]> jfr.im git - yt-dlp.git/blame - yt_dlp/YoutubeDL.py
[outtmpl] Curly braces to filter keys
[yt-dlp.git] / yt_dlp / YoutubeDL.py
CommitLineData
26e63931 1import collections
31bd3925 2import contextlib
9d2ecdbc 3import datetime
c1c9a79c 4import errno
31bd3925 5import fileinput
b5ae35ee 6import functools
8222d8de 7import io
b82f815f 8import itertools
8694c600 9import json
62fec3b2 10import locale
083c9df9 11import operator
8222d8de 12import os
f8271158 13import random
8222d8de
JMF
14import re
15import shutil
dca08720 16import subprocess
8222d8de 17import sys
21cd8fae 18import tempfile
8222d8de 19import time
67134eab 20import tokenize
8222d8de 21import traceback
524e2e4f 22import unicodedata
f9934b96 23import urllib.request
961ea474
S
24from string import ascii_letters
25
f8271158 26from .cache import Cache
14f25df2 27from .compat import compat_os_name, compat_shlex_quote
982ee69a 28from .cookies import load_cookies
f8271158 29from .downloader import FFmpegFD, get_suitable_downloader, shorten_protocol_name
30from .downloader.rtmp import rtmpdump_version
f8271158 31from .extractor import gen_extractor_classes, get_info_extractor
fe7866d0 32from .extractor.common import UnsupportedURLIE
f8271158 33from .extractor.openload import PhantomJSwrapper
34from .minicurses import format_text
35from .postprocessor import _PLUGIN_CLASSES as plugin_postprocessors
36from .postprocessor import (
37 EmbedThumbnailPP,
38 FFmpegFixupDuplicateMoovPP,
39 FFmpegFixupDurationPP,
40 FFmpegFixupM3u8PP,
41 FFmpegFixupM4aPP,
42 FFmpegFixupStretchedPP,
43 FFmpegFixupTimestampPP,
44 FFmpegMergerPP,
45 FFmpegPostProcessor,
ca9def71 46 FFmpegVideoConvertorPP,
f8271158 47 MoveFilesAfterDownloadPP,
48 get_postprocessor,
49)
ca9def71 50from .postprocessor.ffmpeg import resolve_mapping as resolve_recode_mapping
b5e7a2e6 51from .update import REPOSITORY, current_git_head, detect_variant
8c25f81b 52from .utils import (
f8271158 53 DEFAULT_OUTTMPL,
7b2c3f47 54 IDENTITY,
f8271158 55 LINK_TEMPLATES,
8dc59305 56 MEDIA_EXTENSIONS,
f8271158 57 NO_DEFAULT,
1d485a1a 58 NUMBER_RE,
f8271158 59 OUTTMPL_TYPES,
60 POSTPROCESS_WHEN,
61 STR_FORMAT_RE_TMPL,
62 STR_FORMAT_TYPES,
63 ContentTooShortError,
64 DateRange,
65 DownloadCancelled,
66 DownloadError,
67 EntryNotInPlaylist,
68 ExistingVideoReached,
69 ExtractorError,
70 GeoRestrictedError,
71 HEADRequest,
f8271158 72 ISO3166Utils,
73 LazyList,
74 MaxDownloadsReached,
19a03940 75 Namespace,
f8271158 76 PagedList,
77 PerRequestProxyHandler,
7e88d7d7 78 PlaylistEntries,
f8271158 79 Popen,
80 PostProcessingError,
81 ReExtractInfo,
82 RejectedVideoReached,
83 SameFileError,
84 UnavailableVideoError,
693f0600 85 UserNotLive,
f8271158 86 YoutubeDLCookieProcessor,
87 YoutubeDLHandler,
88 YoutubeDLRedirectHandler,
eedb7ba5
S
89 age_restricted,
90 args_to_str,
cb794ee0 91 bug_reports_message,
ce02ed60 92 date_from_str,
da4db748 93 deprecation_warning,
ce02ed60 94 determine_ext,
b5559424 95 determine_protocol,
c0384f22 96 encode_compat_str,
ce02ed60 97 encodeFilename,
a06916d9 98 error_to_compat_str,
47cdc68e 99 escapeHTML,
590bc6f6 100 expand_path,
90137ca4 101 filter_dict,
e29663c6 102 float_or_none,
02dbf93f 103 format_bytes,
e0fd9573 104 format_decimal_suffix,
f8271158 105 format_field,
525ef922 106 formatSeconds,
fc61aff4 107 get_compatible_ext,
0bb322b9 108 get_domain,
c9969434 109 int_or_none,
732044af 110 iri_to_uri,
34921b43 111 join_nonempty,
ce02ed60 112 locked_file,
0647d925 113 make_archive_id,
0202b52a 114 make_dir,
dca08720 115 make_HTTPS_handler,
8b7539d2 116 merge_headers,
3158150c 117 network_exceptions,
ec11a9f4 118 number_of_digits,
cd6fc19e 119 orderedSet,
5314b521 120 orderedSet_from_options,
083c9df9 121 parse_filesize,
ce02ed60 122 preferredencoding,
eedb7ba5 123 prepend_extension,
51fb4995 124 register_socks_protocols,
3efb96a6 125 remove_terminal_sequences,
cfb56d1a 126 render_table,
eedb7ba5 127 replace_extension,
ce02ed60 128 sanitize_filename,
1bb5c511 129 sanitize_path,
dcf77cf1 130 sanitize_url,
67dda517 131 sanitized_Request,
e5660ee6 132 std_headers,
1211bb6d 133 str_or_none,
e29663c6 134 strftime_or_none,
ce02ed60 135 subtitles_filename,
819e0531 136 supports_terminal_sequences,
b1f94422 137 system_identifier,
f2ebc5c7 138 timetuple_from_msec,
732044af 139 to_high_limit_path,
324ad820 140 traverse_obj,
fc61aff4 141 try_call,
6033d980 142 try_get,
29eb5174 143 url_basename,
7d1eb38a 144 variadic,
58b1f00d 145 version_tuple,
53973b4d 146 windows_enable_vt_mode,
ce02ed60
PH
147 write_json_file,
148 write_string,
4f026faf 149)
70b23409 150from .version import RELEASE_GIT_HEAD, VARIANT, __version__
8222d8de 151
e9c0cdd3
YCH
152if compat_os_name == 'nt':
153 import ctypes
154
2459b6e1 155
86e5f3ed 156class YoutubeDL:
8222d8de
JMF
157 """YoutubeDL class.
158
159 YoutubeDL objects are the ones responsible of downloading the
160 actual video file and writing it to disk if the user has requested
161 it, among some other tasks. In most cases there should be one per
162 program. As, given a video URL, the downloader doesn't know how to
163 extract all the needed information, task that InfoExtractors do, it
164 has to pass the URL to one of them.
165
166 For this, YoutubeDL objects have a method that allows
167 InfoExtractors to be registered in a given order. When it is passed
168 a URL, the YoutubeDL object handles it to the first InfoExtractor it
169 finds that reports being able to handle it. The InfoExtractor extracts
170 all the information about the video or videos the URL refers to, and
171 YoutubeDL process the extracted information, possibly using a File
172 Downloader to download the video.
173
174 YoutubeDL objects accept a lot of parameters. In order not to saturate
175 the object constructor with arguments, it receives a dictionary of
176 options instead. These options are available through the params
177 attribute for the InfoExtractors to use. The YoutubeDL also
178 registers itself as the downloader in charge for the InfoExtractors
179 that are added to it, so this is a "mutual registration".
180
181 Available options:
182
183 username: Username for authentication purposes.
184 password: Password for authentication purposes.
180940e0 185 videopassword: Password for accessing a video.
1da50aa3
S
186 ap_mso: Adobe Pass multiple-system operator identifier.
187 ap_username: Multiple-system operator account username.
188 ap_password: Multiple-system operator account password.
8222d8de
JMF
189 usenetrc: Use netrc for authentication instead.
190 verbose: Print additional info to stdout.
191 quiet: Do not print messages to stdout.
ad8915b7 192 no_warnings: Do not print out anything for warnings.
bb66c247 193 forceprint: A dict with keys WHEN mapped to a list of templates to
194 print to stdout. The allowed keys are video or any of the
195 items in utils.POSTPROCESS_WHEN.
ca30f449 196 For compatibility, a single list is also accepted
bb66c247 197 print_to_file: A dict with keys WHEN (same as forceprint) mapped to
198 a list of tuples with (template, filename)
8694c600 199 forcejson: Force printing info_dict as JSON.
63e0be34
PH
200 dump_single_json: Force printing the info_dict of the whole playlist
201 (or video) as a single JSON line.
c25228e5 202 force_write_download_archive: Force writing download archive regardless
203 of 'skip_download' or 'simulate'.
b7b04c78 204 simulate: Do not download the video files. If unset (or None),
205 simulate only if listsubtitles, listformats or list_thumbnails is used
eb8a4433 206 format: Video format code. see "FORMAT SELECTION" for more details.
093a1710 207 You can also pass a function. The function takes 'ctx' as
208 argument and returns the formats to download.
209 See "build_format_selector" for an implementation
63ad4d43 210 allow_unplayable_formats: Allow unplayable formats to be extracted and downloaded.
b7da73eb 211 ignore_no_formats_error: Ignore "No video formats" error. Usefull for
212 extracting metadata even if the video is not actually
213 available for download (experimental)
0930b11f 214 format_sort: A list of fields by which to sort the video formats.
215 See "Sorting Formats" for more details.
c25228e5 216 format_sort_force: Force the given format_sort. see "Sorting Formats"
217 for more details.
08d30158 218 prefer_free_formats: Whether to prefer video formats with free containers
219 over non-free ones of same quality.
c25228e5 220 allow_multiple_video_streams: Allow multiple video streams to be merged
221 into a single file
222 allow_multiple_audio_streams: Allow multiple audio streams to be merged
223 into a single file
0ba692ac 224 check_formats Whether to test if the formats are downloadable.
9f1a1c36 225 Can be True (check all), False (check none),
226 'selected' (check selected formats),
0ba692ac 227 or None (check only if requested by extractor)
4524baf0 228 paths: Dictionary of output paths. The allowed keys are 'home'
229 'temp' and the keys of OUTTMPL_TYPES (in utils.py)
de6000d9 230 outtmpl: Dictionary of templates for output names. Allowed keys
4524baf0 231 are 'default' and the keys of OUTTMPL_TYPES (in utils.py).
34488702 232 For compatibility with youtube-dl, a single string can also be used
a820dc72
RA
233 outtmpl_na_placeholder: Placeholder for unavailable meta fields.
234 restrictfilenames: Do not allow "&" and spaces in file names
235 trim_file_name: Limit length of filename (extension excluded)
4524baf0 236 windowsfilenames: Force the filenames to be windows compatible
b1940459 237 ignoreerrors: Do not stop on download/postprocessing errors.
238 Can be 'only_download' to ignore only download errors.
239 Default is 'only_download' for CLI, but False for API
26e2805c 240 skip_playlist_after_errors: Number of allowed failures until the rest of
241 the playlist is skipped
fe7866d0 242 allowed_extractors: List of regexes to match against extractor names that are allowed
0c3d0f51 243 overwrites: Overwrite all video and metadata files if True,
244 overwrite only non-video files if None
245 and don't overwrite any file if False
34488702 246 For compatibility with youtube-dl,
247 "nooverwrites" may also be used instead
c14e88f0 248 playlist_items: Specific indices of playlist to download.
75822ca7 249 playlistrandom: Download playlist items in random order.
7e9a6125 250 lazy_playlist: Process playlist entries as they are received.
8222d8de
JMF
251 matchtitle: Download only matching titles.
252 rejecttitle: Reject downloads for matching titles.
8bf9319e 253 logger: Log messages to a logging.Logger instance.
8222d8de 254 logtostderr: Log messages to stderr instead of stdout.
819e0531 255 consoletitle: Display progress in console window's titlebar.
8222d8de
JMF
256 writedescription: Write the video description to a .description file
257 writeinfojson: Write the video description to a .info.json file
75d43ca0 258 clean_infojson: Remove private fields from the infojson
34488702 259 getcomments: Extract video comments. This will not be written to disk
06167fbb 260 unless writeinfojson is also given
1fb07d10 261 writeannotations: Write the video annotations to a .annotations.xml file
8222d8de 262 writethumbnail: Write the thumbnail image to a file
c25228e5 263 allow_playlist_files: Whether to write playlists' description, infojson etc
264 also to disk when using the 'write*' options
ec82d85a 265 write_all_thumbnails: Write all thumbnail formats to files
732044af 266 writelink: Write an internet shortcut file, depending on the
267 current platform (.url/.webloc/.desktop)
268 writeurllink: Write a Windows internet shortcut file (.url)
269 writewebloclink: Write a macOS internet shortcut file (.webloc)
270 writedesktoplink: Write a Linux internet shortcut file (.desktop)
8222d8de 271 writesubtitles: Write the video subtitles to a file
741dd8ea 272 writeautomaticsub: Write the automatically generated subtitles to a file
8222d8de 273 listsubtitles: Lists all available subtitles for the video
a504ced0 274 subtitlesformat: The format code for subtitles
c32b0aab 275 subtitleslangs: List of languages of the subtitles to download (can be regex).
276 The list may contain "all" to refer to all the available
277 subtitles. The language can be prefixed with a "-" to
62b58c09 278 exclude it from the requested languages, e.g. ['all', '-live_chat']
8222d8de
JMF
279 keepvideo: Keep the video file after post-processing
280 daterange: A DateRange object, download only if the upload_date is in the range.
281 skip_download: Skip the actual download of the video file
c35f9e72 282 cachedir: Location of the cache files in the filesystem.
a0e07d31 283 False to disable filesystem cache.
47192f92 284 noplaylist: Download single video instead of a playlist if in doubt.
8dbe9899
PH
285 age_limit: An integer representing the user's age in years.
286 Unsuitable videos for the given age are skipped.
5fe18bdb
PH
287 min_views: An integer representing the minimum view count the video
288 must have in order to not be skipped.
289 Videos without view count information are always
290 downloaded. None for no limit.
291 max_views: An integer representing the maximum view count.
292 Videos that are more popular than that are not
293 downloaded.
294 Videos without view count information are always
295 downloaded. None for no limit.
296 download_archive: File name of a file where all downloads are recorded.
c1c9a79c
PH
297 Videos already present in the file are not downloaded
298 again.
8a51f564 299 break_on_existing: Stop the download process after attempting to download a
300 file that is in the archive.
301 break_on_reject: Stop the download process when encountering a video that
302 has been filtered out.
b222c271 303 break_per_url: Whether break_on_reject and break_on_existing
304 should act on each input URL as opposed to for the entire queue
d76fa1f3 305 cookiefile: File name or text stream from where cookies should be read and dumped to
f59f5ef8 306 cookiesfrombrowser: A tuple containing the name of the browser, the profile
9bd13fe5 307 name/path from where cookies are loaded, the name of the keyring,
308 and the container name, e.g. ('chrome', ) or
309 ('vivaldi', 'default', 'BASICTEXT') or ('firefox', 'default', None, 'Meta')
f81c62a6 310 legacyserverconnect: Explicitly allow HTTPS connection to servers that do not
311 support RFC 5746 secure renegotiation
f59f5ef8 312 nocheckcertificate: Do not verify SSL certificates
bb58c9ed 313 client_certificate: Path to client certificate file in PEM format. May include the private key
314 client_certificate_key: Path to private key file for client certificate
315 client_certificate_password: Password for client certificate private key, if encrypted.
316 If not provided and the key is encrypted, yt-dlp will ask interactively
7e8c0af0 317 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
c6e07cf1 318 (Only supported by some extractors)
8b7539d2 319 http_headers: A dictionary of custom headers to be used for all requests
a1ee09e8 320 proxy: URL of the proxy server to use
38cce791 321 geo_verification_proxy: URL of the proxy to use for IP address verification
504f20dd 322 on geo-restricted sites.
e344693b 323 socket_timeout: Time to wait for unresponsive hosts, in seconds
0783b09b
PH
324 bidi_workaround: Work around buggy terminals without bidirectional text
325 support, using fridibi
a0ddb8a2 326 debug_printtraffic:Print out sent and received HTTP traffic
04b4d394
PH
327 default_search: Prepend this string if an input url is not valid.
328 'auto' for elaborate guessing
62fec3b2 329 encoding: Use this encoding instead of the system-specified.
134c913c 330 extract_flat: Whether to resolve and process url_results further
331 * False: Always process (default)
332 * True: Never process
333 * 'in_playlist': Do not process inside playlist/multi_video
334 * 'discard': Always process, but don't return the result
335 from inside playlist/multi_video
336 * 'discard_in_playlist': Same as "discard", but only for
337 playlists (not multi_video)
f2ebc5c7 338 wait_for_video: If given, wait for scheduled streams to become available.
339 The value should be a tuple containing the range
340 (min_secs, max_secs) to wait between retries
4f026faf 341 postprocessors: A list of dictionaries, each with an entry
71b640cc 342 * key: The name of the postprocessor. See
7a5c1cfe 343 yt_dlp/postprocessor/__init__.py for a list.
bb66c247 344 * when: When to run the postprocessor. Allowed values are
345 the entries of utils.POSTPROCESS_WHEN
56d868db 346 Assumed to be 'post_process' if not given
71b640cc
PH
347 progress_hooks: A list of functions that get called on download
348 progress, with a dictionary with the entries
5cda4eda 349 * status: One of "downloading", "error", or "finished".
ee69b99a 350 Check this first and ignore unknown values.
3ba7740d 351 * info_dict: The extracted info_dict
71b640cc 352
5cda4eda 353 If status is one of "downloading", or "finished", the
ee69b99a
PH
354 following properties may also be present:
355 * filename: The final filename (always present)
5cda4eda 356 * tmpfilename: The filename we're currently writing to
71b640cc
PH
357 * downloaded_bytes: Bytes on disk
358 * total_bytes: Size of the whole file, None if unknown
5cda4eda
PH
359 * total_bytes_estimate: Guess of the eventual file size,
360 None if unavailable.
361 * elapsed: The number of seconds since download started.
71b640cc
PH
362 * eta: The estimated time in seconds, None if unknown
363 * speed: The download speed in bytes/second, None if
364 unknown
5cda4eda
PH
365 * fragment_index: The counter of the currently
366 downloaded video fragment.
367 * fragment_count: The number of fragments (= individual
368 files that will be merged)
71b640cc
PH
369
370 Progress hooks are guaranteed to be called at least once
371 (with status "finished") if the download is successful.
819e0531 372 postprocessor_hooks: A list of functions that get called on postprocessing
373 progress, with a dictionary with the entries
374 * status: One of "started", "processing", or "finished".
375 Check this first and ignore unknown values.
376 * postprocessor: Name of the postprocessor
377 * info_dict: The extracted info_dict
378
379 Progress hooks are guaranteed to be called at least twice
380 (with status "started" and "finished") if the processing is successful.
fc61aff4 381 merge_output_format: "/" separated list of extensions to use when merging formats.
6b591b29 382 final_ext: Expected final extension; used to detect when the file was
59a7a13e 383 already downloaded and converted
6271f1ca
PH
384 fixup: Automatically correct known faults of the file.
385 One of:
386 - "never": do nothing
387 - "warn": only emit a warning
388 - "detect_or_warn": check whether we can do anything
62cd676c 389 about it, warn otherwise (default)
504f20dd 390 source_address: Client-side IP address to bind to.
1cf376f5 391 sleep_interval_requests: Number of seconds to sleep between requests
392 during extraction
7aa589a5
S
393 sleep_interval: Number of seconds to sleep before each download when
394 used alone or a lower bound of a range for randomized
395 sleep before each download (minimum possible number
396 of seconds to sleep) when used along with
397 max_sleep_interval.
398 max_sleep_interval:Upper bound of a range for randomized sleep before each
399 download (maximum possible number of seconds to sleep).
400 Must only be used along with sleep_interval.
401 Actual sleep time will be a random float from range
402 [sleep_interval; max_sleep_interval].
1cf376f5 403 sleep_interval_subtitles: Number of seconds to sleep before each subtitle download
cfb56d1a
PH
404 listformats: Print an overview of available video formats and exit.
405 list_thumbnails: Print a table of all thumbnails and exit.
0a41f331 406 match_filter: A function that gets called for every video with the signature
407 (info_dict, *, incomplete: bool) -> Optional[str]
408 For backward compatibility with youtube-dl, the signature
409 (info_dict) -> Optional[str] is also allowed.
410 - If it returns a message, the video is ignored.
411 - If it returns None, the video is downloaded.
412 - If it returns utils.NO_DEFAULT, the user is interactively
413 asked whether to download the video.
347de493 414 match_filter_func in utils.py is one example for this.
7e5db8c9 415 no_color: Do not emit color codes in output.
0a840f58 416 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
504f20dd 417 HTTP header
0a840f58 418 geo_bypass_country:
773f291d
S
419 Two-letter ISO 3166-2 country code that will be used for
420 explicit geographic restriction bypassing via faking
504f20dd 421 X-Forwarded-For HTTP header
5f95927a
S
422 geo_bypass_ip_block:
423 IP range in CIDR notation that will be used similarly to
504f20dd 424 geo_bypass_country
52a8a1e1 425 external_downloader: A dictionary of protocol keys and the executable of the
426 external downloader to use for it. The allowed protocols
427 are default|http|ftp|m3u8|dash|rtsp|rtmp|mms.
428 Set the value to 'native' to use the native downloader
53ed7066 429 compat_opts: Compatibility options. See "Differences in default behavior".
3acf6d38 430 The following options do not work when used through the API:
b5ae35ee 431 filename, abort-on-error, multistreams, no-live-chat, format-sort
dac5df5a 432 no-clean-infojson, no-playlist-metafiles, no-keep-subs, no-attach-info-json.
e4f02757 433 Refer __init__.py for their implementation
819e0531 434 progress_template: Dictionary of templates for progress outputs.
435 Allowed keys are 'download', 'postprocess',
436 'download-title' (console title) and 'postprocess-title'.
437 The template is mapped on a dictionary with keys 'progress' and 'info'
23326151 438 retry_sleep_functions: Dictionary of functions that takes the number of attempts
439 as argument and returns the time to sleep in seconds.
440 Allowed keys are 'http', 'fragment', 'file_access'
0f446365
SW
441 download_ranges: A callback function that gets called for every video with
442 the signature (info_dict, ydl) -> Iterable[Section].
443 Only the returned sections will be downloaded.
444 Each Section is a dict with the following keys:
5ec1b6b7 445 * start_time: Start time of the section in seconds
446 * end_time: End time of the section in seconds
447 * title: Section title (Optional)
448 * index: Section number (Optional)
0f446365 449 force_keyframes_at_cuts: Re-encode the video when downloading ranges to get precise cuts
a7dc6a89 450 noprogress: Do not print the progress bar
a831c2ea 451 live_from_start: Whether to download livestreams videos from the start
fe7e0c98 452
8222d8de 453 The following parameters are not used by YoutubeDL itself, they are used by
7a5c1cfe 454 the downloader (see yt_dlp/downloader/common.py):
51d9739f 455 nopart, updatetime, buffersize, ratelimit, throttledratelimit, min_filesize,
205a0654 456 max_filesize, test, noresizebuffer, retries, file_access_retries, fragment_retries,
a7dc6a89 457 continuedl, xattr_set_filesize, hls_use_mpegts, http_chunk_size,
59a7a13e 458 external_downloader_args, concurrent_fragment_downloads.
76b1bd67
JMF
459
460 The following options are used by the post processors:
c0b7d117
S
461 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
462 to the binary or its containing directory.
43820c03 463 postprocessor_args: A dictionary of postprocessor/executable keys (in lower case)
34488702 464 and a list of additional command-line arguments for the
465 postprocessor/executable. The dict can also have "PP+EXE" keys
466 which are used when the given exe is used by the given PP.
467 Use 'default' as the name for arguments to passed to all PP
468 For compatibility with youtube-dl, a single list of args
469 can also be used
e409895f 470
471 The following options are used by the extractors:
62bff2c1 472 extractor_retries: Number of times to retry for known errors
473 dynamic_mpd: Whether to process dynamic DASH manifests (default: True)
e409895f 474 hls_split_discontinuity: Split HLS playlists to different formats at
62bff2c1 475 discontinuities such as ad breaks (default: False)
5d3a0e79 476 extractor_args: A dictionary of arguments to be passed to the extractors.
477 See "EXTRACTOR ARGUMENTS" for details.
62b58c09 478 E.g. {'youtube': {'skip': ['dash', 'hls']}}
88f23a18 479 mark_watched: Mark videos watched (even with --simulate). Only for YouTube
1890fc63 480
481 The following options are deprecated and may be removed in the future:
482
fe7866d0 483 force_generic_extractor: Force downloader to use the generic extractor
484 - Use allowed_extractors = ['generic', 'default']
7e9a6125 485 playliststart: - Use playlist_items
486 Playlist item to start at.
487 playlistend: - Use playlist_items
488 Playlist item to end at.
489 playlistreverse: - Use playlist_items
490 Download playlist items in reverse order.
1890fc63 491 forceurl: - Use forceprint
492 Force printing final URL.
493 forcetitle: - Use forceprint
494 Force printing title.
495 forceid: - Use forceprint
496 Force printing ID.
497 forcethumbnail: - Use forceprint
498 Force printing thumbnail URL.
499 forcedescription: - Use forceprint
500 Force printing description.
501 forcefilename: - Use forceprint
502 Force printing final filename.
503 forceduration: - Use forceprint
504 Force printing duration.
505 allsubtitles: - Use subtitleslangs = ['all']
506 Downloads all the subtitles of the video
507 (requires writesubtitles or writeautomaticsub)
508 include_ads: - Doesn't work
509 Download ads as well
510 call_home: - Not implemented
511 Boolean, true iff we are allowed to contact the
512 yt-dlp servers for debugging.
