]> jfr.im git - yt-dlp.git/blame - yt_dlp/YoutubeDL.py
[extractor/crunchyroll:beta] Use anonymous access (#4704)
[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
1e4fca9a 307 name/path from where cookies are loaded, and the name of the
62b58c09 308 keyring, e.g. ('chrome', ) or ('vivaldi', 'default', 'BASICTEXT')
f81c62a6 309 legacyserverconnect: Explicitly allow HTTPS connection to servers that do not
310 support RFC 5746 secure renegotiation
f59f5ef8 311 nocheckcertificate: Do not verify SSL certificates
bb58c9ed 312 client_certificate: Path to client certificate file in PEM format. May include the private key
313 client_certificate_key: Path to private key file for client certificate
314 client_certificate_password: Password for client certificate private key, if encrypted.
315 If not provided and the key is encrypted, yt-dlp will ask interactively
7e8c0af0 316 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
c6e07cf1 317 (Only supported by some extractors)
8b7539d2 318 http_headers: A dictionary of custom headers to be used for all requests
a1ee09e8 319 proxy: URL of the proxy server to use
38cce791 320 geo_verification_proxy: URL of the proxy to use for IP address verification
504f20dd 321 on geo-restricted sites.
e344693b 322 socket_timeout: Time to wait for unresponsive hosts, in seconds
0783b09b
PH
323 bidi_workaround: Work around buggy terminals without bidirectional text
324 support, using fridibi
a0ddb8a2 325 debug_printtraffic:Print out sent and received HTTP traffic
04b4d394
PH
326 default_search: Prepend this string if an input url is not valid.
327 'auto' for elaborate guessing
62fec3b2 328 encoding: Use this encoding instead of the system-specified.
134c913c 329 extract_flat: Whether to resolve and process url_results further
330 * False: Always process (default)
331 * True: Never process
332 * 'in_playlist': Do not process inside playlist/multi_video
333 * 'discard': Always process, but don't return the result
334 from inside playlist/multi_video
335 * 'discard_in_playlist': Same as "discard", but only for
336 playlists (not multi_video)
f2ebc5c7 337 wait_for_video: If given, wait for scheduled streams to become available.
338 The value should be a tuple containing the range
339 (min_secs, max_secs) to wait between retries
4f026faf 340 postprocessors: A list of dictionaries, each with an entry
71b640cc 341 * key: The name of the postprocessor. See
7a5c1cfe 342 yt_dlp/postprocessor/__init__.py for a list.
bb66c247 343 * when: When to run the postprocessor. Allowed values are
344 the entries of utils.POSTPROCESS_WHEN
56d868db 345 Assumed to be 'post_process' if not given
71b640cc
PH
346 progress_hooks: A list of functions that get called on download
347 progress, with a dictionary with the entries
5cda4eda 348 * status: One of "downloading", "error", or "finished".
ee69b99a 349 Check this first and ignore unknown values.
3ba7740d 350 * info_dict: The extracted info_dict
71b640cc 351
5cda4eda 352 If status is one of "downloading", or "finished", the
ee69b99a
PH
353 following properties may also be present:
354 * filename: The final filename (always present)
5cda4eda 355 * tmpfilename: The filename we're currently writing to
71b640cc
PH
356 * downloaded_bytes: Bytes on disk
357 * total_bytes: Size of the whole file, None if unknown
5cda4eda
PH
358 * total_bytes_estimate: Guess of the eventual file size,
359 None if unavailable.
360 * elapsed: The number of seconds since download started.
71b640cc
PH
361 * eta: The estimated time in seconds, None if unknown
362 * speed: The download speed in bytes/second, None if
363 unknown
5cda4eda
PH
364 * fragment_index: The counter of the currently
365 downloaded video fragment.
366 * fragment_count: The number of fragments (= individual
367 files that will be merged)
71b640cc
PH
368
369 Progress hooks are guaranteed to be called at least once
370 (with status "finished") if the download is successful.
819e0531 371 postprocessor_hooks: A list of functions that get called on postprocessing
372 progress, with a dictionary with the entries
373 * status: One of "started", "processing", or "finished".
374 Check this first and ignore unknown values.
375 * postprocessor: Name of the postprocessor
376 * info_dict: The extracted info_dict
377
378 Progress hooks are guaranteed to be called at least twice
379 (with status "started" and "finished") if the processing is successful.
fc61aff4 380 merge_output_format: "/" separated list of extensions to use when merging formats.
6b591b29 381 final_ext: Expected final extension; used to detect when the file was
59a7a13e 382 already downloaded and converted
6271f1ca
PH
383 fixup: Automatically correct known faults of the file.
384 One of:
385 - "never": do nothing
386 - "warn": only emit a warning
387 - "detect_or_warn": check whether we can do anything
62cd676c 388 about it, warn otherwise (default)
504f20dd 389 source_address: Client-side IP address to bind to.
1cf376f5 390 sleep_interval_requests: Number of seconds to sleep between requests
391 during extraction
7aa589a5
S
392 sleep_interval: Number of seconds to sleep before each download when
393 used alone or a lower bound of a range for randomized
394 sleep before each download (minimum possible number
395 of seconds to sleep) when used along with
396 max_sleep_interval.
397 max_sleep_interval:Upper bound of a range for randomized sleep before each
398 download (maximum possible number of seconds to sleep).
399 Must only be used along with sleep_interval.
400 Actual sleep time will be a random float from range
401 [sleep_interval; max_sleep_interval].
1cf376f5 402 sleep_interval_subtitles: Number of seconds to sleep before each subtitle download
cfb56d1a
PH
403 listformats: Print an overview of available video formats and exit.
404 list_thumbnails: Print a table of all thumbnails and exit.
0a41f331 405 match_filter: A function that gets called for every video with the signature
406 (info_dict, *, incomplete: bool) -> Optional[str]
407 For backward compatibility with youtube-dl, the signature
408 (info_dict) -> Optional[str] is also allowed.
409 - If it returns a message, the video is ignored.
410 - If it returns None, the video is downloaded.
411 - If it returns utils.NO_DEFAULT, the user is interactively
412 asked whether to download the video.
347de493 413 match_filter_func in utils.py is one example for this.
7e5db8c9 414 no_color: Do not emit color codes in output.
0a840f58 415 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
504f20dd 416 HTTP header
0a840f58 417 geo_bypass_country:
773f291d
S
418 Two-letter ISO 3166-2 country code that will be used for
419 explicit geographic restriction bypassing via faking
504f20dd 420 X-Forwarded-For HTTP header
5f95927a
S
421 geo_bypass_ip_block:
422 IP range in CIDR notation that will be used similarly to
504f20dd 423 geo_bypass_country
52a8a1e1 424 external_downloader: A dictionary of protocol keys and the executable of the
425 external downloader to use for it. The allowed protocols
426 are default|http|ftp|m3u8|dash|rtsp|rtmp|mms.
427 Set the value to 'native' to use the native downloader
53ed7066 428 compat_opts: Compatibility options. See "Differences in default behavior".
3acf6d38 429 The following options do not work when used through the API:
b5ae35ee 430 filename, abort-on-error, multistreams, no-live-chat, format-sort
dac5df5a 431 no-clean-infojson, no-playlist-metafiles, no-keep-subs, no-attach-info-json.
e4f02757 432 Refer __init__.py for their implementation
819e0531 433 progress_template: Dictionary of templates for progress outputs.
434 Allowed keys are 'download', 'postprocess',
435 'download-title' (console title) and 'postprocess-title'.
436 The template is mapped on a dictionary with keys 'progress' and 'info'
23326151 437 retry_sleep_functions: Dictionary of functions that takes the number of attempts
438 as argument and returns the time to sleep in seconds.
439 Allowed keys are 'http', 'fragment', 'file_access'
0f446365
SW
440 download_ranges: A callback function that gets called for every video with
441 the signature (info_dict, ydl) -> Iterable[Section].
442 Only the returned sections will be downloaded.
443 Each Section is a dict with the following keys:
5ec1b6b7 444 * start_time: Start time of the section in seconds
445 * end_time: End time of the section in seconds
446 * title: Section title (Optional)
447 * index: Section number (Optional)
0f446365 448 force_keyframes_at_cuts: Re-encode the video when downloading ranges to get precise cuts
a7dc6a89 449 noprogress: Do not print the progress bar
a831c2ea 450 live_from_start: Whether to download livestreams videos from the start
fe7e0c98 451
8222d8de 452 The following parameters are not used by YoutubeDL itself, they are used by
7a5c1cfe 453 the downloader (see yt_dlp/downloader/common.py):
51d9739f 454 nopart, updatetime, buffersize, ratelimit, throttledratelimit, min_filesize,
205a0654 455 max_filesize, test, noresizebuffer, retries, file_access_retries, fragment_retries,
a7dc6a89 456 continuedl, xattr_set_filesize, hls_use_mpegts, http_chunk_size,
59a7a13e 457 external_downloader_args, concurrent_fragment_downloads.
76b1bd67
JMF
458
459 The following options are used by the post processors:
c0b7d117
S
460 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
461 to the binary or its containing directory.
43820c03 462 postprocessor_args: A dictionary of postprocessor/executable keys (in lower case)
34488702 463 and a list of additional command-line arguments for the
464 postprocessor/executable. The dict can also have "PP+EXE" keys
465 which are used when the given exe is used by the given PP.
466 Use 'default' as the name for arguments to passed to all PP
467 For compatibility with youtube-dl, a single list of args
468 can also be used
e409895f 469
470 The following options are used by the extractors:
62bff2c1 471 extractor_retries: Number of times to retry for known errors
472 dynamic_mpd: Whether to process dynamic DASH manifests (default: True)
e409895f 473 hls_split_discontinuity: Split HLS playlists to different formats at
62bff2c1 474 discontinuities such as ad breaks (default: False)
5d3a0e79 475 extractor_args: A dictionary of arguments to be passed to the extractors.
476 See "EXTRACTOR ARGUMENTS" for details.
62b58c09 477 E.g. {'youtube': {'skip': ['dash', 'hls']}}
88f23a18 478 mark_watched: Mark videos watched (even with --simulate). Only for YouTube
1890fc63 479
480 The following options are deprecated and may be removed in the future:
481
fe7866d0 482 force_generic_extractor: Force downloader to use the generic extractor
483 - Use allowed_extractors = ['generic', 'default']
7e9a6125 484 playliststart: - Use playlist_items
485 Playlist item to start at.
486 playlistend: - Use playlist_items
487 Playlist item to end at.
488 playlistreverse: - Use playlist_items
489 Download playlist items in reverse order.
1890fc63 490 forceurl: - Use forceprint
491 Force printing final URL.
492 forcetitle: - Use forceprint
493 Force printing title.
494 forceid: - Use forceprint
495 Force printing ID.
496 forcethumbnail: - Use forceprint
497 Force printing thumbnail URL.
498 forcedescription: - Use forceprint
499 Force printing description.
500 forcefilename: - Use forceprint
501 Force printing final filename.
502 forceduration: - Use forceprint
503 Force printing duration.
504 allsubtitles: - Use subtitleslangs = ['all']
505 Downloads all the subtitles of the video
506 (requires writesubtitles or writeautomaticsub)
507 include_ads: - Doesn't work
508 Download ads as well
509 call_home: - Not implemented
510 Boolean, true iff we are allowed to contact the
511 yt-dlp servers for debugging.
512 post_hooks: - Register a custom postprocessor
513 A list of functions that get called as the final step
514 for each video file, after all postprocessors have been
515 called. The filename will be passed as the only argument.
516 hls_prefer_native: - Use external_downloader = {'m3u8': 'native'} or {'m3u8': 'ffmpeg'}.
517 Use the native HLS downloader instead of ffmpeg/avconv
518 if True, otherwise use ffmpeg/avconv if False, otherwise
519 use downloader suggested by extractor if None.
520 prefer_ffmpeg: - avconv support is deprecated
521 If False, use avconv instead of ffmpeg if both are available,
522 otherwise prefer ffmpeg.
523 youtube_include_dash_manifest: - Use extractor_args
5d3a0e79 524 If True (default), DASH manifests and related
62bff2c1 525 data will be downloaded and processed by extractor.
526 You can reduce network I/O by disabling it if you don't
527 care about DASH. (only for youtube)
1890fc63 528 youtube_include_hls_manifest: - Use extractor_args
5d3a0e79 529 If True (default), HLS manifests and related
62bff2c1 530 data will be downloaded and processed by extractor.
531 You can reduce network I/O by disabling it if you don't
532 care about HLS. (only for youtube)
8222d8de
JMF
533 """
534
86e5f3ed 535 _NUMERIC_FIELDS = {
b8ed0f15 536 'width', 'height', 'asr', 'audio_channels', 'fps',
537 'tbr', 'abr', 'vbr', 'filesize', 'filesize_approx',
e6f21b3d 538 'timestamp', 'release_timestamp',
c9969434
S
539 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
540 'average_rating', 'comment_count', 'age_limit',
541 'start_time', 'end_time',
542 'chapter_number', 'season_number', 'episode_number',
543 'track_number', 'disc_number', 'release_year',
86e5f3ed 544 }
c9969434 545
6db9c4d5 546 _format_fields = {
547 # NB: Keep in sync with the docstring of extractor/common.py
a44ca5a4 548 'url', 'manifest_url', 'manifest_stream_number', 'ext', 'format', 'format_id', 'format_note',
b8ed0f15 549 'width', 'height', 'resolution', 'dynamic_range', 'tbr', 'abr', 'acodec', 'asr', 'audio_channels',
6db9c4d5 550 'vbr', 'fps', 'vcodec', 'container', 'filesize', 'filesize_approx',
551 'player_url', 'protocol', 'fragment_base_url', 'fragments', 'is_from_start',
552 'preference', 'language', 'language_preference', 'quality', 'source_preference',
553 'http_headers', 'stretched_ratio', 'no_resume', 'has_drm', 'downloader_options',
554 'page_url', 'app', 'play_path', 'tc_url', 'flash_version', 'rtmp_live', 'rtmp_conn', 'rtmp_protocol', 'rtmp_real_time'
555 }
48ee10ee 556 _format_selection_exts = {
8dc59305 557 'audio': set(MEDIA_EXTENSIONS.common_audio),
558 'video': set(MEDIA_EXTENSIONS.common_video + ('3gp', )),
559 'storyboards': set(MEDIA_EXTENSIONS.storyboards),
48ee10ee 560 }
561
3511266b 562 def __init__(self, params=None, auto_init=True):
883d4b1e 563 """Create a FileDownloader object with the given options.
564 @param auto_init Whether to load the default extractors and print header (if verbose).
49a57e70 565 Set to 'no_verbose_header' to not print the header
883d4b1e 566 """
e9f9a10f
JMF
567 if params is None:
568 params = {}
592b7485 569 self.params = params
8b7491c8 570 self._ies = {}
56c73665 571 self._ies_instances = {}
1e43a6f7 572 self._pps = {k: [] for k in POSTPROCESS_WHEN}
b35496d8 573 self._printed_messages = set()
1cf376f5 574 self._first_webpage_request = True
ab8e5e51 575 self._post_hooks = []
933605d7 576 self._progress_hooks = []
819e0531 577 self._postprocessor_hooks = []
8222d8de
JMF
578 self._download_retcode = 0
579 self._num_downloads = 0
9c906919 580 self._num_videos = 0
592b7485 581 self._playlist_level = 0
582 self._playlist_urls = set()
a0e07d31 583 self.cache = Cache(self)
34308b30 584
819e0531 585 windows_enable_vt_mode()
591bb9d3 586 stdout = sys.stderr if self.params.get('logtostderr') else sys.stdout
587 self._out_files = Namespace(
588 out=stdout,
589 error=sys.stderr,
590 screen=sys.stderr if self.params.get('quiet') else stdout,
591 console=None if compat_os_name == 'nt' else next(
cf4f42cb 592 filter(supports_terminal_sequences, (sys.stderr, sys.stdout)), None)
591bb9d3 593 )
594 self._allow_colors = Namespace(**{
595 type_: not self.params.get('no_color') and supports_terminal_sequences(stream)
64fa820c 596 for type_, stream in self._out_files.items_ if type_ != 'console'
591bb9d3 597 })
819e0531 598
6929b41a 599 # The code is left like this to be reused for future deprecations
600 MIN_SUPPORTED, MIN_RECOMMENDED = (3, 7), (3, 7)
eff42759 601 current_version = sys.version_info[:2]
602 if current_version < MIN_RECOMMENDED:
9d339c41 603 msg = ('Support for Python version %d.%d has been deprecated. '
24093d52 604 'See https://github.com/yt-dlp/yt-dlp/issues/3764 for more details.'
c6e07cf1 605 '\n You will no longer receive updates on this version')
eff42759 606 if current_version < MIN_SUPPORTED:
607 msg = 'Python version %d.%d is no longer supported'
608 self.deprecation_warning(
609 f'{msg}! Please update to Python %d.%d or above' % (*current_version, *MIN_RECOMMENDED))
a61f4b28 610
88acdbc2 611 if self.params.get('allow_unplayable_formats'):
612 self.report_warning(
ec11a9f4 613 f'You have asked for {self._format_err("UNPLAYABLE", self.Styles.EMPHASIS)} formats to be listed/downloaded. '
819e0531 614 'This is a developer option intended for debugging. \n'
615 ' If you experience any issues while using this option, '
ec11a9f4 616 f'{self._format_err("DO NOT", self.Styles.ERROR)} open a bug report')
88acdbc2 617
be5df5ee
S
618 def check_deprecated(param, option, suggestion):
619 if self.params.get(param) is not None:
86e5f3ed 620 self.report_warning(f'{option} is deprecated. Use {suggestion} instead')
be5df5ee
S
621 return True
622 return False
623
624 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
38cce791
YCH
625 if self.params.get('geo_verification_proxy') is None:
626 self.params['geo_verification_proxy'] = self.params['cn_verification_proxy']
627
0d1bb027 628 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
629 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
53ed7066 630 check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"')
0d1bb027 631
49a57e70 632 for msg in self.params.get('_warnings', []):
0d1bb027 633 self.report_warning(msg)
ee8dd27a 634 for msg in self.params.get('_deprecation_warnings', []):
da4db748 635 self.deprecated_feature(msg)
0d1bb027 636
8a82af35 637 self.params['compat_opts'] = set(self.params.get('compat_opts', ()))
8a82af35 638 if 'list-formats' in self.params['compat_opts']:
ec11a9f4 639 self.params['listformats_table'] = False
640
b5ae35ee 641 if 'overwrites' not in self.params and self.params.get('nooverwrites') is not None:
b868936c 642 # nooverwrites was unnecessarily changed to overwrites
643 # in 0c3d0f51778b153f65c21906031c2e091fcfb641
644 # This ensures compatibility with both keys
645 self.params['overwrites'] = not self.params['nooverwrites']
b5ae35ee 646 elif self.params.get('overwrites') is None:
647 self.params.pop('overwrites', None)
b868936c 648 else:
649 self.params['nooverwrites'] = not self.params['overwrites']
b9d973be 650
455a15e2 651 self.params.setdefault('forceprint', {})
652 self.params.setdefault('print_to_file', {})
bb66c247 653
654 # Compatibility with older syntax
ca30f449 655 if not isinstance(params['forceprint'], dict):
455a15e2 656 self.params['forceprint'] = {'video': params['forceprint']}
ca30f449 657
455a15e2 658 if self.params.get('bidi_workaround', False):
1c088fa8
PH
659 try:
660 import pty
661 master, slave = pty.openpty()
ac668111 662 width = shutil.get_terminal_size().columns
591bb9d3 663 width_args = [] if width is None else ['-w', str(width)]
664 sp_kwargs = {'stdin': subprocess.PIPE, 'stdout': slave, 'stderr': self._out_files.error}
5d681e96 665 try:
d3c93ec2 666 self._output_process = Popen(['bidiv'] + width_args, **sp_kwargs)
5d681e96 667 except OSError:
d3c93ec2 668 self._output_process = Popen(['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
5d681e96 669 self._output_channel = os.fdopen(master, 'rb')
1c088fa8 670 except OSError as ose:
66e7ace1 671 if ose.errno == errno.ENOENT:
49a57e70 672 self.report_warning(
673 'Could not find fribidi executable, ignoring --bidi-workaround. '
674 'Make sure that fribidi is an executable file in one of the directories in your $PATH.')
