header_line, table, hide_empty=True,
591bb9d3 3632 delim=self._format_out('\u2500', self.Styles.DELIM, '-', test_encoding=True))
8130779d 3633
3634 def render_thumbnails_table(self, info_dict):
88f23a18 3635 thumbnails = list(info_dict.get('thumbnails') or [])
cfb56d1a 3636 if not thumbnails:
8130779d 3637 return None
3638 return render_table(
ec11a9f4 3639 self._list_format_headers('ID', 'Width', 'Height', 'URL'),
6970b600 3640 [[t.get('id'), t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails])
2412044c 3641
8130779d 3642 def render_subtitles_table(self, video_id, subtitles):
2412044c 3643 def _row(lang, formats):
49c258e1 3644 exts, names = zip(*((f['ext'], f.get('name') or 'unknown') for f in reversed(formats)))
2412044c 3645 if len(set(names)) == 1:
7aee40c1 3646 names = [] if names[0] == 'unknown' else names[:1]
2412044c 3647 return [lang, ', '.join(names), ', '.join(exts)]
3648
8130779d 3649 if not subtitles:
3650 return None
3651 return render_table(
ec11a9f4 3652 self._list_format_headers('Language', 'Name', 'Formats'),
2412044c 3653 [_row(lang, formats) for lang, formats in subtitles.items()],
8130779d 3654 hide_empty=True)
3655
3656 def __list_table(self, video_id, name, func, *args):
3657 table = func(*args)
3658 if not table:
3659 self.to_screen(f'{video_id} has no {name}')
3660 return
3661 self.to_screen(f'[info] Available {name} for {video_id}:')
3662 self.to_stdout(table)
3663
3664 def list_formats(self, info_dict):
3665 self.__list_table(info_dict['id'], 'formats', self.render_formats_table, info_dict)
3666
3667 def list_thumbnails(self, info_dict):
3668 self.__list_table(info_dict['id'], 'thumbnails', self.render_thumbnails_table, info_dict)
3669
3670 def list_subtitles(self, video_id, subtitles, name='subtitles'):
3671 self.__list_table(video_id, name, self.render_subtitles_table, video_id, subtitles)
a504ced0 3672
dca08720
PH
3673 def urlopen(self, req):
3674 """ Start an HTTP download """
f9934b96 3675 if isinstance(req, str):
67dda517 3676 req = sanitized_Request(req)
19a41fc6 3677 return self._opener.open(req, timeout=self._socket_timeout)
dca08720
PH
3678
3679 def print_debug_header(self):
3680 if not self.params.get('verbose'):
3681 return
49a57e70 3682
560738f3 3683 # These imports can be slow. So import them only as needed
3684 from .extractor.extractors import _LAZY_LOADER
3685 from .extractor.extractors import _PLUGIN_CLASSES as plugin_extractors
3686
49a57e70 3687 def get_encoding(stream):
2a938746 3688 ret = str(getattr(stream, 'encoding', 'missing (%s)' % type(stream).__name__))
49a57e70 3689 if not supports_terminal_sequences(stream):
53973b4d 3690 from .utils import WINDOWS_VT_MODE # Must be imported locally
e3c7d495 3691 ret += ' (No VT)' if WINDOWS_VT_MODE is False else ' (No ANSI)'
49a57e70 3692 return ret
3693
591bb9d3 3694 encoding_str = 'Encodings: locale %s, fs %s, pref %s, %s' % (
49a57e70 3695 locale.getpreferredencoding(),
3696 sys.getfilesystemencoding(),
591bb9d3 3697 self.get_encoding(),
3698 ', '.join(
64fa820c 3699 f'{key} {get_encoding(stream)}' for key, stream in self._out_files.items_