513 post_hooks: - Register a custom postprocessor
514 A list of functions that get called as the final step
515 for each video file, after all postprocessors have been
516 called. The filename will be passed as the only argument.
517 hls_prefer_native: - Use external_downloader = {'m3u8': 'native'} or {'m3u8': 'ffmpeg'}.
518 Use the native HLS downloader instead of ffmpeg/avconv
519 if True, otherwise use ffmpeg/avconv if False, otherwise
520 use downloader suggested by extractor if None.
521 prefer_ffmpeg: - avconv support is deprecated
522 If False, use avconv instead of ffmpeg if both are available,
523 otherwise prefer ffmpeg.
524 youtube_include_dash_manifest: - Use extractor_args
5d3a0e79 525 If True (default), DASH manifests and related
62bff2c1 526 data will be downloaded and processed by extractor.
527 You can reduce network I/O by disabling it if you don't
528 care about DASH. (only for youtube)
1890fc63 529 youtube_include_hls_manifest: - Use extractor_args
5d3a0e79 530 If True (default), HLS manifests and related
62bff2c1 531 data will be downloaded and processed by extractor.
532 You can reduce network I/O by disabling it if you don't
533 care about HLS. (only for youtube)
8222d8de
JMF
534 """
535
86e5f3ed 536 _NUMERIC_FIELDS = {
b8ed0f15 537 'width', 'height', 'asr', 'audio_channels', 'fps',
538 'tbr', 'abr', 'vbr', 'filesize', 'filesize_approx',
e6f21b3d 539 'timestamp', 'release_timestamp',
c9969434
S
540 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
541 'average_rating', 'comment_count', 'age_limit',
542 'start_time', 'end_time',
543 'chapter_number', 'season_number', 'episode_number',
544 'track_number', 'disc_number', 'release_year',
86e5f3ed 545 }
c9969434 546
6db9c4d5 547 _format_fields = {
548 # NB: Keep in sync with the docstring of extractor/common.py
a44ca5a4 549 'url', 'manifest_url', 'manifest_stream_number', 'ext', 'format', 'format_id', 'format_note',
b8ed0f15 550 'width', 'height', 'resolution', 'dynamic_range', 'tbr', 'abr', 'acodec', 'asr', 'audio_channels',
6db9c4d5 551 'vbr', 'fps', 'vcodec', 'container', 'filesize', 'filesize_approx',
552 'player_url', 'protocol', 'fragment_base_url', 'fragments', 'is_from_start',
553 'preference', 'language', 'language_preference', 'quality', 'source_preference',
554 'http_headers', 'stretched_ratio', 'no_resume', 'has_drm', 'downloader_options',
555 'page_url', 'app', 'play_path', 'tc_url', 'flash_version', 'rtmp_live', 'rtmp_conn', 'rtmp_protocol', 'rtmp_real_time'
556 }
48ee10ee 557 _format_selection_exts = {
8dc59305 558 'audio': set(MEDIA_EXTENSIONS.common_audio),
559 'video': set(MEDIA_EXTENSIONS.common_video + ('3gp', )),
560 'storyboards': set(MEDIA_EXTENSIONS.storyboards),
48ee10ee 561 }
562
3511266b 563 def __init__(self, params=None, auto_init=True):
883d4b1e 564 """Create a FileDownloader object with the given options.
565 @param auto_init Whether to load the default extractors and print header (if verbose).
49a57e70 566 Set to 'no_verbose_header' to not print the header
883d4b1e 567 """
e9f9a10f
JMF
568 if params is None:
569 params = {}
592b7485 570 self.params = params
8b7491c8 571 self._ies = {}
56c73665 572 self._ies_instances = {}
1e43a6f7 573 self._pps = {k: [] for k in POSTPROCESS_WHEN}
b35496d8 574 self._printed_messages = set()
1cf376f5 575 self._first_webpage_request = True
ab8e5e51 576 self._post_hooks = []
933605d7 577 self._progress_hooks = []
819e0531 578 self._postprocessor_hooks = []
8222d8de
JMF
579 self._download_retcode = 0
580 self._num_downloads = 0
9c906919 581 self._num_videos = 0
592b7485 582 self._playlist_level = 0
583 self._playlist_urls = set()
a0e07d31 584 self.cache = Cache(self)
34308b30 585
819e0531 586 windows_enable_vt_mode()
591bb9d3 587 stdout = sys.stderr if self.params.get('logtostderr') else sys.stdout
588 self._out_files = Namespace(
589 out=stdout,
590 error=sys.stderr,
591 screen=sys.stderr if self.params.get('quiet') else stdout,
592 console=None if compat_os_name == 'nt' else next(
cf4f42cb 593 filter(supports_terminal_sequences, (sys.stderr, sys.stdout)), None)
591bb9d3 594 )
595 self._allow_colors = Namespace(**{
596 type_: not self.params.get('no_color') and supports_terminal_sequences(stream)
64fa820c 597 for type_, stream in self._out_files.items_ if type_ != 'console'
591bb9d3 598 })
819e0531 599
6929b41a 600 # The code is left like this to be reused for future deprecations
601 MIN_SUPPORTED, MIN_RECOMMENDED = (3, 7), (3, 7)
eff42759 602 current_version = sys.version_info[:2]
603 if current_version < MIN_RECOMMENDED:
9d339c41 604 msg = ('Support for Python version %d.%d has been deprecated. '
24093d52 605 'See https://github.com/yt-dlp/yt-dlp/issues/3764 for more details.'
c6e07cf1 606 '\n You will no longer receive updates on this version')
eff42759 607 if current_version < MIN_SUPPORTED:
608 msg = 'Python version %d.%d is no longer supported'
609 self.deprecation_warning(
610 f'{msg}! Please update to Python %d.%d or above' % (*current_version, *MIN_RECOMMENDED))
a61f4b28 611
88acdbc2 612 if self.params.get('allow_unplayable_formats'):
613 self.report_warning(
ec11a9f4 614 f'You have asked for {self._format_err("UNPLAYABLE", self.Styles.EMPHASIS)} formats to be listed/downloaded. '
819e0531 615 'This is a developer option intended for debugging. \n'
616 ' If you experience any issues while using this option, '
ec11a9f4 617 f'{self._format_err("DO NOT", self.Styles.ERROR)} open a bug report')
88acdbc2 618
be5df5ee
S
619 def check_deprecated(param, option, suggestion):
620 if self.params.get(param) is not None:
86e5f3ed 621 self.report_warning(f'{option} is deprecated. Use {suggestion} instead')
be5df5ee
S
622 return True
623 return False
624
625 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
38cce791
YCH
626 if self.params.get('geo_verification_proxy') is None:
627 self.params['geo_verification_proxy'] = self.params['cn_verification_proxy']
628
0d1bb027 629 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
630 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
53ed7066 631 check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"')
0d1bb027 632
49a57e70 633 for msg in self.params.get('_warnings', []):
0d1bb027 634 self.report_warning(msg)
ee8dd27a 635 for msg in self.params.get('_deprecation_warnings', []):
da4db748 636 self.deprecated_feature(msg)
0d1bb027 637
8a82af35 638 self.params['compat_opts'] = set(self.params.get('compat_opts', ()))
8a82af35 639 if 'list-formats' in self.params['compat_opts']:
ec11a9f4 640 self.params['listformats_table'] = False
641
b5ae35ee 642 if 'overwrites' not in self.params and self.params.get('nooverwrites') is not None:
b868936c 643 # nooverwrites was unnecessarily changed to overwrites
644 # in 0c3d0f51778b153f65c21906031c2e091fcfb641
645 # This ensures compatibility with both keys
646 self.params['overwrites'] = not self.params['nooverwrites']
b5ae35ee 647 elif self.params.get('overwrites') is None:
648 self.params.pop('overwrites', None)
b868936c 649 else:
650 self.params['nooverwrites'] = not self.params['overwrites']
b9d973be 651
455a15e2 652 self.params.setdefault('forceprint', {})
653 self.params.setdefault('print_to_file', {})
bb66c247 654
655 # Compatibility with older syntax
ca30f449 656 if not isinstance(params['forceprint'], dict):
455a15e2 657 self.params['forceprint'] = {'video': params['forceprint']}
ca30f449 658
455a15e2 659 if self.params.get('bidi_workaround', False):
1c088fa8
PH
660 try:
661 import pty
662 master, slave = pty.openpty()
ac668111 663 width = shutil.get_terminal_size().columns
591bb9d3 664 width_args = [] if width is None else ['-w', str(width)]
665 sp_kwargs = {'stdin': subprocess.PIPE, 'stdout': slave, 'stderr': self._out_files.error}
5d681e96 666 try:
d3c93ec2 667 self._output_process = Popen(['bidiv'] + width_args, **sp_kwargs)
5d681e96 668 except OSError:
d3c93ec2 669 self._output_process = Popen(['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
5d681e96 670 self._output_channel = os.fdopen(master, 'rb')
1c088fa8 671 except OSError as ose:
66e7ace1 672 if ose.errno == errno.ENOENT:
49a57e70 673 self.report_warning(
674 'Could not find fribidi executable, ignoring --bidi-workaround. '
675 'Make sure that fribidi is an executable file in one of the directories in your $PATH.')
1c088fa8
PH
676 else:
677 raise
0783b09b 678
97ec5bc5 679 if auto_init:
680 if auto_init != 'no_verbose_header':
681 self.print_debug_header()
682 self.add_default_info_extractors()
683
3089bc74
S
684 if (sys.platform != 'win32'
685 and sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
455a15e2 686 and not self.params.get('restrictfilenames', False)):
e9137224 687 # Unicode filesystem API will throw errors (#1474, #13027)
34308b30 688 self.report_warning(
6febd1c1 689 'Assuming --restrict-filenames since file system encoding '
1b725173 690 'cannot encode all characters. '
6febd1c1 691 'Set the LC_ALL environment variable to fix this.')
4a98cdbf 692 self.params['restrictfilenames'] = True
34308b30 693
bf1824b3 694 self._parse_outtmpl()
486dd09e 695
187986a8 696 # Creating format selector here allows us to catch syntax errors before the extraction
697 self.format_selector = (
fa9f30b8 698 self.params.get('format') if self.params.get('format') in (None, '-')
093a1710 699 else self.params['format'] if callable(self.params['format'])
187986a8 700 else self.build_format_selector(self.params['format']))
701
8b7539d2 702 # Set http_headers defaults according to std_headers
703 self.params['http_headers'] = merge_headers(std_headers, self.params.get('http_headers', {}))
704
013b50b7 705 hooks = {
706 'post_hooks': self.add_post_hook,
707 'progress_hooks': self.add_progress_hook,
708 'postprocessor_hooks': self.add_postprocessor_hook,
709 }
710 for opt, fn in hooks.items():
711 for ph in self.params.get(opt, []):
712 fn(ph)
71b640cc 713
5bfc8bee 714 for pp_def_raw in self.params.get('postprocessors', []):
715 pp_def = dict(pp_def_raw)
716 when = pp_def.pop('when', 'post_process')
717 self.add_post_processor(
f9934b96 718 get_postprocessor(pp_def.pop('key'))(self, **pp_def),
5bfc8bee 719 when=when)
720
97ec5bc5 721 self._setup_opener()
51fb4995
YCH
722 register_socks_protocols()
723
ed39cac5 724 def preload_download_archive(fn):
725 """Preload the archive, if any is specified"""
726 if fn is None:
727 return False
49a57e70 728 self.write_debug(f'Loading archive file {fn!r}')
ed39cac5 729 try:
730 with locked_file(fn, 'r', encoding='utf-8') as archive_file:
731 for line in archive_file:
732 self.archive.add(line.strip())
86e5f3ed 733 except OSError as ioe:
ed39cac5 734 if ioe.errno != errno.ENOENT:
735 raise
736 return False
737 return True
738
739 self.archive = set()
740 preload_download_archive(self.params.get('download_archive'))
741
7d4111ed
PH
742 def warn_if_short_id(self, argv):
743 # short YouTube ID starting with dash?
744 idxs = [
745 i for i, a in enumerate(argv)
746 if re.match(r'^-[0-9A-Za-z_-]{10}$', a)]
747 if idxs:
748 correct_argv = (
7a5c1cfe 749 ['yt-dlp']
3089bc74
S
750 + [a for i, a in enumerate(argv) if i not in idxs]
751 + ['--'] + [argv[i] for i in idxs]
7d4111ed
PH
752 )
753 self.report_warning(
754 'Long argument string detected. '
49a57e70 755 'Use -- to separate parameters and URLs, like this:\n%s' %
7d4111ed
PH
756 args_to_str(correct_argv))
757
8222d8de
JMF
758 def add_info_extractor(self, ie):
759 """Add an InfoExtractor object to the end of the list."""
8b7491c8 760 ie_key = ie.ie_key()
761 self._ies[ie_key] = ie
e52d7f85 762 if not isinstance(ie, type):
8b7491c8 763 self._ies_instances[ie_key] = ie
e52d7f85 764 ie.set_downloader(self)
8222d8de 765
56c73665
JMF
766 def get_info_extractor(self, ie_key):
767 """
768 Get an instance of an IE with name ie_key, it will try to get one from
769 the _ies list, if there's no instance it will create a new one and add
770 it to the extractor list.
771 """
772 ie = self._ies_instances.get(ie_key)
773 if ie is None:
774 ie = get_info_extractor(ie_key)()
775 self.add_info_extractor(ie)
776 return ie
777
023fa8c4
JMF
778 def add_default_info_extractors(self):
779 """
780 Add the InfoExtractors returned by gen_extractors to the end of the list
781 """
fe7866d0 782 all_ies = {ie.IE_NAME.lower(): ie for ie in gen_extractor_classes()}
783 all_ies['end'] = UnsupportedURLIE()
784 try:
785 ie_names = orderedSet_from_options(
786 self.params.get('allowed_extractors', ['default']), {
787 'all': list(all_ies),
788 'default': [name for name, ie in all_ies.items() if ie._ENABLED],
789 }, use_regex=True)
790 except re.error as e:
791 raise ValueError(f'Wrong regex for allowed_extractors: {e.pattern}')
792 for name in ie_names:
793 self.add_info_extractor(all_ies[name])
794 self.write_debug(f'Loaded {len(ie_names)} extractors')
023fa8c4 795
56d868db 796 def add_post_processor(self, pp, when='post_process'):
8222d8de 797 """Add a PostProcessor object to the end of the chain."""
8aa0e7cd 798 assert when in POSTPROCESS_WHEN, f'Invalid when={when}'
5bfa4862 799 self._pps[when].append(pp)
8222d8de
JMF
800 pp.set_downloader(self)
801
ab8e5e51
AM
802 def add_post_hook(self, ph):
803 """Add the post hook"""
804 self._post_hooks.append(ph)
805
933605d7 806 def add_progress_hook(self, ph):
819e0531 807 """Add the download progress hook"""
933605d7 808 self._progress_hooks.append(ph)
8ab470f1 809
819e0531 810 def add_postprocessor_hook(self, ph):
811 """Add the postprocessing progress hook"""
812 self._postprocessor_hooks.append(ph)
5bfc8bee 813 for pps in self._pps.values():
814 for pp in pps:
815 pp.add_progress_hook(ph)
819e0531 816
1c088fa8 817 def _bidi_workaround(self, message):
5d681e96 818 if not hasattr(self, '_output_channel'):
1c088fa8
PH
819 return message
820
5d681e96 821 assert hasattr(self, '_output_process')
14f25df2 822 assert isinstance(message, str)
6febd1c1 823 line_count = message.count('\n') + 1
0f06bcd7 824 self._output_process.stdin.write((message + '\n').encode())
5d681e96 825 self._output_process.stdin.flush()
0f06bcd7 826 res = ''.join(self._output_channel.readline().decode()
9e1a5b84 827 for _ in range(line_count))
6febd1c1 828 return res[:-len('\n')]
1c088fa8 829
b35496d8 830 def _write_string(self, message, out=None, only_once=False):
831 if only_once:
832 if message in self._printed_messages:
833 return
834 self._printed_messages.add(message)
835 write_string(message, out=out, encoding=self.params.get('encoding'))
734f90bb 836
cf4f42cb 837 def to_stdout(self, message, skip_eol=False, quiet=None):
0760b0a7 838 """Print message to stdout"""
cf4f42cb 839 if quiet is not None:
da4db748 840 self.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument quiet. '
841 'Use "YoutubeDL.to_screen" instead')
8a82af35 842 if skip_eol is not False:
da4db748 843 self.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument skip_eol. '
844 'Use "YoutubeDL.to_screen" instead')
0bf9dc1e 845 self._write_string(f'{self._bidi_workaround(message)}\n', self._out_files.out)
cf4f42cb 846
847 def to_screen(self, message, skip_eol=False, quiet=None):
848 """Print message to screen if not in quiet mode"""
8bf9319e 849 if self.params.get('logger'):
43afe285 850 self.params['logger'].debug(message)
cf4f42cb 851 return
852 if (self.params.get('quiet') if quiet is None else quiet) and not self.params.get('verbose'):
853 return
854 self._write_string(
855 '%s%s' % (self._bidi_workaround(message), ('' if skip_eol else '\n')),
591bb9d3 856 self._out_files.screen)
8222d8de 857
b35496d8 858 def to_stderr(self, message, only_once=False):
0760b0a7 859 """Print message to stderr"""
14f25df2 860 assert isinstance(message, str)
8bf9319e 861 if self.params.get('logger'):
43afe285
IB
862 self.params['logger'].error(message)
863 else:
5792c950 864 self._write_string(f'{self._bidi_workaround(message)}\n', self._out_files.error, only_once=only_once)
cf4f42cb 865
866 def _send_console_code(self, code):
591bb9d3 867 if compat_os_name == 'nt' or not self._out_files.console:
cf4f42cb 868 return
591bb9d3 869 self._write_string(code, self._out_files.console)
8222d8de 870
1e5b9a95
PH
871 def to_console_title(self, message):
872 if not self.params.get('consoletitle', False):
873 return
3efb96a6 874 message = remove_terminal_sequences(message)
4bede0d8
C
875 if compat_os_name == 'nt':
876 if ctypes.windll.kernel32.GetConsoleWindow():
877 # c_wchar_p() might not be necessary if `message` is
878 # already of type unicode()
879 ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
cf4f42cb 880 else:
881 self._send_console_code(f'\033]0;{message}\007')
1e5b9a95 882
bdde425c 883 def save_console_title(self):
cf4f42cb 884 if not self.params.get('consoletitle') or self.params.get('simulate'):
bdde425c 885 return
592b7485 886 self._send_console_code('\033[22;0t') # Save the title on stack
bdde425c
PH
887
888 def restore_console_title(self):
cf4f42cb 889 if not self.params.get('consoletitle') or self.params.get('simulate'):
bdde425c 890 return
592b7485 891 self._send_console_code('\033[23;0t') # Restore the title from stack
bdde425c
PH
892
893 def __enter__(self):
894 self.save_console_title()
895 return self
896
897 def __exit__(self, *args):
898 self.restore_console_title()
f89197d7 899
dca08720 900 if self.params.get('cookiefile') is not None:
1bab3437 901 self.cookiejar.save(ignore_discard=True, ignore_expires=True)
bdde425c 902
fa9f30b8 903 def trouble(self, message=None, tb=None, is_error=True):
8222d8de
JMF
904 """Determine action to take when a download problem appears.
905
906 Depending on if the downloader has been configured to ignore
907 download errors or not, this method may throw an exception or
908 not when errors are found, after printing the message.
909
fa9f30b8 910 @param tb If given, is additional traceback information
911 @param is_error Whether to raise error according to ignorerrors
8222d8de
JMF
912 """
913 if message is not None:
914 self.to_stderr(message)
915 if self.params.get('verbose'):
916 if tb is None:
917 if sys.exc_info()[0]: # if .trouble has been called from an except block
6febd1c1 918 tb = ''
8222d8de 919 if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
6febd1c1 920 tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
c0384f22 921 tb += encode_compat_str(traceback.format_exc())
8222d8de
JMF
922 else:
923 tb_data = traceback.format_list(traceback.extract_stack())
6febd1c1 924 tb = ''.join(tb_data)
c19bc311 925 if tb:
926 self.to_stderr(tb)
fa9f30b8 927 if not is_error:
928 return
b1940459 929 if not self.params.get('ignoreerrors'):
8222d8de
JMF
930 if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
931 exc_info = sys.exc_info()[1].exc_info
932 else:
933 exc_info = sys.exc_info()
934 raise DownloadError(message, exc_info)
935 self._download_retcode = 1
936
19a03940 937 Styles = Namespace(
938 HEADERS='yellow',
939 EMPHASIS='light blue',
492272fe 940 FILENAME='green',
19a03940 941 ID='green',
942 DELIM='blue',
943 ERROR='red',
944 WARNING='yellow',
945 SUPPRESS='light black',
946 )
ec11a9f4 947
7578d77d 948 def _format_text(self, handle, allow_colors, text, f, fallback=None, *, test_encoding=False):
e5a998f3 949 text = str(text)
ec11a9f4 950 if test_encoding:
951 original_text = text
5c104538 952 # handle.encoding can be None. See https://github.com/yt-dlp/yt-dlp/issues/2711
953 encoding = self.params.get('encoding') or getattr(handle, 'encoding', None) or 'ascii'
ec11a9f4 954 text = text.encode(encoding, 'ignore').decode(encoding)
955 if fallback is not None and text != original_text:
956 text = fallback
7578d77d 957 return format_text(text, f) if allow_colors else text if fallback is None else fallback
ec11a9f4 958
591bb9d3 959 def _format_out(self, *args, **kwargs):
960 return self._format_text(self._out_files.out, self._allow_colors.out, *args, **kwargs)
961
ec11a9f4 962 def _format_screen(self, *args, **kwargs):
591bb9d3 963 return self._format_text(self._out_files.screen, self._allow_colors.screen, *args, **kwargs)
ec11a9f4 964
965 def _format_err(self, *args, **kwargs):
591bb9d3 966 return self._format_text(self._out_files.error, self._allow_colors.error, *args, **kwargs)
819e0531 967
c84aeac6 968 def report_warning(self, message, only_once=False):
8222d8de
JMF
969 '''
970 Print the message to stderr, it will be prefixed with 'WARNING:'
971 If stderr is a tty file the 'WARNING:' will be colored
972 '''
6d07ce01
JMF
973 if self.params.get('logger') is not None:
974 self.params['logger'].warning(message)
8222d8de 975 else:
ad8915b7
PH
976 if self.params.get('no_warnings'):
977 return
ec11a9f4 978 self.to_stderr(f'{self._format_err("WARNING:", self.Styles.WARNING)} {message}', only_once)
8222d8de 979
da4db748 980 def deprecation_warning(self, message, *, stacklevel=0):
981 deprecation_warning(
982 message, stacklevel=stacklevel + 1, printer=self.report_error, is_error=False)
983
984 def deprecated_feature(self, message):
ee8dd27a 985 if self.params.get('logger') is not None:
da4db748 986 self.params['logger'].warning(f'Deprecated Feature: {message}')
987 self.to_stderr(f'{self._format_err("Deprecated Feature:", self.Styles.ERROR)} {message}', True)
ee8dd27a 988
fa9f30b8 989 def report_error(self, message, *args, **kwargs):
8222d8de
JMF
990 '''
991 Do the same as trouble, but prefixes the message with 'ERROR:', colored
992 in red if stderr is a tty file.
993 '''
fa9f30b8 994 self.trouble(f'{self._format_err("ERROR:", self.Styles.ERROR)} {message}', *args, **kwargs)
8222d8de 995
b35496d8 996 def write_debug(self, message, only_once=False):
0760b0a7 997 '''Log debug message or Print message to stderr'''
998 if not self.params.get('verbose', False):
999 return
8a82af35 1000 message = f'[debug] {message}'
0760b0a7 1001 if self.params.get('logger'):
1002 self.params['logger'].debug(message)
1003 else:
b35496d8 1004 self.to_stderr(message, only_once)
0760b0a7 1005
8222d8de
JMF
1006 def report_file_already_downloaded(self, file_name):
1007 """Report file has already been fully downloaded."""
1008 try:
6febd1c1 1009 self.to_screen('[download] %s has already been downloaded' % file_name)
ce02ed60 1010 except UnicodeEncodeError:
6febd1c1 1011 self.to_screen('[download] The file has already been downloaded')
8222d8de 1012
0c3d0f51 1013 def report_file_delete(self, file_name):
1014 """Report that existing file will be deleted."""