1c088fa8
PH
675 else:
676 raise
0783b09b 677
97ec5bc5 678 if auto_init:
679 if auto_init != 'no_verbose_header':
680 self.print_debug_header()
681 self.add_default_info_extractors()
682
3089bc74
S
683 if (sys.platform != 'win32'
684 and sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
455a15e2 685 and not self.params.get('restrictfilenames', False)):
e9137224 686 # Unicode filesystem API will throw errors (#1474, #13027)
34308b30 687 self.report_warning(
6febd1c1 688 'Assuming --restrict-filenames since file system encoding '
1b725173 689 'cannot encode all characters. '
6febd1c1 690 'Set the LC_ALL environment variable to fix this.')
4a98cdbf 691 self.params['restrictfilenames'] = True
34308b30 692
bf1824b3 693 self._parse_outtmpl()
486dd09e 694
187986a8 695 # Creating format selector here allows us to catch syntax errors before the extraction
696 self.format_selector = (
fa9f30b8 697 self.params.get('format') if self.params.get('format') in (None, '-')
093a1710 698 else self.params['format'] if callable(self.params['format'])
187986a8 699 else self.build_format_selector(self.params['format']))
700
8b7539d2 701 # Set http_headers defaults according to std_headers
702 self.params['http_headers'] = merge_headers(std_headers, self.params.get('http_headers', {}))
703
013b50b7 704 hooks = {
705 'post_hooks': self.add_post_hook,
706 'progress_hooks': self.add_progress_hook,
707 'postprocessor_hooks': self.add_postprocessor_hook,
708 }
709 for opt, fn in hooks.items():
710 for ph in self.params.get(opt, []):
711 fn(ph)
71b640cc 712
5bfc8bee 713 for pp_def_raw in self.params.get('postprocessors', []):
714 pp_def = dict(pp_def_raw)
715 when = pp_def.pop('when', 'post_process')
716 self.add_post_processor(
f9934b96 717 get_postprocessor(pp_def.pop('key'))(self, **pp_def),
5bfc8bee 718 when=when)
719
97ec5bc5 720 self._setup_opener()
51fb4995
YCH
721 register_socks_protocols()
722
ed39cac5 723 def preload_download_archive(fn):
724 """Preload the archive, if any is specified"""
725 if fn is None:
726 return False
49a57e70 727 self.write_debug(f'Loading archive file {fn!r}')
ed39cac5 728 try:
729 with locked_file(fn, 'r', encoding='utf-8') as archive_file:
730 for line in archive_file:
731 self.archive.add(line.strip())
86e5f3ed 732 except OSError as ioe:
ed39cac5 733 if ioe.errno != errno.ENOENT:
734 raise
735 return False
736 return True
737
738 self.archive = set()
739 preload_download_archive(self.params.get('download_archive'))
740
7d4111ed
PH
741 def warn_if_short_id(self, argv):
742 # short YouTube ID starting with dash?
743 idxs = [
744 i for i, a in enumerate(argv)
745 if re.match(r'^-[0-9A-Za-z_-]{10}$', a)]
746 if idxs:
747 correct_argv = (
7a5c1cfe 748 ['yt-dlp']
3089bc74
S
749 + [a for i, a in enumerate(argv) if i not in idxs]
750 + ['--'] + [argv[i] for i in idxs]
7d4111ed
PH
751 )
752 self.report_warning(
753 'Long argument string detected. '
49a57e70 754 'Use -- to separate parameters and URLs, like this:\n%s' %
7d4111ed
PH
755 args_to_str(correct_argv))
756
8222d8de
JMF
757 def add_info_extractor(self, ie):
758 """Add an InfoExtractor object to the end of the list."""
8b7491c8 759 ie_key = ie.ie_key()
760 self._ies[ie_key] = ie
e52d7f85 761 if not isinstance(ie, type):
8b7491c8 762 self._ies_instances[ie_key] = ie
e52d7f85 763 ie.set_downloader(self)
8222d8de 764
56c73665
JMF
765 def get_info_extractor(self, ie_key):
766 """
767 Get an instance of an IE with name ie_key, it will try to get one from
768 the _ies list, if there's no instance it will create a new one and add
769 it to the extractor list.
770 """
771 ie = self._ies_instances.get(ie_key)
772 if ie is None:
773 ie = get_info_extractor(ie_key)()
774 self.add_info_extractor(ie)
775 return ie
776
023fa8c4
JMF
777 def add_default_info_extractors(self):
778 """
779 Add the InfoExtractors returned by gen_extractors to the end of the list
780 """
fe7866d0 781 all_ies = {ie.IE_NAME.lower(): ie for ie in gen_extractor_classes()}
782 all_ies['end'] = UnsupportedURLIE()
783 try:
784 ie_names = orderedSet_from_options(
785 self.params.get('allowed_extractors', ['default']), {
786 'all': list(all_ies),
787 'default': [name for name, ie in all_ies.items() if ie._ENABLED],
788 }, use_regex=True)
789 except re.error as e:
790 raise ValueError(f'Wrong regex for allowed_extractors: {e.pattern}')
791 for name in ie_names:
792 self.add_info_extractor(all_ies[name])
793 self.write_debug(f'Loaded {len(ie_names)} extractors')
023fa8c4 794
56d868db 795 def add_post_processor(self, pp, when='post_process'):
8222d8de 796 """Add a PostProcessor object to the end of the chain."""
8aa0e7cd 797 assert when in POSTPROCESS_WHEN, f'Invalid when={when}'
5bfa4862 798 self._pps[when].append(pp)
8222d8de
JMF
799 pp.set_downloader(self)
800
ab8e5e51
AM
801 def add_post_hook(self, ph):
802 """Add the post hook"""
803 self._post_hooks.append(ph)
804
933605d7 805 def add_progress_hook(self, ph):
819e0531 806 """Add the download progress hook"""
933605d7 807 self._progress_hooks.append(ph)
8ab470f1 808
819e0531 809 def add_postprocessor_hook(self, ph):
810 """Add the postprocessing progress hook"""
811 self._postprocessor_hooks.append(ph)
5bfc8bee 812 for pps in self._pps.values():
813 for pp in pps:
814 pp.add_progress_hook(ph)
819e0531 815
1c088fa8 816 def _bidi_workaround(self, message):
5d681e96 817 if not hasattr(self, '_output_channel'):
1c088fa8
PH
818 return message
819
5d681e96 820 assert hasattr(self, '_output_process')
14f25df2 821 assert isinstance(message, str)
6febd1c1 822 line_count = message.count('\n') + 1
0f06bcd7 823 self._output_process.stdin.write((message + '\n').encode())
5d681e96 824 self._output_process.stdin.flush()
0f06bcd7 825 res = ''.join(self._output_channel.readline().decode()
9e1a5b84 826 for _ in range(line_count))
6febd1c1 827 return res[:-len('\n')]
1c088fa8 828
b35496d8 829 def _write_string(self, message, out=None, only_once=False):
830 if only_once:
831 if message in self._printed_messages:
832 return
833 self._printed_messages.add(message)
834 write_string(message, out=out, encoding=self.params.get('encoding'))
734f90bb 835
cf4f42cb 836 def to_stdout(self, message, skip_eol=False, quiet=None):
0760b0a7 837 """Print message to stdout"""
cf4f42cb 838 if quiet is not None:
da4db748 839 self.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument quiet. '
840 'Use "YoutubeDL.to_screen" instead')
8a82af35 841 if skip_eol is not False:
da4db748 842 self.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument skip_eol. '
843 'Use "YoutubeDL.to_screen" instead')
0bf9dc1e 844 self._write_string(f'{self._bidi_workaround(message)}\n', self._out_files.out)
cf4f42cb 845
846 def to_screen(self, message, skip_eol=False, quiet=None):
847 """Print message to screen if not in quiet mode"""
8bf9319e 848 if self.params.get('logger'):
43afe285 849 self.params['logger'].debug(message)
cf4f42cb 850 return
851 if (self.params.get('quiet') if quiet is None else quiet) and not self.params.get('verbose'):
852 return
853 self._write_string(
854 '%s%s' % (self._bidi_workaround(message), ('' if skip_eol else '\n')),
591bb9d3 855 self._out_files.screen)
8222d8de 856
b35496d8 857 def to_stderr(self, message, only_once=False):
0760b0a7 858 """Print message to stderr"""
14f25df2 859 assert isinstance(message, str)
8bf9319e 860 if self.params.get('logger'):
43afe285
IB
861 self.params['logger'].error(message)
862 else:
5792c950 863 self._write_string(f'{self._bidi_workaround(message)}\n', self._out_files.error, only_once=only_once)
cf4f42cb 864
865 def _send_console_code(self, code):
591bb9d3 866 if compat_os_name == 'nt' or not self._out_files.console:
cf4f42cb 867 return
591bb9d3 868 self._write_string(code, self._out_files.console)
8222d8de 869
1e5b9a95
PH
870 def to_console_title(self, message):
871 if not self.params.get('consoletitle', False):
872 return
3efb96a6 873 message = remove_terminal_sequences(message)
4bede0d8
C
874 if compat_os_name == 'nt':
875 if ctypes.windll.kernel32.GetConsoleWindow():
876 # c_wchar_p() might not be necessary if `message` is
877 # already of type unicode()
878 ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
cf4f42cb 879 else:
880 self._send_console_code(f'\033]0;{message}\007')
1e5b9a95 881
bdde425c 882 def save_console_title(self):
cf4f42cb 883 if not self.params.get('consoletitle') or self.params.get('simulate'):
bdde425c 884 return
592b7485 885 self._send_console_code('\033[22;0t') # Save the title on stack
bdde425c
PH
886
887 def restore_console_title(self):
cf4f42cb 888 if not self.params.get('consoletitle') or self.params.get('simulate'):
bdde425c 889 return
592b7485 890 self._send_console_code('\033[23;0t') # Restore the title from stack
bdde425c
PH
891
892 def __enter__(self):
893 self.save_console_title()
894 return self
895
896 def __exit__(self, *args):
897 self.restore_console_title()
f89197d7 898
dca08720 899 if self.params.get('cookiefile') is not None:
1bab3437 900 self.cookiejar.save(ignore_discard=True, ignore_expires=True)
bdde425c 901
fa9f30b8 902 def trouble(self, message=None, tb=None, is_error=True):
8222d8de
JMF
903 """Determine action to take when a download problem appears.
904
905 Depending on if the downloader has been configured to ignore
906 download errors or not, this method may throw an exception or
907 not when errors are found, after printing the message.
908
fa9f30b8 909 @param tb If given, is additional traceback information
910 @param is_error Whether to raise error according to ignorerrors
8222d8de
JMF
911 """
912 if message is not None:
913 self.to_stderr(message)
914 if self.params.get('verbose'):
915 if tb is None:
916 if sys.exc_info()[0]: # if .trouble has been called from an except block
6febd1c1 917 tb = ''
8222d8de 918 if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
6febd1c1 919 tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
c0384f22 920 tb += encode_compat_str(traceback.format_exc())
8222d8de
JMF
921 else:
922 tb_data = traceback.format_list(traceback.extract_stack())
6febd1c1 923 tb = ''.join(tb_data)
c19bc311 924 if tb:
925 self.to_stderr(tb)
fa9f30b8 926 if not is_error:
927 return
b1940459 928 if not self.params.get('ignoreerrors'):
8222d8de
JMF
929 if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
930 exc_info = sys.exc_info()[1].exc_info
931 else:
932 exc_info = sys.exc_info()
933 raise DownloadError(message, exc_info)
934 self._download_retcode = 1
935
19a03940 936 Styles = Namespace(
937 HEADERS='yellow',
938 EMPHASIS='light blue',
492272fe 939 FILENAME='green',
19a03940 940 ID='green',
941 DELIM='blue',
942 ERROR='red',
943 WARNING='yellow',
944 SUPPRESS='light black',
945 )
ec11a9f4 946
7578d77d 947 def _format_text(self, handle, allow_colors, text, f, fallback=None, *, test_encoding=False):
e5a998f3 948 text = str(text)
ec11a9f4 949 if test_encoding:
950 original_text = text
5c104538 951 # handle.encoding can be None. See https://github.com/yt-dlp/yt-dlp/issues/2711
952 encoding = self.params.get('encoding') or getattr(handle, 'encoding', None) or 'ascii'
ec11a9f4 953 text = text.encode(encoding, 'ignore').decode(encoding)
954 if fallback is not None and text != original_text:
955 text = fallback
7578d77d 956 return format_text(text, f) if allow_colors else text if fallback is None else fallback
ec11a9f4 957
591bb9d3 958 def _format_out(self, *args, **kwargs):
959 return self._format_text(self._out_files.out, self._allow_colors.out, *args, **kwargs)
960
ec11a9f4 961 def _format_screen(self, *args, **kwargs):
591bb9d3 962 return self._format_text(self._out_files.screen, self._allow_colors.screen, *args, **kwargs)
ec11a9f4 963
964 def _format_err(self, *args, **kwargs):
591bb9d3 965 return self._format_text(self._out_files.error, self._allow_colors.error, *args, **kwargs)
819e0531 966
c84aeac6 967 def report_warning(self, message, only_once=False):
8222d8de
JMF
968 '''
969 Print the message to stderr, it will be prefixed with 'WARNING:'
970 If stderr is a tty file the 'WARNING:' will be colored
971 '''
6d07ce01
JMF
972 if self.params.get('logger') is not None:
973 self.params['logger'].warning(message)
8222d8de 974 else:
ad8915b7
PH
975 if self.params.get('no_warnings'):
976 return
ec11a9f4 977 self.to_stderr(f'{self._format_err("WARNING:", self.Styles.WARNING)} {message}', only_once)
8222d8de 978
da4db748 979 def deprecation_warning(self, message, *, stacklevel=0):
980 deprecation_warning(
981 message, stacklevel=stacklevel + 1, printer=self.report_error, is_error=False)
982
983 def deprecated_feature(self, message):
ee8dd27a 984 if self.params.get('logger') is not None:
da4db748 985 self.params['logger'].warning(f'Deprecated Feature: {message}')
986 self.to_stderr(f'{self._format_err("Deprecated Feature:", self.Styles.ERROR)} {message}', True)
ee8dd27a 987
fa9f30b8 988 def report_error(self, message, *args, **kwargs):
8222d8de
JMF
989 '''
990 Do the same as trouble, but prefixes the message with 'ERROR:', colored
991 in red if stderr is a tty file.
992 '''
fa9f30b8 993 self.trouble(f'{self._format_err("ERROR:", self.Styles.ERROR)} {message}', *args, **kwargs)
8222d8de 994
b35496d8 995 def write_debug(self, message, only_once=False):
0760b0a7 996 '''Log debug message or Print message to stderr'''
997 if not self.params.get('verbose', False):
998 return
8a82af35 999 message = f'[debug] {message}'
0760b0a7 1000 if self.params.get('logger'):
1001 self.params['logger'].debug(message)
1002 else:
b35496d8 1003 self.to_stderr(message, only_once)
0760b0a7 1004
8222d8de
JMF
1005 def report_file_already_downloaded(self, file_name):
1006 """Report file has already been fully downloaded."""
1007 try:
6febd1c1 1008 self.to_screen('[download] %s has already been downloaded' % file_name)
ce02ed60 1009 except UnicodeEncodeError:
6febd1c1 1010 self.to_screen('[download] The file has already been downloaded')
8222d8de 1011
0c3d0f51 1012 def report_file_delete(self, file_name):
1013 """Report that existing file will be deleted."""
1014 try:
c25228e5 1015 self.to_screen('Deleting existing file %s' % file_name)
0c3d0f51 1016 except UnicodeEncodeError:
c25228e5 1017 self.to_screen('Deleting existing file')
0c3d0f51 1018
319b6059 1019 def raise_no_formats(self, info, forced=False, *, msg=None):
0a5a191a 1020 has_drm = info.get('_has_drm')
319b6059 1021 ignored, expected = self.params.get('ignore_no_formats_error'), bool(msg)
1022 msg = msg or has_drm and 'This video is DRM protected' or 'No video formats found!'
1023 if forced or not ignored:
1151c407 1024 raise ExtractorError(msg, video_id=info['id'], ie=info['extractor'],
319b6059 1025 expected=has_drm or ignored or expected)
88acdbc2 1026 else:
1027 self.report_warning(msg)
1028
de6000d9 1029 def parse_outtmpl(self):
bf1824b3 1030 self.deprecation_warning('"YoutubeDL.parse_outtmpl" is deprecated and may be removed in a future version')
1031 self._parse_outtmpl()
1032 return self.params['outtmpl']
1033
1034 def _parse_outtmpl(self):
7b2c3f47 1035 sanitize = IDENTITY
bf1824b3 1036 if self.params.get('restrictfilenames'): # Remove spaces in the default template
71ce444a 1037 sanitize = lambda x: x.replace(' - ', ' ').replace(' ', '-')
bf1824b3 1038
1039 outtmpl = self.params.setdefault('outtmpl', {})
1040 if not isinstance(outtmpl, dict):
1041 self.params['outtmpl'] = outtmpl = {'default': outtmpl}
1042 outtmpl.update({k: sanitize(v) for k, v in DEFAULT_OUTTMPL.items() if outtmpl.get(k) is None})
de6000d9 1043
21cd8fae 1044 def get_output_path(self, dir_type='', filename=None):
1045 paths = self.params.get('paths', {})
1046 assert isinstance(paths, dict)
1047 path = os.path.join(
1048 expand_path(paths.get('home', '').strip()),
1049 expand_path(paths.get(dir_type, '').strip()) if dir_type else '',
1050 filename or '')
21cd8fae 1051 return sanitize_path(path, force=self.params.get('windowsfilenames'))
1052
76a264ac 1053 @staticmethod
901130bb 1054 def _outtmpl_expandpath(outtmpl):
1055 # expand_path translates '%%' into '%' and '$$' into '$'
1056 # correspondingly that is not what we want since we need to keep
1057 # '%%' intact for template dict substitution step. Working around
1058 # with boundary-alike separator hack.
1059 sep = ''.join([random.choice(ascii_letters) for _ in range(32)])
86e5f3ed 1060 outtmpl = outtmpl.replace('%%', f'%{sep}%').replace('$$', f'${sep}$')
901130bb 1061
1062 # outtmpl should be expand_path'ed before template dict substitution
1063 # because meta fields may contain env variables we don't want to
62b58c09 1064 # be expanded. E.g. for outtmpl "%(title)s.%(ext)s" and
901130bb 1065 # title "Hello $PATH", we don't want `$PATH` to be expanded.
1066 return expand_path(outtmpl).replace(sep, '')
1067
1068 @staticmethod
1069 def escape_outtmpl(outtmpl):
1070 ''' Escape any remaining strings like %s, %abc% etc. '''
1071 return re.sub(
1072 STR_FORMAT_RE_TMPL.format('', '(?![%(\0])'),
1073 lambda mobj: ('' if mobj.group('has_key') else '%') + mobj.group(0),
1074 outtmpl)
1075
1076 @classmethod
1077 def validate_outtmpl(cls, outtmpl):
76a264ac 1078 ''' @return None or Exception object '''
7d1eb38a 1079 outtmpl = re.sub(
47cdc68e 1080 STR_FORMAT_RE_TMPL.format('[^)]*', '[ljhqBUDS]'),
7d1eb38a 1081 lambda mobj: f'{mobj.group(0)[:-1]}s',
1082 cls._outtmpl_expandpath(outtmpl))
76a264ac 1083 try:
7d1eb38a 1084 cls.escape_outtmpl(outtmpl) % collections.defaultdict(int)
76a264ac 1085 return None
1086 except ValueError as err:
1087 return err
1088
03b4de72 1089 @staticmethod
1090 def _copy_infodict(info_dict):
1091 info_dict = dict(info_dict)
09b49e1f 1092 info_dict.pop('__postprocessors', None)
415f8d51 1093 info_dict.pop('__pending_error', None)
03b4de72 1094 return info_dict
1095
e0fd9573 1096 def prepare_outtmpl(self, outtmpl, info_dict, sanitize=False):
1097 """ Make the outtmpl and info_dict suitable for substitution: ydl.escape_outtmpl(outtmpl) % info_dict
1098 @param sanitize Whether to sanitize the output as a filename.