591bb9d3 3700 if stream is not None and key != 'console')
3701 )
883d4b1e 3702
3703 logger = self.params.get('logger')
3704 if logger:
3705 write_debug = lambda msg: logger.debug(f'[debug] {msg}')
3706 write_debug(encoding_str)
3707 else:
96565c7e 3708 write_string(f'[debug] {encoding_str}\n', encoding=None)
49a57e70 3709 write_debug = lambda msg: self._write_string(f'[debug] {msg}\n')
734f90bb 3710
4c88ff87 3711 source = detect_variant()
70b23409 3712 if VARIANT not in (None, 'pip'):
3713 source += '*'
36eaf303 3714 write_debug(join_nonempty(
b5e7a2e6 3715 f'{"yt-dlp" if REPOSITORY == "yt-dlp/yt-dlp" else REPOSITORY} version',
3716 __version__,
36eaf303 3717 f'[{RELEASE_GIT_HEAD}]' if RELEASE_GIT_HEAD else '',
3718 '' if source == 'unknown' else f'({source})',
3719 delim=' '))
6e21fdd2 3720 if not _LAZY_LOADER:
3721 if os.environ.get('YTDLP_NO_LAZY_EXTRACTORS'):
49a57e70 3722 write_debug('Lazy loading extractors is forcibly disabled')
6e21fdd2 3723 else:
49a57e70 3724 write_debug('Lazy loading extractors is disabled')
3ae5e797 3725 if plugin_extractors or plugin_postprocessors:
49a57e70 3726 write_debug('Plugins: %s' % [
3ae5e797 3727 '%s%s' % (klass.__name__, '' if klass.__name__ == name else f' as {name}')
3728 for name, klass in itertools.chain(plugin_extractors.items(), plugin_postprocessors.items())])
8a82af35 3729 if self.params['compat_opts']:
3730 write_debug('Compatibility options: %s' % ', '.join(self.params['compat_opts']))
36eaf303 3731
b5e7a2e6 3732 if current_git_head():
3733 write_debug(f'Git HEAD: {current_git_head()}')
b1f94422 3734 write_debug(system_identifier())
d28b5171 3735
8913ef74 3736 exe_versions, ffmpeg_features = FFmpegPostProcessor.get_versions_and_features(self)
3737 ffmpeg_features = {key for key, val in ffmpeg_features.items() if val}
3738 if ffmpeg_features:
19a03940 3739 exe_versions['ffmpeg'] += ' (%s)' % ','.join(sorted(ffmpeg_features))
8913ef74 3740
4c83c967 3741 exe_versions['rtmpdump'] = rtmpdump_version()
feee8d32 3742 exe_versions['phantomjs'] = PhantomJSwrapper._version()
d28b5171 3743 exe_str = ', '.join(
2831b468 3744 f'{exe} {v}' for exe, v in sorted(exe_versions.items()) if v
3745 ) or 'none'
49a57e70 3746 write_debug('exe versions: %s' % exe_str)
dca08720 3747
1d485a1a 3748 from .compat.compat_utils import get_package_info
9b8ee23b 3749 from .dependencies import available_dependencies
3750
3751 write_debug('Optional libraries: %s' % (', '.join(sorted({
1d485a1a 3752 join_nonempty(*get_package_info(m)) for m in available_dependencies.values()
9b8ee23b 3753 })) or 'none'))
2831b468 3754
97ec5bc5 3755 self._setup_opener()
dca08720
PH
3756 proxy_map = {}
3757 for handler in self._opener.handlers:
3758 if hasattr(handler, 'proxies'):
3759 proxy_map.update(handler.proxies)
49a57e70 3760 write_debug(f'Proxy map: {proxy_map}')
dca08720 3761
49a57e70 3762 # Not implemented
3763 if False and self.params.get('call_home'):
0f06bcd7 3764 ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode()
49a57e70 3765 write_debug('Public IP address: %s' % ipaddr)