1015 try:
c25228e5 1016 self.to_screen('Deleting existing file %s' % file_name)
0c3d0f51 1017 except UnicodeEncodeError:
c25228e5 1018 self.to_screen('Deleting existing file')
0c3d0f51 1019
319b6059 1020 def raise_no_formats(self, info, forced=False, *, msg=None):
0a5a191a 1021 has_drm = info.get('_has_drm')
319b6059 1022 ignored, expected = self.params.get('ignore_no_formats_error'), bool(msg)
1023 msg = msg or has_drm and 'This video is DRM protected' or 'No video formats found!'
1024 if forced or not ignored:
1151c407 1025 raise ExtractorError(msg, video_id=info['id'], ie=info['extractor'],
319b6059 1026 expected=has_drm or ignored or expected)
88acdbc2 1027 else:
1028 self.report_warning(msg)
1029
de6000d9 1030 def parse_outtmpl(self):
bf1824b3 1031 self.deprecation_warning('"YoutubeDL.parse_outtmpl" is deprecated and may be removed in a future version')
1032 self._parse_outtmpl()
1033 return self.params['outtmpl']
1034
1035 def _parse_outtmpl(self):
7b2c3f47 1036 sanitize = IDENTITY
bf1824b3 1037 if self.params.get('restrictfilenames'): # Remove spaces in the default template
71ce444a 1038 sanitize = lambda x: x.replace(' - ', ' ').replace(' ', '-')
bf1824b3 1039
1040 outtmpl = self.params.setdefault('outtmpl', {})
1041 if not isinstance(outtmpl, dict):
1042 self.params['outtmpl'] = outtmpl = {'default': outtmpl}
1043 outtmpl.update({k: sanitize(v) for k, v in DEFAULT_OUTTMPL.items() if outtmpl.get(k) is None})
de6000d9 1044
21cd8fae 1045 def get_output_path(self, dir_type='', filename=None):
1046 paths = self.params.get('paths', {})
d2c8aadf 1047 assert isinstance(paths, dict), '"paths" parameter must be a dictionary'
21cd8fae 1048 path = os.path.join(
1049 expand_path(paths.get('home', '').strip()),
1050 expand_path(paths.get(dir_type, '').strip()) if dir_type else '',
1051 filename or '')
21cd8fae 1052 return sanitize_path(path, force=self.params.get('windowsfilenames'))
1053
76a264ac 1054 @staticmethod
901130bb 1055 def _outtmpl_expandpath(outtmpl):
1056 # expand_path translates '%%' into '%' and '$$' into '$'
1057 # correspondingly that is not what we want since we need to keep
1058 # '%%' intact for template dict substitution step. Working around
1059 # with boundary-alike separator hack.
1060 sep = ''.join([random.choice(ascii_letters) for _ in range(32)])
86e5f3ed 1061 outtmpl = outtmpl.replace('%%', f'%{sep}%').replace('$$', f'${sep}$')
901130bb 1062
1063 # outtmpl should be expand_path'ed before template dict substitution
1064 # because meta fields may contain env variables we don't want to
62b58c09 1065 # be expanded. E.g. for outtmpl "%(title)s.%(ext)s" and
901130bb 1066 # title "Hello $PATH", we don't want `$PATH` to be expanded.
1067 return expand_path(outtmpl).replace(sep, '')
1068
1069 @staticmethod
1070 def escape_outtmpl(outtmpl):
1071 ''' Escape any remaining strings like %s, %abc% etc. '''
1072 return re.sub(
1073 STR_FORMAT_RE_TMPL.format('', '(?![%(\0])'),
1074 lambda mobj: ('' if mobj.group('has_key') else '%') + mobj.group(0),
1075 outtmpl)
1076
1077 @classmethod
1078 def validate_outtmpl(cls, outtmpl):
76a264ac 1079 ''' @return None or Exception object '''
7d1eb38a 1080 outtmpl = re.sub(
47cdc68e 1081 STR_FORMAT_RE_TMPL.format('[^)]*', '[ljhqBUDS]'),
7d1eb38a 1082 lambda mobj: f'{mobj.group(0)[:-1]}s',
1083 cls._outtmpl_expandpath(outtmpl))
76a264ac 1084 try:
7d1eb38a 1085 cls.escape_outtmpl(outtmpl) % collections.defaultdict(int)
76a264ac 1086 return None
1087 except ValueError as err:
1088 return err
1089
03b4de72 1090 @staticmethod
1091 def _copy_infodict(info_dict):
1092 info_dict = dict(info_dict)
09b49e1f 1093 info_dict.pop('__postprocessors', None)
415f8d51 1094 info_dict.pop('__pending_error', None)
03b4de72 1095 return info_dict
1096
e0fd9573 1097 def prepare_outtmpl(self, outtmpl, info_dict, sanitize=False):
1098 """ Make the outtmpl and info_dict suitable for substitution: ydl.escape_outtmpl(outtmpl) % info_dict
1099 @param sanitize Whether to sanitize the output as a filename.
1100 For backward compatibility, a function can also be passed
1101 """
1102
6e84b215 1103 info_dict.setdefault('epoch', int(time.time())) # keep epoch consistent once set
143db31d 1104
03b4de72 1105 info_dict = self._copy_infodict(info_dict)
752cda38 1106 info_dict['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
53c18592 1107 formatSeconds(info_dict['duration'], '-' if sanitize else ':')
143db31d 1108 if info_dict.get('duration', None) is not None
1109 else None)
1d485a1a 1110 info_dict['autonumber'] = int(self.params.get('autonumber_start', 1) - 1 + self._num_downloads)
9c906919 1111 info_dict['video_autonumber'] = self._num_videos
752cda38 1112 if info_dict.get('resolution') is None:
1113 info_dict['resolution'] = self.format_resolution(info_dict, default=None)
143db31d 1114
e6f21b3d 1115 # For fields playlist_index, playlist_autonumber and autonumber convert all occurrences
143db31d 1116 # of %(field)s to %(field)0Nd for backward compatibility
1117 field_size_compat_map = {
0a5a191a 1118 'playlist_index': number_of_digits(info_dict.get('__last_playlist_index') or 0),
ec11a9f4 1119 'playlist_autonumber': number_of_digits(info_dict.get('n_entries') or 0),
752cda38 1120 'autonumber': self.params.get('autonumber_size') or 5,
143db31d 1121 }
752cda38 1122
385a27fa 1123 TMPL_DICT = {}
47cdc68e 1124 EXTERNAL_FORMAT_RE = re.compile(STR_FORMAT_RE_TMPL.format('[^)]*', f'[{STR_FORMAT_TYPES}ljhqBUDS]'))
385a27fa 1125 MATH_FUNCTIONS = {
1126 '+': float.__add__,
1127 '-': float.__sub__,
1128 }
e625be0d 1129 # Field is of the form key1.key2...
07a1250e 1130 # where keys (except first) can be string, int, slice or "{field, ...}"
1131 FIELD_INNER_RE = r'(?:\w+|%(num)s|%(num)s?(?::%(num)s?){1,2})' % {'num': r'(?:-?\d+)'}
1132 FIELD_RE = r'\w*(?:\.(?:%(inner)s|{%(field)s(?:,%(field)s)*}))*' % {
1133 'inner': FIELD_INNER_RE,
1134 'field': rf'\w*(?:\.{FIELD_INNER_RE})*'
1135 }
1d485a1a 1136 MATH_FIELD_RE = rf'(?:{FIELD_RE}|-?{NUMBER_RE})'
385a27fa 1137 MATH_OPERATORS_RE = r'(?:%s)' % '|'.join(map(re.escape, MATH_FUNCTIONS.keys()))
1d485a1a 1138 INTERNAL_FORMAT_RE = re.compile(rf'''(?x)
e625be0d 1139 (?P<negate>-)?
1d485a1a 1140 (?P<fields>{FIELD_RE})
1141 (?P<maths>(?:{MATH_OPERATORS_RE}{MATH_FIELD_RE})*)
e625be0d 1142 (?:>(?P<strf_format>.+?))?
34baa9fd 1143 (?P<remaining>
1144 (?P<alternate>(?<!\\),[^|&)]+)?
1145 (?:&(?P<replacement>.*?))?
1146 (?:\|(?P<default>.*?))?
1d485a1a 1147 )$''')
752cda38 1148
07a1250e 1149 def _traverse_infodict(fields):
1150 fields = [f for x in re.split(r'\.({.+?})\.?', fields)
1151 for f in ([x] if x.startswith('{') else x.split('.'))]
1152 for i in (0, -1):
1153 if fields and not fields[i]:
1154 fields.pop(i)
1155
1156 for i, f in enumerate(fields):
1157 if not f.startswith('{'):
1158 continue
1159 assert f.endswith('}'), f'No closing brace for {f} in {fields}'
1160 fields[i] = {k: k.split('.') for k in f[1:-1].split(',')}
1161
1162 return traverse_obj(info_dict, fields, is_user_input=True, traverse_string=True)
76a264ac 1163
752cda38 1164 def get_value(mdict):
1165 # Object traversal
2b8a2973 1166 value = _traverse_infodict(mdict['fields'])
752cda38 1167 # Negative
1168 if mdict['negate']:
1169 value = float_or_none(value)
1170 if value is not None:
1171 value *= -1
1172 # Do maths
385a27fa 1173 offset_key = mdict['maths']
1174 if offset_key:
752cda38 1175 value = float_or_none(value)
1176 operator = None
385a27fa 1177 while offset_key:
1178 item = re.match(
1179 MATH_FIELD_RE if operator else MATH_OPERATORS_RE,
1180 offset_key).group(0)
1181 offset_key = offset_key[len(item):]
1182 if operator is None:
752cda38 1183 operator = MATH_FUNCTIONS[item]
385a27fa 1184 continue
1185 item, multiplier = (item[1:], -1) if item[0] == '-' else (item, 1)
1186 offset = float_or_none(item)
1187 if offset is None:
2b8a2973 1188 offset = float_or_none(_traverse_infodict(item))
385a27fa 1189 try:
1190 value = operator(value, multiplier * offset)
1191 except (TypeError, ZeroDivisionError):
1192 return None
1193 operator = None
752cda38 1194 # Datetime formatting
1195 if mdict['strf_format']:
7c37ff97 1196 value = strftime_or_none(value, mdict['strf_format'].replace('\\,', ','))
752cda38 1197
a6bcaf71 1198 # XXX: Workaround for https://github.com/yt-dlp/yt-dlp/issues/4485
1199 if sanitize and value == '':
1200 value = None
752cda38 1201 return value
1202
b868936c 1203 na = self.params.get('outtmpl_na_placeholder', 'NA')
1204
e0fd9573 1205 def filename_sanitizer(key, value, restricted=self.params.get('restrictfilenames')):
5c3895ff 1206 return sanitize_filename(str(value), restricted=restricted, is_id=(
1207 bool(re.search(r'(^|[_.])id(\.|$)', key))
8a82af35 1208 if 'filename-sanitization' in self.params['compat_opts']
5c3895ff 1209 else NO_DEFAULT))
e0fd9573 1210
1211 sanitizer = sanitize if callable(sanitize) else filename_sanitizer
1212 sanitize = bool(sanitize)
1213
6e84b215 1214 def _dumpjson_default(obj):
1215 if isinstance(obj, (set, LazyList)):
1216 return list(obj)
adbc4ec4 1217 return repr(obj)
6e84b215 1218
752cda38 1219 def create_key(outer_mobj):
1220 if not outer_mobj.group('has_key'):
b836dc94 1221 return outer_mobj.group(0)
752cda38 1222 key = outer_mobj.group('key')
752cda38 1223 mobj = re.match(INTERNAL_FORMAT_RE, key)
e0fd9573 1224 initial_field = mobj.group('fields') if mobj else ''
e978789f 1225 value, replacement, default = None, None, na
7c37ff97 1226 while mobj:
e625be0d 1227 mobj = mobj.groupdict()
7c37ff97 1228 default = mobj['default'] if mobj['default'] is not None else default
752cda38 1229 value = get_value(mobj)
e978789f 1230 replacement = mobj['replacement']
7c37ff97 1231 if value is None and mobj['alternate']:
34baa9fd 1232 mobj = re.match(INTERNAL_FORMAT_RE, mobj['remaining'][1:])
7c37ff97 1233 else:
1234 break
752cda38 1235
b868936c 1236 fmt = outer_mobj.group('format')
752cda38 1237 if fmt == 's' and value is not None and key in field_size_compat_map.keys():
86e5f3ed 1238 fmt = f'0{field_size_compat_map[key]:d}d'
752cda38 1239
e978789f 1240 value = default if value is None else value if replacement is None else replacement
752cda38 1241
4476d2c7 1242 flags = outer_mobj.group('conversion') or ''
7d1eb38a 1243 str_fmt = f'{fmt[:-1]}s'
524e2e4f 1244 if fmt[-1] == 'l': # list
4476d2c7 1245 delim = '\n' if '#' in flags else ', '
9e907ebd 1246 value, fmt = delim.join(map(str, variadic(value, allowed_types=(str, bytes)))), str_fmt
524e2e4f 1247 elif fmt[-1] == 'j': # json
4476d2c7 1248 value, fmt = json.dumps(value, default=_dumpjson_default, indent=4 if '#' in flags else None), str_fmt
47cdc68e 1249 elif fmt[-1] == 'h': # html
1250 value, fmt = escapeHTML(value), str_fmt
524e2e4f 1251 elif fmt[-1] == 'q': # quoted
4476d2c7 1252 value = map(str, variadic(value) if '#' in flags else [value])
1253 value, fmt = ' '.join(map(compat_shlex_quote, value)), str_fmt
524e2e4f 1254 elif fmt[-1] == 'B': # bytes
0f06bcd7 1255 value = f'%{str_fmt}'.encode() % str(value).encode()
f5aa5cfb 1256 value, fmt = value.decode('utf-8', 'ignore'), 's'
524e2e4f 1257 elif fmt[-1] == 'U': # unicode normalized
524e2e4f 1258 value, fmt = unicodedata.normalize(
1259 # "+" = compatibility equivalence, "#" = NFD
4476d2c7 1260 'NF%s%s' % ('K' if '+' in flags else '', 'D' if '#' in flags else 'C'),
524e2e4f 1261 value), str_fmt
e0fd9573 1262 elif fmt[-1] == 'D': # decimal suffix
abbeeebc 1263 num_fmt, fmt = fmt[:-1].replace('#', ''), 's'
1264 value = format_decimal_suffix(value, f'%{num_fmt}f%s' if num_fmt else '%d%s',
1265 factor=1024 if '#' in flags else 1000)
37893bb0 1266 elif fmt[-1] == 'S': # filename sanitization
e0fd9573 1267 value, fmt = filename_sanitizer(initial_field, value, restricted='#' in flags), str_fmt
7d1eb38a 1268 elif fmt[-1] == 'c':
524e2e4f 1269 if value:
1270 value = str(value)[0]
76a264ac 1271 else:
524e2e4f 1272 fmt = str_fmt
76a264ac 1273 elif fmt[-1] not in 'rs': # numeric
a439a3a4 1274 value = float_or_none(value)
752cda38 1275 if value is None:
1276 value, fmt = default, 's'
901130bb 1277
752cda38 1278 if sanitize:
1279 if fmt[-1] == 'r':
1280 # If value is an object, sanitize might convert it to a string
1281 # So we convert it to repr first
7d1eb38a 1282 value, fmt = repr(value), str_fmt
639f1cea 1283 if fmt[-1] in 'csr':
e0fd9573 1284 value = sanitizer(initial_field, value)
901130bb 1285
b868936c 1286 key = '%s\0%s' % (key.replace('%', '%\0'), outer_mobj.group('format'))
385a27fa 1287 TMPL_DICT[key] = value
b868936c 1288 return '{prefix}%({key}){fmt}'.format(key=key, fmt=fmt, prefix=outer_mobj.group('prefix'))
752cda38 1289
385a27fa 1290 return EXTERNAL_FORMAT_RE.sub(create_key, outtmpl), TMPL_DICT
143db31d 1291
819e0531 1292 def evaluate_outtmpl(self, outtmpl, info_dict, *args, **kwargs):
1293 outtmpl, info_dict = self.prepare_outtmpl(outtmpl, info_dict, *args, **kwargs)
1294 return self.escape_outtmpl(outtmpl) % info_dict
1295
5127e92a 1296 def _prepare_filename(self, info_dict, *, outtmpl=None, tmpl_type=None):
1297 assert None in (outtmpl, tmpl_type), 'outtmpl and tmpl_type are mutually exclusive'
1298 if outtmpl is None:
bf1824b3 1299 outtmpl = self.params['outtmpl'].get(tmpl_type or 'default', self.params['outtmpl']['default'])
8222d8de 1300 try:
5127e92a 1301 outtmpl = self._outtmpl_expandpath(outtmpl)
e0fd9573 1302 filename = self.evaluate_outtmpl(outtmpl, info_dict, True)
6a0546e3 1303 if not filename:
1304 return None
15da37c7 1305
5127e92a 1306 if tmpl_type in ('', 'temp'):
6a0546e3 1307 final_ext, ext = self.params.get('final_ext'), info_dict.get('ext')
1308 if final_ext and ext and final_ext != ext and filename.endswith(f'.{final_ext}'):
1309 filename = replace_extension(filename, ext, final_ext)
5127e92a 1310 elif tmpl_type:
6a0546e3 1311 force_ext = OUTTMPL_TYPES[tmpl_type]
1312 if force_ext:
1313 filename = replace_extension(filename, force_ext, info_dict.get('ext'))
de6000d9 1314
bdc3fd2f
U
1315 # https://github.com/blackjack4494/youtube-dlc/issues/85
1316 trim_file_name = self.params.get('trim_file_name', False)
1317 if trim_file_name:
5c22c63d 1318 no_ext, *ext = filename.rsplit('.', 2)
1319 filename = join_nonempty(no_ext[:trim_file_name], *ext, delim='.')
bdc3fd2f 1320
0202b52a 1321 return filename
8222d8de 1322 except ValueError as err:
6febd1c1 1323 self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')')
8222d8de
JMF
1324 return None
1325
5127e92a 1326 def prepare_filename(self, info_dict, dir_type='', *, outtmpl=None, warn=False):
1327 """Generate the output filename"""
1328 if outtmpl:
1329 assert not dir_type, 'outtmpl and dir_type are mutually exclusive'
1330 dir_type = None
1331 filename = self._prepare_filename(info_dict, tmpl_type=dir_type, outtmpl=outtmpl)
80c03fa9 1332 if not filename and dir_type not in ('', 'temp'):
1333 return ''
de6000d9 1334
c84aeac6 1335 if warn:
21cd8fae 1336 if not self.params.get('paths'):
de6000d9 1337 pass
1338 elif filename == '-':
c84aeac6 1339 self.report_warning('--paths is ignored when an outputting to stdout', only_once=True)
de6000d9 1340 elif os.path.isabs(filename):
c84aeac6 1341 self.report_warning('--paths is ignored since an absolute path is given in output template', only_once=True)
de6000d9 1342 if filename == '-' or not filename:
1343 return filename
1344
21cd8fae 1345 return self.get_output_path(dir_type, filename)
0202b52a 1346
120fe513 1347 def _match_entry(self, info_dict, incomplete=False, silent=False):
ecdec191 1348 """ Returns None if the file should be downloaded """
8222d8de 1349
3bec830a 1350 video_title = info_dict.get('title', info_dict.get('id', 'entry'))
c77495e3 1351
8b0d7497 1352 def check_filter():
8b0d7497 1353 if 'title' in info_dict:
1354 # This can happen when we're just evaluating the playlist
1355 title = info_dict['title']
1356 matchtitle = self.params.get('matchtitle', False)
1357 if matchtitle:
1358 if not re.search(matchtitle, title, re.IGNORECASE):
1359 return '"' + title + '" title did not match pattern "' + matchtitle + '"'
1360 rejecttitle = self.params.get('rejecttitle', False)
1361 if rejecttitle:
1362 if re.search(rejecttitle, title, re.IGNORECASE):
1363 return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
1364 date = info_dict.get('upload_date')
1365 if date is not None:
1366 dateRange = self.params.get('daterange', DateRange())
1367 if date not in dateRange:
86e5f3ed 1368 return f'{date_from_str(date).isoformat()} upload date is not in range {dateRange}'
8b0d7497 1369 view_count = info_dict.get('view_count')
1370 if view_count is not None:
1371 min_views = self.params.get('min_views')
1372 if min_views is not None and view_count < min_views:
1373 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
1374 max_views = self.params.get('max_views')
1375 if max_views is not None and view_count > max_views:
1376 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
1377 if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')):
1378 return 'Skipping "%s" because it is age restricted' % video_title
8b0d7497 1379
8f18aca8 1380 match_filter = self.params.get('match_filter')
1381 if match_filter is not None:
1382 try:
1383 ret = match_filter(info_dict, incomplete=incomplete)
1384 except TypeError:
1385 # For backward compatibility
1386 ret = None if incomplete else match_filter(info_dict)
492272fe 1387 if ret is NO_DEFAULT:
1388 while True:
1389 filename = self._format_screen(self.prepare_filename(info_dict), self.Styles.FILENAME)
1390 reply = input(self._format_screen(
1391 f'Download "{filename}"? (Y/n): ', self.Styles.EMPHASIS)).lower().strip()
1392 if reply in {'y', ''}:
1393 return None
1394 elif reply == 'n':
1395 return f'Skipping {video_title}'
492272fe 1396 elif ret is not None:
8f18aca8 1397 return ret
8b0d7497 1398 return None
1399
c77495e3 1400 if self.in_download_archive(info_dict):
1401 reason = '%s has already been recorded in the archive' % video_title
1402 break_opt, break_err = 'break_on_existing', ExistingVideoReached
1403 else:
1404 reason = check_filter()
1405 break_opt, break_err = 'break_on_reject', RejectedVideoReached
8b0d7497 1406 if reason is not None:
120fe513 1407 if not silent:
1408 self.to_screen('[download] ' + reason)
c77495e3 1409 if self.params.get(break_opt, False):
1410 raise break_err()
8b0d7497 1411 return reason
fe7e0c98 1412
b6c45014
JMF
1413 @staticmethod
1414 def add_extra_info(info_dict, extra_info):
1415 '''Set the keys from extra_info in info dict if they are missing'''
1416 for key, value in extra_info.items():
1417 info_dict.setdefault(key, value)
1418
409e1828 1419 def extract_info(self, url, download=True, ie_key=None, extra_info=None,
61aa5ba3 1420 process=True, force_generic_extractor=False):
41d1cca3 1421 """
1422 Return a list with a dictionary for each video extracted.
1423
1424 Arguments:
1425 url -- URL to extract
1426
1427 Keyword arguments:
1428 download -- whether to download videos during extraction
1429 ie_key -- extractor key hint
1430 extra_info -- dictionary containing the extra values to add to each result
1431 process -- whether to resolve all unresolved references (URLs, playlist items),
1432 must be True for download to work.
1433 force_generic_extractor -- force using the generic extractor
1434 """
fe7e0c98 1435
409e1828 1436 if extra_info is None:
1437 extra_info = {}
1438
61aa5ba3 1439 if not ie_key and force_generic_extractor:
d22dec74
S
1440 ie_key = 'Generic'
1441
8222d8de 1442 if ie_key:
fe7866d0 1443 ies = {ie_key: self._ies[ie_key]} if ie_key in self._ies else {}
8222d8de
JMF
1444 else:
1445 ies = self._ies
1446
fe7866d0 1447 for key, ie in ies.items():
8222d8de
JMF
1448 if not ie.suitable(url):
1449 continue
1450
1451 if not ie.working():
6febd1c1
PH
1452 self.report_warning('The program functionality for this site has been marked as broken, '
1453 'and will probably not work.')