1099 For backward compatibility, a function can also be passed
1100 """
1101
6e84b215 1102 info_dict.setdefault('epoch', int(time.time())) # keep epoch consistent once set
143db31d 1103
03b4de72 1104 info_dict = self._copy_infodict(info_dict)
752cda38 1105 info_dict['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
53c18592 1106 formatSeconds(info_dict['duration'], '-' if sanitize else ':')
143db31d 1107 if info_dict.get('duration', None) is not None
1108 else None)
1d485a1a 1109 info_dict['autonumber'] = int(self.params.get('autonumber_start', 1) - 1 + self._num_downloads)
9c906919 1110 info_dict['video_autonumber'] = self._num_videos
752cda38 1111 if info_dict.get('resolution') is None:
1112 info_dict['resolution'] = self.format_resolution(info_dict, default=None)
143db31d 1113
e6f21b3d 1114 # For fields playlist_index, playlist_autonumber and autonumber convert all occurrences
143db31d 1115 # of %(field)s to %(field)0Nd for backward compatibility
1116 field_size_compat_map = {
0a5a191a 1117 'playlist_index': number_of_digits(info_dict.get('__last_playlist_index') or 0),
ec11a9f4 1118 'playlist_autonumber': number_of_digits(info_dict.get('n_entries') or 0),
752cda38 1119 'autonumber': self.params.get('autonumber_size') or 5,
143db31d 1120 }
752cda38 1121
385a27fa 1122 TMPL_DICT = {}
47cdc68e 1123 EXTERNAL_FORMAT_RE = re.compile(STR_FORMAT_RE_TMPL.format('[^)]*', f'[{STR_FORMAT_TYPES}ljhqBUDS]'))
385a27fa 1124 MATH_FUNCTIONS = {
1125 '+': float.__add__,
1126 '-': float.__sub__,
1127 }
e625be0d 1128 # Field is of the form key1.key2...
1129 # where keys (except first) can be string, int or slice
2b8a2973 1130 FIELD_RE = r'\w*(?:\.(?:\w+|{num}|{num}?(?::{num}?){{1,2}}))*'.format(num=r'(?:-?\d+)')
1d485a1a 1131 MATH_FIELD_RE = rf'(?:{FIELD_RE}|-?{NUMBER_RE})'
385a27fa 1132 MATH_OPERATORS_RE = r'(?:%s)' % '|'.join(map(re.escape, MATH_FUNCTIONS.keys()))
1d485a1a 1133 INTERNAL_FORMAT_RE = re.compile(rf'''(?x)
e625be0d 1134 (?P<negate>-)?
1d485a1a 1135 (?P<fields>{FIELD_RE})
1136 (?P<maths>(?:{MATH_OPERATORS_RE}{MATH_FIELD_RE})*)
e625be0d 1137 (?:>(?P<strf_format>.+?))?
34baa9fd 1138 (?P<remaining>
1139 (?P<alternate>(?<!\\),[^|&)]+)?
1140 (?:&(?P<replacement>.*?))?
1141 (?:\|(?P<default>.*?))?
1d485a1a 1142 )$''')
752cda38 1143
2b8a2973 1144 def _traverse_infodict(k):
1145 k = k.split('.')
1146 if k[0] == '':
1147 k.pop(0)
1148 return traverse_obj(info_dict, k, is_user_input=True, traverse_string=True)
76a264ac 1149
752cda38 1150 def get_value(mdict):
1151 # Object traversal
2b8a2973 1152 value = _traverse_infodict(mdict['fields'])
752cda38 1153 # Negative
1154 if mdict['negate']:
1155 value = float_or_none(value)
1156 if value is not None:
1157 value *= -1
1158 # Do maths
385a27fa 1159 offset_key = mdict['maths']
1160 if offset_key:
752cda38 1161 value = float_or_none(value)
1162 operator = None
385a27fa 1163 while offset_key:
1164 item = re.match(
1165 MATH_FIELD_RE if operator else MATH_OPERATORS_RE,
1166 offset_key).group(0)
1167 offset_key = offset_key[len(item):]
1168 if operator is None:
752cda38 1169 operator = MATH_FUNCTIONS[item]
385a27fa 1170 continue
1171 item, multiplier = (item[1:], -1) if item[0] == '-' else (item, 1)
1172 offset = float_or_none(item)
1173 if offset is None:
2b8a2973 1174 offset = float_or_none(_traverse_infodict(item))
385a27fa 1175 try:
1176 value = operator(value, multiplier * offset)
1177 except (TypeError, ZeroDivisionError):
1178 return None
1179 operator = None
752cda38 1180 # Datetime formatting
1181 if mdict['strf_format']:
7c37ff97 1182 value = strftime_or_none(value, mdict['strf_format'].replace('\\,', ','))
752cda38 1183
a6bcaf71 1184 # XXX: Workaround for https://github.com/yt-dlp/yt-dlp/issues/4485
1185 if sanitize and value == '':
1186 value = None
752cda38 1187 return value
1188
b868936c 1189 na = self.params.get('outtmpl_na_placeholder', 'NA')
1190
e0fd9573 1191 def filename_sanitizer(key, value, restricted=self.params.get('restrictfilenames')):
5c3895ff 1192 return sanitize_filename(str(value), restricted=restricted, is_id=(
1193 bool(re.search(r'(^|[_.])id(\.|$)', key))
8a82af35 1194 if 'filename-sanitization' in self.params['compat_opts']
5c3895ff 1195 else NO_DEFAULT))
e0fd9573 1196
1197 sanitizer = sanitize if callable(sanitize) else filename_sanitizer
1198 sanitize = bool(sanitize)
1199
6e84b215 1200 def _dumpjson_default(obj):
1201 if isinstance(obj, (set, LazyList)):
1202 return list(obj)
adbc4ec4 1203 return repr(obj)
6e84b215 1204
752cda38 1205 def create_key(outer_mobj):
1206 if not outer_mobj.group('has_key'):
b836dc94 1207 return outer_mobj.group(0)
752cda38 1208 key = outer_mobj.group('key')
752cda38 1209 mobj = re.match(INTERNAL_FORMAT_RE, key)
e0fd9573 1210 initial_field = mobj.group('fields') if mobj else ''
e978789f 1211 value, replacement, default = None, None, na
7c37ff97 1212 while mobj:
e625be0d 1213 mobj = mobj.groupdict()
7c37ff97 1214 default = mobj['default'] if mobj['default'] is not None else default
752cda38 1215 value = get_value(mobj)
e978789f 1216 replacement = mobj['replacement']
7c37ff97 1217 if value is None and mobj['alternate']:
34baa9fd 1218 mobj = re.match(INTERNAL_FORMAT_RE, mobj['remaining'][1:])
7c37ff97 1219 else:
1220 break
752cda38 1221
b868936c 1222 fmt = outer_mobj.group('format')
752cda38 1223 if fmt == 's' and value is not None and key in field_size_compat_map.keys():
86e5f3ed 1224 fmt = f'0{field_size_compat_map[key]:d}d'
752cda38 1225
e978789f 1226 value = default if value is None else value if replacement is None else replacement
752cda38 1227
4476d2c7 1228 flags = outer_mobj.group('conversion') or ''
7d1eb38a 1229 str_fmt = f'{fmt[:-1]}s'
524e2e4f 1230 if fmt[-1] == 'l': # list
4476d2c7 1231 delim = '\n' if '#' in flags else ', '
9e907ebd 1232 value, fmt = delim.join(map(str, variadic(value, allowed_types=(str, bytes)))), str_fmt
524e2e4f 1233 elif fmt[-1] == 'j': # json
4476d2c7 1234 value, fmt = json.dumps(value, default=_dumpjson_default, indent=4 if '#' in flags else None), str_fmt
47cdc68e 1235 elif fmt[-1] == 'h': # html
1236 value, fmt = escapeHTML(value), str_fmt
524e2e4f 1237 elif fmt[-1] == 'q': # quoted
4476d2c7 1238 value = map(str, variadic(value) if '#' in flags else [value])
1239 value, fmt = ' '.join(map(compat_shlex_quote, value)), str_fmt
524e2e4f 1240 elif fmt[-1] == 'B': # bytes
0f06bcd7 1241 value = f'%{str_fmt}'.encode() % str(value).encode()
f5aa5cfb 1242 value, fmt = value.decode('utf-8', 'ignore'), 's'
524e2e4f 1243 elif fmt[-1] == 'U': # unicode normalized
524e2e4f 1244 value, fmt = unicodedata.normalize(
1245 # "+" = compatibility equivalence, "#" = NFD
4476d2c7 1246 'NF%s%s' % ('K' if '+' in flags else '', 'D' if '#' in flags else 'C'),
524e2e4f 1247 value), str_fmt
e0fd9573 1248 elif fmt[-1] == 'D': # decimal suffix
abbeeebc 1249 num_fmt, fmt = fmt[:-1].replace('#', ''), 's'
1250 value = format_decimal_suffix(value, f'%{num_fmt}f%s' if num_fmt else '%d%s',
1251 factor=1024 if '#' in flags else 1000)
37893bb0 1252 elif fmt[-1] == 'S': # filename sanitization
e0fd9573 1253 value, fmt = filename_sanitizer(initial_field, value, restricted='#' in flags), str_fmt
7d1eb38a 1254 elif fmt[-1] == 'c':
524e2e4f 1255 if value:
1256 value = str(value)[0]
76a264ac 1257 else:
524e2e4f 1258 fmt = str_fmt
76a264ac 1259 elif fmt[-1] not in 'rs': # numeric
a439a3a4 1260 value = float_or_none(value)
752cda38 1261 if value is None:
1262 value, fmt = default, 's'
901130bb 1263
752cda38 1264 if sanitize:
1265 if fmt[-1] == 'r':
1266 # If value is an object, sanitize might convert it to a string
1267 # So we convert it to repr first
7d1eb38a 1268 value, fmt = repr(value), str_fmt
639f1cea 1269 if fmt[-1] in 'csr':
e0fd9573 1270 value = sanitizer(initial_field, value)
901130bb 1271
b868936c 1272 key = '%s\0%s' % (key.replace('%', '%\0'), outer_mobj.group('format'))
385a27fa 1273 TMPL_DICT[key] = value
b868936c 1274 return '{prefix}%({key}){fmt}'.format(key=key, fmt=fmt, prefix=outer_mobj.group('prefix'))
752cda38 1275
385a27fa 1276 return EXTERNAL_FORMAT_RE.sub(create_key, outtmpl), TMPL_DICT
143db31d 1277
819e0531 1278 def evaluate_outtmpl(self, outtmpl, info_dict, *args, **kwargs):
1279 outtmpl, info_dict = self.prepare_outtmpl(outtmpl, info_dict, *args, **kwargs)
1280 return self.escape_outtmpl(outtmpl) % info_dict
1281
5127e92a 1282 def _prepare_filename(self, info_dict, *, outtmpl=None, tmpl_type=None):
1283 assert None in (outtmpl, tmpl_type), 'outtmpl and tmpl_type are mutually exclusive'
1284 if outtmpl is None:
bf1824b3 1285 outtmpl = self.params['outtmpl'].get(tmpl_type or 'default', self.params['outtmpl']['default'])
8222d8de 1286 try:
5127e92a 1287 outtmpl = self._outtmpl_expandpath(outtmpl)
e0fd9573 1288 filename = self.evaluate_outtmpl(outtmpl, info_dict, True)
6a0546e3 1289 if not filename:
1290 return None
15da37c7 1291
5127e92a 1292 if tmpl_type in ('', 'temp'):
6a0546e3 1293 final_ext, ext = self.params.get('final_ext'), info_dict.get('ext')
1294 if final_ext and ext and final_ext != ext and filename.endswith(f'.{final_ext}'):
1295 filename = replace_extension(filename, ext, final_ext)
5127e92a 1296 elif tmpl_type:
6a0546e3 1297 force_ext = OUTTMPL_TYPES[tmpl_type]
1298 if force_ext:
1299 filename = replace_extension(filename, force_ext, info_dict.get('ext'))
de6000d9 1300
bdc3fd2f
U
1301 # https://github.com/blackjack4494/youtube-dlc/issues/85
1302 trim_file_name = self.params.get('trim_file_name', False)
1303 if trim_file_name:
5c22c63d 1304 no_ext, *ext = filename.rsplit('.', 2)
1305 filename = join_nonempty(no_ext[:trim_file_name], *ext, delim='.')
bdc3fd2f 1306
0202b52a 1307 return filename
8222d8de 1308 except ValueError as err:
6febd1c1 1309 self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')')
8222d8de
JMF
1310 return None
1311
5127e92a 1312 def prepare_filename(self, info_dict, dir_type='', *, outtmpl=None, warn=False):
1313 """Generate the output filename"""
1314 if outtmpl:
1315 assert not dir_type, 'outtmpl and dir_type are mutually exclusive'
1316 dir_type = None
1317 filename = self._prepare_filename(info_dict, tmpl_type=dir_type, outtmpl=outtmpl)
80c03fa9 1318 if not filename and dir_type not in ('', 'temp'):
1319 return ''
de6000d9 1320
c84aeac6 1321 if warn:
21cd8fae 1322 if not self.params.get('paths'):
de6000d9 1323 pass
1324 elif filename == '-':
c84aeac6 1325 self.report_warning('--paths is ignored when an outputting to stdout', only_once=True)
de6000d9 1326 elif os.path.isabs(filename):
c84aeac6 1327 self.report_warning('--paths is ignored since an absolute path is given in output template', only_once=True)
de6000d9 1328 if filename == '-' or not filename:
1329 return filename
1330
21cd8fae 1331 return self.get_output_path(dir_type, filename)
0202b52a 1332
120fe513 1333 def _match_entry(self, info_dict, incomplete=False, silent=False):
ecdec191 1334 """ Returns None if the file should be downloaded """
8222d8de 1335
3bec830a 1336 video_title = info_dict.get('title', info_dict.get('id', 'entry'))
c77495e3 1337
8b0d7497 1338 def check_filter():
8b0d7497 1339 if 'title' in info_dict:
1340 # This can happen when we're just evaluating the playlist
1341 title = info_dict['title']
1342 matchtitle = self.params.get('matchtitle', False)
1343 if matchtitle:
1344 if not re.search(matchtitle, title, re.IGNORECASE):
1345 return '"' + title + '" title did not match pattern "' + matchtitle + '"'
1346 rejecttitle = self.params.get('rejecttitle', False)
1347 if rejecttitle:
1348 if re.search(rejecttitle, title, re.IGNORECASE):
1349 return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
1350 date = info_dict.get('upload_date')
1351 if date is not None:
1352 dateRange = self.params.get('daterange', DateRange())
1353 if date not in dateRange:
86e5f3ed 1354 return f'{date_from_str(date).isoformat()} upload date is not in range {dateRange}'
8b0d7497 1355 view_count = info_dict.get('view_count')
1356 if view_count is not None:
1357 min_views = self.params.get('min_views')
1358 if min_views is not None and view_count < min_views:
1359 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
1360 max_views = self.params.get('max_views')
1361 if max_views is not None and view_count > max_views:
1362 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
1363 if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')):
1364 return 'Skipping "%s" because it is age restricted' % video_title
8b0d7497 1365
8f18aca8 1366 match_filter = self.params.get('match_filter')
1367 if match_filter is not None:
1368 try:
1369 ret = match_filter(info_dict, incomplete=incomplete)
1370 except TypeError:
1371 # For backward compatibility
1372 ret = None if incomplete else match_filter(info_dict)
492272fe 1373 if ret is NO_DEFAULT:
1374 while True:
1375 filename = self._format_screen(self.prepare_filename(info_dict), self.Styles.FILENAME)
1376 reply = input(self._format_screen(
1377 f'Download "{filename}"? (Y/n): ', self.Styles.EMPHASIS)).lower().strip()
1378 if reply in {'y', ''}:
1379 return None
1380 elif reply == 'n':
1381 return f'Skipping {video_title}'
492272fe 1382 elif ret is not None:
8f18aca8 1383 return ret
8b0d7497 1384 return None
1385
c77495e3 1386 if self.in_download_archive(info_dict):
1387 reason = '%s has already been recorded in the archive' % video_title
1388 break_opt, break_err = 'break_on_existing', ExistingVideoReached
1389 else:
1390 reason = check_filter()
1391 break_opt, break_err = 'break_on_reject', RejectedVideoReached
8b0d7497 1392 if reason is not None:
120fe513 1393 if not silent:
1394 self.to_screen('[download] ' + reason)
c77495e3 1395 if self.params.get(break_opt, False):
1396 raise break_err()
8b0d7497 1397 return reason
fe7e0c98 1398
b6c45014
JMF
1399 @staticmethod
1400 def add_extra_info(info_dict, extra_info):
1401 '''Set the keys from extra_info in info dict if they are missing'''
1402 for key, value in extra_info.items():
1403 info_dict.setdefault(key, value)
1404
409e1828 1405 def extract_info(self, url, download=True, ie_key=None, extra_info=None,
61aa5ba3 1406 process=True, force_generic_extractor=False):
41d1cca3 1407 """
1408 Return a list with a dictionary for each video extracted.
1409
1410 Arguments:
1411 url -- URL to extract
1412
1413 Keyword arguments:
1414 download -- whether to download videos during extraction
1415 ie_key -- extractor key hint
1416 extra_info -- dictionary containing the extra values to add to each result
1417 process -- whether to resolve all unresolved references (URLs, playlist items),
1418 must be True for download to work.
1419 force_generic_extractor -- force using the generic extractor
1420 """
fe7e0c98 1421
409e1828 1422 if extra_info is None:
1423 extra_info = {}
1424
61aa5ba3 1425 if not ie_key and force_generic_extractor:
d22dec74
S
1426 ie_key = 'Generic'
1427
8222d8de 1428 if ie_key:
fe7866d0 1429 ies = {ie_key: self._ies[ie_key]} if ie_key in self._ies else {}
8222d8de
JMF
1430 else:
1431 ies = self._ies
1432
fe7866d0 1433 for key, ie in ies.items():
8222d8de
JMF
1434 if not ie.suitable(url):
1435 continue
1436
1437 if not ie.working():
6febd1c1
PH
1438 self.report_warning('The program functionality for this site has been marked as broken, '
1439 'and will probably not work.')