58b1f00d 3766 latest_version = self.urlopen(
0f06bcd7 3767 'https://yt-dl.org/latest/version').read().decode()
58b1f00d
PH
3768 if version_tuple(latest_version) > version_tuple(__version__):
3769 self.report_warning(
3770 'You are using an outdated version (newest version: %s)! '
3771 'See https://yt-dl.org/update if you need help updating.' %
3772 latest_version)
3773
e344693b 3774 def _setup_opener(self):
97ec5bc5 3775 if hasattr(self, '_opener'):
3776 return
6ad14cab 3777 timeout_val = self.params.get('socket_timeout')
17bddf3e 3778 self._socket_timeout = 20 if timeout_val is None else float(timeout_val)
6ad14cab 3779
982ee69a 3780 opts_cookiesfrombrowser = self.params.get('cookiesfrombrowser')
dca08720
PH
3781 opts_cookiefile = self.params.get('cookiefile')
3782 opts_proxy = self.params.get('proxy')
3783
982ee69a 3784 self.cookiejar = load_cookies(opts_cookiefile, opts_cookiesfrombrowser, self)
dca08720 3785
6a3f4c3f 3786 cookie_processor = YoutubeDLCookieProcessor(self.cookiejar)
dca08720
PH
3787 if opts_proxy is not None:
3788 if opts_proxy == '':
3789 proxies = {}
3790 else:
3791 proxies = {'http': opts_proxy, 'https': opts_proxy}
3792 else:
ac668111 3793 proxies = urllib.request.getproxies()
067aa17e 3794 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
dca08720
PH
3795 if 'http' in proxies and 'https' not in proxies:
3796 proxies['https'] = proxies['http']
91410c9b 3797 proxy_handler = PerRequestProxyHandler(proxies)
a0ddb8a2
PH
3798
3799 debuglevel = 1 if self.params.get('debug_printtraffic') else 0
be4a824d
PH
3800 https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel)
3801 ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel)
fca6dba8 3802 redirect_handler = YoutubeDLRedirectHandler()
f9934b96 3803 data_handler = urllib.request.DataHandler()
6240b0a2
JMF
3804
3805 # When passing our own FileHandler instance, build_opener won't add the
3806 # default FileHandler and allows us to disable the file protocol, which
3807 # can be used for malicious purposes (see
067aa17e 3808 # https://github.com/ytdl-org/youtube-dl/issues/8227)
ac668111 3809 file_handler = urllib.request.FileHandler()
6240b0a2
JMF
3810
3811 def file_open(*args, **kwargs):
ac668111 3812 raise urllib.error.URLError('file:// scheme is explicitly disabled in yt-dlp for security reasons')
6240b0a2
JMF
3813 file_handler.file_open = file_open
3814
ac668111 3815 opener = urllib.request.build_opener(
fca6dba8 3816 proxy_handler, https_handler, cookie_processor, ydlh, redirect_handler, data_handler, file_handler)
2461f79d 3817
dca08720
PH
3818 # Delete the default user-agent header, which would otherwise apply in
3819 # cases where our custom HTTP handler doesn't come into play
067aa17e 3820 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
dca08720
PH
3821 opener.addheaders = []
3822 self._opener = opener
62fec3b2
PH
3823
3824 def encode(self, s):
3825 if isinstance(s, bytes):
3826 return s # Already encoded
3827
3828 try:
3829 return s.encode(self.get_encoding())
3830 except UnicodeEncodeError as err:
3831 err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.'