8222d8de 1454
1151c407 1455 temp_id = ie.get_temp_id(url)
fe7866d0 1456 if temp_id is not None and self.in_download_archive({'id': temp_id, 'ie_key': key}):
1457 self.to_screen(f'[{key}] {temp_id}: has already been recorded in the archive')
5e5be0c0 1458 if self.params.get('break_on_existing', False):
1459 raise ExistingVideoReached()
a0566bbf 1460 break
fe7866d0 1461 return self.__extract_info(url, self.get_info_extractor(key), download, extra_info, process)
a0566bbf 1462 else:
fe7866d0 1463 extractors_restricted = self.params.get('allowed_extractors') not in (None, ['default'])
1464 self.report_error(f'No suitable extractor{format_field(ie_key, None, " (%s)")} found for URL {url}',
1465 tb=False if extractors_restricted else None)
a0566bbf 1466
7e88d7d7 1467 def _handle_extraction_exceptions(func):
b5ae35ee 1468 @functools.wraps(func)
a0566bbf 1469 def wrapper(self, *args, **kwargs):
6da22e7d 1470 while True:
1471 try:
1472 return func(self, *args, **kwargs)
1473 except (DownloadCancelled, LazyList.IndexError, PagedList.IndexError):
8222d8de 1474 raise
6da22e7d 1475 except ReExtractInfo as e:
1476 if e.expected:
1477 self.to_screen(f'{e}; Re-extracting data')
1478 else:
1479 self.to_stderr('\r')
1480 self.report_warning(f'{e}; Re-extracting data')
1481 continue
1482 except GeoRestrictedError as e:
1483 msg = e.msg
1484 if e.countries:
1485 msg += '\nThis video is available in %s.' % ', '.join(
1486 map(ISO3166Utils.short2full, e.countries))
1487 msg += '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1488 self.report_error(msg)
1489 except ExtractorError as e: # An error we somewhat expected
1490 self.report_error(str(e), e.format_traceback())
1491 except Exception as e:
1492 if self.params.get('ignoreerrors'):
1493 self.report_error(str(e), tb=encode_compat_str(traceback.format_exc()))
1494 else:
1495 raise
1496 break
a0566bbf 1497 return wrapper
1498
693f0600 1499 def _wait_for_video(self, ie_result={}):
f2ebc5c7 1500 if (not self.params.get('wait_for_video')
1501 or ie_result.get('_type', 'video') != 'video'
1502 or ie_result.get('formats') or ie_result.get('url')):
1503 return
1504
1505 format_dur = lambda dur: '%02d:%02d:%02d' % timetuple_from_msec(dur * 1000)[:-1]
1506 last_msg = ''
1507
1508 def progress(msg):
1509 nonlocal last_msg
a7dc6a89 1510 full_msg = f'{msg}\n'
1511 if not self.params.get('noprogress'):
1512 full_msg = msg + ' ' * (len(last_msg) - len(msg)) + '\r'
1513 elif last_msg:
1514 return
1515 self.to_screen(full_msg, skip_eol=True)
f2ebc5c7 1516 last_msg = msg
1517
1518 min_wait, max_wait = self.params.get('wait_for_video')
1519 diff = try_get(ie_result, lambda x: x['release_timestamp'] - time.time())
1520 if diff is None and ie_result.get('live_status') == 'is_upcoming':
16c620bc 1521 diff = round(random.uniform(min_wait, max_wait) if (max_wait and min_wait) else (max_wait or min_wait), 0)
f2ebc5c7 1522 self.report_warning('Release time of video is not known')
693f0600 1523 elif ie_result and (diff or 0) <= 0:
f2ebc5c7 1524 self.report_warning('Video should already be available according to extracted info')
38d79fd1 1525 diff = min(max(diff or 0, min_wait or 0), max_wait or float('inf'))
f2ebc5c7 1526 self.to_screen(f'[wait] Waiting for {format_dur(diff)} - Press Ctrl+C to try now')
1527
1528 wait_till = time.time() + diff
1529 try:
1530 while True:
1531 diff = wait_till - time.time()
1532 if diff <= 0:
1533 progress('')
1534 raise ReExtractInfo('[wait] Wait period ended', expected=True)
1535 progress(f'[wait] Remaining time until next attempt: {self._format_screen(format_dur(diff), self.Styles.EMPHASIS)}')
1536 time.sleep(1)
1537 except KeyboardInterrupt:
1538 progress('')
1539 raise ReExtractInfo('[wait] Interrupted by user', expected=True)
1540 except BaseException as e:
1541 if not isinstance(e, ReExtractInfo):
1542 self.to_screen('')
1543 raise
1544
7e88d7d7 1545 @_handle_extraction_exceptions
58f197b7 1546 def __extract_info(self, url, ie, download, extra_info, process):
693f0600 1547 try:
1548 ie_result = ie.extract(url)
1549 except UserNotLive as e:
1550 if process:
1551 if self.params.get('wait_for_video'):
1552 self.report_warning(e)
1553 self._wait_for_video()
1554 raise
a0566bbf 1555 if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
cb794ee0 1556 self.report_warning(f'Extractor {ie.IE_NAME} returned nothing{bug_reports_message()}')
a0566bbf 1557 return
1558 if isinstance(ie_result, list):
1559 # Backwards compatibility: old IE result format
1560 ie_result = {
1561 '_type': 'compat_list',
1562 'entries': ie_result,
1563 }
e37d0efb 1564 if extra_info.get('original_url'):
1565 ie_result.setdefault('original_url', extra_info['original_url'])
a0566bbf 1566 self.add_default_extra_info(ie_result, ie, url)
1567 if process:
f2ebc5c7 1568 self._wait_for_video(ie_result)
a0566bbf 1569 return self.process_ie_result(ie_result, download, extra_info)
8222d8de 1570 else:
a0566bbf 1571 return ie_result
fe7e0c98 1572
ea38e55f 1573 def add_default_extra_info(self, ie_result, ie, url):
6033d980 1574 if url is not None:
1575 self.add_extra_info(ie_result, {
1576 'webpage_url': url,
1577 'original_url': url,
57ebfca3 1578 })
1579 webpage_url = ie_result.get('webpage_url')
1580 if webpage_url:
1581 self.add_extra_info(ie_result, {
1582 'webpage_url_basename': url_basename(webpage_url),
1583 'webpage_url_domain': get_domain(webpage_url),
6033d980 1584 })
1585 if ie is not None:
1586 self.add_extra_info(ie_result, {
1587 'extractor': ie.IE_NAME,
1588 'extractor_key': ie.ie_key(),
1589 })
ea38e55f 1590
58adec46 1591 def process_ie_result(self, ie_result, download=True, extra_info=None):
8222d8de
JMF
1592 """
1593 Take the result of the ie(may be modified) and resolve all unresolved
1594 references (URLs, playlist items).
1595
1596 It will also download the videos if 'download'.
1597 Returns the resolved ie_result.
1598 """
58adec46 1599 if extra_info is None:
1600 extra_info = {}
e8ee972c
PH
1601 result_type = ie_result.get('_type', 'video')
1602
057a5206 1603 if result_type in ('url', 'url_transparent'):
8f97a15d 1604 ie_result['url'] = sanitize_url(
1605 ie_result['url'], scheme='http' if self.params.get('prefer_insecure') else 'https')
e37d0efb 1606 if ie_result.get('original_url'):
1607 extra_info.setdefault('original_url', ie_result['original_url'])
1608
057a5206 1609 extract_flat = self.params.get('extract_flat', False)
3089bc74
S
1610 if ((extract_flat == 'in_playlist' and 'playlist' in extra_info)
1611 or extract_flat is True):
ecb54191 1612 info_copy = ie_result.copy()
6033d980 1613 ie = try_get(ie_result.get('ie_key'), self.get_info_extractor)
360167b9 1614 if ie and not ie_result.get('id'):
4614bc22 1615 info_copy['id'] = ie.get_temp_id(ie_result['url'])
6033d980 1616 self.add_default_extra_info(info_copy, ie, ie_result['url'])
4614bc22 1617 self.add_extra_info(info_copy, extra_info)
b5475f11 1618 info_copy, _ = self.pre_process(info_copy)
ecb54191 1619 self.__forced_printings(info_copy, self.prepare_filename(info_copy), incomplete=True)
415f8d51 1620 self._raise_pending_errors(info_copy)
4614bc22 1621 if self.params.get('force_write_download_archive', False):
1622 self.record_download_archive(info_copy)
e8ee972c
PH
1623 return ie_result
1624
8222d8de 1625 if result_type == 'video':
b6c45014 1626 self.add_extra_info(ie_result, extra_info)
9c2b75b5 1627 ie_result = self.process_video_result(ie_result, download=download)
415f8d51 1628 self._raise_pending_errors(ie_result)
28b0eb0f 1629 additional_urls = (ie_result or {}).get('additional_urls')
9c2b75b5 1630 if additional_urls:
e9f4ccd1 1631 # TODO: Improve MetadataParserPP to allow setting a list
14f25df2 1632 if isinstance(additional_urls, str):
9c2b75b5 1633 additional_urls = [additional_urls]
1634 self.to_screen(
1635 '[info] %s: %d additional URL(s) requested' % (ie_result['id'], len(additional_urls)))
1636 self.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls))
1637 ie_result['additional_entries'] = [
1638 self.extract_info(
b69fd25c 1639 url, download, extra_info=extra_info,
9c2b75b5 1640 force_generic_extractor=self.params.get('force_generic_extractor'))
1641 for url in additional_urls
1642 ]
1643 return ie_result
8222d8de
JMF
1644 elif result_type == 'url':
1645 # We have to add extra_info to the results because it may be
1646 # contained in a playlist
07cce701 1647 return self.extract_info(
1648 ie_result['url'], download,
1649 ie_key=ie_result.get('ie_key'),
1650 extra_info=extra_info)
7fc3fa05
PH
1651 elif result_type == 'url_transparent':
1652 # Use the information from the embedding page
1653 info = self.extract_info(
1654 ie_result['url'], ie_key=ie_result.get('ie_key'),
1655 extra_info=extra_info, download=False, process=False)
1656
1640eb09
S
1657 # extract_info may return None when ignoreerrors is enabled and
1658 # extraction failed with an error, don't crash and return early
1659 # in this case
1660 if not info:
1661 return info
1662
3975b4d2 1663 exempted_fields = {'_type', 'url', 'ie_key'}
1664 if not ie_result.get('section_end') and ie_result.get('section_start') is None:
1665 # For video clips, the id etc of the clip extractor should be used
1666 exempted_fields |= {'id', 'extractor', 'extractor_key'}
1667
412c617d 1668 new_result = info.copy()
3975b4d2 1669 new_result.update(filter_dict(ie_result, lambda k, v: v is not None and k not in exempted_fields))
7fc3fa05 1670
0563f7ac
S
1671 # Extracted info may not be a video result (i.e.
1672 # info.get('_type', 'video') != video) but rather an url or
1673 # url_transparent. In such cases outer metadata (from ie_result)
1674 # should be propagated to inner one (info). For this to happen
1675 # _type of info should be overridden with url_transparent. This
067aa17e 1676 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
0563f7ac
S
1677 if new_result.get('_type') == 'url':
1678 new_result['_type'] = 'url_transparent'
7fc3fa05
PH
1679
1680 return self.process_ie_result(
1681 new_result, download=download, extra_info=extra_info)
40fcba5e 1682 elif result_type in ('playlist', 'multi_video'):
30a074c2 1683 # Protect from infinite recursion due to recursively nested playlists
1684 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
1685 webpage_url = ie_result['webpage_url']
1686 if webpage_url in self._playlist_urls:
7e85e872 1687 self.to_screen(
30a074c2 1688 '[download] Skipping already downloaded playlist: %s'
1689 % ie_result.get('title') or ie_result.get('id'))
1690 return
7e85e872 1691
30a074c2 1692 self._playlist_level += 1
1693 self._playlist_urls.add(webpage_url)
03f83004 1694 self._fill_common_fields(ie_result, False)
bc516a3f 1695 self._sanitize_thumbnails(ie_result)
30a074c2 1696 try:
1697 return self.__process_playlist(ie_result, download)
1698 finally:
1699 self._playlist_level -= 1
1700 if not self._playlist_level:
1701 self._playlist_urls.clear()
8222d8de 1702 elif result_type == 'compat_list':
c9bf4114
PH
1703 self.report_warning(
1704 'Extractor %s returned a compat_list result. '
1705 'It needs to be updated.' % ie_result.get('extractor'))
5f6a1245 1706
8222d8de 1707 def _fixup(r):
b868936c 1708 self.add_extra_info(r, {
1709 'extractor': ie_result['extractor'],
1710 'webpage_url': ie_result['webpage_url'],
1711 'webpage_url_basename': url_basename(ie_result['webpage_url']),
0bb322b9 1712 'webpage_url_domain': get_domain(ie_result['webpage_url']),
b868936c 1713 'extractor_key': ie_result['extractor_key'],
1714 })
8222d8de
JMF
1715 return r
1716 ie_result['entries'] = [
b6c45014 1717 self.process_ie_result(_fixup(r), download, extra_info)
8222d8de
JMF
1718 for r in ie_result['entries']
1719 ]
1720 return ie_result
1721 else:
1722 raise Exception('Invalid result type: %s' % result_type)
1723
e92caff5 1724 def _ensure_dir_exists(self, path):
1725 return make_dir(path, self.report_error)
1726
3b603dbd 1727 @staticmethod
3bec830a 1728 def _playlist_infodict(ie_result, strict=False, **kwargs):
1729 info = {
1730 'playlist_count': ie_result.get('playlist_count'),
3b603dbd 1731 'playlist': ie_result.get('title') or ie_result.get('id'),
1732 'playlist_id': ie_result.get('id'),
1733 'playlist_title': ie_result.get('title'),
1734 'playlist_uploader': ie_result.get('uploader'),
1735 'playlist_uploader_id': ie_result.get('uploader_id'),
3b603dbd 1736 **kwargs,
1737 }
3bec830a 1738 if strict:
1739 return info
1740 return {
1741 **info,
1742 'playlist_index': 0,
1743 '__last_playlist_index': max(ie_result['requested_entries'] or (0, 0)),
1744 'extractor': ie_result['extractor'],
1745 'webpage_url': ie_result['webpage_url'],
1746 'webpage_url_basename': url_basename(ie_result['webpage_url']),
1747 'webpage_url_domain': get_domain(ie_result['webpage_url']),
1748 'extractor_key': ie_result['extractor_key'],
1749 }
3b603dbd 1750
30a074c2 1751 def __process_playlist(self, ie_result, download):
7e88d7d7 1752 """Process each entry in the playlist"""
f5ea4748 1753 assert ie_result['_type'] in ('playlist', 'multi_video')
1754
3bec830a 1755 common_info = self._playlist_infodict(ie_result, strict=True)
3955b207 1756 title = common_info.get('playlist') or '<Untitled>'
3bec830a 1757 if self._match_entry(common_info, incomplete=True) is not None:
1758 return
c6e07cf1 1759 self.to_screen(f'[download] Downloading {ie_result["_type"]}: {title}')
f0d785d3 1760
7e88d7d7 1761 all_entries = PlaylistEntries(self, ie_result)
7e9a6125 1762 entries = orderedSet(all_entries.get_requested_items(), lazy=True)
1763
1764 lazy = self.params.get('lazy_playlist')
1765 if lazy:
1766 resolved_entries, n_entries = [], 'N/A'
1767 ie_result['requested_entries'], ie_result['entries'] = None, None
1768 else:
1769 entries = resolved_entries = list(entries)
1770 n_entries = len(resolved_entries)
1771 ie_result['requested_entries'], ie_result['entries'] = tuple(zip(*resolved_entries)) or ([], [])
1772 if not ie_result.get('playlist_count'):
1773 # Better to do this after potentially exhausting entries
1774 ie_result['playlist_count'] = all_entries.get_full_count()
498f5606 1775
0647d925 1776 extra = self._playlist_infodict(ie_result, n_entries=int_or_none(n_entries))
1777 ie_copy = collections.ChainMap(ie_result, extra)
3bec830a 1778
e08a85d8 1779 _infojson_written = False
0bfc53d0 1780 write_playlist_files = self.params.get('allow_playlist_files', True)
1781 if write_playlist_files and self.params.get('list_thumbnails'):
1782 self.list_thumbnails(ie_result)
1783 if write_playlist_files and not self.params.get('simulate'):
e08a85d8 1784 _infojson_written = self._write_info_json(
1785 'playlist', ie_result, self.prepare_filename(ie_copy, 'pl_infojson'))
1786 if _infojson_written is None:
80c03fa9 1787 return
1788 if self._write_description('playlist', ie_result,
1789 self.prepare_filename(ie_copy, 'pl_description')) is None:
1790 return
681de68e 1791 # TODO: This should be passed to ThumbnailsConvertor if necessary
3bec830a 1792 self._write_thumbnails('playlist', ie_result, self.prepare_filename(ie_copy, 'pl_thumbnail'))
30a074c2 1793
7e9a6125 1794 if lazy:
1795 if self.params.get('playlistreverse') or self.params.get('playlistrandom'):
1796 self.report_warning('playlistreverse and playlistrandom are not supported with lazy_playlist', only_once=True)
1797 elif self.params.get('playlistreverse'):
1798 entries.reverse()
1799 elif self.params.get('playlistrandom'):
30a074c2 1800 random.shuffle(entries)
1801
7e88d7d7 1802 self.to_screen(f'[{ie_result["extractor"]}] Playlist {title}: Downloading {n_entries} videos'
1803 f'{format_field(ie_result, "playlist_count", " of %s")}')
30a074c2 1804
134c913c 1805 keep_resolved_entries = self.params.get('extract_flat') != 'discard'
1806 if self.params.get('extract_flat') == 'discard_in_playlist':
1807 keep_resolved_entries = ie_result['_type'] != 'playlist'
1808 if keep_resolved_entries:
1809 self.write_debug('The information of all playlist entries will be held in memory')
1810
26e2805c 1811 failures = 0
1812 max_failures = self.params.get('skip_playlist_after_errors') or float('inf')
7e9a6125 1813 for i, (playlist_index, entry) in enumerate(entries):
1814 if lazy:
1815 resolved_entries.append((playlist_index, entry))
3bec830a 1816 if not entry:
7e88d7d7 1817 continue
1818
7e88d7d7 1819 entry['__x_forwarded_for_ip'] = ie_result.get('__x_forwarded_for_ip')
7e9a6125 1820 if not lazy and 'playlist-index' in self.params.get('compat_opts', []):
1821 playlist_index = ie_result['requested_entries'][i]
1822
0647d925 1823 entry_copy = collections.ChainMap(entry, {
3bec830a 1824 **common_info,
3955b207 1825 'n_entries': int_or_none(n_entries),
71729754 1826 'playlist_index': playlist_index,
7e9a6125 1827 'playlist_autonumber': i + 1,
0647d925 1828 })
3bec830a 1829
0647d925 1830 if self._match_entry(entry_copy, incomplete=True) is not None:
f0ad6f8c 1831 # For compatabilty with youtube-dl. See https://github.com/yt-dlp/yt-dlp/issues/4369
1832 resolved_entries[i] = (playlist_index, NO_DEFAULT)
3bec830a 1833 continue
1834
1835 self.to_screen('[download] Downloading video %s of %s' % (
1836 self._format_screen(i + 1, self.Styles.ID), self._format_screen(n_entries, self.Styles.EMPHASIS)))
1837
a6ca61d4 1838 extra.update({
1839 'playlist_index': playlist_index,
1840 'playlist_autonumber': i + 1,
1841 })
3bec830a 1842 entry_result = self.__process_iterable_entry(entry, download, extra)
26e2805c 1843 if not entry_result:
1844 failures += 1
1845 if failures >= max_failures:
1846 self.report_error(
7e88d7d7 1847 f'Skipping the remaining entries in playlist "{title}" since {failures} items failed extraction')
26e2805c 1848 break
134c913c 1849 if keep_resolved_entries:
1850 resolved_entries[i] = (playlist_index, entry_result)
7e88d7d7 1851
1852 # Update with processed data
f0ad6f8c 1853 ie_result['requested_entries'] = [i for i, e in resolved_entries if e is not NO_DEFAULT]
1854 ie_result['entries'] = [e for _, e in resolved_entries if e is not NO_DEFAULT]
e08a85d8 1855
1856 # Write the updated info to json
cb96c5be 1857 if _infojson_written is True and self._write_info_json(
e08a85d8 1858 'updated playlist', ie_result,
1859 self.prepare_filename(ie_copy, 'pl_infojson'), overwrite=True) is None:
1860 return
ca30f449 1861
ed5835b4 1862 ie_result = self.run_all_pps('playlist', ie_result)
7e88d7d7 1863 self.to_screen(f'[download] Finished downloading playlist: {title}')
30a074c2 1864 return ie_result
1865
7e88d7d7 1866 @_handle_extraction_exceptions
a0566bbf 1867 def __process_iterable_entry(self, entry, download, extra_info):
1868 return self.process_ie_result(
1869 entry, download=download, extra_info=extra_info)
1870
67134eab
JMF
1871 def _build_format_filter(self, filter_spec):
1872 " Returns a function to filter the formats according to the filter_spec "
083c9df9
PH
1873
1874 OPERATORS = {
1875 '<': operator.lt,
1876 '<=': operator.le,
1877 '>': operator.gt,
1878 '>=': operator.ge,
1879 '=': operator.eq,
1880 '!=': operator.ne,
1881 }
67134eab 1882 operator_rex = re.compile(r'''(?x)\s*
187986a8 1883 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)\s*
1884 (?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1885 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)\s*
083c9df9 1886 ''' % '|'.join(map(re.escape, OPERATORS.keys())))
187986a8 1887 m = operator_rex.fullmatch(filter_spec)
9ddb6925
S
1888 if m:
1889 try:
1890 comparison_value = int(m.group('value'))
1891 except ValueError:
1892 comparison_value = parse_filesize(m.group('value'))
1893 if comparison_value is None:
1894 comparison_value = parse_filesize(m.group('value') + 'B')
1895 if comparison_value is None:
1896 raise ValueError(
1897 'Invalid value %r in format specification %r' % (
67134eab 1898 m.group('value'), filter_spec))
9ddb6925
S
1899 op = OPERATORS[m.group('op')]
1900
083c9df9 1901 if not m:
9ddb6925
S
1902 STR_OPERATORS = {
1903 '=': operator.eq,
10d33b34
YCH
1904 '^=': lambda attr, value: attr.startswith(value),
1905 '$=': lambda attr, value: attr.endswith(value),
1906 '*=': lambda attr, value: value in attr,
1ce9a3cb 1907 '~=': lambda attr, value: value.search(attr) is not None
9ddb6925 1908 }
187986a8 1909 str_operator_rex = re.compile(r'''(?x)\s*
1910 (?P<key>[a-zA-Z0-9._-]+)\s*
1ce9a3cb
LF
1911 (?P<negation>!\s*)?(?P<op>%s)\s*(?P<none_inclusive>\?\s*)?
1912 (?P<quote>["'])?
1913 (?P<value>(?(quote)(?:(?!(?P=quote))[^\\]|\\.)+|[\w.-]+))
1914 (?(quote)(?P=quote))\s*
9ddb6925 1915 ''' % '|'.join(map(re.escape, STR_OPERATORS.keys())))
187986a8 1916 m = str_operator_rex.fullmatch(filter_spec)
9ddb6925 1917 if m:
1ce9a3cb
LF
1918 if m.group('op') == '~=':
1919 comparison_value = re.compile(m.group('value'))
1920 else:
1921 comparison_value = re.sub(r'''\\([\\"'])''', r'\1', m.group('value'))
2cc779f4
S
1922 str_op = STR_OPERATORS[m.group('op')]
1923 if m.group('negation'):
e118a879 1924 op = lambda attr, value: not str_op(attr, value)
2cc779f4
S
1925 else:
1926 op = str_op
083c9df9 1927
9ddb6925 1928 if not m:
187986a8 1929 raise SyntaxError('Invalid filter specification %r' % filter_spec)
083c9df9
PH
1930
1931 def _filter(f):
1932 actual_value = f.get(m.group('key'))
1933 if actual_value is None:
1934 return m.group('none_inclusive')
1935 return op(actual_value, comparison_value)
67134eab
JMF
1936 return _filter
1937
9f1a1c36 1938 def _check_formats(self, formats):
1939 for f in formats:
1940 self.to_screen('[info] Testing format %s' % f['format_id'])
75689fe5 1941 path = self.get_output_path('temp')
1942 if not self._ensure_dir_exists(f'{path}/'):
1943 continue
1944 temp_file = tempfile.NamedTemporaryFile(suffix='.tmp', delete=False, dir=path or None)
9f1a1c36 1945 temp_file.close()
1946 try:
1947 success, _ = self.dl(temp_file.name, f, test=True)
8a82af35 1948 except (DownloadError, OSError, ValueError) + network_exceptions:
9f1a1c36 1949 success = False
1950 finally:
1951 if os.path.exists(temp_file.name):
1952 try:
1953 os.remove(temp_file.name)
1954 except OSError:
1955 self.report_warning('Unable to delete temporary file "%s"' % temp_file.name)
1956 if success:
1957 yield f
1958 else:
1959 self.to_screen('[info] Unable to download format %s. Skipping...' % f['format_id'])
1960
0017d9ad 1961 def _default_format_spec(self, info_dict, download=True):
0017d9ad 1962
af0f7428
S
1963 def can_merge():
1964 merger = FFmpegMergerPP(self)
1965 return merger.available and merger.can_merge()
1966
91ebc640 1967 prefer_best = (
b7b04c78 1968 not self.params.get('simulate')
91ebc640 1969 and download
1970 and (
1971 not can_merge()
21633673 1972 or info_dict.get('is_live') and not self.params.get('live_from_start')
bf1824b3 1973 or self.params['outtmpl']['default'] == '-'))
53ed7066 1974 compat = (
1975 prefer_best
1976 or self.params.get('allow_multiple_audio_streams', False)
8a82af35 1977 or 'format-spec' in self.params['compat_opts'])
91ebc640 1978
1979 return (
53ed7066 1980 'best/bestvideo+bestaudio' if prefer_best
1981 else 'bestvideo*+bestaudio/best' if not compat
91ebc640 1982 else 'bestvideo+bestaudio/best')
0017d9ad 1983
67134eab
JMF
1984 def build_format_selector(self, format_spec):
1985 def syntax_error(note, start):
1986 message = (
1987 'Invalid format specification: '
86e5f3ed 1988 '{}\n\t{}\n\t{}^'.format(note, format_spec, ' ' * start[1]))
67134eab
JMF
1989 return SyntaxError(message)
1990
1991 PICKFIRST = 'PICKFIRST'
1992 MERGE = 'MERGE'
1993 SINGLE = 'SINGLE'
0130afb7 1994 GROUP = 'GROUP'
67134eab
JMF
1995 FormatSelector = collections.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
1996
91ebc640 1997 allow_multiple_streams = {'audio': self.params.get('allow_multiple_audio_streams', False),
1998 'video': self.params.get('allow_multiple_video_streams', False)}
909d24dd 1999
9f1a1c36 2000 check_formats = self.params.get('check_formats') == 'selected'
e8e73840 2001
67134eab
JMF
2002 def _parse_filter(tokens):
2003 filter_parts = []
2004 for type, string, start, _, _ in tokens:
2005 if type == tokenize.OP and string == ']':
2006 return ''.join(filter_parts)
2007 else:
2008 filter_parts.append(string)
2009
232541df 2010 def _remove_unused_ops(tokens):
62b58c09
L
2011 # Remove operators that we don't use and join them with the surrounding strings.