8222d8de 1440
1151c407 1441 temp_id = ie.get_temp_id(url)
fe7866d0 1442 if temp_id is not None and self.in_download_archive({'id': temp_id, 'ie_key': key}):
1443 self.to_screen(f'[{key}] {temp_id}: has already been recorded in the archive')
5e5be0c0 1444 if self.params.get('break_on_existing', False):
1445 raise ExistingVideoReached()
a0566bbf 1446 break
fe7866d0 1447 return self.__extract_info(url, self.get_info_extractor(key), download, extra_info, process)
a0566bbf 1448 else:
fe7866d0 1449 extractors_restricted = self.params.get('allowed_extractors') not in (None, ['default'])
1450 self.report_error(f'No suitable extractor{format_field(ie_key, None, " (%s)")} found for URL {url}',
1451 tb=False if extractors_restricted else None)
a0566bbf 1452
7e88d7d7 1453 def _handle_extraction_exceptions(func):
b5ae35ee 1454 @functools.wraps(func)
a0566bbf 1455 def wrapper(self, *args, **kwargs):
6da22e7d 1456 while True:
1457 try:
1458 return func(self, *args, **kwargs)
1459 except (DownloadCancelled, LazyList.IndexError, PagedList.IndexError):
8222d8de 1460 raise
6da22e7d 1461 except ReExtractInfo as e:
1462 if e.expected:
1463 self.to_screen(f'{e}; Re-extracting data')
1464 else:
1465 self.to_stderr('\r')
1466 self.report_warning(f'{e}; Re-extracting data')
1467 continue
1468 except GeoRestrictedError as e:
1469 msg = e.msg
1470 if e.countries:
1471 msg += '\nThis video is available in %s.' % ', '.join(
1472 map(ISO3166Utils.short2full, e.countries))
1473 msg += '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1474 self.report_error(msg)
1475 except ExtractorError as e: # An error we somewhat expected
1476 self.report_error(str(e), e.format_traceback())
1477 except Exception as e:
1478 if self.params.get('ignoreerrors'):
1479 self.report_error(str(e), tb=encode_compat_str(traceback.format_exc()))
1480 else:
1481 raise
1482 break
a0566bbf 1483 return wrapper
1484
693f0600 1485 def _wait_for_video(self, ie_result={}):
f2ebc5c7 1486 if (not self.params.get('wait_for_video')
1487 or ie_result.get('_type', 'video') != 'video'
1488 or ie_result.get('formats') or ie_result.get('url')):
1489 return
1490
1491 format_dur = lambda dur: '%02d:%02d:%02d' % timetuple_from_msec(dur * 1000)[:-1]
1492 last_msg = ''
1493
1494 def progress(msg):
1495 nonlocal last_msg
a7dc6a89 1496 full_msg = f'{msg}\n'
1497 if not self.params.get('noprogress'):
1498 full_msg = msg + ' ' * (len(last_msg) - len(msg)) + '\r'
1499 elif last_msg:
1500 return
1501 self.to_screen(full_msg, skip_eol=True)
f2ebc5c7 1502 last_msg = msg
1503
1504 min_wait, max_wait = self.params.get('wait_for_video')
1505 diff = try_get(ie_result, lambda x: x['release_timestamp'] - time.time())
1506 if diff is None and ie_result.get('live_status') == 'is_upcoming':
16c620bc 1507 diff = round(random.uniform(min_wait, max_wait) if (max_wait and min_wait) else (max_wait or min_wait), 0)
f2ebc5c7 1508 self.report_warning('Release time of video is not known')
693f0600 1509 elif ie_result and (diff or 0) <= 0:
f2ebc5c7 1510 self.report_warning('Video should already be available according to extracted info')
38d79fd1 1511 diff = min(max(diff or 0, min_wait or 0), max_wait or float('inf'))
f2ebc5c7 1512 self.to_screen(f'[wait] Waiting for {format_dur(diff)} - Press Ctrl+C to try now')
1513
1514 wait_till = time.time() + diff
1515 try:
1516 while True:
1517 diff = wait_till - time.time()
1518 if diff <= 0:
1519 progress('')
1520 raise ReExtractInfo('[wait] Wait period ended', expected=True)
1521 progress(f'[wait] Remaining time until next attempt: {self._format_screen(format_dur(diff), self.Styles.EMPHASIS)}')
1522 time.sleep(1)
1523 except KeyboardInterrupt:
1524 progress('')
1525 raise ReExtractInfo('[wait] Interrupted by user', expected=True)
1526 except BaseException as e:
1527 if not isinstance(e, ReExtractInfo):
1528 self.to_screen('')
1529 raise
1530
7e88d7d7 1531 @_handle_extraction_exceptions
58f197b7 1532 def __extract_info(self, url, ie, download, extra_info, process):
693f0600 1533 try:
1534 ie_result = ie.extract(url)
1535 except UserNotLive as e:
1536 if process:
1537 if self.params.get('wait_for_video'):
1538 self.report_warning(e)
1539 self._wait_for_video()
1540 raise
a0566bbf 1541 if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
cb794ee0 1542 self.report_warning(f'Extractor {ie.IE_NAME} returned nothing{bug_reports_message()}')
a0566bbf 1543 return
1544 if isinstance(ie_result, list):
1545 # Backwards compatibility: old IE result format
1546 ie_result = {
1547 '_type': 'compat_list',
1548 'entries': ie_result,
1549 }
e37d0efb 1550 if extra_info.get('original_url'):
1551 ie_result.setdefault('original_url', extra_info['original_url'])
a0566bbf 1552 self.add_default_extra_info(ie_result, ie, url)
1553 if process:
f2ebc5c7 1554 self._wait_for_video(ie_result)
a0566bbf 1555 return self.process_ie_result(ie_result, download, extra_info)
8222d8de 1556 else:
a0566bbf 1557 return ie_result
fe7e0c98 1558
ea38e55f 1559 def add_default_extra_info(self, ie_result, ie, url):
6033d980 1560 if url is not None:
1561 self.add_extra_info(ie_result, {
1562 'webpage_url': url,
1563 'original_url': url,
57ebfca3 1564 })
1565 webpage_url = ie_result.get('webpage_url')
1566 if webpage_url:
1567 self.add_extra_info(ie_result, {
1568 'webpage_url_basename': url_basename(webpage_url),
1569 'webpage_url_domain': get_domain(webpage_url),
6033d980 1570 })
1571 if ie is not None:
1572 self.add_extra_info(ie_result, {
1573 'extractor': ie.IE_NAME,
1574 'extractor_key': ie.ie_key(),
1575 })
ea38e55f 1576
58adec46 1577 def process_ie_result(self, ie_result, download=True, extra_info=None):
8222d8de
JMF
1578 """
1579 Take the result of the ie(may be modified) and resolve all unresolved
1580 references (URLs, playlist items).
1581
1582 It will also download the videos if 'download'.
1583 Returns the resolved ie_result.
1584 """
58adec46 1585 if extra_info is None:
1586 extra_info = {}
e8ee972c
PH
1587 result_type = ie_result.get('_type', 'video')
1588
057a5206 1589 if result_type in ('url', 'url_transparent'):
8f97a15d 1590 ie_result['url'] = sanitize_url(
1591 ie_result['url'], scheme='http' if self.params.get('prefer_insecure') else 'https')
e37d0efb 1592 if ie_result.get('original_url'):
1593 extra_info.setdefault('original_url', ie_result['original_url'])
1594
057a5206 1595 extract_flat = self.params.get('extract_flat', False)
3089bc74
S
1596 if ((extract_flat == 'in_playlist' and 'playlist' in extra_info)
1597 or extract_flat is True):
ecb54191 1598 info_copy = ie_result.copy()
6033d980 1599 ie = try_get(ie_result.get('ie_key'), self.get_info_extractor)
360167b9 1600 if ie and not ie_result.get('id'):
4614bc22 1601 info_copy['id'] = ie.get_temp_id(ie_result['url'])
6033d980 1602 self.add_default_extra_info(info_copy, ie, ie_result['url'])
4614bc22 1603 self.add_extra_info(info_copy, extra_info)
b5475f11 1604 info_copy, _ = self.pre_process(info_copy)
ecb54191 1605 self.__forced_printings(info_copy, self.prepare_filename(info_copy), incomplete=True)
415f8d51 1606 self._raise_pending_errors(info_copy)
4614bc22 1607 if self.params.get('force_write_download_archive', False):
1608 self.record_download_archive(info_copy)
e8ee972c
PH
1609 return ie_result
1610
8222d8de 1611 if result_type == 'video':
b6c45014 1612 self.add_extra_info(ie_result, extra_info)
9c2b75b5 1613 ie_result = self.process_video_result(ie_result, download=download)
415f8d51 1614 self._raise_pending_errors(ie_result)
28b0eb0f 1615 additional_urls = (ie_result or {}).get('additional_urls')
9c2b75b5 1616 if additional_urls:
e9f4ccd1 1617 # TODO: Improve MetadataParserPP to allow setting a list
14f25df2 1618 if isinstance(additional_urls, str):
9c2b75b5 1619 additional_urls = [additional_urls]
1620 self.to_screen(
1621 '[info] %s: %d additional URL(s) requested' % (ie_result['id'], len(additional_urls)))
1622 self.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls))
1623 ie_result['additional_entries'] = [
1624 self.extract_info(
b69fd25c 1625 url, download, extra_info=extra_info,
9c2b75b5 1626 force_generic_extractor=self.params.get('force_generic_extractor'))
1627 for url in additional_urls
1628 ]
1629 return ie_result
8222d8de
JMF
1630 elif result_type == 'url':
1631 # We have to add extra_info to the results because it may be
1632 # contained in a playlist
07cce701 1633 return self.extract_info(
1634 ie_result['url'], download,
1635 ie_key=ie_result.get('ie_key'),
1636 extra_info=extra_info)
7fc3fa05
PH
1637 elif result_type == 'url_transparent':
1638 # Use the information from the embedding page
1639 info = self.extract_info(
1640 ie_result['url'], ie_key=ie_result.get('ie_key'),
1641 extra_info=extra_info, download=False, process=False)
1642
1640eb09
S
1643 # extract_info may return None when ignoreerrors is enabled and
1644 # extraction failed with an error, don't crash and return early
1645 # in this case
1646 if not info:
1647 return info
1648
3975b4d2 1649 exempted_fields = {'_type', 'url', 'ie_key'}
1650 if not ie_result.get('section_end') and ie_result.get('section_start') is None:
1651 # For video clips, the id etc of the clip extractor should be used
1652 exempted_fields |= {'id', 'extractor', 'extractor_key'}
1653
412c617d 1654 new_result = info.copy()
3975b4d2 1655 new_result.update(filter_dict(ie_result, lambda k, v: v is not None and k not in exempted_fields))
7fc3fa05 1656
0563f7ac
S
1657 # Extracted info may not be a video result (i.e.
1658 # info.get('_type', 'video') != video) but rather an url or
1659 # url_transparent. In such cases outer metadata (from ie_result)
1660 # should be propagated to inner one (info). For this to happen
1661 # _type of info should be overridden with url_transparent. This
067aa17e 1662 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
0563f7ac
S
1663 if new_result.get('_type') == 'url':
1664 new_result['_type'] = 'url_transparent'
7fc3fa05
PH
1665
1666 return self.process_ie_result(
1667 new_result, download=download, extra_info=extra_info)
40fcba5e 1668 elif result_type in ('playlist', 'multi_video'):
30a074c2 1669 # Protect from infinite recursion due to recursively nested playlists
1670 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
1671 webpage_url = ie_result['webpage_url']
1672 if webpage_url in self._playlist_urls:
7e85e872 1673 self.to_screen(
30a074c2 1674 '[download] Skipping already downloaded playlist: %s'
1675 % ie_result.get('title') or ie_result.get('id'))
1676 return
7e85e872 1677
30a074c2 1678 self._playlist_level += 1
1679 self._playlist_urls.add(webpage_url)
03f83004 1680 self._fill_common_fields(ie_result, False)
bc516a3f 1681 self._sanitize_thumbnails(ie_result)
30a074c2 1682 try:
1683 return self.__process_playlist(ie_result, download)
1684 finally:
1685 self._playlist_level -= 1
1686 if not self._playlist_level:
1687 self._playlist_urls.clear()
8222d8de 1688 elif result_type == 'compat_list':
c9bf4114
PH
1689 self.report_warning(
1690 'Extractor %s returned a compat_list result. '
1691 'It needs to be updated.' % ie_result.get('extractor'))
5f6a1245 1692
8222d8de 1693 def _fixup(r):
b868936c 1694 self.add_extra_info(r, {
1695 'extractor': ie_result['extractor'],
1696 'webpage_url': ie_result['webpage_url'],
1697 'webpage_url_basename': url_basename(ie_result['webpage_url']),
0bb322b9 1698 'webpage_url_domain': get_domain(ie_result['webpage_url']),
b868936c 1699 'extractor_key': ie_result['extractor_key'],
1700 })
8222d8de
JMF
1701 return r
1702 ie_result['entries'] = [
b6c45014 1703 self.process_ie_result(_fixup(r), download, extra_info)
8222d8de
JMF
1704 for r in ie_result['entries']
1705 ]
1706 return ie_result
1707 else:
1708 raise Exception('Invalid result type: %s' % result_type)
1709
e92caff5 1710 def _ensure_dir_exists(self, path):
1711 return make_dir(path, self.report_error)
1712
3b603dbd 1713 @staticmethod
3bec830a 1714 def _playlist_infodict(ie_result, strict=False, **kwargs):
1715 info = {
1716 'playlist_count': ie_result.get('playlist_count'),
3b603dbd 1717 'playlist': ie_result.get('title') or ie_result.get('id'),
1718 'playlist_id': ie_result.get('id'),
1719 'playlist_title': ie_result.get('title'),
1720 'playlist_uploader': ie_result.get('uploader'),
1721 'playlist_uploader_id': ie_result.get('uploader_id'),
3b603dbd 1722 **kwargs,
1723 }
3bec830a 1724 if strict:
1725 return info
1726 return {
1727 **info,
1728 'playlist_index': 0,
1729 '__last_playlist_index': max(ie_result['requested_entries'] or (0, 0)),
1730 'extractor': ie_result['extractor'],
1731 'webpage_url': ie_result['webpage_url'],
1732 'webpage_url_basename': url_basename(ie_result['webpage_url']),
1733 'webpage_url_domain': get_domain(ie_result['webpage_url']),
1734 'extractor_key': ie_result['extractor_key'],
1735 }
3b603dbd 1736
30a074c2 1737 def __process_playlist(self, ie_result, download):
7e88d7d7 1738 """Process each entry in the playlist"""
f5ea4748 1739 assert ie_result['_type'] in ('playlist', 'multi_video')
1740
3bec830a 1741 common_info = self._playlist_infodict(ie_result, strict=True)
3955b207 1742 title = common_info.get('playlist') or '<Untitled>'
3bec830a 1743 if self._match_entry(common_info, incomplete=True) is not None:
1744 return
c6e07cf1 1745 self.to_screen(f'[download] Downloading {ie_result["_type"]}: {title}')
f0d785d3 1746
7e88d7d7 1747 all_entries = PlaylistEntries(self, ie_result)
7e9a6125 1748 entries = orderedSet(all_entries.get_requested_items(), lazy=True)
1749
1750 lazy = self.params.get('lazy_playlist')
1751 if lazy:
1752 resolved_entries, n_entries = [], 'N/A'
1753 ie_result['requested_entries'], ie_result['entries'] = None, None
1754 else:
1755 entries = resolved_entries = list(entries)
1756 n_entries = len(resolved_entries)
1757 ie_result['requested_entries'], ie_result['entries'] = tuple(zip(*resolved_entries)) or ([], [])
1758 if not ie_result.get('playlist_count'):
1759 # Better to do this after potentially exhausting entries
1760 ie_result['playlist_count'] = all_entries.get_full_count()
498f5606 1761
0647d925 1762 extra = self._playlist_infodict(ie_result, n_entries=int_or_none(n_entries))
1763 ie_copy = collections.ChainMap(ie_result, extra)
3bec830a 1764
e08a85d8 1765 _infojson_written = False
0bfc53d0 1766 write_playlist_files = self.params.get('allow_playlist_files', True)
1767 if write_playlist_files and self.params.get('list_thumbnails'):
1768 self.list_thumbnails(ie_result)
1769 if write_playlist_files and not self.params.get('simulate'):
e08a85d8 1770 _infojson_written = self._write_info_json(
1771 'playlist', ie_result, self.prepare_filename(ie_copy, 'pl_infojson'))
1772 if _infojson_written is None:
80c03fa9 1773 return
1774 if self._write_description('playlist', ie_result,
1775 self.prepare_filename(ie_copy, 'pl_description')) is None:
1776 return
681de68e 1777 # TODO: This should be passed to ThumbnailsConvertor if necessary
3bec830a 1778 self._write_thumbnails('playlist', ie_result, self.prepare_filename(ie_copy, 'pl_thumbnail'))
30a074c2 1779
7e9a6125 1780 if lazy:
1781 if self.params.get('playlistreverse') or self.params.get('playlistrandom'):
1782 self.report_warning('playlistreverse and playlistrandom are not supported with lazy_playlist', only_once=True)
1783 elif self.params.get('playlistreverse'):
1784 entries.reverse()
1785 elif self.params.get('playlistrandom'):
30a074c2 1786 random.shuffle(entries)
1787
7e88d7d7 1788 self.to_screen(f'[{ie_result["extractor"]}] Playlist {title}: Downloading {n_entries} videos'
1789 f'{format_field(ie_result, "playlist_count", " of %s")}')
30a074c2 1790
134c913c 1791 keep_resolved_entries = self.params.get('extract_flat') != 'discard'
1792 if self.params.get('extract_flat') == 'discard_in_playlist':
1793 keep_resolved_entries = ie_result['_type'] != 'playlist'
1794 if keep_resolved_entries:
1795 self.write_debug('The information of all playlist entries will be held in memory')
1796
26e2805c 1797 failures = 0
1798 max_failures = self.params.get('skip_playlist_after_errors') or float('inf')
7e9a6125 1799 for i, (playlist_index, entry) in enumerate(entries):
1800 if lazy:
1801 resolved_entries.append((playlist_index, entry))
3bec830a 1802 if not entry:
7e88d7d7 1803 continue
1804
7e88d7d7 1805 entry['__x_forwarded_for_ip'] = ie_result.get('__x_forwarded_for_ip')
7e9a6125 1806 if not lazy and 'playlist-index' in self.params.get('compat_opts', []):
1807 playlist_index = ie_result['requested_entries'][i]
1808
0647d925 1809 entry_copy = collections.ChainMap(entry, {
3bec830a 1810 **common_info,
3955b207 1811 'n_entries': int_or_none(n_entries),
71729754 1812 'playlist_index': playlist_index,
7e9a6125 1813 'playlist_autonumber': i + 1,
0647d925 1814 })
3bec830a 1815
0647d925 1816 if self._match_entry(entry_copy, incomplete=True) is not None:
f0ad6f8c 1817 # For compatabilty with youtube-dl. See https://github.com/yt-dlp/yt-dlp/issues/4369
1818 resolved_entries[i] = (playlist_index, NO_DEFAULT)
3bec830a 1819 continue
1820
1821 self.to_screen('[download] Downloading video %s of %s' % (
1822 self._format_screen(i + 1, self.Styles.ID), self._format_screen(n_entries, self.Styles.EMPHASIS)))
1823
a6ca61d4 1824 extra.update({
1825 'playlist_index': playlist_index,
1826 'playlist_autonumber': i + 1,
1827 })
3bec830a 1828 entry_result = self.__process_iterable_entry(entry, download, extra)
26e2805c 1829 if not entry_result:
1830 failures += 1
1831 if failures >= max_failures:
1832 self.report_error(
7e88d7d7 1833 f'Skipping the remaining entries in playlist "{title}" since {failures} items failed extraction')
26e2805c 1834 break
134c913c 1835 if keep_resolved_entries:
1836 resolved_entries[i] = (playlist_index, entry_result)
7e88d7d7 1837
1838 # Update with processed data
f0ad6f8c 1839 ie_result['requested_entries'] = [i for i, e in resolved_entries if e is not NO_DEFAULT]
1840 ie_result['entries'] = [e for _, e in resolved_entries if e is not NO_DEFAULT]
e08a85d8 1841
1842 # Write the updated info to json
cb96c5be 1843 if _infojson_written is True and self._write_info_json(
e08a85d8 1844 'updated playlist', ie_result,
1845 self.prepare_filename(ie_copy, 'pl_infojson'), overwrite=True) is None:
1846 return
ca30f449 1847
ed5835b4 1848 ie_result = self.run_all_pps('playlist', ie_result)
7e88d7d7 1849 self.to_screen(f'[download] Finished downloading playlist: {title}')
30a074c2 1850 return ie_result
1851
7e88d7d7 1852 @_handle_extraction_exceptions
a0566bbf 1853 def __process_iterable_entry(self, entry, download, extra_info):
1854 return self.process_ie_result(
1855 entry, download=download, extra_info=extra_info)
1856
67134eab
JMF
1857 def _build_format_filter(self, filter_spec):
1858 " Returns a function to filter the formats according to the filter_spec "
083c9df9
PH
1859
1860 OPERATORS = {
1861 '<': operator.lt,
1862 '<=': operator.le,
1863 '>': operator.gt,
1864 '>=': operator.ge,
1865 '=': operator.eq,
1866 '!=': operator.ne,
1867 }
67134eab 1868 operator_rex = re.compile(r'''(?x)\s*
187986a8 1869 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)\s*
1870 (?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1871 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)\s*
083c9df9 1872 ''' % '|'.join(map(re.escape, OPERATORS.keys())))
187986a8 1873 m = operator_rex.fullmatch(filter_spec)
9ddb6925
S
1874 if m:
1875 try:
1876 comparison_value = int(m.group('value'))
1877 except ValueError:
1878 comparison_value = parse_filesize(m.group('value'))
1879 if comparison_value is None:
1880 comparison_value = parse_filesize(m.group('value') + 'B')
1881 if comparison_value is None:
1882 raise ValueError(
1883 'Invalid value %r in format specification %r' % (
67134eab 1884 m.group('value'), filter_spec))
9ddb6925
S
1885 op = OPERATORS[m.group('op')]
1886
083c9df9 1887 if not m:
9ddb6925
S
1888 STR_OPERATORS = {
1889 '=': operator.eq,
10d33b34
YCH
1890 '^=': lambda attr, value: attr.startswith(value),
1891 '$=': lambda attr, value: attr.endswith(value),
1892 '*=': lambda attr, value: value in attr,
1ce9a3cb 1893 '~=': lambda attr, value: value.search(attr) is not None
9ddb6925 1894 }
187986a8 1895 str_operator_rex = re.compile(r'''(?x)\s*
1896 (?P<key>[a-zA-Z0-9._-]+)\s*
1ce9a3cb
LF
1897 (?P<negation>!\s*)?(?P<op>%s)\s*(?P<none_inclusive>\?\s*)?
1898 (?P<quote>["'])?