3832 raise
3833
3834 def get_encoding(self):
3835 encoding = self.params.get('encoding')
3836 if encoding is None:
3837 encoding = preferredencoding()
3838 return encoding
ec82d85a 3839
e08a85d8 3840 def _write_info_json(self, label, ie_result, infofn, overwrite=None):
cb96c5be 3841 ''' Write infojson and returns True = written, 'exists' = Already exists, False = skip, None = error '''
e08a85d8 3842 if overwrite is None:
3843 overwrite = self.params.get('overwrites', True)
80c03fa9 3844 if not self.params.get('writeinfojson'):
3845 return False
3846 elif not infofn:
3847 self.write_debug(f'Skipping writing {label} infojson')
3848 return False
3849 elif not self._ensure_dir_exists(infofn):
3850 return None
e08a85d8 3851 elif not overwrite and os.path.exists(infofn):
80c03fa9 3852 self.to_screen(f'[info] {label.title()} metadata is already present')
cb96c5be 3853 return 'exists'
3854
3855 self.to_screen(f'[info] Writing {label} metadata as JSON to: {infofn}')
3856 try:
3857 write_json_file(self.sanitize_info(ie_result, self.params.get('clean_infojson', True)), infofn)
3858 return True
86e5f3ed 3859 except OSError:
cb96c5be 3860 self.report_error(f'Cannot write {label} metadata to JSON file {infofn}')
3861 return None
80c03fa9 3862
3863 def _write_description(self, label, ie_result, descfn):
3864 ''' Write description and returns True = written, False = skip, None = error '''
3865 if not self.params.get('writedescription'):
3866 return False
3867 elif not descfn:
3868 self.write_debug(f'Skipping writing {label} description')
3869 return False
3870 elif not self._ensure_dir_exists(descfn):
3871 return None
3872 elif not self.params.get('overwrites', True) and os.path.exists(descfn):
3873 self.to_screen(f'[info] {label.title()} description is already present')
3874 elif ie_result.get('description') is None:
3875 self.report_warning(f'There\'s no {label} description to write')
3876 return False
3877 else:
3878 try:
3879 self.to_screen(f'[info] Writing {label} description to: {descfn}')
86e5f3ed 3880 with open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
80c03fa9 3881 descfile.write(ie_result['description'])
86e5f3ed 3882 except OSError:
80c03fa9 3883 self.report_error(f'Cannot write {label} description file {descfn}')
3884 return None
3885 return True
3886
3887 def _write_subtitles(self, info_dict, filename):
3888 ''' Write subtitles to file and return list of (sub_filename, final_sub_filename); or None if error'''
3889 ret = []
3890 subtitles = info_dict.get('requested_subtitles')
3891 if not subtitles or not (self.params.get('writesubtitles') or self.params.get('writeautomaticsub')):
3892 # subtitles download errors are already managed as troubles in relevant IE
3893 # that way it will silently go on when used with unsupporting IE
3894 return ret
3895
3896 sub_filename_base = self.prepare_filename(info_dict, 'subtitle')
3897 if not sub_filename_base:
3898 self.to_screen('[info] Skipping writing video subtitles')
3899 return ret
3900 for sub_lang, sub_info in subtitles.items():
3901 sub_format = sub_info['ext']
3902 sub_filename = subtitles_filename(filename, sub_lang, sub_format, info_dict.get('ext'))
3903 sub_filename_final = subtitles_filename(sub_filename_base, sub_lang, sub_format, info_dict.get('ext'))
e04938ab 3904 existing_sub = self.existing_file((sub_filename_final, sub_filename))
3905 if existing_sub:
80c03fa9 3906 self.to_screen(f'[info] Video subtitle {sub_lang}.{sub_format} is already present')
e04938ab 3907 sub_info['filepath'] = existing_sub
3908 ret.append((existing_sub, sub_filename_final))
80c03fa9 3909 continue
3910