2012 # E.g. 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
232541df
JMF
2013 ALLOWED_OPS = ('/', '+', ',', '(', ')')
2014 last_string, last_start, last_end, last_line = None, None, None, None
2015 for type, string, start, end, line in tokens:
2016 if type == tokenize.OP and string == '[':
2017 if last_string:
2018 yield tokenize.NAME, last_string, last_start, last_end, last_line
2019 last_string = None
2020 yield type, string, start, end, line
2021 # everything inside brackets will be handled by _parse_filter
2022 for type, string, start, end, line in tokens:
2023 yield type, string, start, end, line
2024 if type == tokenize.OP and string == ']':
2025 break
2026 elif type == tokenize.OP and string in ALLOWED_OPS:
2027 if last_string:
2028 yield tokenize.NAME, last_string, last_start, last_end, last_line
2029 last_string = None
2030 yield type, string, start, end, line
2031 elif type in [tokenize.NAME, tokenize.NUMBER, tokenize.OP]:
2032 if not last_string:
2033 last_string = string
2034 last_start = start
2035 last_end = end
2036 else:
2037 last_string += string
2038 if last_string:
2039 yield tokenize.NAME, last_string, last_start, last_end, last_line
2040
cf2ac6df 2041 def _parse_format_selection(tokens, inside_merge=False, inside_choice=False, inside_group=False):
67134eab
JMF
2042 selectors = []
2043 current_selector = None
2044 for type, string, start, _, _ in tokens:
2045 # ENCODING is only defined in python 3.x
2046 if type == getattr(tokenize, 'ENCODING', None):
2047 continue
2048 elif type in [tokenize.NAME, tokenize.NUMBER]:
2049 current_selector = FormatSelector(SINGLE, string, [])
2050 elif type == tokenize.OP:
cf2ac6df
JMF
2051 if string == ')':
2052 if not inside_group:
2053 # ')' will be handled by the parentheses group
2054 tokens.restore_last_token()
67134eab 2055 break
cf2ac6df 2056 elif inside_merge and string in ['/', ',']:
0130afb7
JMF
2057 tokens.restore_last_token()
2058 break
cf2ac6df
JMF
2059 elif inside_choice and string == ',':
2060 tokens.restore_last_token()
2061 break
2062 elif string == ',':
0a31a350
JMF
2063 if not current_selector:
2064 raise syntax_error('"," must follow a format selector', start)
67134eab
JMF
2065 selectors.append(current_selector)
2066 current_selector = None
2067 elif string == '/':
d96d604e
JMF
2068 if not current_selector:
2069 raise syntax_error('"/" must follow a format selector', start)
67134eab 2070 first_choice = current_selector
cf2ac6df 2071 second_choice = _parse_format_selection(tokens, inside_choice=True)
f5f4a27a 2072 current_selector = FormatSelector(PICKFIRST, (first_choice, second_choice), [])
67134eab
JMF
2073 elif string == '[':
2074 if not current_selector:
2075 current_selector = FormatSelector(SINGLE, 'best', [])
2076 format_filter = _parse_filter(tokens)
2077 current_selector.filters.append(format_filter)
0130afb7
JMF
2078 elif string == '(':
2079 if current_selector:
2080 raise syntax_error('Unexpected "("', start)
cf2ac6df
JMF
2081 group = _parse_format_selection(tokens, inside_group=True)
2082 current_selector = FormatSelector(GROUP, group, [])
67134eab 2083 elif string == '+':
d03cfdce 2084 if not current_selector:
2085 raise syntax_error('Unexpected "+"', start)
2086 selector_1 = current_selector
2087 selector_2 = _parse_format_selection(tokens, inside_merge=True)
2088 if not selector_2:
2089 raise syntax_error('Expected a selector', start)
2090 current_selector = FormatSelector(MERGE, (selector_1, selector_2), [])
67134eab 2091 else:
86e5f3ed 2092 raise syntax_error(f'Operator not recognized: "{string}"', start)
67134eab
JMF
2093 elif type == tokenize.ENDMARKER:
2094 break
2095 if current_selector:
2096 selectors.append(current_selector)
2097 return selectors
2098
f8d4ad9a 2099 def _merge(formats_pair):
2100 format_1, format_2 = formats_pair
2101
2102 formats_info = []
2103 formats_info.extend(format_1.get('requested_formats', (format_1,)))
2104 formats_info.extend(format_2.get('requested_formats', (format_2,)))
2105
2106 if not allow_multiple_streams['video'] or not allow_multiple_streams['audio']:
551f9388 2107 get_no_more = {'video': False, 'audio': False}
f8d4ad9a 2108 for (i, fmt_info) in enumerate(formats_info):
551f9388 2109 if fmt_info.get('acodec') == fmt_info.get('vcodec') == 'none':
2110 formats_info.pop(i)
2111 continue
2112 for aud_vid in ['audio', 'video']:
f8d4ad9a 2113 if not allow_multiple_streams[aud_vid] and fmt_info.get(aud_vid[0] + 'codec') != 'none':
2114 if get_no_more[aud_vid]:
2115 formats_info.pop(i)
f5510afe 2116 break
f8d4ad9a 2117 get_no_more[aud_vid] = True
2118
2119 if len(formats_info) == 1:
2120 return formats_info[0]
2121
2122 video_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('vcodec') != 'none']
2123 audio_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('acodec') != 'none']
2124
2125 the_only_video = video_fmts[0] if len(video_fmts) == 1 else None
2126 the_only_audio = audio_fmts[0] if len(audio_fmts) == 1 else None
2127
fc61aff4
LL
2128 output_ext = get_compatible_ext(
2129 vcodecs=[f.get('vcodec') for f in video_fmts],
2130 acodecs=[f.get('acodec') for f in audio_fmts],
2131 vexts=[f['ext'] for f in video_fmts],
2132 aexts=[f['ext'] for f in audio_fmts],
2133 preferences=(try_call(lambda: self.params['merge_output_format'].split('/'))
2134 or self.params.get('prefer_free_formats') and ('webm', 'mkv')))
f8d4ad9a 2135
975a0d0d 2136 filtered = lambda *keys: filter(None, (traverse_obj(fmt, *keys) for fmt in formats_info))
2137
f8d4ad9a 2138 new_dict = {
2139 'requested_formats': formats_info,
975a0d0d 2140 'format': '+'.join(filtered('format')),
2141 'format_id': '+'.join(filtered('format_id')),
f8d4ad9a 2142 'ext': output_ext,
975a0d0d 2143 'protocol': '+'.join(map(determine_protocol, formats_info)),
093a1710 2144 'language': '+'.join(orderedSet(filtered('language'))) or None,
2145 'format_note': '+'.join(orderedSet(filtered('format_note'))) or None,
2146 'filesize_approx': sum(filtered('filesize', 'filesize_approx')) or None,
975a0d0d 2147 'tbr': sum(filtered('tbr', 'vbr', 'abr')),
f8d4ad9a 2148 }
2149
2150 if the_only_video:
2151 new_dict.update({
2152 'width': the_only_video.get('width'),
2153 'height': the_only_video.get('height'),
2154 'resolution': the_only_video.get('resolution') or self.format_resolution(the_only_video),
2155 'fps': the_only_video.get('fps'),
49a57e70 2156 'dynamic_range': the_only_video.get('dynamic_range'),
f8d4ad9a 2157 'vcodec': the_only_video.get('vcodec'),
2158 'vbr': the_only_video.get('vbr'),
2159 'stretched_ratio': the_only_video.get('stretched_ratio'),
2160 })
2161
2162 if the_only_audio:
2163 new_dict.update({
2164 'acodec': the_only_audio.get('acodec'),
2165 'abr': the_only_audio.get('abr'),
975a0d0d 2166 'asr': the_only_audio.get('asr'),
b8ed0f15 2167 'audio_channels': the_only_audio.get('audio_channels')
f8d4ad9a 2168 })
2169
2170 return new_dict
2171
e8e73840 2172 def _check_formats(formats):
981052c9 2173 if not check_formats:
2174 yield from formats
b5ac45b1 2175 return
9f1a1c36 2176 yield from self._check_formats(formats)
e8e73840 2177
67134eab 2178 def _build_selector_function(selector):
909d24dd 2179 if isinstance(selector, list): # ,
67134eab
JMF
2180 fs = [_build_selector_function(s) for s in selector]
2181
317f7ab6 2182 def selector_function(ctx):
67134eab 2183 for f in fs:
981052c9 2184 yield from f(ctx)
67134eab 2185 return selector_function
909d24dd 2186
2187 elif selector.type == GROUP: # ()
0130afb7 2188 selector_function = _build_selector_function(selector.selector)
909d24dd 2189
2190 elif selector.type == PICKFIRST: # /
67134eab
JMF
2191 fs = [_build_selector_function(s) for s in selector.selector]
2192
317f7ab6 2193 def selector_function(ctx):
67134eab 2194 for f in fs:
317f7ab6 2195 picked_formats = list(f(ctx))
67134eab
JMF
2196 if picked_formats:
2197 return picked_formats
2198 return []
67134eab 2199
981052c9 2200 elif selector.type == MERGE: # +
2201 selector_1, selector_2 = map(_build_selector_function, selector.selector)
2202
2203 def selector_function(ctx):
adbc4ec4 2204 for pair in itertools.product(selector_1(ctx), selector_2(ctx)):
981052c9 2205 yield _merge(pair)
2206
909d24dd 2207 elif selector.type == SINGLE: # atom
598d185d 2208 format_spec = selector.selector or 'best'
909d24dd 2209
f8d4ad9a 2210 # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
909d24dd 2211 if format_spec == 'all':
2212 def selector_function(ctx):
9222c381 2213 yield from _check_formats(ctx['formats'][::-1])
f8d4ad9a 2214 elif format_spec == 'mergeall':
2215 def selector_function(ctx):
316f2650 2216 formats = list(_check_formats(
2217 f for f in ctx['formats'] if f.get('vcodec') != 'none' or f.get('acodec') != 'none'))
e01d6aa4 2218 if not formats:
2219 return
921b76ca 2220 merged_format = formats[-1]
2221 for f in formats[-2::-1]:
f8d4ad9a 2222 merged_format = _merge((merged_format, f))
2223 yield merged_format
909d24dd 2224
2225 else:
85e801a9 2226 format_fallback, seperate_fallback, format_reverse, format_idx = False, None, True, 1
eff63539 2227 mobj = re.match(
2228 r'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
2229 format_spec)
2230 if mobj is not None:
2231 format_idx = int_or_none(mobj.group('n'), default=1)
e8e73840 2232 format_reverse = mobj.group('bw')[0] == 'b'
eff63539 2233 format_type = (mobj.group('type') or [None])[0]
2234 not_format_type = {'v': 'a', 'a': 'v'}.get(format_type)
2235 format_modified = mobj.group('mod') is not None
909d24dd 2236
2237 format_fallback = not format_type and not format_modified # for b, w
8326b00a 2238 _filter_f = (
eff63539 2239 (lambda f: f.get('%scodec' % format_type) != 'none')
2240 if format_type and format_modified # bv*, ba*, wv*, wa*
2241 else (lambda f: f.get('%scodec' % not_format_type) == 'none')
2242 if format_type # bv, ba, wv, wa
2243 else (lambda f: f.get('vcodec') != 'none' and f.get('acodec') != 'none')
2244 if not format_modified # b, w
8326b00a 2245 else lambda f: True) # b*, w*
2246 filter_f = lambda f: _filter_f(f) and (
2247 f.get('vcodec') != 'none' or f.get('acodec') != 'none')
67134eab 2248 else:
48ee10ee 2249 if format_spec in self._format_selection_exts['audio']:
b11c04a8 2250 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none'
48ee10ee 2251 elif format_spec in self._format_selection_exts['video']:
b11c04a8 2252 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none' and f.get('vcodec') != 'none'
85e801a9 2253 seperate_fallback = lambda f: f.get('ext') == format_spec and f.get('vcodec') != 'none'
48ee10ee 2254 elif format_spec in self._format_selection_exts['storyboards']:
b11c04a8 2255 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') == 'none' and f.get('vcodec') == 'none'
2256 else:
b5ae35ee 2257 filter_f = lambda f: f.get('format_id') == format_spec # id
909d24dd 2258
2259 def selector_function(ctx):
2260 formats = list(ctx['formats'])
909d24dd 2261 matches = list(filter(filter_f, formats)) if filter_f is not None else formats
85e801a9 2262 if not matches:
2263 if format_fallback and ctx['incomplete_formats']:
2264 # for extractors with incomplete formats (audio only (soundcloud)
2265 # or video only (imgur)) best/worst will fallback to
2266 # best/worst {video,audio}-only format
2267 matches = formats
2268 elif seperate_fallback and not ctx['has_merged_format']:
2269 # for compatibility with youtube-dl when there is no pre-merged format
2270 matches = list(filter(seperate_fallback, formats))
981052c9 2271 matches = LazyList(_check_formats(matches[::-1 if format_reverse else 1]))
2272 try:
e8e73840 2273 yield matches[format_idx - 1]
4abea8ca 2274 except LazyList.IndexError:
981052c9 2275 return
083c9df9 2276
67134eab 2277 filters = [self._build_format_filter(f) for f in selector.filters]
083c9df9 2278
317f7ab6 2279 def final_selector(ctx):
adbc4ec4 2280 ctx_copy = dict(ctx)
67134eab 2281 for _filter in filters:
317f7ab6
S
2282 ctx_copy['formats'] = list(filter(_filter, ctx_copy['formats']))
2283 return selector_function(ctx_copy)
67134eab 2284 return final_selector
083c9df9 2285
0f06bcd7 2286 stream = io.BytesIO(format_spec.encode())
0130afb7 2287 try:
f9934b96 2288 tokens = list(_remove_unused_ops(tokenize.tokenize(stream.readline)))
0130afb7
JMF
2289 except tokenize.TokenError:
2290 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec)))
2291
86e5f3ed 2292 class TokenIterator:
0130afb7
JMF
2293 def __init__(self, tokens):
2294 self.tokens = tokens
2295 self.counter = 0
2296
2297 def __iter__(self):
2298 return self
2299
2300 def __next__(self):
2301 if self.counter >= len(self.tokens):
2302 raise StopIteration()
2303 value = self.tokens[self.counter]
2304 self.counter += 1
2305 return value
2306
2307 next = __next__
2308
2309 def restore_last_token(self):
2310 self.counter -= 1
2311
2312 parsed_selector = _parse_format_selection(iter(TokenIterator(tokens)))
67134eab 2313 return _build_selector_function(parsed_selector)
a9c58ad9 2314
e5660ee6 2315 def _calc_headers(self, info_dict):
8b7539d2 2316 res = merge_headers(self.params['http_headers'], info_dict.get('http_headers') or {})
e5660ee6 2317
c487cf00 2318 cookies = self._calc_cookies(info_dict['url'])
e5660ee6
JMF
2319 if cookies:
2320 res['Cookie'] = cookies
2321
0016b84e
S
2322 if 'X-Forwarded-For' not in res:
2323 x_forwarded_for_ip = info_dict.get('__x_forwarded_for_ip')
2324 if x_forwarded_for_ip:
2325 res['X-Forwarded-For'] = x_forwarded_for_ip
2326
e5660ee6
JMF
2327 return res
2328
c487cf00 2329 def _calc_cookies(self, url):
2330 pr = sanitized_Request(url)
e5660ee6 2331 self.cookiejar.add_cookie_header(pr)
662435f7 2332 return pr.get_header('Cookie')
e5660ee6 2333
9f1a1c36 2334 def _sort_thumbnails(self, thumbnails):
2335 thumbnails.sort(key=lambda t: (
2336 t.get('preference') if t.get('preference') is not None else -1,
2337 t.get('width') if t.get('width') is not None else -1,
2338 t.get('height') if t.get('height') is not None else -1,
2339 t.get('id') if t.get('id') is not None else '',
2340 t.get('url')))
2341
b0249bca 2342 def _sanitize_thumbnails(self, info_dict):
bc516a3f 2343 thumbnails = info_dict.get('thumbnails')
2344 if thumbnails is None:
2345 thumbnail = info_dict.get('thumbnail')
2346 if thumbnail:
2347 info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}]
9f1a1c36 2348 if not thumbnails:
2349 return
2350
2351 def check_thumbnails(thumbnails):
2352 for t in thumbnails:
2353 self.to_screen(f'[info] Testing thumbnail {t["id"]}')
2354 try:
2355 self.urlopen(HEADRequest(t['url']))
2356 except network_exceptions as err:
2357 self.to_screen(f'[info] Unable to connect to thumbnail {t["id"]} URL {t["url"]!r} - {err}. Skipping...')
2358 continue
2359 yield t
2360
2361 self._sort_thumbnails(thumbnails)
2362 for i, t in enumerate(thumbnails):
2363 if t.get('id') is None:
2364 t['id'] = '%d' % i
2365 if t.get('width') and t.get('height'):
2366 t['resolution'] = '%dx%d' % (t['width'], t['height'])
2367 t['url'] = sanitize_url(t['url'])
2368
2369 if self.params.get('check_formats') is True:
282f5709 2370 info_dict['thumbnails'] = LazyList(check_thumbnails(thumbnails[::-1]), reverse=True)
9f1a1c36 2371 else:
2372 info_dict['thumbnails'] = thumbnails
bc516a3f 2373
03f83004
LNO
2374 def _fill_common_fields(self, info_dict, is_video=True):
2375 # TODO: move sanitization here
2376 if is_video:
2377 # playlists are allowed to lack "title"
d4736fdb 2378 title = info_dict.get('title', NO_DEFAULT)
2379 if title is NO_DEFAULT:
03f83004
LNO
2380 raise ExtractorError('Missing "title" field in extractor result',
2381 video_id=info_dict['id'], ie=info_dict['extractor'])
d4736fdb 2382 info_dict['fulltitle'] = title
2383 if not title:
2384 if title == '':
2385 self.write_debug('Extractor gave empty title. Creating a generic title')
2386 else:
2387 self.report_warning('Extractor failed to obtain "title". Creating a generic title instead')
1d485a1a 2388 info_dict['title'] = f'{info_dict["extractor"].replace(":", "-")} video #{info_dict["id"]}'
03f83004
LNO
2389
2390 if info_dict.get('duration') is not None:
2391 info_dict['duration_string'] = formatSeconds(info_dict['duration'])
2392
2393 for ts_key, date_key in (
2394 ('timestamp', 'upload_date'),
2395 ('release_timestamp', 'release_date'),
2396 ('modified_timestamp', 'modified_date'),
2397 ):
2398 if info_dict.get(date_key) is None and info_dict.get(ts_key) is not None:
2399 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
2400 # see http://bugs.python.org/issue1646728)
19a03940 2401 with contextlib.suppress(ValueError, OverflowError, OSError):
03f83004
LNO
2402 upload_date = datetime.datetime.utcfromtimestamp(info_dict[ts_key])
2403 info_dict[date_key] = upload_date.strftime('%Y%m%d')
03f83004
LNO
2404
2405 live_keys = ('is_live', 'was_live')
2406 live_status = info_dict.get('live_status')
2407 if live_status is None:
2408 for key in live_keys:
2409 if info_dict.get(key) is False:
2410 continue
2411 if info_dict.get(key):
2412 live_status = key
2413 break
2414 if all(info_dict.get(key) is False for key in live_keys):
2415 live_status = 'not_live'
2416 if live_status:
2417 info_dict['live_status'] = live_status
2418 for key in live_keys:
2419 if info_dict.get(key) is None:
2420 info_dict[key] = (live_status == key)
2421
2422 # Auto generate title fields corresponding to the *_number fields when missing
2423 # in order to always have clean titles. This is very common for TV series.