1899 (?P<value>(?(quote)(?:(?!(?P=quote))[^\\]|\\.)+|[\w.-]+))
1900 (?(quote)(?P=quote))\s*
9ddb6925 1901 ''' % '|'.join(map(re.escape, STR_OPERATORS.keys())))
187986a8 1902 m = str_operator_rex.fullmatch(filter_spec)
9ddb6925 1903 if m:
1ce9a3cb
LF
1904 if m.group('op') == '~=':
1905 comparison_value = re.compile(m.group('value'))
1906 else:
1907 comparison_value = re.sub(r'''\\([\\"'])''', r'\1', m.group('value'))
2cc779f4
S
1908 str_op = STR_OPERATORS[m.group('op')]
1909 if m.group('negation'):
e118a879 1910 op = lambda attr, value: not str_op(attr, value)
2cc779f4
S
1911 else:
1912 op = str_op
083c9df9 1913
9ddb6925 1914 if not m:
187986a8 1915 raise SyntaxError('Invalid filter specification %r' % filter_spec)
083c9df9
PH
1916
1917 def _filter(f):
1918 actual_value = f.get(m.group('key'))
1919 if actual_value is None:
1920 return m.group('none_inclusive')
1921 return op(actual_value, comparison_value)
67134eab
JMF
1922 return _filter
1923
9f1a1c36 1924 def _check_formats(self, formats):
1925 for f in formats:
1926 self.to_screen('[info] Testing format %s' % f['format_id'])
75689fe5 1927 path = self.get_output_path('temp')
1928 if not self._ensure_dir_exists(f'{path}/'):
1929 continue
1930 temp_file = tempfile.NamedTemporaryFile(suffix='.tmp', delete=False, dir=path or None)
9f1a1c36 1931 temp_file.close()
1932 try:
1933 success, _ = self.dl(temp_file.name, f, test=True)
8a82af35 1934 except (DownloadError, OSError, ValueError) + network_exceptions:
9f1a1c36 1935 success = False
1936 finally:
1937 if os.path.exists(temp_file.name):
1938 try:
1939 os.remove(temp_file.name)
1940 except OSError:
1941 self.report_warning('Unable to delete temporary file "%s"' % temp_file.name)
1942 if success:
1943 yield f
1944 else:
1945 self.to_screen('[info] Unable to download format %s. Skipping...' % f['format_id'])
1946
0017d9ad 1947 def _default_format_spec(self, info_dict, download=True):
0017d9ad 1948
af0f7428
S
1949 def can_merge():
1950 merger = FFmpegMergerPP(self)
1951 return merger.available and merger.can_merge()
1952
91ebc640 1953 prefer_best = (
b7b04c78 1954 not self.params.get('simulate')
91ebc640 1955 and download
1956 and (
1957 not can_merge()
21633673 1958 or info_dict.get('is_live') and not self.params.get('live_from_start')
bf1824b3 1959 or self.params['outtmpl']['default'] == '-'))
53ed7066 1960 compat = (
1961 prefer_best
1962 or self.params.get('allow_multiple_audio_streams', False)
8a82af35 1963 or 'format-spec' in self.params['compat_opts'])
91ebc640 1964
1965 return (
53ed7066 1966 'best/bestvideo+bestaudio' if prefer_best
1967 else 'bestvideo*+bestaudio/best' if not compat
91ebc640 1968 else 'bestvideo+bestaudio/best')
0017d9ad 1969
67134eab
JMF
1970 def build_format_selector(self, format_spec):
1971 def syntax_error(note, start):
1972 message = (
1973 'Invalid format specification: '
86e5f3ed 1974 '{}\n\t{}\n\t{}^'.format(note, format_spec, ' ' * start[1]))
67134eab
JMF
1975 return SyntaxError(message)
1976
1977 PICKFIRST = 'PICKFIRST'
1978 MERGE = 'MERGE'
1979 SINGLE = 'SINGLE'
0130afb7 1980 GROUP = 'GROUP'
67134eab
JMF
1981 FormatSelector = collections.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
1982
91ebc640 1983 allow_multiple_streams = {'audio': self.params.get('allow_multiple_audio_streams', False),
1984 'video': self.params.get('allow_multiple_video_streams', False)}
909d24dd 1985
9f1a1c36 1986 check_formats = self.params.get('check_formats') == 'selected'
e8e73840 1987
67134eab
JMF
1988 def _parse_filter(tokens):
1989 filter_parts = []
1990 for type, string, start, _, _ in tokens:
1991 if type == tokenize.OP and string == ']':
1992 return ''.join(filter_parts)
1993 else:
1994 filter_parts.append(string)
1995
232541df 1996 def _remove_unused_ops(tokens):
62b58c09
L
1997 # Remove operators that we don't use and join them with the surrounding strings.
1998 # E.g. 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
232541df
JMF
1999 ALLOWED_OPS = ('/', '+', ',', '(', ')')
2000 last_string, last_start, last_end, last_line = None, None, None, None
2001 for type, string, start, end, line in tokens:
2002 if type == tokenize.OP and string == '[':
2003 if last_string:
2004 yield tokenize.NAME, last_string, last_start, last_end, last_line
2005 last_string = None
2006 yield type, string, start, end, line
2007 # everything inside brackets will be handled by _parse_filter
2008 for type, string, start, end, line in tokens:
2009 yield type, string, start, end, line
2010 if type == tokenize.OP and string == ']':
2011 break
2012 elif type == tokenize.OP and string in ALLOWED_OPS:
2013 if last_string:
2014 yield tokenize.NAME, last_string, last_start, last_end, last_line
2015 last_string = None
2016 yield type, string, start, end, line
2017 elif type in [tokenize.NAME, tokenize.NUMBER, tokenize.OP]:
2018 if not last_string:
2019 last_string = string
2020 last_start = start
2021 last_end = end
2022 else:
2023 last_string += string
2024 if last_string:
2025 yield tokenize.NAME, last_string, last_start, last_end, last_line
2026
cf2ac6df 2027 def _parse_format_selection(tokens, inside_merge=False, inside_choice=False, inside_group=False):
67134eab
JMF
2028 selectors = []
2029 current_selector = None
2030 for type, string, start, _, _ in tokens:
2031 # ENCODING is only defined in python 3.x
2032 if type == getattr(tokenize, 'ENCODING', None):
2033 continue
2034 elif type in [tokenize.NAME, tokenize.NUMBER]:
2035 current_selector = FormatSelector(SINGLE, string, [])
2036 elif type == tokenize.OP:
cf2ac6df
JMF
2037 if string == ')':
2038 if not inside_group:
2039 # ')' will be handled by the parentheses group
2040 tokens.restore_last_token()
67134eab 2041 break
cf2ac6df 2042 elif inside_merge and string in ['/', ',']:
0130afb7
JMF
2043 tokens.restore_last_token()
2044 break
cf2ac6df
JMF
2045 elif inside_choice and string == ',':
2046 tokens.restore_last_token()
2047 break
2048 elif string == ',':
0a31a350
JMF
2049 if not current_selector:
2050 raise syntax_error('"," must follow a format selector', start)
67134eab
JMF
2051 selectors.append(current_selector)
2052 current_selector = None
2053 elif string == '/':
d96d604e
JMF
2054 if not current_selector:
2055 raise syntax_error('"/" must follow a format selector', start)
67134eab 2056 first_choice = current_selector
cf2ac6df 2057 second_choice = _parse_format_selection(tokens, inside_choice=True)
f5f4a27a 2058 current_selector = FormatSelector(PICKFIRST, (first_choice, second_choice), [])
67134eab
JMF
2059 elif string == '[':
2060 if not current_selector:
2061 current_selector = FormatSelector(SINGLE, 'best', [])
2062 format_filter = _parse_filter(tokens)
2063 current_selector.filters.append(format_filter)
0130afb7
JMF
2064 elif string == '(':
2065 if current_selector:
2066 raise syntax_error('Unexpected "("', start)
cf2ac6df
JMF
2067 group = _parse_format_selection(tokens, inside_group=True)
2068 current_selector = FormatSelector(GROUP, group, [])
67134eab 2069 elif string == '+':
d03cfdce 2070 if not current_selector:
2071 raise syntax_error('Unexpected "+"', start)
2072 selector_1 = current_selector
2073 selector_2 = _parse_format_selection(tokens, inside_merge=True)
2074 if not selector_2:
2075 raise syntax_error('Expected a selector', start)
2076 current_selector = FormatSelector(MERGE, (selector_1, selector_2), [])
67134eab 2077 else:
86e5f3ed 2078 raise syntax_error(f'Operator not recognized: "{string}"', start)
67134eab
JMF
2079 elif type == tokenize.ENDMARKER:
2080 break
2081 if current_selector:
2082 selectors.append(current_selector)
2083 return selectors
2084
f8d4ad9a 2085 def _merge(formats_pair):
2086 format_1, format_2 = formats_pair
2087
2088 formats_info = []
2089 formats_info.extend(format_1.get('requested_formats', (format_1,)))
2090 formats_info.extend(format_2.get('requested_formats', (format_2,)))
2091
2092 if not allow_multiple_streams['video'] or not allow_multiple_streams['audio']:
551f9388 2093 get_no_more = {'video': False, 'audio': False}
f8d4ad9a 2094 for (i, fmt_info) in enumerate(formats_info):
551f9388 2095 if fmt_info.get('acodec') == fmt_info.get('vcodec') == 'none':
2096 formats_info.pop(i)
2097 continue
2098 for aud_vid in ['audio', 'video']:
f8d4ad9a 2099 if not allow_multiple_streams[aud_vid] and fmt_info.get(aud_vid[0] + 'codec') != 'none':
2100 if get_no_more[aud_vid]:
2101 formats_info.pop(i)
f5510afe 2102 break
f8d4ad9a 2103 get_no_more[aud_vid] = True
2104
2105 if len(formats_info) == 1:
2106 return formats_info[0]
2107
2108 video_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('vcodec') != 'none']
2109 audio_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('acodec') != 'none']
2110
2111 the_only_video = video_fmts[0] if len(video_fmts) == 1 else None
2112 the_only_audio = audio_fmts[0] if len(audio_fmts) == 1 else None
2113
fc61aff4
LL
2114 output_ext = get_compatible_ext(
2115 vcodecs=[f.get('vcodec') for f in video_fmts],
2116 acodecs=[f.get('acodec') for f in audio_fmts],
2117 vexts=[f['ext'] for f in video_fmts],
2118 aexts=[f['ext'] for f in audio_fmts],
2119 preferences=(try_call(lambda: self.params['merge_output_format'].split('/'))
2120 or self.params.get('prefer_free_formats') and ('webm', 'mkv')))
f8d4ad9a 2121
975a0d0d 2122 filtered = lambda *keys: filter(None, (traverse_obj(fmt, *keys) for fmt in formats_info))
2123
f8d4ad9a 2124 new_dict = {
2125 'requested_formats': formats_info,
975a0d0d 2126 'format': '+'.join(filtered('format')),
2127 'format_id': '+'.join(filtered('format_id')),
f8d4ad9a 2128 'ext': output_ext,
975a0d0d 2129 'protocol': '+'.join(map(determine_protocol, formats_info)),
093a1710 2130 'language': '+'.join(orderedSet(filtered('language'))) or None,
2131 'format_note': '+'.join(orderedSet(filtered('format_note'))) or None,
2132 'filesize_approx': sum(filtered('filesize', 'filesize_approx')) or None,
975a0d0d 2133 'tbr': sum(filtered('tbr', 'vbr', 'abr')),
f8d4ad9a 2134 }
2135
2136 if the_only_video:
2137 new_dict.update({
2138 'width': the_only_video.get('width'),
2139 'height': the_only_video.get('height'),
2140 'resolution': the_only_video.get('resolution') or self.format_resolution(the_only_video),
2141 'fps': the_only_video.get('fps'),
49a57e70 2142 'dynamic_range': the_only_video.get('dynamic_range'),
f8d4ad9a 2143 'vcodec': the_only_video.get('vcodec'),
2144 'vbr': the_only_video.get('vbr'),
2145 'stretched_ratio': the_only_video.get('stretched_ratio'),
2146 })
2147
2148 if the_only_audio:
2149 new_dict.update({
2150 'acodec': the_only_audio.get('acodec'),
2151 'abr': the_only_audio.get('abr'),
975a0d0d 2152 'asr': the_only_audio.get('asr'),
b8ed0f15 2153 'audio_channels': the_only_audio.get('audio_channels')
f8d4ad9a 2154 })
2155
2156 return new_dict
2157
e8e73840 2158 def _check_formats(formats):
981052c9 2159 if not check_formats:
2160 yield from formats
b5ac45b1 2161 return
9f1a1c36 2162 yield from self._check_formats(formats)
e8e73840 2163
67134eab 2164 def _build_selector_function(selector):
909d24dd 2165 if isinstance(selector, list): # ,
67134eab
JMF
2166 fs = [_build_selector_function(s) for s in selector]
2167
317f7ab6 2168 def selector_function(ctx):
67134eab 2169 for f in fs:
981052c9 2170 yield from f(ctx)
67134eab 2171 return selector_function
909d24dd 2172
2173 elif selector.type == GROUP: # ()
0130afb7 2174 selector_function = _build_selector_function(selector.selector)
909d24dd 2175
2176 elif selector.type == PICKFIRST: # /
67134eab
JMF
2177 fs = [_build_selector_function(s) for s in selector.selector]
2178
317f7ab6 2179 def selector_function(ctx):
67134eab 2180 for f in fs:
317f7ab6 2181 picked_formats = list(f(ctx))
67134eab
JMF
2182 if picked_formats:
2183 return picked_formats
2184 return []
67134eab 2185
981052c9 2186 elif selector.type == MERGE: # +
2187 selector_1, selector_2 = map(_build_selector_function, selector.selector)
2188
2189 def selector_function(ctx):
adbc4ec4 2190 for pair in itertools.product(selector_1(ctx), selector_2(ctx)):
981052c9 2191 yield _merge(pair)
2192
909d24dd 2193 elif selector.type == SINGLE: # atom
598d185d 2194 format_spec = selector.selector or 'best'
909d24dd 2195
f8d4ad9a 2196 # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
909d24dd 2197 if format_spec == 'all':
2198 def selector_function(ctx):
9222c381 2199 yield from _check_formats(ctx['formats'][::-1])
f8d4ad9a 2200 elif format_spec == 'mergeall':
2201 def selector_function(ctx):
316f2650 2202 formats = list(_check_formats(
2203 f for f in ctx['formats'] if f.get('vcodec') != 'none' or f.get('acodec') != 'none'))
e01d6aa4 2204 if not formats:
2205 return
921b76ca 2206 merged_format = formats[-1]
2207 for f in formats[-2::-1]:
f8d4ad9a 2208 merged_format = _merge((merged_format, f))
2209 yield merged_format
909d24dd 2210
2211 else:
85e801a9 2212 format_fallback, seperate_fallback, format_reverse, format_idx = False, None, True, 1
eff63539 2213 mobj = re.match(
2214 r'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
2215 format_spec)
2216 if mobj is not None:
2217 format_idx = int_or_none(mobj.group('n'), default=1)
e8e73840 2218 format_reverse = mobj.group('bw')[0] == 'b'
eff63539 2219 format_type = (mobj.group('type') or [None])[0]
2220 not_format_type = {'v': 'a', 'a': 'v'}.get(format_type)
2221 format_modified = mobj.group('mod') is not None
909d24dd 2222
2223 format_fallback = not format_type and not format_modified # for b, w
8326b00a 2224 _filter_f = (
eff63539 2225 (lambda f: f.get('%scodec' % format_type) != 'none')
2226 if format_type and format_modified # bv*, ba*, wv*, wa*
2227 else (lambda f: f.get('%scodec' % not_format_type) == 'none')
2228 if format_type # bv, ba, wv, wa
2229 else (lambda f: f.get('vcodec') != 'none' and f.get('acodec') != 'none')
2230 if not format_modified # b, w
8326b00a 2231 else lambda f: True) # b*, w*
2232 filter_f = lambda f: _filter_f(f) and (
2233 f.get('vcodec') != 'none' or f.get('acodec') != 'none')
67134eab 2234 else:
48ee10ee 2235 if format_spec in self._format_selection_exts['audio']:
b11c04a8 2236 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none'
48ee10ee 2237 elif format_spec in self._format_selection_exts['video']:
b11c04a8 2238 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none' and f.get('vcodec') != 'none'
85e801a9 2239 seperate_fallback = lambda f: f.get('ext') == format_spec and f.get('vcodec') != 'none'
48ee10ee 2240 elif format_spec in self._format_selection_exts['storyboards']:
b11c04a8 2241 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') == 'none' and f.get('vcodec') == 'none'
2242 else:
b5ae35ee 2243 filter_f = lambda f: f.get('format_id') == format_spec # id
909d24dd 2244
2245 def selector_function(ctx):
2246 formats = list(ctx['formats'])
909d24dd 2247 matches = list(filter(filter_f, formats)) if filter_f is not None else formats
85e801a9 2248 if not matches:
2249 if format_fallback and ctx['incomplete_formats']:
2250 # for extractors with incomplete formats (audio only (soundcloud)
2251 # or video only (imgur)) best/worst will fallback to
2252 # best/worst {video,audio}-only format
2253 matches = formats
2254 elif seperate_fallback and not ctx['has_merged_format']:
2255 # for compatibility with youtube-dl when there is no pre-merged format
2256 matches = list(filter(seperate_fallback, formats))
981052c9 2257 matches = LazyList(_check_formats(matches[::-1 if format_reverse else 1]))
2258 try:
e8e73840 2259 yield matches[format_idx - 1]
4abea8ca 2260 except LazyList.IndexError:
981052c9 2261 return
083c9df9 2262
67134eab 2263 filters = [self._build_format_filter(f) for f in selector.filters]
083c9df9 2264
317f7ab6 2265 def final_selector(ctx):
adbc4ec4 2266 ctx_copy = dict(ctx)
67134eab 2267 for _filter in filters:
317f7ab6
S
2268 ctx_copy['formats'] = list(filter(_filter, ctx_copy['formats']))
2269 return selector_function(ctx_copy)
67134eab 2270 return final_selector
083c9df9 2271
0f06bcd7 2272 stream = io.BytesIO(format_spec.encode())
0130afb7 2273 try:
f9934b96 2274 tokens = list(_remove_unused_ops(tokenize.tokenize(stream.readline)))
0130afb7
JMF
2275 except tokenize.TokenError:
2276 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec)))
2277
86e5f3ed 2278 class TokenIterator:
0130afb7
JMF
2279 def __init__(self, tokens):
2280 self.tokens = tokens
2281 self.counter = 0
2282
2283 def __iter__(self):
2284 return self
2285
2286 def __next__(self):
2287 if self.counter >= len(self.tokens):
2288 raise StopIteration()
2289 value = self.tokens[self.counter]
2290 self.counter += 1
2291 return value
2292
2293 next = __next__
2294
2295 def restore_last_token(self):
2296 self.counter -= 1
2297
2298 parsed_selector = _parse_format_selection(iter(TokenIterator(tokens)))
67134eab 2299 return _build_selector_function(parsed_selector)
a9c58ad9 2300
e5660ee6 2301 def _calc_headers(self, info_dict):
8b7539d2 2302 res = merge_headers(self.params['http_headers'], info_dict.get('http_headers') or {})
e5660ee6 2303
c487cf00 2304 cookies = self._calc_cookies(info_dict['url'])
e5660ee6
JMF
2305 if cookies:
2306 res['Cookie'] = cookies
2307
0016b84e
S
2308 if 'X-Forwarded-For' not in res:
2309 x_forwarded_for_ip = info_dict.get('__x_forwarded_for_ip')
2310 if x_forwarded_for_ip:
2311 res['X-Forwarded-For'] = x_forwarded_for_ip
2312
e5660ee6
JMF
2313 return res
2314
c487cf00 2315 def _calc_cookies(self, url):
2316 pr = sanitized_Request(url)
e5660ee6 2317 self.cookiejar.add_cookie_header(pr)
662435f7 2318 return pr.get_header('Cookie')
e5660ee6 2319
9f1a1c36 2320 def _sort_thumbnails(self, thumbnails):
2321 thumbnails.sort(key=lambda t: (
2322 t.get('preference') if t.get('preference') is not None else -1,
2323 t.get('width') if t.get('width') is not None else -1,
2324 t.get('height') if t.get('height') is not None else -1,
2325 t.get('id') if t.get('id') is not None else '',
2326 t.get('url')))
2327
b0249bca 2328 def _sanitize_thumbnails(self, info_dict):
bc516a3f 2329 thumbnails = info_dict.get('thumbnails')
2330 if thumbnails is None:
2331 thumbnail = info_dict.get('thumbnail')
2332 if thumbnail:
2333 info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}]
9f1a1c36 2334 if not thumbnails:
2335 return
2336
2337 def check_thumbnails(thumbnails):
2338 for t in thumbnails:
2339 self.to_screen(f'[info] Testing thumbnail {t["id"]}')
2340 try:
2341 self.urlopen(HEADRequest(t['url']))
2342 except network_exceptions as err:
2343 self.to_screen(f'[info] Unable to connect to thumbnail {t["id"]} URL {t["url"]!r} - {err}. Skipping...')