3911 self.to_screen(f'[info] Writing video subtitles to: {sub_filename}')
3912 if sub_info.get('data') is not None:
3913 try:
3914 # Use newline='' to prevent conversion of newline characters
3915 # See https://github.com/ytdl-org/youtube-dl/issues/10268
86e5f3ed 3916 with open(sub_filename, 'w', encoding='utf-8', newline='') as subfile:
80c03fa9 3917 subfile.write(sub_info['data'])
3918 sub_info['filepath'] = sub_filename
3919 ret.append((sub_filename, sub_filename_final))
3920 continue
86e5f3ed 3921 except OSError:
80c03fa9 3922 self.report_error(f'Cannot write video subtitles file {sub_filename}')
3923 return None
3924
3925 try:
3926 sub_copy = sub_info.copy()
3927 sub_copy.setdefault('http_headers', info_dict.get('http_headers'))
3928 self.dl(sub_filename, sub_copy, subtitle=True)
3929 sub_info['filepath'] = sub_filename
3930 ret.append((sub_filename, sub_filename_final))
6020e05d 3931 except (DownloadError, ExtractorError, IOError, OSError, ValueError) + network_exceptions as err:
c70c418d 3932 msg = f'Unable to download video subtitles for {sub_lang!r}: {err}'
6020e05d 3933 if self.params.get('ignoreerrors') is not True: # False or 'only_download'
c70c418d 3934 if not self.params.get('ignoreerrors'):
3935 self.report_error(msg)
3936 raise DownloadError(msg)
3937 self.report_warning(msg)
519804a9 3938 return ret
80c03fa9 3939
3940 def _write_thumbnails(self, label, info_dict, filename, thumb_filename_base=None):
3941 ''' Write thumbnails to file and return list of (thumb_filename, final_thumb_filename) '''
6c4fd172 3942 write_all = self.params.get('write_all_thumbnails', False)
80c03fa9 3943 thumbnails, ret = [], []
6c4fd172 3944 if write_all or self.params.get('writethumbnail', False):
0202b52a 3945 thumbnails = info_dict.get('thumbnails') or []
6c4fd172 3946 multiple = write_all and len(thumbnails) > 1
ec82d85a 3947
80c03fa9 3948 if thumb_filename_base is None:
3949 thumb_filename_base = filename
3950 if thumbnails and not thumb_filename_base:
3951 self.write_debug(f'Skipping writing {label} thumbnail')
3952 return ret
3953
dd0228ce 3954 for idx, t in list(enumerate(thumbnails))[::-1]:
80c03fa9 3955 thumb_ext = (f'{t["id"]}.' if multiple else '') + determine_ext(t['url'], 'jpg')
aa9369a2 3956 thumb_display_id = f'{label} thumbnail {t["id"]}'
80c03fa9 3957 thumb_filename = replace_extension(filename, thumb_ext, info_dict.get('ext'))
3958 thumb_filename_final = replace_extension(thumb_filename_base, thumb_ext, info_dict.get('ext'))
ec82d85a 3959
e04938ab 3960 existing_thumb = self.existing_file((thumb_filename_final, thumb_filename))
3961 if existing_thumb:
aa9369a2 3962 self.to_screen('[info] %s is already present' % (
3963 thumb_display_id if multiple else f'{label} thumbnail').capitalize())
e04938ab 3964 t['filepath'] = existing_thumb
3965 ret.append((existing_thumb, thumb_filename_final))
ec82d85a 3966 else:
80c03fa9 3967 self.to_screen(f'[info] Downloading {thumb_display_id} ...')
ec82d85a 3968 try:
297e9952 3969 uf = self.urlopen(sanitized_Request(t['url'], headers=t.get('http_headers', {})))
80c03fa9 3970 self.to_screen(f'[info] Writing {thumb_display_id} to: {thumb_filename}')
d3d89c32 3971 with open(encodeFilename(thumb_filename), 'wb') as thumbf:
ec82d85a 3972 shutil.copyfileobj(uf, thumbf)
80c03fa9 3973 ret.append((thumb_filename, thumb_filename_final))
885cc0b7 3974 t['filepath'] = thumb_filename
3158150c 3975 except network_exceptions as err:
dd0228ce 3976 thumbnails.pop(idx)
80c03fa9 3977 self.report_warning(f'Unable to download {thumb_display_id}: {err}')
6c4fd172 3978 if ret and not write_all:
3979 break
0202b52a 3980 return ret