2424 for field in ('chapter', 'season', 'episode'):
2425 if info_dict.get('%s_number' % field) is not None and not info_dict.get(field):
2426 info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field])
2427
415f8d51 2428 def _raise_pending_errors(self, info):
2429 err = info.pop('__pending_error', None)
2430 if err:
2431 self.report_error(err, tb=False)
2432
dd82ffea
JMF
2433 def process_video_result(self, info_dict, download=True):
2434 assert info_dict.get('_type', 'video') == 'video'
9c906919 2435 self._num_videos += 1
dd82ffea 2436
bec1fad2 2437 if 'id' not in info_dict:
fc08bdd6 2438 raise ExtractorError('Missing "id" field in extractor result', ie=info_dict['extractor'])
2439 elif not info_dict.get('id'):
2440 raise ExtractorError('Extractor failed to obtain "id"', ie=info_dict['extractor'])
455a15e2 2441
c9969434
S
2442 def report_force_conversion(field, field_not, conversion):
2443 self.report_warning(
2444 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
2445 % (field, field_not, conversion))
2446
2447 def sanitize_string_field(info, string_field):
2448 field = info.get(string_field)
14f25df2 2449 if field is None or isinstance(field, str):
c9969434
S
2450 return
2451 report_force_conversion(string_field, 'a string', 'string')
14f25df2 2452 info[string_field] = str(field)
c9969434
S
2453
2454 def sanitize_numeric_fields(info):
2455 for numeric_field in self._NUMERIC_FIELDS:
2456 field = info.get(numeric_field)
f9934b96 2457 if field is None or isinstance(field, (int, float)):
c9969434
S
2458 continue
2459 report_force_conversion(numeric_field, 'numeric', 'int')
2460 info[numeric_field] = int_or_none(field)
2461
2462 sanitize_string_field(info_dict, 'id')
2463 sanitize_numeric_fields(info_dict)
3975b4d2 2464 if info_dict.get('section_end') and info_dict.get('section_start') is not None:
2465 info_dict['duration'] = round(info_dict['section_end'] - info_dict['section_start'], 3)
4c3f8c3f 2466 if (info_dict.get('duration') or 0) <= 0 and info_dict.pop('duration', None):
50e93e03 2467 self.report_warning('"duration" field is negative, there is an error in extractor')
be6217b2 2468
9eef7c4e 2469 chapters = info_dict.get('chapters') or []
a3976e07 2470 if chapters and chapters[0].get('start_time'):
2471 chapters.insert(0, {'start_time': 0})
2472
9eef7c4e 2473 dummy_chapter = {'end_time': 0, 'start_time': info_dict.get('duration')}
a3976e07 2474 for idx, (prev, current, next_) in enumerate(zip(
2475 (dummy_chapter, *chapters), chapters, (*chapters[1:], dummy_chapter)), 1):
9eef7c4e 2476 if current.get('start_time') is None:
2477 current['start_time'] = prev.get('end_time')
2478 if not current.get('end_time'):
2479 current['end_time'] = next_.get('start_time')
a3976e07 2480 if not current.get('title'):
2481 current['title'] = f'<Untitled Chapter {idx}>'
9eef7c4e 2482
dd82ffea
JMF
2483 if 'playlist' not in info_dict:
2484 # It isn't part of a playlist
2485 info_dict['playlist'] = None
2486 info_dict['playlist_index'] = None
2487
bc516a3f 2488 self._sanitize_thumbnails(info_dict)
d5519808 2489
536a55da 2490 thumbnail = info_dict.get('thumbnail')
bc516a3f 2491 thumbnails = info_dict.get('thumbnails')
536a55da
S
2492 if thumbnail:
2493 info_dict['thumbnail'] = sanitize_url(thumbnail)
2494 elif thumbnails:
d5519808
PH
2495 info_dict['thumbnail'] = thumbnails[-1]['url']
2496
ae30b840 2497 if info_dict.get('display_id') is None and 'id' in info_dict:
0afef30b
PH
2498 info_dict['display_id'] = info_dict['id']
2499
03f83004 2500 self._fill_common_fields(info_dict)
33d2fc2f 2501
05108a49
S
2502 for cc_kind in ('subtitles', 'automatic_captions'):
2503 cc = info_dict.get(cc_kind)
2504 if cc:
2505 for _, subtitle in cc.items():
2506 for subtitle_format in subtitle:
2507 if subtitle_format.get('url'):
2508 subtitle_format['url'] = sanitize_url(subtitle_format['url'])
2509 if subtitle_format.get('ext') is None:
2510 subtitle_format['ext'] = determine_ext(subtitle_format['url']).lower()
2511
2512 automatic_captions = info_dict.get('automatic_captions')
4bba3716 2513 subtitles = info_dict.get('subtitles')
4bba3716 2514
360e1ca5 2515 info_dict['requested_subtitles'] = self.process_subtitles(
05108a49 2516 info_dict['id'], subtitles, automatic_captions)
a504ced0 2517
dd82ffea
JMF
2518 if info_dict.get('formats') is None:
2519 # There's only one format available
2520 formats = [info_dict]
2521 else:
2522 formats = info_dict['formats']
2523
0a5a191a 2524 # or None ensures --clean-infojson removes it
2525 info_dict['_has_drm'] = any(f.get('has_drm') for f in formats) or None
88acdbc2 2526 if not self.params.get('allow_unplayable_formats'):
2527 formats = [f for f in formats if not f.get('has_drm')]
7356a444 2528 if info_dict['_has_drm'] and formats and all(
c0b6e5c7 2529 f.get('acodec') == f.get('vcodec') == 'none' for f in formats):
2530 self.report_warning(
2531 'This video is DRM protected and only images are available for download. '
2532 'Use --list-formats to see them')
88acdbc2 2533
319b6059 2534 get_from_start = not info_dict.get('is_live') or bool(self.params.get('live_from_start'))
2535 if not get_from_start:
2536 info_dict['title'] += ' ' + datetime.datetime.now().strftime('%Y-%m-%d %H:%M')
2537 if info_dict.get('is_live') and formats:
adbc4ec4 2538 formats = [f for f in formats if bool(f.get('is_from_start')) == get_from_start]
319b6059 2539 if get_from_start and not formats:
a44ca5a4 2540 self.raise_no_formats(info_dict, msg=(
2541 '--live-from-start is passed, but there are no formats that can be downloaded from the start. '
2542 'If you want to download from the current time, use --no-live-from-start'))
adbc4ec4 2543
73af5cc8
S
2544 def is_wellformed(f):
2545 url = f.get('url')
a5ac0c47 2546 if not url:
73af5cc8
S
2547 self.report_warning(
2548 '"url" field is missing or empty - skipping format, '
2549 'there is an error in extractor')
a5ac0c47
S
2550 return False
2551 if isinstance(url, bytes):
2552 sanitize_string_field(f, 'url')
2553 return True
73af5cc8
S
2554
2555 # Filter out malformed formats for better extraction robustness
1ac7f461 2556 formats = list(filter(is_wellformed, formats or []))
2557
2558 if not formats:
2559 self.raise_no_formats(info_dict)
73af5cc8 2560
181c7053
S
2561 formats_dict = {}
2562
dd82ffea 2563 # We check that all the formats have the format and format_id fields
db95dc13 2564 for i, format in enumerate(formats):
c9969434
S
2565 sanitize_string_field(format, 'format_id')
2566 sanitize_numeric_fields(format)
dcf77cf1 2567 format['url'] = sanitize_url(format['url'])
e74e3b63 2568 if not format.get('format_id'):
14f25df2 2569 format['format_id'] = str(i)
e2effb08
S
2570 else:
2571 # Sanitize format_id from characters used in format selector expression
ec85ded8 2572 format['format_id'] = re.sub(r'[\s,/+\[\]()]', '_', format['format_id'])
181c7053
S
2573 format_id = format['format_id']
2574 if format_id not in formats_dict:
2575 formats_dict[format_id] = []
2576 formats_dict[format_id].append(format)
2577
2578 # Make sure all formats have unique format_id
03b4de72 2579 common_exts = set(itertools.chain(*self._format_selection_exts.values()))
181c7053 2580 for format_id, ambiguous_formats in formats_dict.items():
48ee10ee 2581 ambigious_id = len(ambiguous_formats) > 1
2582 for i, format in enumerate(ambiguous_formats):
2583 if ambigious_id:
181c7053 2584 format['format_id'] = '%s-%d' % (format_id, i)
48ee10ee 2585 if format.get('ext') is None:
2586 format['ext'] = determine_ext(format['url']).lower()
2587 # Ensure there is no conflict between id and ext in format selection
2588 # See https://github.com/yt-dlp/yt-dlp/issues/1282
2589 if format['format_id'] != format['ext'] and format['format_id'] in common_exts:
2590 format['format_id'] = 'f%s' % format['format_id']
181c7053
S
2591
2592 for i, format in enumerate(formats):
8c51aa65 2593 if format.get('format') is None:
6febd1c1 2594 format['format'] = '{id} - {res}{note}'.format(
8c51aa65
JMF
2595 id=format['format_id'],
2596 res=self.format_resolution(format),
b868936c 2597 note=format_field(format, 'format_note', ' (%s)'),
8c51aa65 2598 )
6f0be937 2599 if format.get('protocol') is None:
b5559424 2600 format['protocol'] = determine_protocol(format)
239df021 2601 if format.get('resolution') is None:
2602 format['resolution'] = self.format_resolution(format, default=None)
176f1866 2603 if format.get('dynamic_range') is None and format.get('vcodec') != 'none':
2604 format['dynamic_range'] = 'SDR'
f2fe69c7 2605 if (info_dict.get('duration') and format.get('tbr')
2606 and not format.get('filesize') and not format.get('filesize_approx')):
56ba69e4 2607 format['filesize_approx'] = int(info_dict['duration'] * format['tbr'] * (1024 / 8))
f2fe69c7 2608
e5660ee6
JMF
2609 # Add HTTP headers, so that external programs can use them from the
2610 # json output
2611 full_format_info = info_dict.copy()
2612 full_format_info.update(format)
2613 format['http_headers'] = self._calc_headers(full_format_info)
0016b84e
S
2614 # Remove private housekeeping stuff
2615 if '__x_forwarded_for_ip' in info_dict:
2616 del info_dict['__x_forwarded_for_ip']
dd82ffea 2617
9f1a1c36 2618 if self.params.get('check_formats') is True:
282f5709 2619 formats = LazyList(self._check_formats(formats[::-1]), reverse=True)
9f1a1c36 2620
88acdbc2 2621 if not formats or formats[0] is not info_dict:
b3d9ef88
JMF
2622 # only set the 'formats' fields if the original info_dict list them
2623 # otherwise we end up with a circular reference, the first (and unique)
f89197d7 2624 # element in the 'formats' field in info_dict is info_dict itself,
dfb1b146 2625 # which can't be exported to json
b3d9ef88 2626 info_dict['formats'] = formats
4ec82a72 2627
2628 info_dict, _ = self.pre_process(info_dict)
2629
6db9c4d5 2630 if self._match_entry(info_dict, incomplete=self._format_fields) is not None:
09b49e1f 2631 return info_dict
2632
2633 self.post_extract(info_dict)
2634 info_dict, _ = self.pre_process(info_dict, 'after_filter')
2635
093a1710 2636 # The pre-processors may have modified the formats
2637 formats = info_dict.get('formats', [info_dict])
2638
fa9f30b8 2639 list_only = self.params.get('simulate') is None and (
2640 self.params.get('list_thumbnails') or self.params.get('listformats') or self.params.get('listsubtitles'))
2641 interactive_format_selection = not list_only and self.format_selector == '-'
b7b04c78 2642 if self.params.get('list_thumbnails'):
2643 self.list_thumbnails(info_dict)
b7b04c78 2644 if self.params.get('listsubtitles'):
2645 if 'automatic_captions' in info_dict:
2646 self.list_subtitles(
2647 info_dict['id'], automatic_captions, 'automatic captions')
2648 self.list_subtitles(info_dict['id'], subtitles, 'subtitles')
fa9f30b8 2649 if self.params.get('listformats') or interactive_format_selection:
b69fd25c 2650 self.list_formats(info_dict)
169dbde9 2651 if list_only:
b7b04c78 2652 # Without this printing, -F --print-json will not work
169dbde9 2653 self.__forced_printings(info_dict, self.prepare_filename(info_dict), incomplete=True)
c487cf00 2654 return info_dict
bfaae0a7 2655
187986a8 2656 format_selector = self.format_selector
2657 if format_selector is None:
0017d9ad 2658 req_format = self._default_format_spec(info_dict, download=download)
0760b0a7 2659 self.write_debug('Default format spec: %s' % req_format)
187986a8 2660 format_selector = self.build_format_selector(req_format)
317f7ab6 2661
fa9f30b8 2662 while True:
2663 if interactive_format_selection:
2664 req_format = input(
2665 self._format_screen('\nEnter format selector: ', self.Styles.EMPHASIS))
2666 try:
2667 format_selector = self.build_format_selector(req_format)
2668 except SyntaxError as err:
2669 self.report_error(err, tb=False, is_error=False)
2670 continue
2671
85e801a9 2672 formats_to_download = list(format_selector({
fa9f30b8 2673 'formats': formats,
85e801a9 2674 'has_merged_format': any('none' not in (f.get('acodec'), f.get('vcodec')) for f in formats),
2675 'incomplete_formats': (
2676 # All formats are video-only or
2677 all(f.get('vcodec') != 'none' and f.get('acodec') == 'none' for f in formats)
2678 # all formats are audio-only
2679 or all(f.get('vcodec') == 'none' and f.get('acodec') != 'none' for f in formats)),
2680 }))
fa9f30b8 2681 if interactive_format_selection and not formats_to_download:
2682 self.report_error('Requested format is not available', tb=False, is_error=False)
2683 continue
2684 break
317f7ab6 2685
dd82ffea 2686 if not formats_to_download:
b7da73eb 2687 if not self.params.get('ignore_no_formats_error'):
c0b6e5c7 2688 raise ExtractorError(
2689 'Requested format is not available. Use --list-formats for a list of available formats',
2690 expected=True, video_id=info_dict['id'], ie=info_dict['extractor'])
b62fa6d7 2691 self.report_warning('Requested format is not available')
2692 # Process what we can, even without any available formats.
2693 formats_to_download = [{}]
a13e6848 2694
5ec1b6b7 2695 requested_ranges = self.params.get('download_ranges')
2696 if requested_ranges:
2697 requested_ranges = tuple(requested_ranges(info_dict, self))
2698
2699 best_format, downloaded_formats = formats_to_download[-1], []
b62fa6d7 2700 if download:
2701 if best_format:
5ec1b6b7 2702 def to_screen(*msg):
2703 self.to_screen(f'[info] {info_dict["id"]}: {" ".join(", ".join(variadic(m)) for m in msg)}')
2704
2705 to_screen(f'Downloading {len(formats_to_download)} format(s):',
2706 (f['format_id'] for f in formats_to_download))
2707 if requested_ranges:
2708 to_screen(f'Downloading {len(requested_ranges)} time ranges:',
2709 (f'{int(c["start_time"])}-{int(c["end_time"])}' for c in requested_ranges))
a13e6848 2710 max_downloads_reached = False
5ec1b6b7 2711
2712 for fmt, chapter in itertools.product(formats_to_download, requested_ranges or [{}]):
2713 new_info = self._copy_infodict(info_dict)
b7da73eb 2714 new_info.update(fmt)
3975b4d2 2715 offset, duration = info_dict.get('section_start') or 0, info_dict.get('duration') or float('inf')
2716 if chapter or offset:
5ec1b6b7 2717 new_info.update({
3975b4d2 2718 'section_start': offset + chapter.get('start_time', 0),
bc401608 2719 'section_end': offset + min(chapter.get('end_time', duration), duration),
5ec1b6b7 2720 'section_title': chapter.get('title'),
2721 'section_number': chapter.get('index'),
2722 })
2723 downloaded_formats.append(new_info)
a13e6848 2724 try:
2725 self.process_info(new_info)
2726 except MaxDownloadsReached:
2727 max_downloads_reached = True
415f8d51 2728 self._raise_pending_errors(new_info)
f46e2f9d 2729 # Remove copied info
2730 for key, val in tuple(new_info.items()):
2731 if info_dict.get(key) == val:
2732 new_info.pop(key)
a13e6848 2733 if max_downloads_reached:
2734 break
ebed8b37 2735
5ec1b6b7 2736 write_archive = {f.get('__write_download_archive', False) for f in downloaded_formats}
a13e6848 2737 assert write_archive.issubset({True, False, 'ignore'})
2738 if True in write_archive and False not in write_archive:
2739 self.record_download_archive(info_dict)
be72c624 2740
5ec1b6b7 2741 info_dict['requested_downloads'] = downloaded_formats
ed5835b4 2742 info_dict = self.run_all_pps('after_video', info_dict)
a13e6848 2743 if max_downloads_reached:
2744 raise MaxDownloadsReached()
ebed8b37 2745
49a57e70 2746 # We update the info dict with the selected best quality format (backwards compatibility)
be72c624 2747 info_dict.update(best_format)
dd82ffea
JMF
2748 return info_dict
2749
98c70d6f 2750 def process_subtitles(self, video_id, normal_subtitles, automatic_captions):
a504ced0 2751 """Select the requested subtitles and their format"""
d8a58ddc 2752 available_subs, normal_sub_langs = {}, []
98c70d6f
JMF
2753 if normal_subtitles and self.params.get('writesubtitles'):
2754 available_subs.update(normal_subtitles)
d8a58ddc 2755 normal_sub_langs = tuple(normal_subtitles.keys())
98c70d6f
JMF
2756 if automatic_captions and self.params.get('writeautomaticsub'):
2757 for lang, cap_info in automatic_captions.items():
360e1ca5
JMF
2758 if lang not in available_subs:
2759 available_subs[lang] = cap_info
2760
d2c8aadf 2761 if not available_subs or (
2762 not self.params.get('writesubtitles')
2763 and not self.params.get('writeautomaticsub')):
4d171848 2764 return None
a504ced0 2765
d8a58ddc 2766 all_sub_langs = tuple(available_subs.keys())
a504ced0 2767 if self.params.get('allsubtitles', False):
c32b0aab 2768 requested_langs = all_sub_langs
2769 elif self.params.get('subtitleslangs', False):
5314b521 2770 try:
2771 requested_langs = orderedSet_from_options(
2772 self.params.get('subtitleslangs'), {'all': all_sub_langs}, use_regex=True)
2773 except re.error as e:
2774 raise ValueError(f'Wrong regex for subtitlelangs: {e.pattern}')
d8a58ddc 2775 elif normal_sub_langs:
2776 requested_langs = ['en'] if 'en' in normal_sub_langs else normal_sub_langs[:1]
a504ced0 2777 else:
d8a58ddc 2778 requested_langs = ['en'] if 'en' in all_sub_langs else all_sub_langs[:1]
ad3dc496 2779 if requested_langs:
d2c8aadf 2780 self.to_screen(f'[info] {video_id}: Downloading subtitles: {", ".join(requested_langs)}')
a504ced0
JMF
2781
2782 formats_query = self.params.get('subtitlesformat', 'best')
2783 formats_preference = formats_query.split('/') if formats_query else []
2784 subs = {}
2785 for lang in requested_langs:
2786 formats = available_subs.get(lang)
2787 if formats is None:
86e5f3ed 2788 self.report_warning(f'{lang} subtitles not available for {video_id}')
a504ced0 2789 continue
a504ced0
JMF
2790 for ext in formats_preference:
2791 if ext == 'best':
2792 f = formats[-1]
2793 break
2794 matches = list(filter(lambda f: f['ext'] == ext, formats))
2795 if matches:
2796 f = matches[-1]
2797 break
2798 else:
2799 f = formats[-1]
2800 self.report_warning(
2801 'No subtitle format found matching "%s" for language %s, '
2802 'using %s' % (formats_query, lang, f['ext']))
2803 subs[lang] = f
2804 return subs
2805
bb66c247 2806 def _forceprint(self, key, info_dict):
2807 if info_dict is None:
2808 return
2809 info_copy = info_dict.copy()
2810 info_copy['formats_table'] = self.render_formats_table(info_dict)
2811 info_copy['thumbnails_table'] = self.render_thumbnails_table(info_dict)
2812 info_copy['subtitles_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('subtitles'))
2813 info_copy['automatic_captions_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('automatic_captions'))
2814
2815 def format_tmpl(tmpl):
07a1250e 2816 mobj = re.fullmatch(r'([\w.:,-]|(?P<dict>{[\w.:,-]+}))+=', tmpl)
2817 if not mobj:
2818 return tmpl
2819 elif not mobj.group('dict'):
2820 return '\n'.join(f'{f} = %({f})r' for f in tmpl[:-1].split(','))
2821 tmpl = f'.{tmpl[:-1]}' if tmpl.startswith('{') else tmpl[:-1]
2822 return f'{tmpl} = %({tmpl})#j'
8130779d 2823
bb66c247 2824 for tmpl in self.params['forceprint'].get(key, []):
2825 self.to_stdout(self.evaluate_outtmpl(format_tmpl(tmpl), info_copy))
2826
2827 for tmpl, file_tmpl in self.params['print_to_file'].get(key, []):
5127e92a 2828 filename = self.prepare_filename(info_dict, outtmpl=file_tmpl)
bb66c247 2829 tmpl = format_tmpl(tmpl)
2830 self.to_screen(f'[info] Writing {tmpl!r} to: {filename}')
8d93e69d 2831 if self._ensure_dir_exists(filename):
86e5f3ed 2832 with open(filename, 'a', encoding='utf-8') as f:
8d93e69d 2833 f.write(self.evaluate_outtmpl(tmpl, info_copy) + '\n')
ca30f449 2834
d06daf23 2835 def __forced_printings(self, info_dict, filename, incomplete):
53c18592 2836 def print_mandatory(field, actual_field=None):
2837 if actual_field is None:
2838 actual_field = field
d06daf23 2839 if (self.params.get('force%s' % field, False)
53c18592 2840 and (not incomplete or info_dict.get(actual_field) is not None)):
2841 self.to_stdout(info_dict[actual_field])
d06daf23
S
2842
2843 def print_optional(field):
2844 if (self.params.get('force%s' % field, False)
2845 and info_dict.get(field) is not None):
2846 self.to_stdout(info_dict[field])
2847
53c18592 2848 info_dict = info_dict.copy()
2849 if filename is not None:
2850 info_dict['filename'] = filename
2851 if info_dict.get('requested_formats') is not None:
2852 # For RTMP URLs, also include the playpath
2853 info_dict['urls'] = '\n'.join(f['url'] + f.get('play_path', '') for f in info_dict['requested_formats'])
10331a26 2854 elif info_dict.get('url'):
53c18592 2855 info_dict['urls'] = info_dict['url'] + info_dict.get('play_path', '')
2856
bb66c247 2857 if (self.params.get('forcejson')
2858 or self.params['forceprint'].get('video')
2859 or self.params['print_to_file'].get('video')):
2b8a2973 2860 self.post_extract(info_dict)
bb66c247 2861 self._forceprint('video', info_dict)
53c18592 2862
d06daf23
S
2863 print_mandatory('title')
2864 print_mandatory('id')
53c18592 2865 print_mandatory('url', 'urls')
d06daf23
S
2866 print_optional('thumbnail')
2867 print_optional('description')
53c18592 2868 print_optional('filename')
b868936c 2869 if self.params.get('forceduration') and info_dict.get('duration') is not None:
d06daf23
S
2870 self.to_stdout(formatSeconds(info_dict['duration']))
2871 print_mandatory('format')
53c18592 2872
2b8a2973 2873 if self.params.get('forcejson'):
6e84b215 2874 self.to_stdout(json.dumps(self.sanitize_info(info_dict)))
d06daf23 2875
e8e73840 2876 def dl(self, name, info, subtitle=False, test=False):
88acdbc2 2877 if not info.get('url'):
1151c407 2878 self.raise_no_formats(info, True)
e8e73840 2879
2880 if test:
2881 verbose = self.params.get('verbose')
2882 params = {
2883 'test': True,
a169858f 2884 'quiet': self.params.get('quiet') or not verbose,
e8e73840 2885 'verbose': verbose,
2886 'noprogress': not verbose,
2887 'nopart': True,
2888 'skip_unavailable_fragments': False,
2889 'keep_fragments': False,
2890 'overwrites': True,
2891 '_no_ytdl_file': True,
2892 }
2893 else:
2894 params = self.params
96fccc10 2895 fd = get_suitable_downloader(info, params, to_stdout=(name == '-'))(self, params)
e8e73840 2896 if not test:
2897 for ph in self._progress_hooks:
2898 fd.add_progress_hook(ph)
42676437
M
2899 urls = '", "'.join(
2900 (f['url'].split(',')[0] + ',<data>' if f['url'].startswith('data:') else f['url'])
2901 for f in info.get('requested_formats', []) or [info])
3a408f9d 2902 self.write_debug(f'Invoking {fd.FD_NAME} downloader on "{urls}"')
03b4de72 2903
adbc4ec4
THD
2904 # Note: Ideally info should be a deep-copied so that hooks cannot modify it.
2905 # But it may contain objects that are not deep-copyable
2906 new_info = self._copy_infodict(info)
e8e73840 2907 if new_info.get('http_headers') is None:
2908 new_info['http_headers'] = self._calc_headers(new_info)
2909 return fd.download(name, new_info, subtitle)
2910
e04938ab 2911 def existing_file(self, filepaths, *, default_overwrite=True):
2912 existing_files = list(filter(os.path.exists, orderedSet(filepaths)))
2913 if existing_files and not self.params.get('overwrites', default_overwrite):
2914 return existing_files[0]
2915
2916 for file in existing_files:
2917 self.report_file_delete(file)
2918 os.remove(file)
2919 return None
2920
8222d8de 2921 def process_info(self, info_dict):
09b49e1f 2922 """Process a single resolved IE result. (Modifies it in-place)"""
8222d8de
JMF
2923
2924 assert info_dict.get('_type', 'video') == 'video'
f46e2f9d 2925 original_infodict = info_dict
fd288278 2926
4513a41a 2927 if 'format' not in info_dict and 'ext' in info_dict:
8222d8de
JMF
2928 info_dict['format'] = info_dict['ext']
2929
09b49e1f 2930 # This is mostly just for backward compatibility of process_info
2931 # As a side-effect, this allows for format-specific filters
c77495e3 2932 if self._match_entry(info_dict) is not None:
9e907ebd 2933 info_dict['__write_download_archive'] = 'ignore'
8222d8de
JMF
2934 return
2935
09b49e1f 2936 # Does nothing under normal operation - for backward compatibility of process_info
277d6ff5 2937 self.post_extract(info_dict)
0c14d66a 2938 self._num_downloads += 1
8222d8de 2939
dcf64d43 2940 # info_dict['_filename'] needs to be set for backward compatibility
de6000d9 2941 info_dict['_filename'] = full_filename = self.prepare_filename(info_dict, warn=True)
2942 temp_filename = self.prepare_filename(info_dict, 'temp')
0202b52a 2943 files_to_move = {}
8222d8de
JMF
2944
2945 # Forced printings
4513a41a 2946 self.__forced_printings(info_dict, full_filename, incomplete=('format' not in info_dict))
8222d8de 2947
ca6d59d2 2948 def check_max_downloads():
2949 if self._num_downloads >= float(self.params.get('max_downloads') or 'inf'):
2950 raise MaxDownloadsReached()
2951
b7b04c78 2952 if self.params.get('simulate'):
9e907ebd 2953 info_dict['__write_download_archive'] = self.params.get('force_write_download_archive')
ca6d59d2 2954 check_max_downloads()
8222d8de
JMF
2955 return
2956
de6000d9 2957 if full_filename is None:
8222d8de 2958 return
e92caff5 2959 if not self._ensure_dir_exists(encodeFilename(full_filename)):
0202b52a 2960 return
e92caff5 2961 if not self._ensure_dir_exists(encodeFilename(temp_filename)):
8222d8de
JMF
2962 return
2963
80c03fa9 2964 if self._write_description('video', info_dict,
2965 self.prepare_filename(info_dict, 'description')) is None:
2966 return
2967
2968 sub_files = self._write_subtitles(info_dict, temp_filename)
2969 if sub_files is None:
2970 return
2971 files_to_move.update(dict(sub_files))
2972
2973 thumb_files = self._write_thumbnails(
2974 'video', info_dict, temp_filename, self.prepare_filename(info_dict, 'thumbnail'))
2975 if thumb_files is None:
2976 return
2977 files_to_move.update(dict(thumb_files))
8222d8de 2978
80c03fa9 2979 infofn = self.prepare_filename(info_dict, 'infojson')
2980 _infojson_written = self._write_info_json('video', info_dict, infofn)
2981 if _infojson_written:
dac5df5a 2982 info_dict['infojson_filename'] = infofn
e75bb0d6 2983 # For backward compatibility, even though it was a private field
80c03fa9 2984 info_dict['__infojson_filename'] = infofn
2985 elif _infojson_written is None:
2986 return
2987
2988 # Note: Annotations are deprecated
2989 annofn = None
1fb07d10 2990 if self.params.get('writeannotations', False):
de6000d9 2991 annofn = self.prepare_filename(info_dict, 'annotation')
80c03fa9 2992 if annofn:
e92caff5 2993 if not self._ensure_dir_exists(encodeFilename(annofn)):
0202b52a 2994 return
0c3d0f51 2995 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(annofn)):
6febd1c1 2996 self.to_screen('[info] Video annotations are already present')
ffddb112
RA
2997 elif not info_dict.get('annotations'):
2998 self.report_warning('There are no annotations to write.')
7b6fefc9
PH
2999 else:
3000 try:
6febd1c1 3001 self.to_screen('[info] Writing video annotations to: ' + annofn)
86e5f3ed 3002 with open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
7b6fefc9
PH
3003 annofile.write(info_dict['annotations'])
3004 except (KeyError, TypeError):
6febd1c1 3005 self.report_warning('There are no annotations to write.')