2344 continue
2345 yield t
2346
2347 self._sort_thumbnails(thumbnails)
2348 for i, t in enumerate(thumbnails):
2349 if t.get('id') is None:
2350 t['id'] = '%d' % i
2351 if t.get('width') and t.get('height'):
2352 t['resolution'] = '%dx%d' % (t['width'], t['height'])
2353 t['url'] = sanitize_url(t['url'])
2354
2355 if self.params.get('check_formats') is True:
282f5709 2356 info_dict['thumbnails'] = LazyList(check_thumbnails(thumbnails[::-1]), reverse=True)
9f1a1c36 2357 else:
2358 info_dict['thumbnails'] = thumbnails
bc516a3f 2359
03f83004
LNO
2360 def _fill_common_fields(self, info_dict, is_video=True):
2361 # TODO: move sanitization here
2362 if is_video:
2363 # playlists are allowed to lack "title"
d4736fdb 2364 title = info_dict.get('title', NO_DEFAULT)
2365 if title is NO_DEFAULT:
03f83004
LNO
2366 raise ExtractorError('Missing "title" field in extractor result',
2367 video_id=info_dict['id'], ie=info_dict['extractor'])
d4736fdb 2368 info_dict['fulltitle'] = title
2369 if not title:
2370 if title == '':
2371 self.write_debug('Extractor gave empty title. Creating a generic title')
2372 else:
2373 self.report_warning('Extractor failed to obtain "title". Creating a generic title instead')
1d485a1a 2374 info_dict['title'] = f'{info_dict["extractor"].replace(":", "-")} video #{info_dict["id"]}'
03f83004
LNO
2375
2376 if info_dict.get('duration') is not None:
2377 info_dict['duration_string'] = formatSeconds(info_dict['duration'])
2378
2379 for ts_key, date_key in (
2380 ('timestamp', 'upload_date'),
2381 ('release_timestamp', 'release_date'),
2382 ('modified_timestamp', 'modified_date'),
2383 ):
2384 if info_dict.get(date_key) is None and info_dict.get(ts_key) is not None:
2385 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
2386 # see http://bugs.python.org/issue1646728)
19a03940 2387 with contextlib.suppress(ValueError, OverflowError, OSError):
03f83004
LNO
2388 upload_date = datetime.datetime.utcfromtimestamp(info_dict[ts_key])
2389 info_dict[date_key] = upload_date.strftime('%Y%m%d')
03f83004
LNO
2390
2391 live_keys = ('is_live', 'was_live')
2392 live_status = info_dict.get('live_status')
2393 if live_status is None:
2394 for key in live_keys:
2395 if info_dict.get(key) is False:
2396 continue
2397 if info_dict.get(key):
2398 live_status = key
2399 break
2400 if all(info_dict.get(key) is False for key in live_keys):
2401 live_status = 'not_live'
2402 if live_status:
2403 info_dict['live_status'] = live_status
2404 for key in live_keys:
2405 if info_dict.get(key) is None:
2406 info_dict[key] = (live_status == key)
2407
2408 # Auto generate title fields corresponding to the *_number fields when missing
2409 # in order to always have clean titles. This is very common for TV series.
2410 for field in ('chapter', 'season', 'episode'):
2411 if info_dict.get('%s_number' % field) is not None and not info_dict.get(field):
2412 info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field])
2413
415f8d51 2414 def _raise_pending_errors(self, info):
2415 err = info.pop('__pending_error', None)
2416 if err:
2417 self.report_error(err, tb=False)
2418
dd82ffea
JMF
2419 def process_video_result(self, info_dict, download=True):
2420 assert info_dict.get('_type', 'video') == 'video'
9c906919 2421 self._num_videos += 1
dd82ffea 2422
bec1fad2 2423 if 'id' not in info_dict:
fc08bdd6 2424 raise ExtractorError('Missing "id" field in extractor result', ie=info_dict['extractor'])
2425 elif not info_dict.get('id'):
2426 raise ExtractorError('Extractor failed to obtain "id"', ie=info_dict['extractor'])
455a15e2 2427
c9969434
S
2428 def report_force_conversion(field, field_not, conversion):
2429 self.report_warning(
2430 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
2431 % (field, field_not, conversion))
2432
2433 def sanitize_string_field(info, string_field):
2434 field = info.get(string_field)
14f25df2 2435 if field is None or isinstance(field, str):
c9969434
S
2436 return
2437 report_force_conversion(string_field, 'a string', 'string')
14f25df2 2438 info[string_field] = str(field)
c9969434
S
2439
2440 def sanitize_numeric_fields(info):
2441 for numeric_field in self._NUMERIC_FIELDS:
2442 field = info.get(numeric_field)
f9934b96 2443 if field is None or isinstance(field, (int, float)):
c9969434
S
2444 continue
2445 report_force_conversion(numeric_field, 'numeric', 'int')
2446 info[numeric_field] = int_or_none(field)
2447
2448 sanitize_string_field(info_dict, 'id')
2449 sanitize_numeric_fields(info_dict)
3975b4d2 2450 if info_dict.get('section_end') and info_dict.get('section_start') is not None:
2451 info_dict['duration'] = round(info_dict['section_end'] - info_dict['section_start'], 3)
4c3f8c3f 2452 if (info_dict.get('duration') or 0) <= 0 and info_dict.pop('duration', None):
50e93e03 2453 self.report_warning('"duration" field is negative, there is an error in extractor')
be6217b2 2454
9eef7c4e 2455 chapters = info_dict.get('chapters') or []
a3976e07 2456 if chapters and chapters[0].get('start_time'):
2457 chapters.insert(0, {'start_time': 0})
2458
9eef7c4e 2459 dummy_chapter = {'end_time': 0, 'start_time': info_dict.get('duration')}
a3976e07 2460 for idx, (prev, current, next_) in enumerate(zip(
2461 (dummy_chapter, *chapters), chapters, (*chapters[1:], dummy_chapter)), 1):
9eef7c4e 2462 if current.get('start_time') is None:
2463 current['start_time'] = prev.get('end_time')
2464 if not current.get('end_time'):
2465 current['end_time'] = next_.get('start_time')
a3976e07 2466 if not current.get('title'):
2467 current['title'] = f'<Untitled Chapter {idx}>'
9eef7c4e 2468
dd82ffea
JMF
2469 if 'playlist' not in info_dict:
2470 # It isn't part of a playlist
2471 info_dict['playlist'] = None
2472 info_dict['playlist_index'] = None
2473
bc516a3f 2474 self._sanitize_thumbnails(info_dict)
d5519808 2475
536a55da 2476 thumbnail = info_dict.get('thumbnail')
bc516a3f 2477 thumbnails = info_dict.get('thumbnails')
536a55da
S
2478 if thumbnail:
2479 info_dict['thumbnail'] = sanitize_url(thumbnail)
2480 elif thumbnails:
d5519808
PH
2481 info_dict['thumbnail'] = thumbnails[-1]['url']
2482
ae30b840 2483 if info_dict.get('display_id') is None and 'id' in info_dict:
0afef30b
PH
2484 info_dict['display_id'] = info_dict['id']
2485
03f83004 2486 self._fill_common_fields(info_dict)
33d2fc2f 2487
05108a49
S
2488 for cc_kind in ('subtitles', 'automatic_captions'):
2489 cc = info_dict.get(cc_kind)
2490 if cc:
2491 for _, subtitle in cc.items():
2492 for subtitle_format in subtitle:
2493 if subtitle_format.get('url'):
2494 subtitle_format['url'] = sanitize_url(subtitle_format['url'])
2495 if subtitle_format.get('ext') is None:
2496 subtitle_format['ext'] = determine_ext(subtitle_format['url']).lower()
2497
2498 automatic_captions = info_dict.get('automatic_captions')
4bba3716 2499 subtitles = info_dict.get('subtitles')
4bba3716 2500
360e1ca5 2501 info_dict['requested_subtitles'] = self.process_subtitles(
05108a49 2502 info_dict['id'], subtitles, automatic_captions)
a504ced0 2503
dd82ffea
JMF
2504 if info_dict.get('formats') is None:
2505 # There's only one format available
2506 formats = [info_dict]
2507 else:
2508 formats = info_dict['formats']
2509
0a5a191a 2510 # or None ensures --clean-infojson removes it
2511 info_dict['_has_drm'] = any(f.get('has_drm') for f in formats) or None
88acdbc2 2512 if not self.params.get('allow_unplayable_formats'):
2513 formats = [f for f in formats if not f.get('has_drm')]
7356a444 2514 if info_dict['_has_drm'] and formats and all(
c0b6e5c7 2515 f.get('acodec') == f.get('vcodec') == 'none' for f in formats):
2516 self.report_warning(
2517 'This video is DRM protected and only images are available for download. '
2518 'Use --list-formats to see them')
88acdbc2 2519
319b6059 2520 get_from_start = not info_dict.get('is_live') or bool(self.params.get('live_from_start'))
2521 if not get_from_start:
2522 info_dict['title'] += ' ' + datetime.datetime.now().strftime('%Y-%m-%d %H:%M')
2523 if info_dict.get('is_live') and formats:
adbc4ec4 2524 formats = [f for f in formats if bool(f.get('is_from_start')) == get_from_start]
319b6059 2525 if get_from_start and not formats:
a44ca5a4 2526 self.raise_no_formats(info_dict, msg=(
2527 '--live-from-start is passed, but there are no formats that can be downloaded from the start. '
2528 'If you want to download from the current time, use --no-live-from-start'))
adbc4ec4 2529
db95dc13 2530 if not formats:
1151c407 2531 self.raise_no_formats(info_dict)
db95dc13 2532
73af5cc8
S
2533 def is_wellformed(f):
2534 url = f.get('url')
a5ac0c47 2535 if not url:
73af5cc8
S
2536 self.report_warning(
2537 '"url" field is missing or empty - skipping format, '
2538 'there is an error in extractor')
a5ac0c47
S
2539 return False
2540 if isinstance(url, bytes):
2541 sanitize_string_field(f, 'url')
2542 return True
73af5cc8
S
2543
2544 # Filter out malformed formats for better extraction robustness
2545 formats = list(filter(is_wellformed, formats))
2546
181c7053
S
2547 formats_dict = {}
2548
dd82ffea 2549 # We check that all the formats have the format and format_id fields
db95dc13 2550 for i, format in enumerate(formats):
c9969434
S
2551 sanitize_string_field(format, 'format_id')
2552 sanitize_numeric_fields(format)
dcf77cf1 2553 format['url'] = sanitize_url(format['url'])
e74e3b63 2554 if not format.get('format_id'):
14f25df2 2555 format['format_id'] = str(i)
e2effb08
S
2556 else:
2557 # Sanitize format_id from characters used in format selector expression
ec85ded8 2558 format['format_id'] = re.sub(r'[\s,/+\[\]()]', '_', format['format_id'])
181c7053
S
2559 format_id = format['format_id']
2560 if format_id not in formats_dict:
2561 formats_dict[format_id] = []
2562 formats_dict[format_id].append(format)
2563
2564 # Make sure all formats have unique format_id
03b4de72 2565 common_exts = set(itertools.chain(*self._format_selection_exts.values()))
181c7053 2566 for format_id, ambiguous_formats in formats_dict.items():
48ee10ee 2567 ambigious_id = len(ambiguous_formats) > 1
2568 for i, format in enumerate(ambiguous_formats):
2569 if ambigious_id:
181c7053 2570 format['format_id'] = '%s-%d' % (format_id, i)
48ee10ee 2571 if format.get('ext') is None:
2572 format['ext'] = determine_ext(format['url']).lower()
2573 # Ensure there is no conflict between id and ext in format selection
2574 # See https://github.com/yt-dlp/yt-dlp/issues/1282
2575 if format['format_id'] != format['ext'] and format['format_id'] in common_exts:
2576 format['format_id'] = 'f%s' % format['format_id']
181c7053
S
2577
2578 for i, format in enumerate(formats):
8c51aa65 2579 if format.get('format') is None:
6febd1c1 2580 format['format'] = '{id} - {res}{note}'.format(
8c51aa65
JMF
2581 id=format['format_id'],
2582 res=self.format_resolution(format),
b868936c 2583 note=format_field(format, 'format_note', ' (%s)'),
8c51aa65 2584 )
6f0be937 2585 if format.get('protocol') is None:
b5559424 2586 format['protocol'] = determine_protocol(format)
239df021 2587 if format.get('resolution') is None:
2588 format['resolution'] = self.format_resolution(format, default=None)
176f1866 2589 if format.get('dynamic_range') is None and format.get('vcodec') != 'none':
2590 format['dynamic_range'] = 'SDR'
f2fe69c7 2591 if (info_dict.get('duration') and format.get('tbr')
2592 and not format.get('filesize') and not format.get('filesize_approx')):
56ba69e4 2593 format['filesize_approx'] = int(info_dict['duration'] * format['tbr'] * (1024 / 8))
f2fe69c7 2594
e5660ee6
JMF
2595 # Add HTTP headers, so that external programs can use them from the
2596 # json output
2597 full_format_info = info_dict.copy()
2598 full_format_info.update(format)
2599 format['http_headers'] = self._calc_headers(full_format_info)
0016b84e
S
2600 # Remove private housekeeping stuff
2601 if '__x_forwarded_for_ip' in info_dict:
2602 del info_dict['__x_forwarded_for_ip']
dd82ffea 2603
9f1a1c36 2604 if self.params.get('check_formats') is True:
282f5709 2605 formats = LazyList(self._check_formats(formats[::-1]), reverse=True)
9f1a1c36 2606
88acdbc2 2607 if not formats or formats[0] is not info_dict:
b3d9ef88
JMF
2608 # only set the 'formats' fields if the original info_dict list them
2609 # otherwise we end up with a circular reference, the first (and unique)
f89197d7 2610 # element in the 'formats' field in info_dict is info_dict itself,
dfb1b146 2611 # which can't be exported to json
b3d9ef88 2612 info_dict['formats'] = formats
4ec82a72 2613
2614 info_dict, _ = self.pre_process(info_dict)
2615
6db9c4d5 2616 if self._match_entry(info_dict, incomplete=self._format_fields) is not None:
09b49e1f 2617 return info_dict
2618
2619 self.post_extract(info_dict)
2620 info_dict, _ = self.pre_process(info_dict, 'after_filter')
2621
093a1710 2622 # The pre-processors may have modified the formats
2623 formats = info_dict.get('formats', [info_dict])
2624
fa9f30b8 2625 list_only = self.params.get('simulate') is None and (
2626 self.params.get('list_thumbnails') or self.params.get('listformats') or self.params.get('listsubtitles'))
2627 interactive_format_selection = not list_only and self.format_selector == '-'
b7b04c78 2628 if self.params.get('list_thumbnails'):
2629 self.list_thumbnails(info_dict)
b7b04c78 2630 if self.params.get('listsubtitles'):
2631 if 'automatic_captions' in info_dict:
2632 self.list_subtitles(
2633 info_dict['id'], automatic_captions, 'automatic captions')
2634 self.list_subtitles(info_dict['id'], subtitles, 'subtitles')
fa9f30b8 2635 if self.params.get('listformats') or interactive_format_selection:
b69fd25c 2636 self.list_formats(info_dict)
169dbde9 2637 if list_only:
b7b04c78 2638 # Without this printing, -F --print-json will not work
169dbde9 2639 self.__forced_printings(info_dict, self.prepare_filename(info_dict), incomplete=True)
c487cf00 2640 return info_dict
bfaae0a7 2641
187986a8 2642 format_selector = self.format_selector
2643 if format_selector is None:
0017d9ad 2644 req_format = self._default_format_spec(info_dict, download=download)
0760b0a7 2645 self.write_debug('Default format spec: %s' % req_format)
187986a8 2646 format_selector = self.build_format_selector(req_format)
317f7ab6 2647
fa9f30b8 2648 while True:
2649 if interactive_format_selection:
2650 req_format = input(
2651 self._format_screen('\nEnter format selector: ', self.Styles.EMPHASIS))
2652 try:
2653 format_selector = self.build_format_selector(req_format)
2654 except SyntaxError as err:
2655 self.report_error(err, tb=False, is_error=False)
2656 continue
2657
85e801a9 2658 formats_to_download = list(format_selector({
fa9f30b8 2659 'formats': formats,
85e801a9 2660 'has_merged_format': any('none' not in (f.get('acodec'), f.get('vcodec')) for f in formats),
2661 'incomplete_formats': (
2662 # All formats are video-only or
2663 all(f.get('vcodec') != 'none' and f.get('acodec') == 'none' for f in formats)
2664 # all formats are audio-only
2665 or all(f.get('vcodec') == 'none' and f.get('acodec') != 'none' for f in formats)),
2666 }))
fa9f30b8 2667 if interactive_format_selection and not formats_to_download:
2668 self.report_error('Requested format is not available', tb=False, is_error=False)
2669 continue
2670 break
317f7ab6 2671
dd82ffea 2672 if not formats_to_download:
b7da73eb 2673 if not self.params.get('ignore_no_formats_error'):
c0b6e5c7 2674 raise ExtractorError(
2675 'Requested format is not available. Use --list-formats for a list of available formats',
2676 expected=True, video_id=info_dict['id'], ie=info_dict['extractor'])
b62fa6d7 2677 self.report_warning('Requested format is not available')
2678 # Process what we can, even without any available formats.
2679 formats_to_download = [{}]
a13e6848 2680
5ec1b6b7 2681 requested_ranges = self.params.get('download_ranges')
2682 if requested_ranges:
2683 requested_ranges = tuple(requested_ranges(info_dict, self))
2684
2685 best_format, downloaded_formats = formats_to_download[-1], []
b62fa6d7 2686 if download:
2687 if best_format:
5ec1b6b7 2688 def to_screen(*msg):
2689 self.to_screen(f'[info] {info_dict["id"]}: {" ".join(", ".join(variadic(m)) for m in msg)}')
2690
2691 to_screen(f'Downloading {len(formats_to_download)} format(s):',
2692 (f['format_id'] for f in formats_to_download))
2693 if requested_ranges:
2694 to_screen(f'Downloading {len(requested_ranges)} time ranges:',
2695 (f'{int(c["start_time"])}-{int(c["end_time"])}' for c in requested_ranges))
a13e6848 2696 max_downloads_reached = False
5ec1b6b7 2697
2698 for fmt, chapter in itertools.product(formats_to_download, requested_ranges or [{}]):
2699 new_info = self._copy_infodict(info_dict)
b7da73eb 2700 new_info.update(fmt)
3975b4d2 2701 offset, duration = info_dict.get('section_start') or 0, info_dict.get('duration') or float('inf')
2702 if chapter or offset:
5ec1b6b7 2703 new_info.update({
3975b4d2 2704 'section_start': offset + chapter.get('start_time', 0),
bc401608 2705 'section_end': offset + min(chapter.get('end_time', duration), duration),
5ec1b6b7 2706 'section_title': chapter.get('title'),
2707 'section_number': chapter.get('index'),
2708 })
2709 downloaded_formats.append(new_info)
a13e6848 2710 try:
2711 self.process_info(new_info)
2712 except MaxDownloadsReached:
2713 max_downloads_reached = True
415f8d51 2714 self._raise_pending_errors(new_info)
f46e2f9d 2715 # Remove copied info
2716 for key, val in tuple(new_info.items()):
2717 if info_dict.get(key) == val:
2718 new_info.pop(key)
a13e6848 2719 if max_downloads_reached:
2720 break
ebed8b37 2721
5ec1b6b7 2722 write_archive = {f.get('__write_download_archive', False) for f in downloaded_formats}
a13e6848 2723 assert write_archive.issubset({True, False, 'ignore'})
2724 if True in write_archive and False not in write_archive:
2725 self.record_download_archive(info_dict)
be72c624 2726
5ec1b6b7 2727 info_dict['requested_downloads'] = downloaded_formats
ed5835b4 2728 info_dict = self.run_all_pps('after_video', info_dict)
a13e6848 2729 if max_downloads_reached:
2730 raise MaxDownloadsReached()
ebed8b37 2731
49a57e70 2732 # We update the info dict with the selected best quality format (backwards compatibility)
be72c624 2733 info_dict.update(best_format)
dd82ffea
JMF
2734 return info_dict
2735
98c70d6f 2736 def process_subtitles(self, video_id, normal_subtitles, automatic_captions):
a504ced0 2737 """Select the requested subtitles and their format"""
d8a58ddc 2738 available_subs, normal_sub_langs = {}, []
98c70d6f
JMF
2739 if normal_subtitles and self.params.get('writesubtitles'):
2740 available_subs.update(normal_subtitles)
d8a58ddc 2741 normal_sub_langs = tuple(normal_subtitles.keys())
98c70d6f
JMF
2742 if automatic_captions and self.params.get('writeautomaticsub'):
2743 for lang, cap_info in automatic_captions.items():
360e1ca5
JMF
2744 if lang not in available_subs:
2745 available_subs[lang] = cap_info
2746
4d171848
JMF
2747 if (not self.params.get('writesubtitles') and not
2748 self.params.get('writeautomaticsub') or not
2749 available_subs):
2750 return None
a504ced0 2751
d8a58ddc 2752 all_sub_langs = tuple(available_subs.keys())
a504ced0 2753 if self.params.get('allsubtitles', False):
c32b0aab 2754 requested_langs = all_sub_langs
2755 elif self.params.get('subtitleslangs', False):
5314b521 2756 try:
2757 requested_langs = orderedSet_from_options(
2758 self.params.get('subtitleslangs'), {'all': all_sub_langs}, use_regex=True)
2759 except re.error as e:
2760 raise ValueError(f'Wrong regex for subtitlelangs: {e.pattern}')
d8a58ddc 2761 elif normal_sub_langs:
2762 requested_langs = ['en'] if 'en' in normal_sub_langs else normal_sub_langs[:1]
a504ced0 2763 else:
d8a58ddc 2764 requested_langs = ['en'] if 'en' in all_sub_langs else all_sub_langs[:1]
ad3dc496 2765 if requested_langs:
2766 self.write_debug('Downloading subtitles: %s' % ', '.join(requested_langs))
a504ced0
JMF
2767
2768 formats_query = self.params.get('subtitlesformat', 'best')
2769 formats_preference = formats_query.split('/') if formats_query else []
2770 subs = {}
2771 for lang in requested_langs:
2772 formats = available_subs.get(lang)
2773 if formats is None:
86e5f3ed 2774 self.report_warning(f'{lang} subtitles not available for {video_id}')
a504ced0 2775 continue
a504ced0
JMF
2776 for ext in formats_preference:
2777 if ext == 'best':
2778 f = formats[-1]
2779 break
2780 matches = list(filter(lambda f: f['ext'] == ext, formats))
2781 if matches:
2782 f = matches[-1]
2783 break
2784 else:
2785 f = formats[-1]
2786 self.report_warning(
2787 'No subtitle format found matching "%s" for language %s, '
2788 'using %s' % (formats_query, lang, f['ext']))
2789 subs[lang] = f
2790 return subs
2791
bb66c247 2792 def _forceprint(self, key, info_dict):
2793 if info_dict is None:
2794 return
2795 info_copy = info_dict.copy()
2796 info_copy['formats_table'] = self.render_formats_table(info_dict)
2797 info_copy['thumbnails_table'] = self.render_thumbnails_table(info_dict)
2798 info_copy['subtitles_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('subtitles'))
2799 info_copy['automatic_captions_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('automatic_captions'))
2800
2801 def format_tmpl(tmpl):
2802 mobj = re.match(r'\w+(=?)$', tmpl)
2803 if mobj and mobj.group(1):
2804 return f'{tmpl[:-1]} = %({tmpl[:-1]})r'
2805 elif mobj:
2806 return f'%({tmpl})s'
2807 return tmpl
8130779d 2808
bb66c247 2809 for tmpl in self.params['forceprint'].get(key, []):
2810 self.to_stdout(self.evaluate_outtmpl(format_tmpl(tmpl), info_copy))
2811
2812 for tmpl, file_tmpl in self.params['print_to_file'].get(key, []):
5127e92a 2813 filename = self.prepare_filename(info_dict, outtmpl=file_tmpl)
bb66c247 2814 tmpl = format_tmpl(tmpl)
2815 self.to_screen(f'[info] Writing {tmpl!r} to: {filename}')
8d93e69d 2816 if self._ensure_dir_exists(filename):
86e5f3ed 2817 with open(filename, 'a', encoding='utf-8') as f:
8d93e69d 2818 f.write(self.evaluate_outtmpl(tmpl, info_copy) + '\n')
ca30f449 2819
d06daf23 2820 def __forced_printings(self, info_dict, filename, incomplete):
53c18592 2821 def print_mandatory(field, actual_field=None):
2822 if actual_field is None:
2823 actual_field = field
d06daf23 2824 if (self.params.get('force%s' % field, False)
53c18592 2825 and (not incomplete or info_dict.get(actual_field) is not None)):
2826 self.to_stdout(info_dict[actual_field])
d06daf23
S
2827
2828 def print_optional(field):
2829 if (self.params.get('force%s' % field, False)
2830 and info_dict.get(field) is not None):
2831 self.to_stdout(info_dict[field])
2832
53c18592 2833 info_dict = info_dict.copy()
2834 if filename is not None:
2835 info_dict['filename'] = filename
2836 if info_dict.get('requested_formats') is not None:
2837 # For RTMP URLs, also include the playpath
2838 info_dict['urls'] = '\n'.join(f['url'] + f.get('play_path', '') for f in info_dict['requested_formats'])
10331a26 2839 elif info_dict.get('url'):
53c18592 2840 info_dict['urls'] = info_dict['url'] + info_dict.get('play_path', '')
2841
bb66c247 2842 if (self.params.get('forcejson')
2843 or self.params['forceprint'].get('video')
2844 or self.params['print_to_file'].get('video')):
2b8a2973 2845 self.post_extract(info_dict)
bb66c247 2846 self._forceprint('video', info_dict)
53c18592 2847
d06daf23
S
2848 print_mandatory('title')
2849 print_mandatory('id')
53c18592 2850 print_mandatory('url', 'urls')
d06daf23
S
2851 print_optional('thumbnail')
2852 print_optional('description')
53c18592 2853 print_optional('filename')
b868936c 2854 if self.params.get('forceduration') and info_dict.get('duration') is not None:
d06daf23
S
2855 self.to_stdout(formatSeconds(info_dict['duration']))
2856 print_mandatory('format')
53c18592 2857
2b8a2973 2858 if self.params.get('forcejson'):
6e84b215 2859 self.to_stdout(json.dumps(self.sanitize_info(info_dict)))
d06daf23 2860
e8e73840 2861 def dl(self, name, info, subtitle=False, test=False):
88acdbc2 2862 if not info.get('url'):
1151c407 2863 self.raise_no_formats(info, True)
e8e73840 2864
2865 if test:
2866 verbose = self.params.get('verbose')
2867 params = {
2868 'test': True,
a169858f 2869 'quiet': self.params.get('quiet') or not verbose,
e8e73840 2870 'verbose': verbose,
2871 'noprogress': not verbose,
2872 'nopart': True,
2873 'skip_unavailable_fragments': False,
2874 'keep_fragments': False,
2875 'overwrites': True,
2876 '_no_ytdl_file': True,
2877 }
2878 else:
2879 params = self.params
96fccc10 2880 fd = get_suitable_downloader(info, params, to_stdout=(name == '-'))(self, params)
e8e73840 2881 if not test:
2882 for ph in self._progress_hooks:
2883 fd.add_progress_hook(ph)
42676437
M
2884 urls = '", "'.join(
2885 (f['url'].split(',')[0] + ',<data>' if f['url'].startswith('data:') else f['url'])
2886 for f in info.get('requested_formats', []) or [info])
3a408f9d 2887 self.write_debug(f'Invoking {fd.FD_NAME} downloader on "{urls}"')
03b4de72 2888
adbc4ec4
THD
2889 # Note: Ideally info should be a deep-copied so that hooks cannot modify it.
2890 # But it may contain objects that are not deep-copyable
2891 new_info = self._copy_infodict(info)
e8e73840 2892 if new_info.get('http_headers') is None:
2893 new_info['http_headers'] = self._calc_headers(new_info)
2894 return fd.download(name, new_info, subtitle)
2895
e04938ab 2896 def existing_file(self, filepaths, *, default_overwrite=True):
2897 existing_files = list(filter(os.path.exists, orderedSet(filepaths)))
2898 if existing_files and not self.params.get('overwrites', default_overwrite):
2899 return existing_files[0]
2900
2901 for file in existing_files:
2902 self.report_file_delete(file)
2903 os.remove(file)
2904 return None
2905
8222d8de 2906 def process_info(self, info_dict):
09b49e1f 2907 """Process a single resolved IE result. (Modifies it in-place)"""
8222d8de
JMF
2908
2909 assert info_dict.get('_type', 'video') == 'video'
f46e2f9d 2910 original_infodict = info_dict
fd288278 2911
4513a41a 2912 if 'format' not in info_dict and 'ext' in info_dict:
8222d8de
JMF
2913 info_dict['format'] = info_dict['ext']
2914
09b49e1f 2915 # This is mostly just for backward compatibility of process_info
2916 # As a side-effect, this allows for format-specific filters
c77495e3 2917 if self._match_entry(info_dict) is not None:
9e907ebd 2918 info_dict['__write_download_archive'] = 'ignore'
8222d8de
JMF
2919 return
2920
09b49e1f 2921 # Does nothing under normal operation - for backward compatibility of process_info
277d6ff5 2922 self.post_extract(info_dict)
0c14d66a 2923 self._num_downloads += 1
8222d8de 2924
dcf64d43 2925 # info_dict['_filename'] needs to be set for backward compatibility
de6000d9 2926 info_dict['_filename'] = full_filename = self.prepare_filename(info_dict, warn=True)
2927 temp_filename = self.prepare_filename(info_dict, 'temp')
0202b52a 2928 files_to_move = {}
8222d8de
JMF
2929
2930 # Forced printings
4513a41a 2931 self.__forced_printings(info_dict, full_filename, incomplete=('format' not in info_dict))
8222d8de 2932
ca6d59d2 2933 def check_max_downloads():
2934 if self._num_downloads >= float(self.params.get('max_downloads') or 'inf'):
2935 raise MaxDownloadsReached()
2936
b7b04c78 2937 if self.params.get('simulate'):
9e907ebd 2938 info_dict['__write_download_archive'] = self.params.get('force_write_download_archive')
ca6d59d2 2939 check_max_downloads()
8222d8de
JMF
2940 return
2941
de6000d9 2942 if full_filename is None:
8222d8de 2943 return
e92caff5 2944 if not self._ensure_dir_exists(encodeFilename(full_filename)):
0202b52a 2945 return
e92caff5 2946 if not self._ensure_dir_exists(encodeFilename(temp_filename)):
8222d8de
JMF
2947 return
2948
80c03fa9 2949 if self._write_description('video', info_dict,
2950 self.prepare_filename(info_dict, 'description')) is None:
2951 return
2952
2953 sub_files = self._write_subtitles(info_dict, temp_filename)
2954 if sub_files is None:
2955 return
2956 files_to_move.update(dict(sub_files))
2957
2958 thumb_files = self._write_thumbnails(
2959 'video', info_dict, temp_filename, self.prepare_filename(info_dict, 'thumbnail'))
2960 if thumb_files is None:
2961 return
2962 files_to_move.update(dict(thumb_files))
8222d8de 2963
80c03fa9 2964 infofn = self.prepare_filename(info_dict, 'infojson')
2965 _infojson_written = self._write_info_json('video', info_dict, infofn)
2966 if _infojson_written:
dac5df5a 2967 info_dict['infojson_filename'] = infofn
e75bb0d6 2968 # For backward compatibility, even though it was a private field
80c03fa9 2969 info_dict['__infojson_filename'] = infofn
2970 elif _infojson_written is None:
2971 return
2972
2973 # Note: Annotations are deprecated
2974 annofn = None
1fb07d10 2975 if self.params.get('writeannotations', False):
de6000d9 2976 annofn = self.prepare_filename(info_dict, 'annotation')
80c03fa9 2977 if annofn:
e92caff5 2978 if not self._ensure_dir_exists(encodeFilename(annofn)):
0202b52a 2979 return
0c3d0f51 2980 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(annofn)):
6febd1c1 2981 self.to_screen('[info] Video annotations are already present')
ffddb112
RA
2982 elif not info_dict.get('annotations'):
2983 self.report_warning('There are no annotations to write.')
7b6fefc9
PH
2984 else:
2985 try:
6febd1c1 2986 self.to_screen('[info] Writing video annotations to: ' + annofn)
86e5f3ed 2987 with open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
7b6fefc9
PH
2988 annofile.write(info_dict['annotations'])
2989 except (KeyError, TypeError):
6febd1c1 2990 self.report_warning('There are no annotations to write.')
86e5f3ed 2991 except OSError:
6febd1c1 2992 self.report_error('Cannot write annotations file: ' + annofn)
7b6fefc9 2993 return
1fb07d10 2994
732044af 2995 # Write internet shortcut files
08438d2c 2996 def _write_link_file(link_type):
60f3e995 2997 url = try_get(info_dict['webpage_url'], iri_to_uri)
2998 if not url:
2999 self.report_warning(
3000 f'Cannot write internet shortcut file because the actual URL of "{info_dict["webpage_url"]}" is unknown')
3001 return True
08438d2c 3002 linkfn = replace_extension(self.prepare_filename(info_dict, 'link'), link_type, info_dict.get('ext'))
0e6b018a
Z
3003 if not self._ensure_dir_exists(encodeFilename(linkfn)):
3004 return False
10e3742e 3005 if self.params.get('overwrites', True) and os.path.exists(encodeFilename(linkfn)):
08438d2c 3006 self.to_screen(f'[info] Internet shortcut (.{link_type}) is already present')
3007 return True
3008 try:
3009 self.to_screen(f'[info] Writing internet shortcut (.{link_type}) to: {linkfn}')
86e5f3ed 3010 with open(encodeFilename(to_high_limit_path(linkfn)), 'w', encoding='utf-8',
3011 newline='\r\n' if link_type == 'url' else '\n') as linkfile:
60f3e995 3012 template_vars = {'url': url}
08438d2c 3013 if link_type == 'desktop':
3014 template_vars['filename'] = linkfn[:-(len(link_type) + 1)]
3015 linkfile.write(LINK_TEMPLATES[link_type] % template_vars)
86e5f3ed 3016 except OSError:
08438d2c 3017 self.report_error(f'Cannot write internet shortcut {linkfn}')
3018 return False
732044af 3019 return True
3020
08438d2c 3021 write_links = {
3022 'url': self.params.get('writeurllink'),
3023 'webloc': self.params.get('writewebloclink'),
3024 'desktop': self.params.get('writedesktoplink'),
3025 }
3026 if self.params.get('writelink'):
3027 link_type = ('webloc' if sys.platform == 'darwin'
3028 else 'desktop' if sys.platform.startswith('linux')
3029 else 'url')
3030 write_links[link_type] = True
3031
3032 if any(should_write and not _write_link_file(link_type)
3033 for link_type, should_write in write_links.items()):
3034 return
732044af 3035
f46e2f9d 3036 def replace_info_dict(new_info):
3037 nonlocal info_dict
3038 if new_info == info_dict:
3039 return
3040 info_dict.clear()
3041 info_dict.update(new_info)
3042
415f8d51 3043 new_info, files_to_move = self.pre_process(info_dict, 'before_dl', files_to_move)
3044 replace_info_dict(new_info)
56d868db 3045
a13e6848 3046 if self.params.get('skip_download'):
56d868db 3047 info_dict['filepath'] = temp_filename
3048 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
3049 info_dict['__files_to_move'] = files_to_move
f46e2f9d 3050 replace_info_dict(self.run_pp(MoveFilesAfterDownloadPP(self, False), info_dict))
9e907ebd 3051 info_dict['__write_download_archive'] = self.params.get('force_write_download_archive')
56d868db 3052 else:
3053 # Download
b868936c 3054 info_dict.setdefault('__postprocessors', [])
4340deca 3055 try:
0202b52a 3056
e04938ab 3057 def existing_video_file(*filepaths):
6b591b29 3058 ext = info_dict.get('ext')
e04938ab 3059 converted = lambda file: replace_extension(file, self.params.get('final_ext') or ext, ext)
3060 file = self.existing_file(itertools.chain(*zip(map(converted, filepaths), filepaths)),
3061 default_overwrite=False)
3062 if file:
3063 info_dict['ext'] = os.path.splitext(file)[1][1:]
3064 return file
0202b52a 3065
7b2c3f47 3066 fd, success = None, True
fccf90e7 3067 if info_dict.get('protocol') or info_dict.get('url'):
56ba69e4 3068 fd = get_suitable_downloader(info_dict, self.params, to_stdout=temp_filename == '-')
3069 if fd is not FFmpegFD and (
3070 info_dict.get('section_start') or info_dict.get('section_end')):
7b2c3f47 3071 msg = ('This format cannot be partially downloaded' if FFmpegFD.available()
56ba69e4 3072 else 'You have requested downloading the video partially, but ffmpeg is not installed')
3073 self.report_error(f'{msg}. Aborting')
5ec1b6b7 3074 return
5ec1b6b7 3075
4340deca 3076 if info_dict.get('requested_formats') is not None:
81cd954a 3077 requested_formats = info_dict['requested_formats']
0202b52a 3078 old_ext = info_dict['ext']
4e3b637d 3079 if self.params.get('merge_output_format') is None:
4e3b637d 3080 if (info_dict['ext'] == 'webm'
3081 and info_dict.get('thumbnails')
3082 # check with type instead of pp_key, __name__, or isinstance
3083 # since we dont want any custom PPs to trigger this
c487cf00 3084 and any(type(pp) == EmbedThumbnailPP for pp in self._pps['post_process'])): # noqa: E721
4e3b637d 3085 info_dict['ext'] = 'mkv'
3086 self.report_warning(
3087 'webm doesn\'t support embedding a thumbnail, mkv will be used')
124bc071 3088 new_ext = info_dict['ext']
0202b52a 3089
124bc071 3090 def correct_ext(filename, ext=new_ext):
96fccc10 3091 if filename == '-':
3092 return filename
0202b52a 3093 filename_real_ext = os.path.splitext(filename)[1][1:]
3094 filename_wo_ext = (
3095 os.path.splitext(filename)[0]
124bc071 3096 if filename_real_ext in (old_ext, new_ext)
0202b52a 3097 else filename)
86e5f3ed 3098 return f'{filename_wo_ext}.{ext}'
0202b52a 3099
38c6902b 3100 # Ensure filename always has a correct extension for successful merge
0202b52a 3101 full_filename = correct_ext(full_filename)
3102 temp_filename = correct_ext(temp_filename)
e04938ab 3103 dl_filename = existing_video_file(full_filename, temp_filename)
1ea24129 3104 info_dict['__real_download'] = False
18e674b4 3105
7b2c3f47 3106 merger = FFmpegMergerPP(self)
adbc4ec4 3107 downloaded = []
dbf5416a 3108 if dl_filename is not None:
6c7274ec 3109 self.report_file_already_downloaded(dl_filename)
adbc4ec4
THD
3110 elif fd:
3111 for f in requested_formats if fd != FFmpegFD else []:
3112 f['filepath'] = fname = prepend_extension(
3113 correct_ext(temp_filename, info_dict['ext']),
3114 'f%s' % f['format_id'], info_dict['ext'])
3115 downloaded.append(fname)
dbf5416a 3116 info_dict['url'] = '\n'.join(f['url'] for f in requested_formats)
3117 success, real_download = self.dl(temp_filename, info_dict)
3118 info_dict['__real_download'] = real_download
18e674b4 3119 else:
18e674b4 3120 if self.params.get('allow_unplayable_formats'):
3121 self.report_warning(
3122 'You have requested merging of multiple formats '
3123 'while also allowing unplayable formats to be downloaded. '
3124 'The formats won\'t be merged to prevent data corruption.')