86e5f3ed 3006 except OSError:
6febd1c1 3007 self.report_error('Cannot write annotations file: ' + annofn)
7b6fefc9 3008 return
1fb07d10 3009
732044af 3010 # Write internet shortcut files
08438d2c 3011 def _write_link_file(link_type):
60f3e995 3012 url = try_get(info_dict['webpage_url'], iri_to_uri)
3013 if not url:
3014 self.report_warning(
3015 f'Cannot write internet shortcut file because the actual URL of "{info_dict["webpage_url"]}" is unknown')
3016 return True
08438d2c 3017 linkfn = replace_extension(self.prepare_filename(info_dict, 'link'), link_type, info_dict.get('ext'))
0e6b018a
Z
3018 if not self._ensure_dir_exists(encodeFilename(linkfn)):
3019 return False
10e3742e 3020 if self.params.get('overwrites', True) and os.path.exists(encodeFilename(linkfn)):
08438d2c 3021 self.to_screen(f'[info] Internet shortcut (.{link_type}) is already present')
3022 return True
3023 try:
3024 self.to_screen(f'[info] Writing internet shortcut (.{link_type}) to: {linkfn}')
86e5f3ed 3025 with open(encodeFilename(to_high_limit_path(linkfn)), 'w', encoding='utf-8',
3026 newline='\r\n' if link_type == 'url' else '\n') as linkfile:
60f3e995 3027 template_vars = {'url': url}
08438d2c 3028 if link_type == 'desktop':
3029 template_vars['filename'] = linkfn[:-(len(link_type) + 1)]
3030 linkfile.write(LINK_TEMPLATES[link_type] % template_vars)
86e5f3ed 3031 except OSError:
08438d2c 3032 self.report_error(f'Cannot write internet shortcut {linkfn}')
3033 return False
732044af 3034 return True
3035
08438d2c 3036 write_links = {
3037 'url': self.params.get('writeurllink'),
3038 'webloc': self.params.get('writewebloclink'),
3039 'desktop': self.params.get('writedesktoplink'),
3040 }
3041 if self.params.get('writelink'):
3042 link_type = ('webloc' if sys.platform == 'darwin'
3043 else 'desktop' if sys.platform.startswith('linux')
3044 else 'url')
3045 write_links[link_type] = True
3046
3047 if any(should_write and not _write_link_file(link_type)
3048 for link_type, should_write in write_links.items()):
3049 return
732044af 3050
f46e2f9d 3051 def replace_info_dict(new_info):
3052 nonlocal info_dict
3053 if new_info == info_dict:
3054 return
3055 info_dict.clear()
3056 info_dict.update(new_info)
3057
415f8d51 3058 new_info, files_to_move = self.pre_process(info_dict, 'before_dl', files_to_move)
3059 replace_info_dict(new_info)
56d868db 3060
a13e6848 3061 if self.params.get('skip_download'):
56d868db 3062 info_dict['filepath'] = temp_filename
3063 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
3064 info_dict['__files_to_move'] = files_to_move
f46e2f9d 3065 replace_info_dict(self.run_pp(MoveFilesAfterDownloadPP(self, False), info_dict))
9e907ebd 3066 info_dict['__write_download_archive'] = self.params.get('force_write_download_archive')
56d868db 3067 else:
3068 # Download
b868936c 3069 info_dict.setdefault('__postprocessors', [])
4340deca 3070 try:
0202b52a 3071
e04938ab 3072 def existing_video_file(*filepaths):
6b591b29 3073 ext = info_dict.get('ext')
e04938ab 3074 converted = lambda file: replace_extension(file, self.params.get('final_ext') or ext, ext)
3075 file = self.existing_file(itertools.chain(*zip(map(converted, filepaths), filepaths)),
3076 default_overwrite=False)
3077 if file:
3078 info_dict['ext'] = os.path.splitext(file)[1][1:]
3079 return file
0202b52a 3080
7b2c3f47 3081 fd, success = None, True
fccf90e7 3082 if info_dict.get('protocol') or info_dict.get('url'):
56ba69e4 3083 fd = get_suitable_downloader(info_dict, self.params, to_stdout=temp_filename == '-')
3084 if fd is not FFmpegFD and (
3085 info_dict.get('section_start') or info_dict.get('section_end')):
7b2c3f47 3086 msg = ('This format cannot be partially downloaded' if FFmpegFD.available()
56ba69e4 3087 else 'You have requested downloading the video partially, but ffmpeg is not installed')
3088 self.report_error(f'{msg}. Aborting')
5ec1b6b7 3089 return
5ec1b6b7 3090
4340deca 3091 if info_dict.get('requested_formats') is not None:
81cd954a 3092 requested_formats = info_dict['requested_formats']
0202b52a 3093 old_ext = info_dict['ext']
4e3b637d 3094 if self.params.get('merge_output_format') is None:
4e3b637d 3095 if (info_dict['ext'] == 'webm'
3096 and info_dict.get('thumbnails')
3097 # check with type instead of pp_key, __name__, or isinstance
3098 # since we dont want any custom PPs to trigger this
c487cf00 3099 and any(type(pp) == EmbedThumbnailPP for pp in self._pps['post_process'])): # noqa: E721
4e3b637d 3100 info_dict['ext'] = 'mkv'
3101 self.report_warning(
3102 'webm doesn\'t support embedding a thumbnail, mkv will be used')
124bc071 3103 new_ext = info_dict['ext']
0202b52a 3104
124bc071 3105 def correct_ext(filename, ext=new_ext):
96fccc10 3106 if filename == '-':
3107 return filename
0202b52a 3108 filename_real_ext = os.path.splitext(filename)[1][1:]
3109 filename_wo_ext = (
3110 os.path.splitext(filename)[0]
124bc071 3111 if filename_real_ext in (old_ext, new_ext)
0202b52a 3112 else filename)
86e5f3ed 3113 return f'{filename_wo_ext}.{ext}'
0202b52a 3114
38c6902b 3115 # Ensure filename always has a correct extension for successful merge
0202b52a 3116 full_filename = correct_ext(full_filename)
3117 temp_filename = correct_ext(temp_filename)
e04938ab 3118 dl_filename = existing_video_file(full_filename, temp_filename)
1ea24129 3119 info_dict['__real_download'] = False
18e674b4 3120
7b2c3f47 3121 merger = FFmpegMergerPP(self)
adbc4ec4 3122 downloaded = []
dbf5416a 3123 if dl_filename is not None:
6c7274ec 3124 self.report_file_already_downloaded(dl_filename)
adbc4ec4
THD
3125 elif fd:
3126 for f in requested_formats if fd != FFmpegFD else []:
3127 f['filepath'] = fname = prepend_extension(
3128 correct_ext(temp_filename, info_dict['ext']),
3129 'f%s' % f['format_id'], info_dict['ext'])
3130 downloaded.append(fname)
dbf5416a 3131 info_dict['url'] = '\n'.join(f['url'] for f in requested_formats)
3132 success, real_download = self.dl(temp_filename, info_dict)
3133 info_dict['__real_download'] = real_download
18e674b4 3134 else:
18e674b4 3135 if self.params.get('allow_unplayable_formats'):
3136 self.report_warning(
3137 'You have requested merging of multiple formats '
3138 'while also allowing unplayable formats to be downloaded. '
3139 'The formats won\'t be merged to prevent data corruption.')
3140 elif not merger.available:
e8969bda 3141 msg = 'You have requested merging of multiple formats but ffmpeg is not installed'
3142 if not self.params.get('ignoreerrors'):
3143 self.report_error(f'{msg}. Aborting due to --abort-on-error')
3144 return
3145 self.report_warning(f'{msg}. The formats won\'t be merged')
18e674b4 3146
96fccc10 3147 if temp_filename == '-':
adbc4ec4 3148 reason = ('using a downloader other than ffmpeg' if FFmpegFD.can_merge_formats(info_dict, self.params)
96fccc10 3149 else 'but the formats are incompatible for simultaneous download' if merger.available
3150 else 'but ffmpeg is not installed')
3151 self.report_warning(
3152 f'You have requested downloading multiple formats to stdout {reason}. '
3153 'The formats will be streamed one after the other')
3154 fname = temp_filename
dbf5416a 3155 for f in requested_formats:
3156 new_info = dict(info_dict)
3157 del new_info['requested_formats']
3158 new_info.update(f)
96fccc10 3159 if temp_filename != '-':
124bc071 3160 fname = prepend_extension(
3161 correct_ext(temp_filename, new_info['ext']),
3162 'f%s' % f['format_id'], new_info['ext'])
96fccc10 3163 if not self._ensure_dir_exists(fname):
3164 return
a21e0ab1 3165 f['filepath'] = fname
96fccc10 3166 downloaded.append(fname)
dbf5416a 3167 partial_success, real_download = self.dl(fname, new_info)
3168 info_dict['__real_download'] = info_dict['__real_download'] or real_download
3169 success = success and partial_success
adbc4ec4
THD
3170
3171 if downloaded and merger.available and not self.params.get('allow_unplayable_formats'):
3172 info_dict['__postprocessors'].append(merger)
3173 info_dict['__files_to_merge'] = downloaded
3174 # Even if there were no downloads, it is being merged only now
3175 info_dict['__real_download'] = True
3176 else:
3177 for file in downloaded:
3178 files_to_move[file] = None
4340deca
P
3179 else:
3180 # Just a single file
e04938ab 3181 dl_filename = existing_video_file(full_filename, temp_filename)
6c7274ec 3182 if dl_filename is None or dl_filename == temp_filename:
3183 # dl_filename == temp_filename could mean that the file was partially downloaded with --no-part.
3184 # So we should try to resume the download
e8e73840 3185 success, real_download = self.dl(temp_filename, info_dict)
0202b52a 3186 info_dict['__real_download'] = real_download
6c7274ec 3187 else:
3188 self.report_file_already_downloaded(dl_filename)
0202b52a 3189
0202b52a 3190 dl_filename = dl_filename or temp_filename
c571435f 3191 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
0202b52a 3192
3158150c 3193 except network_exceptions as err:
7960b056 3194 self.report_error('unable to download video data: %s' % error_to_compat_str(err))
4340deca 3195 return
86e5f3ed 3196 except OSError as err:
4340deca
P
3197 raise UnavailableVideoError(err)
3198 except (ContentTooShortError, ) as err:
86e5f3ed 3199 self.report_error(f'content too short (expected {err.expected} bytes and served {err.downloaded})')
4340deca 3200 return
8222d8de 3201
415f8d51 3202 self._raise_pending_errors(info_dict)
de6000d9 3203 if success and full_filename != '-':
f17f8651 3204
fd7cfb64 3205 def fixup():
3206 do_fixup = True
3207 fixup_policy = self.params.get('fixup')
3208 vid = info_dict['id']
3209
3210 if fixup_policy in ('ignore', 'never'):
3211 return
3212 elif fixup_policy == 'warn':
3fe75fdc 3213 do_fixup = 'warn'
f89b3e2d 3214 elif fixup_policy != 'force':
3215 assert fixup_policy in ('detect_or_warn', None)
3216 if not info_dict.get('__real_download'):
3217 do_fixup = False
fd7cfb64 3218
3219 def ffmpeg_fixup(cndn, msg, cls):
3fe75fdc 3220 if not (do_fixup and cndn):
fd7cfb64 3221 return
3fe75fdc 3222 elif do_fixup == 'warn':
fd7cfb64 3223 self.report_warning(f'{vid}: {msg}')
3224 return
3225 pp = cls(self)
3226 if pp.available:
3227 info_dict['__postprocessors'].append(pp)
3228 else:
3229 self.report_warning(f'{vid}: {msg}. Install ffmpeg to fix this automatically')
3230
3231 stretched_ratio = info_dict.get('stretched_ratio')
ca9def71
LNO
3232 ffmpeg_fixup(stretched_ratio not in (1, None),
3233 f'Non-uniform pixel ratio {stretched_ratio}',
3234 FFmpegFixupStretchedPP)
fd7cfb64 3235
993191c0 3236 downloader = get_suitable_downloader(info_dict, self.params) if 'protocol' in info_dict else None
24146491 3237 downloader = downloader.FD_NAME if downloader else None
adbc4ec4 3238
ca9def71
LNO
3239 ext = info_dict.get('ext')
3240 postprocessed_by_ffmpeg = info_dict.get('requested_formats') or any((
3241 isinstance(pp, FFmpegVideoConvertorPP)
3242 and resolve_recode_mapping(ext, pp.mapping)[0] not in (ext, None)
3243 ) for pp in self._pps['post_process'])
3244
3245 if not postprocessed_by_ffmpeg:
3246 ffmpeg_fixup(ext == 'm4a' and info_dict.get('container') == 'm4a_dash',
f2df4071 3247 'writing DASH m4a. Only some players support this container',
3248 FFmpegFixupM4aPP)
24146491 3249 ffmpeg_fixup(downloader == 'hlsnative' and not self.params.get('hls_use_mpegts')
494f5230 3250 or info_dict.get('is_live') and self.params.get('hls_use_mpegts') is None,
adbc4ec4
THD
3251 'Possible MPEG-TS in MP4 container or malformed AAC timestamps',
3252 FFmpegFixupM3u8PP)
3253 ffmpeg_fixup(info_dict.get('is_live') and downloader == 'DashSegmentsFD',
3254 'Possible duplicate MOOV atoms', FFmpegFixupDuplicateMoovPP)
3255
24146491 3256 ffmpeg_fixup(downloader == 'web_socket_fragment', 'Malformed timestamps detected', FFmpegFixupTimestampPP)
3257 ffmpeg_fixup(downloader == 'web_socket_fragment', 'Malformed duration detected', FFmpegFixupDurationPP)
fd7cfb64 3258
3259 fixup()
8222d8de 3260 try:
f46e2f9d 3261 replace_info_dict(self.post_process(dl_filename, info_dict, files_to_move))
af819c21 3262 except PostProcessingError as err:
3263 self.report_error('Postprocessing: %s' % str(err))
8222d8de 3264 return
ab8e5e51
AM
3265 try:
3266 for ph in self._post_hooks:
23c1a667 3267 ph(info_dict['filepath'])
ab8e5e51
AM
3268 except Exception as err:
3269 self.report_error('post hooks: %s' % str(err))
3270 return
9e907ebd 3271 info_dict['__write_download_archive'] = True
2d30509f 3272
c487cf00 3273 assert info_dict is original_infodict # Make sure the info_dict was modified in-place
a13e6848 3274 if self.params.get('force_write_download_archive'):
9e907ebd 3275 info_dict['__write_download_archive'] = True
ca6d59d2 3276 check_max_downloads()
8222d8de 3277
aa9369a2 3278 def __download_wrapper(self, func):
3279 @functools.wraps(func)
3280 def wrapper(*args, **kwargs):
3281 try:
3282 res = func(*args, **kwargs)
3283 except UnavailableVideoError as e:
3284 self.report_error(e)
b222c271 3285 except DownloadCancelled as e:
3286 self.to_screen(f'[info] {e}')
3287 if not self.params.get('break_per_url'):
3288 raise
fd404bec 3289 self._num_downloads = 0
aa9369a2 3290 else:
3291 if self.params.get('dump_single_json', False):
3292 self.post_extract(res)
3293 self.to_stdout(json.dumps(self.sanitize_info(res)))
3294 return wrapper
3295
8222d8de
JMF
3296 def download(self, url_list):
3297 """Download a given list of URLs."""
aa9369a2 3298 url_list = variadic(url_list) # Passing a single URL is a common mistake
bf1824b3 3299 outtmpl = self.params['outtmpl']['default']
3089bc74
S
3300 if (len(url_list) > 1
3301 and outtmpl != '-'
3302 and '%' not in outtmpl
3303 and self.params.get('max_downloads') != 1):
acd69589 3304 raise SameFileError(outtmpl)
8222d8de
JMF
3305
3306 for url in url_list:
aa9369a2 3307 self.__download_wrapper(self.extract_info)(
3308 url, force_generic_extractor=self.params.get('force_generic_extractor', False))
8222d8de
JMF
3309
3310 return self._download_retcode
3311
1dcc4c0c 3312 def download_with_info_file(self, info_filename):
31bd3925
JMF
3313 with contextlib.closing(fileinput.FileInput(
3314 [info_filename], mode='r',
3315 openhook=fileinput.hook_encoded('utf-8'))) as f:
3316 # FileInput doesn't have a read method, we can't call json.load
8012d892 3317 info = self.sanitize_info(json.loads('\n'.join(f)), self.params.get('clean_infojson', True))
d4943898 3318 try:
aa9369a2 3319 self.__download_wrapper(self.process_ie_result)(info, download=True)
f2ebc5c7 3320 except (DownloadError, EntryNotInPlaylist, ReExtractInfo) as e:
bf5f605e 3321 if not isinstance(e, EntryNotInPlaylist):
3322 self.to_stderr('\r')
d4943898
JMF
3323 webpage_url = info.get('webpage_url')
3324 if webpage_url is not None:
aa9369a2 3325 self.report_warning(f'The info failed to download: {e}; trying with URL {webpage_url}')
d4943898
JMF
3326 return self.download([webpage_url])
3327 else:
3328 raise
3329 return self._download_retcode
1dcc4c0c 3330
cb202fd2 3331 @staticmethod
8012d892 3332 def sanitize_info(info_dict, remove_private_keys=False):
3333 ''' Sanitize the infodict for converting to json '''
3ad56b42 3334 if info_dict is None:
3335 return info_dict
6e84b215 3336 info_dict.setdefault('epoch', int(time.time()))
6a5a30f9 3337 info_dict.setdefault('_type', 'video')
b5e7a2e6 3338 info_dict.setdefault('_version', {
3339 'version': __version__,
3340 'current_git_head': current_git_head(),
3341 'release_git_head': RELEASE_GIT_HEAD,
3342 'repository': REPOSITORY,
3343 })
09b49e1f 3344
8012d892 3345 if remove_private_keys:
0a5a191a 3346 reject = lambda k, v: v is None or k.startswith('__') or k in {
f46e2f9d 3347 'requested_downloads', 'requested_formats', 'requested_subtitles', 'requested_entries',
0a5a191a 3348 'entries', 'filepath', '_filename', 'infojson_filename', 'original_url', 'playlist_autonumber',
6e84b215 3349 }
ae8f99e6 3350 else:
09b49e1f 3351 reject = lambda k, v: False
adbc4ec4
THD
3352
3353 def filter_fn(obj):
3354 if isinstance(obj, dict):
3355 return {k: filter_fn(v) for k, v in obj.items() if not reject(k, v)}
3356 elif isinstance(obj, (list, tuple, set, LazyList)):
3357 return list(map(filter_fn, obj))
3358 elif obj is None or isinstance(obj, (str, int, float, bool)):
3359 return obj
3360 else:
3361 return repr(obj)
3362
5226731e 3363 return filter_fn(info_dict)
cb202fd2 3364
8012d892 3365 @staticmethod
3366 def filter_requested_info(info_dict, actually_filter=True):
3367 ''' Alias of sanitize_info for backward compatibility '''
3368 return YoutubeDL.sanitize_info(info_dict, actually_filter)
3369
43d7f5a5 3370 def _delete_downloaded_files(self, *files_to_delete, info={}, msg=None):
3371 for filename in set(filter(None, files_to_delete)):
3372 if msg:
3373 self.to_screen(msg % filename)
3374 try:
3375 os.remove(filename)
3376 except OSError:
3377 self.report_warning(f'Unable to delete file {filename}')
3378 if filename in info.get('__files_to_move', []): # NB: Delete even if None
3379 del info['__files_to_move'][filename]
3380
ed5835b4 3381 @staticmethod
3382 def post_extract(info_dict):
3383 def actual_post_extract(info_dict):
3384 if info_dict.get('_type') in ('playlist', 'multi_video'):
3385 for video_dict in info_dict.get('entries', {}):
3386 actual_post_extract(video_dict or {})
3387 return
3388
09b49e1f 3389 post_extractor = info_dict.pop('__post_extractor', None) or (lambda: {})
3390 info_dict.update(post_extractor())
ed5835b4 3391
3392 actual_post_extract(info_dict or {})
3393
dcf64d43 3394 def run_pp(self, pp, infodict):
5bfa4862 3395 files_to_delete = []
dcf64d43 3396 if '__files_to_move' not in infodict:
3397 infodict['__files_to_move'] = {}
b1940459 3398 try:
3399 files_to_delete, infodict = pp.run(infodict)
3400 except PostProcessingError as e:
3401 # Must be True and not 'only_download'
3402 if self.params.get('ignoreerrors') is True:
3403 self.report_error(e)
3404 return infodict
3405 raise
3406
5bfa4862 3407 if not files_to_delete:
dcf64d43 3408 return infodict
5bfa4862 3409 if self.params.get('keepvideo', False):
3410 for f in files_to_delete:
dcf64d43 3411 infodict['__files_to_move'].setdefault(f, '')
5bfa4862 3412 else:
43d7f5a5 3413 self._delete_downloaded_files(
3414 *files_to_delete, info=infodict, msg='Deleting original file %s (pass -k to keep)')
dcf64d43 3415 return infodict
5bfa4862 3416
ed5835b4 3417 def run_all_pps(self, key, info, *, additional_pps=None):
bb66c247 3418 self._forceprint(key, info)
ed5835b4 3419 for pp in (additional_pps or []) + self._pps[key]:
dc5f409c 3420 info = self.run_pp(pp, info)
ed5835b4 3421 return info
277d6ff5 3422
56d868db 3423 def pre_process(self, ie_info, key='pre_process', files_to_move=None):
5bfa4862 3424 info = dict(ie_info)
56d868db 3425 info['__files_to_move'] = files_to_move or {}
415f8d51 3426 try:
3427 info = self.run_all_pps(key, info)
3428 except PostProcessingError as err:
3429 msg = f'Preprocessing: {err}'
3430 info.setdefault('__pending_error', msg)
3431 self.report_error(msg, is_error=False)
56d868db 3432 return info, info.pop('__files_to_move', None)
5bfa4862 3433
f46e2f9d 3434 def post_process(self, filename, info, files_to_move=None):
8222d8de 3435 """Run all the postprocessors on the given file."""
8222d8de 3436 info['filepath'] = filename
dcf64d43 3437 info['__files_to_move'] = files_to_move or {}
ed5835b4 3438 info = self.run_all_pps('post_process', info, additional_pps=info.get('__postprocessors'))
dcf64d43 3439 info = self.run_pp(MoveFilesAfterDownloadPP(self), info)
3440 del info['__files_to_move']
ed5835b4 3441 return self.run_all_pps('after_move', info)
c1c9a79c 3442
5db07df6 3443 def _make_archive_id(self, info_dict):
e9fef7ee
S
3444 video_id = info_dict.get('id')
3445 if not video_id:
3446 return
5db07df6
PH
3447 # Future-proof against any change in case
3448 # and backwards compatibility with prior versions
e9fef7ee 3449 extractor = info_dict.get('extractor_key') or info_dict.get('ie_key') # key in a playlist
7012b23c 3450 if extractor is None:
1211bb6d
S
3451 url = str_or_none(info_dict.get('url'))
3452 if not url:
3453 return
e9fef7ee 3454 # Try to find matching extractor for the URL and take its ie_key
8b7491c8 3455 for ie_key, ie in self._ies.items():
1211bb6d 3456 if ie.suitable(url):
8b7491c8 3457 extractor = ie_key
e9fef7ee
S
3458 break
3459 else:
3460 return
0647d925 3461 return make_archive_id(extractor, video_id)
5db07df6
PH
3462
3463 def in_download_archive(self, info_dict):
3464 fn = self.params.get('download_archive')
3465 if fn is None:
3466 return False
3467
1e8fe57e 3468 vid_ids = [self._make_archive_id(info_dict)]
c200096c 3469 vid_ids.extend(info_dict.get('_old_archive_ids') or [])
1e8fe57e 3470 return any(id_ in self.archive for id_ in vid_ids)
c1c9a79c
PH
3471
3472 def record_download_archive(self, info_dict):
3473 fn = self.params.get('download_archive')
3474 if fn is None:
3475 return
5db07df6
PH
3476 vid_id = self._make_archive_id(info_dict)
3477 assert vid_id
a13e6848 3478 self.write_debug(f'Adding to archive: {vid_id}')
c1c9a79c 3479 with locked_file(fn, 'a', encoding='utf-8') as archive_file:
6febd1c1 3480 archive_file.write(vid_id + '\n')
a45e8619 3481 self.archive.add(vid_id)
dd82ffea 3482
8c51aa65 3483 @staticmethod
8abeeb94 3484 def format_resolution(format, default='unknown'):
9359f3d4 3485 if format.get('vcodec') == 'none' and format.get('acodec') != 'none':
fb04e403 3486 return 'audio only'
f49d89ee
PH
3487 if format.get('resolution') is not None:
3488 return format['resolution']
35615307 3489 if format.get('width') and format.get('height'):
ff51ed58 3490 return '%dx%d' % (format['width'], format['height'])
35615307 3491 elif format.get('height'):
ff51ed58 3492 return '%sp' % format['height']
35615307 3493 elif format.get('width'):
ff51ed58 3494 return '%dx?' % format['width']
3495 return default
8c51aa65 3496
8130779d 3497 def _list_format_headers(self, *headers):
3498 if self.params.get('listformats_table', True) is not False:
591bb9d3 3499 return [self._format_out(header, self.Styles.HEADERS) for header in headers]
8130779d 3500 return headers
3501
c57f7757
PH
3502 def _format_note(self, fdict):
3503 res = ''
3504 if fdict.get('ext') in ['f4f', 'f4m']:
f304da8a 3505 res += '(unsupported)'
32f90364
PH
3506 if fdict.get('language'):
3507 if res:
3508 res += ' '
f304da8a 3509 res += '[%s]' % fdict['language']
c57f7757 3510 if fdict.get('format_note') is not None:
f304da8a 3511 if res:
3512 res += ' '
3513 res += fdict['format_note']
c57f7757 3514 if fdict.get('tbr') is not None:
f304da8a 3515 if res:
3516 res += ', '
3517 res += '%4dk' % fdict['tbr']
c57f7757
PH
3518 if fdict.get('container') is not None:
3519 if res:
3520 res += ', '
3521 res += '%s container' % fdict['container']
3089bc74
S
3522 if (fdict.get('vcodec') is not None
3523 and fdict.get('vcodec') != 'none'):
c57f7757
PH
3524 if res:
3525 res += ', '
3526 res += fdict['vcodec']
91c7271a 3527 if fdict.get('vbr') is not None:
c57f7757
PH
3528 res += '@'
3529 elif fdict.get('vbr') is not None and fdict.get('abr') is not None:
3530 res += 'video@'
3531 if fdict.get('vbr') is not None:
3532 res += '%4dk' % fdict['vbr']
fbb21cf5 3533 if fdict.get('fps') is not None:
5d583bdf
S
3534 if res:
3535 res += ', '
3536 res += '%sfps' % fdict['fps']
c57f7757
PH
3537 if fdict.get('acodec') is not None:
3538 if res:
3539 res += ', '
3540 if fdict['acodec'] == 'none':
3541 res += 'video only'
3542 else:
3543 res += '%-5s' % fdict['acodec']
3544 elif fdict.get('abr') is not None:
3545 if res:
3546 res += ', '
3547 res += 'audio'
3548 if fdict.get('abr') is not None:
3549 res += '@%3dk' % fdict['abr']
3550 if fdict.get('asr') is not None:
3551 res += ' (%5dHz)' % fdict['asr']
3552 if fdict.get('filesize') is not None:
3553 if res:
3554 res += ', '
3555 res += format_bytes(fdict['filesize'])
9732d77e
PH
3556 elif fdict.get('filesize_approx') is not None:
3557 if res:
3558 res += ', '
3559 res += '~' + format_bytes(fdict['filesize_approx'])
c57f7757 3560 return res
91c7271a 3561
8130779d 3562 def render_formats_table(self, info_dict):
b69fd25c 3563 if not info_dict.get('formats') and not info_dict.get('url'):
8130779d 3564 return None
b69fd25c 3565
94badb25 3566 formats = info_dict.get('formats', [info_dict])
8130779d 3567 if not self.params.get('listformats_table', True) is not False:
76d321f6 3568 table = [
3569 [
3570 format_field(f, 'format_id'),
3571 format_field(f, 'ext'),
3572 self.format_resolution(f),
8130779d 3573 self._format_note(f)
3574 ] for f in formats if f.get('preference') is None or f['preference'] >= -1000]
3575 return render_table(['format code', 'extension', 'resolution', 'note'], table, extra_gap=1)
3576
d816f61f 3577 def simplified_codec(f, field):
3578 assert field in ('acodec', 'vcodec')
3579 codec = f.get(field, 'unknown')
f5ea4748 3580 if not codec:
3581 return 'unknown'
3582 elif codec != 'none':
d816f61f 3583 return '.'.join(codec.split('.')[:4])
3584
3585 if field == 'vcodec' and f.get('acodec') == 'none':
3586 return 'images'
3587 elif field == 'acodec' and f.get('vcodec') == 'none':
3588 return ''
3589 return self._format_out('audio only' if field == 'vcodec' else 'video only',
3590 self.Styles.SUPPRESS)
3591
591bb9d3 3592 delim = self._format_out('\u2502', self.Styles.DELIM, '|', test_encoding=True)
8130779d 3593 table = [
3594 [
591bb9d3 3595 self._format_out(format_field(f, 'format_id'), self.Styles.ID),
8130779d 3596 format_field(f, 'ext'),
3597 format_field(f, func=self.format_resolution, ignore=('audio only', 'images')),
563e0bf8 3598 format_field(f, 'fps', '\t%d', func=round),
8130779d 3599 format_field(f, 'dynamic_range', '%s', ignore=(None, 'SDR')).replace('HDR', ''),
b8ed0f15 3600 format_field(f, 'audio_channels', '\t%s'),
8130779d 3601 delim,
3602 format_field(f, 'filesize', ' \t%s', func=format_bytes) + format_field(f, 'filesize_approx', '~\t%s', func=format_bytes),
563e0bf8 3603 format_field(f, 'tbr', '\t%dk', func=round),
8130779d 3604 shorten_protocol_name(f.get('protocol', '')),
3605 delim,
d816f61f 3606 simplified_codec(f, 'vcodec'),
563e0bf8 3607 format_field(f, 'vbr', '\t%dk', func=round),
d816f61f 3608 simplified_codec(f, 'acodec'),
563e0bf8 3609 format_field(f, 'abr', '\t%dk', func=round),
ae61d108 3610 format_field(f, 'asr', '\t%s', func=format_decimal_suffix),
8130779d 3611 join_nonempty(
591bb9d3 3612 self._format_out('UNSUPPORTED', 'light red') if f.get('ext') in ('f4f', 'f4m') else None,
8130779d 3613 format_field(f, 'language', '[%s]'),
3614 join_nonempty(format_field(f, 'format_note'),
3615 format_field(f, 'container', ignore=(None, f.get('ext'))),
3616 delim=', '),
3617 delim=' '),
3618 ] for f in formats if f.get('preference') is None or f['preference'] >= -1000]
3619 header_line = self._list_format_headers(
b8ed0f15 3620 'ID', 'EXT', 'RESOLUTION', '\tFPS', 'HDR', 'CH', delim, '\tFILESIZE', '\tTBR', 'PROTO',
8130779d 3621 delim, 'VCODEC', '\tVBR', 'ACODEC', '\tABR', '\tASR', 'MORE INFO')
3622
3623 return render_table(
3624 header_line, table, hide_empty=True,
591bb9d3 3625 delim=self._format_out('\u2500', self.Styles.DELIM, '-', test_encoding=True))
8130779d 3626
3627 def render_thumbnails_table(self, info_dict):
88f23a18 3628 thumbnails = list(info_dict.get('thumbnails') or [])
cfb56d1a 3629 if not thumbnails:
8130779d 3630 return None
3631 return render_table(
ec11a9f4 3632 self._list_format_headers('ID', 'Width', 'Height', 'URL'),
6970b600 3633 [[t.get('id'), t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails])
2412044c 3634
8130779d 3635 def render_subtitles_table(self, video_id, subtitles):
2412044c 3636 def _row(lang, formats):
49c258e1 3637 exts, names = zip(*((f['ext'], f.get('name') or 'unknown') for f in reversed(formats)))
2412044c 3638 if len(set(names)) == 1:
7aee40c1 3639 names = [] if names[0] == 'unknown' else names[:1]
2412044c 3640 return [lang, ', '.join(names), ', '.join(exts)]
3641
8130779d 3642 if not subtitles:
3643 return None
3644 return render_table(
ec11a9f4 3645 self._list_format_headers('Language', 'Name', 'Formats'),
2412044c 3646 [_row(lang, formats) for lang, formats in subtitles.items()],
8130779d 3647 hide_empty=True)
3648
3649 def __list_table(self, video_id, name, func, *args):
3650 table = func(*args)
3651 if not table:
3652 self.to_screen(f'{video_id} has no {name}')
3653 return
3654 self.to_screen(f'[info] Available {name} for {video_id}:')
3655 self.to_stdout(table)
3656
3657 def list_formats(self, info_dict):
3658 self.__list_table(info_dict['id'], 'formats', self.render_formats_table, info_dict)
3659
3660 def list_thumbnails(self, info_dict):
3661 self.__list_table(info_dict['id'], 'thumbnails', self.render_thumbnails_table, info_dict)
3662
3663 def list_subtitles(self, video_id, subtitles, name='subtitles'):
3664 self.__list_table(video_id, name, self.render_subtitles_table, video_id, subtitles)
a504ced0 3665
dca08720
PH
3666 def urlopen(self, req):
3667 """ Start an HTTP download """
f9934b96 3668 if isinstance(req, str):
67dda517 3669 req = sanitized_Request(req)
19a41fc6 3670 return self._opener.open(req, timeout=self._socket_timeout)
dca08720
PH
3671
3672 def print_debug_header(self):
3673 if not self.params.get('verbose'):
3674 return
49a57e70 3675
560738f3 3676 # These imports can be slow. So import them only as needed
3677 from .extractor.extractors import _LAZY_LOADER
3678 from .extractor.extractors import _PLUGIN_CLASSES as plugin_extractors
3679
49a57e70 3680 def get_encoding(stream):
2a938746 3681 ret = str(getattr(stream, 'encoding', 'missing (%s)' % type(stream).__name__))
49a57e70 3682 if not supports_terminal_sequences(stream):
53973b4d 3683 from .utils import WINDOWS_VT_MODE # Must be imported locally
e3c7d495 3684 ret += ' (No VT)' if WINDOWS_VT_MODE is False else ' (No ANSI)'
49a57e70 3685 return ret
3686
591bb9d3 3687 encoding_str = 'Encodings: locale %s, fs %s, pref %s, %s' % (
49a57e70 3688 locale.getpreferredencoding(),
3689 sys.getfilesystemencoding(),
591bb9d3 3690 self.get_encoding(),
3691 ', '.join(
64fa820c 3692 f'{key} {get_encoding(stream)}' for key, stream in self._out_files.items_
591bb9d3 3693 if stream is not None and key != 'console')
3694 )
883d4b1e 3695
3696 logger = self.params.get('logger')
3697 if logger:
3698 write_debug = lambda msg: logger.debug(f'[debug] {msg}')
3699 write_debug(encoding_str)
3700 else:
96565c7e 3701 write_string(f'[debug] {encoding_str}\n', encoding=None)
49a57e70 3702 write_debug = lambda msg: self._write_string(f'[debug] {msg}\n')
734f90bb 3703
4c88ff87 3704 source = detect_variant()
70b23409 3705 if VARIANT not in (None, 'pip'):
3706 source += '*'
36eaf303 3707 write_debug(join_nonempty(
b5e7a2e6 3708 f'{"yt-dlp" if REPOSITORY == "yt-dlp/yt-dlp" else REPOSITORY} version',
3709 __version__,
36eaf303 3710 f'[{RELEASE_GIT_HEAD}]' if RELEASE_GIT_HEAD else '',
3711 '' if source == 'unknown' else f'({source})',
3712 delim=' '))
6e21fdd2 3713 if not _LAZY_LOADER:
3714 if os.environ.get('YTDLP_NO_LAZY_EXTRACTORS'):
49a57e70 3715 write_debug('Lazy loading extractors is forcibly disabled')
6e21fdd2 3716 else:
49a57e70 3717 write_debug('Lazy loading extractors is disabled')
3ae5e797 3718 if plugin_extractors or plugin_postprocessors:
49a57e70 3719 write_debug('Plugins: %s' % [
3ae5e797 3720 '%s%s' % (klass.__name__, '' if klass.__name__ == name else f' as {name}')
3721 for name, klass in itertools.chain(plugin_extractors.items(), plugin_postprocessors.items())])
8a82af35 3722 if self.params['compat_opts']:
3723 write_debug('Compatibility options: %s' % ', '.join(self.params['compat_opts']))
36eaf303 3724
b5e7a2e6 3725 if current_git_head():
3726 write_debug(f'Git HEAD: {current_git_head()}')
b1f94422 3727 write_debug(system_identifier())
d28b5171 3728
8913ef74 3729 exe_versions, ffmpeg_features = FFmpegPostProcessor.get_versions_and_features(self)
3730 ffmpeg_features = {key for key, val in ffmpeg_features.items() if val}
3731 if ffmpeg_features:
19a03940 3732 exe_versions['ffmpeg'] += ' (%s)' % ','.join(sorted(ffmpeg_features))
8913ef74 3733
4c83c967 3734 exe_versions['rtmpdump'] = rtmpdump_version()
feee8d32 3735 exe_versions['phantomjs'] = PhantomJSwrapper._version()
d28b5171 3736 exe_str = ', '.join(
2831b468 3737 f'{exe} {v}' for exe, v in sorted(exe_versions.items()) if v
3738 ) or 'none'
49a57e70 3739 write_debug('exe versions: %s' % exe_str)
dca08720 3740
1d485a1a 3741 from .compat.compat_utils import get_package_info
9b8ee23b 3742 from .dependencies import available_dependencies
3743
3744 write_debug('Optional libraries: %s' % (', '.join(sorted({
1d485a1a 3745 join_nonempty(*get_package_info(m)) for m in available_dependencies.values()
9b8ee23b 3746 })) or 'none'))
2831b468 3747
97ec5bc5 3748 self._setup_opener()
dca08720
PH
3749 proxy_map = {}
3750 for handler in self._opener.handlers:
3751 if hasattr(handler, 'proxies'):
3752 proxy_map.update(handler.proxies)
49a57e70 3753 write_debug(f'Proxy map: {proxy_map}')
dca08720 3754
49a57e70 3755 # Not implemented
3756 if False and self.params.get('call_home'):
0f06bcd7 3757 ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode()
49a57e70 3758 write_debug('Public IP address: %s' % ipaddr)
58b1f00d 3759 latest_version = self.urlopen(
0f06bcd7 3760 'https://yt-dl.org/latest/version').read().decode()
58b1f00d
PH
3761 if version_tuple(latest_version) > version_tuple(__version__):
3762 self.report_warning(
3763 'You are using an outdated version (newest version: %s)! '
3764 'See https://yt-dl.org/update if you need help updating.' %
3765 latest_version)
3766
e344693b 3767 def _setup_opener(self):
97ec5bc5 3768 if hasattr(self, '_opener'):
3769 return
6ad14cab 3770 timeout_val = self.params.get('socket_timeout')
17bddf3e 3771 self._socket_timeout = 20 if timeout_val is None else float(timeout_val)
6ad14cab 3772
982ee69a 3773 opts_cookiesfrombrowser = self.params.get('cookiesfrombrowser')
dca08720
PH
3774 opts_cookiefile = self.params.get('cookiefile')
3775 opts_proxy = self.params.get('proxy')
3776
982ee69a 3777 self.cookiejar = load_cookies(opts_cookiefile, opts_cookiesfrombrowser, self)
dca08720 3778
6a3f4c3f 3779 cookie_processor = YoutubeDLCookieProcessor(self.cookiejar)
dca08720
PH
3780 if opts_proxy is not None:
3781 if opts_proxy == '':
3782 proxies = {}
3783 else:
3784 proxies = {'http': opts_proxy, 'https': opts_proxy}
3785 else:
ac668111 3786 proxies = urllib.request.getproxies()
067aa17e 3787 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
dca08720
PH
3788 if 'http' in proxies and 'https' not in proxies:
3789 proxies['https'] = proxies['http']
91410c9b 3790 proxy_handler = PerRequestProxyHandler(proxies)
a0ddb8a2
PH
3791
3792 debuglevel = 1 if self.params.get('debug_printtraffic') else 0
be4a824d
PH
3793 https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel)
3794 ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel)
fca6dba8 3795 redirect_handler = YoutubeDLRedirectHandler()
f9934b96 3796 data_handler = urllib.request.DataHandler()
6240b0a2
JMF
3797
3798 # When passing our own FileHandler instance, build_opener won't add the
3799 # default FileHandler and allows us to disable the file protocol, which
3800 # can be used for malicious purposes (see
067aa17e 3801 # https://github.com/ytdl-org/youtube-dl/issues/8227)
ac668111 3802 file_handler = urllib.request.FileHandler()
6240b0a2
JMF
3803
3804 def file_open(*args, **kwargs):
ac668111 3805 raise urllib.error.URLError('file:// scheme is explicitly disabled in yt-dlp for security reasons')
6240b0a2
JMF
3806 file_handler.file_open = file_open
3807
ac668111 3808 opener = urllib.request.build_opener(
fca6dba8 3809 proxy_handler, https_handler, cookie_processor, ydlh, redirect_handler, data_handler, file_handler)
2461f79d 3810
dca08720
PH
3811 # Delete the default user-agent header, which would otherwise apply in
3812 # cases where our custom HTTP handler doesn't come into play
067aa17e 3813 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
dca08720
PH
3814 opener.addheaders = []
3815 self._opener = opener
62fec3b2
PH
3816
3817 def encode(self, s):
3818 if isinstance(s, bytes):
3819 return s # Already encoded
3820
3821 try:
3822 return s.encode(self.get_encoding())
3823 except UnicodeEncodeError as err:
3824 err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.'
3825 raise
3826
3827 def get_encoding(self):
3828 encoding = self.params.get('encoding')
3829 if encoding is None:
3830 encoding = preferredencoding()
3831 return encoding
ec82d85a 3832
e08a85d8 3833 def _write_info_json(self, label, ie_result, infofn, overwrite=None):
cb96c5be 3834 ''' Write infojson and returns True = written, 'exists' = Already exists, False = skip, None = error '''
e08a85d8 3835 if overwrite is None:
3836 overwrite = self.params.get('overwrites', True)
80c03fa9 3837 if not self.params.get('writeinfojson'):
3838 return False
3839 elif not infofn:
3840 self.write_debug(f'Skipping writing {label} infojson')
3841 return False
3842 elif not self._ensure_dir_exists(infofn):
3843 return None
e08a85d8 3844 elif not overwrite and os.path.exists(infofn):
80c03fa9 3845 self.to_screen(f'[info] {label.title()} metadata is already present')
cb96c5be 3846 return 'exists'
3847
3848 self.to_screen(f'[info] Writing {label} metadata as JSON to: {infofn}')
3849 try:
3850 write_json_file(self.sanitize_info(ie_result, self.params.get('clean_infojson', True)), infofn)
3851 return True
86e5f3ed 3852 except OSError:
cb96c5be 3853 self.report_error(f'Cannot write {label} metadata to JSON file {infofn}')
3854 return None
80c03fa9 3855
3856 def _write_description(self, label, ie_result, descfn):
3857 ''' Write description and returns True = written, False = skip, None = error '''
3858 if not self.params.get('writedescription'):
3859 return False
3860 elif not descfn:
3861 self.write_debug(f'Skipping writing {label} description')
3862 return False
3863 elif not self._ensure_dir_exists(descfn):
3864 return None
3865 elif not self.params.get('overwrites', True) and os.path.exists(descfn):
3866 self.to_screen(f'[info] {label.title()} description is already present')
3867 elif ie_result.get('description') is None:
3868 self.report_warning(f'There\'s no {label} description to write')
3869 return False
3870 else:
3871 try:
3872 self.to_screen(f'[info] Writing {label} description to: {descfn}')
86e5f3ed 3873 with open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
80c03fa9 3874 descfile.write(ie_result['description'])
86e5f3ed 3875 except OSError:
80c03fa9 3876 self.report_error(f'Cannot write {label} description file {descfn}')
3877 return None
3878 return True
3879
3880 def _write_subtitles(self, info_dict, filename):
3881 ''' Write subtitles to file and return list of (sub_filename, final_sub_filename); or None if error'''
3882 ret = []
3883 subtitles = info_dict.get('requested_subtitles')
3884 if not subtitles or not (self.params.get('writesubtitles') or self.params.get('writeautomaticsub')):
3885 # subtitles download errors are already managed as troubles in relevant IE
3886 # that way it will silently go on when used with unsupporting IE
3887 return ret
3888
3889 sub_filename_base = self.prepare_filename(info_dict, 'subtitle')
3890 if not sub_filename_base:
3891 self.to_screen('[info] Skipping writing video subtitles')
3892 return ret
3893 for sub_lang, sub_info in subtitles.items():
3894 sub_format = sub_info['ext']
3895 sub_filename = subtitles_filename(filename, sub_lang, sub_format, info_dict.get('ext'))
3896 sub_filename_final = subtitles_filename(sub_filename_base, sub_lang, sub_format, info_dict.get('ext'))
e04938ab 3897 existing_sub = self.existing_file((sub_filename_final, sub_filename))
3898 if existing_sub:
80c03fa9 3899 self.to_screen(f'[info] Video subtitle {sub_lang}.{sub_format} is already present')
e04938ab 3900 sub_info['filepath'] = existing_sub
3901 ret.append((existing_sub, sub_filename_final))
80c03fa9 3902 continue
3903
3904 self.to_screen(f'[info] Writing video subtitles to: {sub_filename}')
3905 if sub_info.get('data') is not None:
3906 try:
3907 # Use newline='' to prevent conversion of newline characters
3908 # See https://github.com/ytdl-org/youtube-dl/issues/10268
86e5f3ed 3909 with open(sub_filename, 'w', encoding='utf-8', newline='') as subfile:
80c03fa9 3910 subfile.write(sub_info['data'])
3911 sub_info['filepath'] = sub_filename
3912 ret.append((sub_filename, sub_filename_final))
3913 continue
86e5f3ed 3914 except OSError:
80c03fa9 3915 self.report_error(f'Cannot write video subtitles file {sub_filename}')
3916 return None
3917
3918 try:
3919 sub_copy = sub_info.copy()
3920 sub_copy.setdefault('http_headers', info_dict.get('http_headers'))
3921 self.dl(sub_filename, sub_copy, subtitle=True)
3922 sub_info['filepath'] = sub_filename
3923 ret.append((sub_filename, sub_filename_final))
6020e05d 3924 except (DownloadError, ExtractorError, IOError, OSError, ValueError) + network_exceptions as err:
c70c418d 3925 msg = f'Unable to download video subtitles for {sub_lang!r}: {err}'
6020e05d 3926 if self.params.get('ignoreerrors') is not True: # False or 'only_download'
c70c418d 3927 if not self.params.get('ignoreerrors'):
3928 self.report_error(msg)
3929 raise DownloadError(msg)
3930 self.report_warning(msg)
519804a9 3931 return ret
80c03fa9 3932
3933 def _write_thumbnails(self, label, info_dict, filename, thumb_filename_base=None):
3934 ''' Write thumbnails to file and return list of (thumb_filename, final_thumb_filename) '''
6c4fd172 3935 write_all = self.params.get('write_all_thumbnails', False)
80c03fa9 3936 thumbnails, ret = [], []
6c4fd172 3937 if write_all or self.params.get('writethumbnail', False):
0202b52a 3938 thumbnails = info_dict.get('thumbnails') or []
6c4fd172 3939 multiple = write_all and len(thumbnails) > 1
ec82d85a 3940
80c03fa9 3941 if thumb_filename_base is None:
3942 thumb_filename_base = filename
3943 if thumbnails and not thumb_filename_base:
3944 self.write_debug(f'Skipping writing {label} thumbnail')
3945 return ret
3946
dd0228ce 3947 for idx, t in list(enumerate(thumbnails))[::-1]:
80c03fa9 3948 thumb_ext = (f'{t["id"]}.' if multiple else '') + determine_ext(t['url'], 'jpg')
aa9369a2 3949 thumb_display_id = f'{label} thumbnail {t["id"]}'
80c03fa9 3950 thumb_filename = replace_extension(filename, thumb_ext, info_dict.get('ext'))
3951 thumb_filename_final = replace_extension(thumb_filename_base, thumb_ext, info_dict.get('ext'))
ec82d85a 3952
e04938ab 3953 existing_thumb = self.existing_file((thumb_filename_final, thumb_filename))
3954 if existing_thumb:
aa9369a2 3955 self.to_screen('[info] %s is already present' % (
3956 thumb_display_id if multiple else f'{label} thumbnail').capitalize())
e04938ab 3957 t['filepath'] = existing_thumb
3958 ret.append((existing_thumb, thumb_filename_final))
ec82d85a 3959 else:
80c03fa9 3960 self.to_screen(f'[info] Downloading {thumb_display_id} ...')
ec82d85a 3961 try:
297e9952 3962 uf = self.urlopen(sanitized_Request(t['url'], headers=t.get('http_headers', {})))
80c03fa9 3963 self.to_screen(f'[info] Writing {thumb_display_id} to: {thumb_filename}')
d3d89c32 3964 with open(encodeFilename(thumb_filename), 'wb') as thumbf:
ec82d85a 3965 shutil.copyfileobj(uf, thumbf)
80c03fa9 3966 ret.append((thumb_filename, thumb_filename_final))
885cc0b7 3967 t['filepath'] = thumb_filename
3158150c 3968 except network_exceptions as err:
dd0228ce 3969 thumbnails.pop(idx)
80c03fa9 3970 self.report_warning(f'Unable to download {thumb_display_id}: {err}')
6c4fd172 3971 if ret and not write_all:
3972 break
0202b52a 3973 return ret