3125 elif not merger.available:
e8969bda 3126 msg = 'You have requested merging of multiple formats but ffmpeg is not installed'
3127 if not self.params.get('ignoreerrors'):
3128 self.report_error(f'{msg}. Aborting due to --abort-on-error')
3129 return
3130 self.report_warning(f'{msg}. The formats won\'t be merged')
18e674b4 3131
96fccc10 3132 if temp_filename == '-':
adbc4ec4 3133 reason = ('using a downloader other than ffmpeg' if FFmpegFD.can_merge_formats(info_dict, self.params)
96fccc10 3134 else 'but the formats are incompatible for simultaneous download' if merger.available
3135 else 'but ffmpeg is not installed')
3136 self.report_warning(
3137 f'You have requested downloading multiple formats to stdout {reason}. '
3138 'The formats will be streamed one after the other')
3139 fname = temp_filename
dbf5416a 3140 for f in requested_formats:
3141 new_info = dict(info_dict)
3142 del new_info['requested_formats']
3143 new_info.update(f)
96fccc10 3144 if temp_filename != '-':
124bc071 3145 fname = prepend_extension(
3146 correct_ext(temp_filename, new_info['ext']),
3147 'f%s' % f['format_id'], new_info['ext'])
96fccc10 3148 if not self._ensure_dir_exists(fname):
3149 return
a21e0ab1 3150 f['filepath'] = fname
96fccc10 3151 downloaded.append(fname)
dbf5416a 3152 partial_success, real_download = self.dl(fname, new_info)
3153 info_dict['__real_download'] = info_dict['__real_download'] or real_download
3154 success = success and partial_success
adbc4ec4
THD
3155
3156 if downloaded and merger.available and not self.params.get('allow_unplayable_formats'):
3157 info_dict['__postprocessors'].append(merger)
3158 info_dict['__files_to_merge'] = downloaded
3159 # Even if there were no downloads, it is being merged only now
3160 info_dict['__real_download'] = True
3161 else:
3162 for file in downloaded:
3163 files_to_move[file] = None
4340deca
P
3164 else:
3165 # Just a single file
e04938ab 3166 dl_filename = existing_video_file(full_filename, temp_filename)
6c7274ec 3167 if dl_filename is None or dl_filename == temp_filename:
3168 # dl_filename == temp_filename could mean that the file was partially downloaded with --no-part.
3169 # So we should try to resume the download
e8e73840 3170 success, real_download = self.dl(temp_filename, info_dict)
0202b52a 3171 info_dict['__real_download'] = real_download
6c7274ec 3172 else:
3173 self.report_file_already_downloaded(dl_filename)
0202b52a 3174
0202b52a 3175 dl_filename = dl_filename or temp_filename
c571435f 3176 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
0202b52a 3177
3158150c 3178 except network_exceptions as err:
7960b056 3179 self.report_error('unable to download video data: %s' % error_to_compat_str(err))
4340deca 3180 return
86e5f3ed 3181 except OSError as err:
4340deca
P
3182 raise UnavailableVideoError(err)
3183 except (ContentTooShortError, ) as err:
86e5f3ed 3184 self.report_error(f'content too short (expected {err.expected} bytes and served {err.downloaded})')
4340deca 3185 return
8222d8de 3186
415f8d51 3187 self._raise_pending_errors(info_dict)
de6000d9 3188 if success and full_filename != '-':
f17f8651 3189
fd7cfb64 3190 def fixup():
3191 do_fixup = True
3192 fixup_policy = self.params.get('fixup')
3193 vid = info_dict['id']
3194
3195 if fixup_policy in ('ignore', 'never'):
3196 return
3197 elif fixup_policy == 'warn':
3fe75fdc 3198 do_fixup = 'warn'
f89b3e2d 3199 elif fixup_policy != 'force':
3200 assert fixup_policy in ('detect_or_warn', None)
3201 if not info_dict.get('__real_download'):
3202 do_fixup = False
fd7cfb64 3203
3204 def ffmpeg_fixup(cndn, msg, cls):
3fe75fdc 3205 if not (do_fixup and cndn):
fd7cfb64 3206 return
3fe75fdc 3207 elif do_fixup == 'warn':
fd7cfb64 3208 self.report_warning(f'{vid}: {msg}')
3209 return
3210 pp = cls(self)
3211 if pp.available:
3212 info_dict['__postprocessors'].append(pp)
3213 else:
3214 self.report_warning(f'{vid}: {msg}. Install ffmpeg to fix this automatically')
3215
3216 stretched_ratio = info_dict.get('stretched_ratio')
ca9def71
LNO
3217 ffmpeg_fixup(stretched_ratio not in (1, None),
3218 f'Non-uniform pixel ratio {stretched_ratio}',
3219 FFmpegFixupStretchedPP)
fd7cfb64 3220
993191c0 3221 downloader = get_suitable_downloader(info_dict, self.params) if 'protocol' in info_dict else None
24146491 3222 downloader = downloader.FD_NAME if downloader else None
adbc4ec4 3223
ca9def71
LNO
3224 ext = info_dict.get('ext')
3225 postprocessed_by_ffmpeg = info_dict.get('requested_formats') or any((
3226 isinstance(pp, FFmpegVideoConvertorPP)
3227 and resolve_recode_mapping(ext, pp.mapping)[0] not in (ext, None)
3228 ) for pp in self._pps['post_process'])
3229
3230 if not postprocessed_by_ffmpeg:
3231 ffmpeg_fixup(ext == 'm4a' and info_dict.get('container') == 'm4a_dash',
f2df4071 3232 'writing DASH m4a. Only some players support this container',
3233 FFmpegFixupM4aPP)
24146491 3234 ffmpeg_fixup(downloader == 'hlsnative' and not self.params.get('hls_use_mpegts')
494f5230 3235 or info_dict.get('is_live') and self.params.get('hls_use_mpegts') is None,
adbc4ec4
THD
3236 'Possible MPEG-TS in MP4 container or malformed AAC timestamps',
3237 FFmpegFixupM3u8PP)
3238 ffmpeg_fixup(info_dict.get('is_live') and downloader == 'DashSegmentsFD',
3239 'Possible duplicate MOOV atoms', FFmpegFixupDuplicateMoovPP)
3240
24146491 3241 ffmpeg_fixup(downloader == 'web_socket_fragment', 'Malformed timestamps detected', FFmpegFixupTimestampPP)
3242 ffmpeg_fixup(downloader == 'web_socket_fragment', 'Malformed duration detected', FFmpegFixupDurationPP)
fd7cfb64 3243
3244 fixup()
8222d8de 3245 try:
f46e2f9d 3246 replace_info_dict(self.post_process(dl_filename, info_dict, files_to_move))
af819c21 3247 except PostProcessingError as err:
3248 self.report_error('Postprocessing: %s' % str(err))
8222d8de 3249 return
ab8e5e51
AM
3250 try:
3251 for ph in self._post_hooks:
23c1a667 3252 ph(info_dict['filepath'])
ab8e5e51
AM
3253 except Exception as err:
3254 self.report_error('post hooks: %s' % str(err))
3255 return
9e907ebd 3256 info_dict['__write_download_archive'] = True
2d30509f 3257
c487cf00 3258 assert info_dict is original_infodict # Make sure the info_dict was modified in-place
a13e6848 3259 if self.params.get('force_write_download_archive'):
9e907ebd 3260 info_dict['__write_download_archive'] = True
ca6d59d2 3261 check_max_downloads()
8222d8de 3262
aa9369a2 3263 def __download_wrapper(self, func):
3264 @functools.wraps(func)
3265 def wrapper(*args, **kwargs):
3266 try:
3267 res = func(*args, **kwargs)
3268 except UnavailableVideoError as e:
3269 self.report_error(e)
b222c271 3270 except DownloadCancelled as e:
3271 self.to_screen(f'[info] {e}')
3272 if not self.params.get('break_per_url'):
3273 raise
fd404bec 3274 self._num_downloads = 0
aa9369a2 3275 else:
3276 if self.params.get('dump_single_json', False):
3277 self.post_extract(res)
3278 self.to_stdout(json.dumps(self.sanitize_info(res)))
3279 return wrapper
3280
8222d8de
JMF
3281 def download(self, url_list):
3282 """Download a given list of URLs."""
aa9369a2 3283 url_list = variadic(url_list) # Passing a single URL is a common mistake
bf1824b3 3284 outtmpl = self.params['outtmpl']['default']
3089bc74
S
3285 if (len(url_list) > 1
3286 and outtmpl != '-'
3287 and '%' not in outtmpl
3288 and self.params.get('max_downloads') != 1):
acd69589 3289 raise SameFileError(outtmpl)
8222d8de
JMF
3290
3291 for url in url_list:
aa9369a2 3292 self.__download_wrapper(self.extract_info)(
3293 url, force_generic_extractor=self.params.get('force_generic_extractor', False))
8222d8de
JMF
3294
3295 return self._download_retcode
3296
1dcc4c0c 3297 def download_with_info_file(self, info_filename):
31bd3925
JMF
3298 with contextlib.closing(fileinput.FileInput(
3299 [info_filename], mode='r',
3300 openhook=fileinput.hook_encoded('utf-8'))) as f:
3301 # FileInput doesn't have a read method, we can't call json.load
8012d892 3302 info = self.sanitize_info(json.loads('\n'.join(f)), self.params.get('clean_infojson', True))
d4943898 3303 try:
aa9369a2 3304 self.__download_wrapper(self.process_ie_result)(info, download=True)
f2ebc5c7 3305 except (DownloadError, EntryNotInPlaylist, ReExtractInfo) as e:
bf5f605e 3306 if not isinstance(e, EntryNotInPlaylist):
3307 self.to_stderr('\r')
d4943898
JMF
3308 webpage_url = info.get('webpage_url')
3309 if webpage_url is not None:
aa9369a2 3310 self.report_warning(f'The info failed to download: {e}; trying with URL {webpage_url}')
d4943898
JMF
3311 return self.download([webpage_url])
3312 else:
3313 raise
3314 return self._download_retcode
1dcc4c0c 3315
cb202fd2 3316 @staticmethod
8012d892 3317 def sanitize_info(info_dict, remove_private_keys=False):
3318 ''' Sanitize the infodict for converting to json '''
3ad56b42 3319 if info_dict is None:
3320 return info_dict
6e84b215 3321 info_dict.setdefault('epoch', int(time.time()))
6a5a30f9 3322 info_dict.setdefault('_type', 'video')
b5e7a2e6 3323 info_dict.setdefault('_version', {
3324 'version': __version__,
3325 'current_git_head': current_git_head(),
3326 'release_git_head': RELEASE_GIT_HEAD,
3327 'repository': REPOSITORY,
3328 })
09b49e1f 3329
8012d892 3330 if remove_private_keys:
0a5a191a 3331 reject = lambda k, v: v is None or k.startswith('__') or k in {
f46e2f9d 3332 'requested_downloads', 'requested_formats', 'requested_subtitles', 'requested_entries',
0a5a191a 3333 'entries', 'filepath', '_filename', 'infojson_filename', 'original_url', 'playlist_autonumber',
6e84b215 3334 }
ae8f99e6 3335 else:
09b49e1f 3336 reject = lambda k, v: False
adbc4ec4
THD
3337
3338 def filter_fn(obj):
3339 if isinstance(obj, dict):
3340 return {k: filter_fn(v) for k, v in obj.items() if not reject(k, v)}
3341 elif isinstance(obj, (list, tuple, set, LazyList)):
3342 return list(map(filter_fn, obj))
3343 elif obj is None or isinstance(obj, (str, int, float, bool)):
3344 return obj
3345 else:
3346 return repr(obj)
3347
5226731e 3348 return filter_fn(info_dict)
cb202fd2 3349
8012d892 3350 @staticmethod
3351 def filter_requested_info(info_dict, actually_filter=True):
3352 ''' Alias of sanitize_info for backward compatibility '''
3353 return YoutubeDL.sanitize_info(info_dict, actually_filter)
3354
43d7f5a5 3355 def _delete_downloaded_files(self, *files_to_delete, info={}, msg=None):
3356 for filename in set(filter(None, files_to_delete)):
3357 if msg:
3358 self.to_screen(msg % filename)
3359 try:
3360 os.remove(filename)
3361 except OSError:
3362 self.report_warning(f'Unable to delete file {filename}')
3363 if filename in info.get('__files_to_move', []): # NB: Delete even if None
3364 del info['__files_to_move'][filename]
3365
ed5835b4 3366 @staticmethod
3367 def post_extract(info_dict):
3368 def actual_post_extract(info_dict):
3369 if info_dict.get('_type') in ('playlist', 'multi_video'):
3370 for video_dict in info_dict.get('entries', {}):
3371 actual_post_extract(video_dict or {})
3372 return
3373
09b49e1f 3374 post_extractor = info_dict.pop('__post_extractor', None) or (lambda: {})
3375 info_dict.update(post_extractor())
ed5835b4 3376
3377 actual_post_extract(info_dict or {})
3378
dcf64d43 3379 def run_pp(self, pp, infodict):
5bfa4862 3380 files_to_delete = []
dcf64d43 3381 if '__files_to_move' not in infodict:
3382 infodict['__files_to_move'] = {}
b1940459 3383 try:
3384 files_to_delete, infodict = pp.run(infodict)
3385 except PostProcessingError as e:
3386 # Must be True and not 'only_download'
3387 if self.params.get('ignoreerrors') is True:
3388 self.report_error(e)
3389 return infodict
3390 raise
3391
5bfa4862 3392 if not files_to_delete:
dcf64d43 3393 return infodict
5bfa4862 3394 if self.params.get('keepvideo', False):
3395 for f in files_to_delete:
dcf64d43 3396 infodict['__files_to_move'].setdefault(f, '')
5bfa4862 3397 else:
43d7f5a5 3398 self._delete_downloaded_files(
3399 *files_to_delete, info=infodict, msg='Deleting original file %s (pass -k to keep)')
dcf64d43 3400 return infodict
5bfa4862 3401
ed5835b4 3402 def run_all_pps(self, key, info, *, additional_pps=None):
bb66c247 3403 self._forceprint(key, info)
ed5835b4 3404 for pp in (additional_pps or []) + self._pps[key]:
dc5f409c 3405 info = self.run_pp(pp, info)
ed5835b4 3406 return info
277d6ff5 3407
56d868db 3408 def pre_process(self, ie_info, key='pre_process', files_to_move=None):
5bfa4862 3409 info = dict(ie_info)
56d868db 3410 info['__files_to_move'] = files_to_move or {}
415f8d51 3411 try:
3412 info = self.run_all_pps(key, info)
3413 except PostProcessingError as err:
3414 msg = f'Preprocessing: {err}'
3415 info.setdefault('__pending_error', msg)
3416 self.report_error(msg, is_error=False)
56d868db 3417 return info, info.pop('__files_to_move', None)
5bfa4862 3418
f46e2f9d 3419 def post_process(self, filename, info, files_to_move=None):
8222d8de 3420 """Run all the postprocessors on the given file."""
8222d8de 3421 info['filepath'] = filename
dcf64d43 3422 info['__files_to_move'] = files_to_move or {}
ed5835b4 3423 info = self.run_all_pps('post_process', info, additional_pps=info.get('__postprocessors'))
dcf64d43 3424 info = self.run_pp(MoveFilesAfterDownloadPP(self), info)
3425 del info['__files_to_move']
ed5835b4 3426 return self.run_all_pps('after_move', info)
c1c9a79c 3427
5db07df6 3428 def _make_archive_id(self, info_dict):
e9fef7ee
S
3429 video_id = info_dict.get('id')
3430 if not video_id:
3431 return
5db07df6
PH
3432 # Future-proof against any change in case
3433 # and backwards compatibility with prior versions
e9fef7ee 3434 extractor = info_dict.get('extractor_key') or info_dict.get('ie_key') # key in a playlist
7012b23c 3435 if extractor is None:
1211bb6d
S
3436 url = str_or_none(info_dict.get('url'))
3437 if not url:
3438 return
e9fef7ee 3439 # Try to find matching extractor for the URL and take its ie_key
8b7491c8 3440 for ie_key, ie in self._ies.items():
1211bb6d 3441 if ie.suitable(url):
8b7491c8 3442 extractor = ie_key
e9fef7ee
S
3443 break
3444 else:
3445 return
0647d925 3446 return make_archive_id(extractor, video_id)
5db07df6
PH
3447
3448 def in_download_archive(self, info_dict):
3449 fn = self.params.get('download_archive')
3450 if fn is None:
3451 return False
3452
1e8fe57e 3453 vid_ids = [self._make_archive_id(info_dict)]
c200096c 3454 vid_ids.extend(info_dict.get('_old_archive_ids') or [])
1e8fe57e 3455 return any(id_ in self.archive for id_ in vid_ids)
c1c9a79c
PH
3456
3457 def record_download_archive(self, info_dict):
3458 fn = self.params.get('download_archive')
3459 if fn is None:
3460 return
5db07df6
PH
3461 vid_id = self._make_archive_id(info_dict)
3462 assert vid_id
a13e6848 3463 self.write_debug(f'Adding to archive: {vid_id}')
c1c9a79c 3464 with locked_file(fn, 'a', encoding='utf-8') as archive_file:
6febd1c1 3465 archive_file.write(vid_id + '\n')
a45e8619 3466 self.archive.add(vid_id)
dd82ffea 3467
8c51aa65 3468 @staticmethod
8abeeb94 3469 def format_resolution(format, default='unknown'):
9359f3d4 3470 if format.get('vcodec') == 'none' and format.get('acodec') != 'none':
fb04e403 3471 return 'audio only'
f49d89ee
PH
3472 if format.get('resolution') is not None:
3473 return format['resolution']
35615307 3474 if format.get('width') and format.get('height'):
ff51ed58 3475 return '%dx%d' % (format['width'], format['height'])
35615307 3476 elif format.get('height'):
ff51ed58 3477 return '%sp' % format['height']
35615307 3478 elif format.get('width'):
ff51ed58 3479 return '%dx?' % format['width']
3480 return default
8c51aa65 3481
8130779d 3482 def _list_format_headers(self, *headers):
3483 if self.params.get('listformats_table', True) is not False:
591bb9d3 3484 return [self._format_out(header, self.Styles.HEADERS) for header in headers]
8130779d 3485 return headers
3486
c57f7757
PH
3487 def _format_note(self, fdict):
3488 res = ''
3489 if fdict.get('ext') in ['f4f', 'f4m']:
f304da8a 3490 res += '(unsupported)'
32f90364
PH
3491 if fdict.get('language'):
3492 if res:
3493 res += ' '
f304da8a 3494 res += '[%s]' % fdict['language']
c57f7757 3495 if fdict.get('format_note') is not None:
f304da8a 3496 if res:
3497 res += ' '
3498 res += fdict['format_note']
c57f7757 3499 if fdict.get('tbr') is not None:
f304da8a 3500 if res:
3501 res += ', '
3502 res += '%4dk' % fdict['tbr']
c57f7757
PH
3503 if fdict.get('container') is not None:
3504 if res:
3505 res += ', '
3506 res += '%s container' % fdict['container']
3089bc74
S
3507 if (fdict.get('vcodec') is not None
3508 and fdict.get('vcodec') != 'none'):
c57f7757
PH
3509 if res:
3510 res += ', '
3511 res += fdict['vcodec']
91c7271a 3512 if fdict.get('vbr') is not None:
c57f7757
PH
3513 res += '@'
3514 elif fdict.get('vbr') is not None and fdict.get('abr') is not None:
3515 res += 'video@'
3516 if fdict.get('vbr') is not None:
3517 res += '%4dk' % fdict['vbr']
fbb21cf5 3518 if fdict.get('fps') is not None:
5d583bdf
S
3519 if res:
3520 res += ', '
3521 res += '%sfps' % fdict['fps']
c57f7757
PH
3522 if fdict.get('acodec') is not None:
3523 if res:
3524 res += ', '
3525 if fdict['acodec'] == 'none':
3526 res += 'video only'
3527 else:
3528 res += '%-5s' % fdict['acodec']
3529 elif fdict.get('abr') is not None:
3530 if res:
3531 res += ', '
3532 res += 'audio'
3533 if fdict.get('abr') is not None:
3534 res += '@%3dk' % fdict['abr']
3535 if fdict.get('asr') is not None:
3536 res += ' (%5dHz)' % fdict['asr']
3537 if fdict.get('filesize') is not None:
3538 if res:
3539 res += ', '
3540 res += format_bytes(fdict['filesize'])
9732d77e
PH
3541 elif fdict.get('filesize_approx') is not None:
3542 if res:
3543 res += ', '
3544 res += '~' + format_bytes(fdict['filesize_approx'])
c57f7757 3545 return res
91c7271a 3546
8130779d 3547 def render_formats_table(self, info_dict):
b69fd25c 3548 if not info_dict.get('formats') and not info_dict.get('url'):
8130779d 3549 return None
b69fd25c 3550
94badb25 3551 formats = info_dict.get('formats', [info_dict])
8130779d 3552 if not self.params.get('listformats_table', True) is not False:
76d321f6 3553 table = [
3554 [
3555 format_field(f, 'format_id'),
3556 format_field(f, 'ext'),
3557 self.format_resolution(f),
8130779d 3558 self._format_note(f)
3559 ] for f in formats if f.get('preference') is None or f['preference'] >= -1000]
3560 return render_table(['format code', 'extension', 'resolution', 'note'], table, extra_gap=1)
3561
d816f61f 3562 def simplified_codec(f, field):
3563 assert field in ('acodec', 'vcodec')
3564 codec = f.get(field, 'unknown')
f5ea4748 3565 if not codec:
3566 return 'unknown'
3567 elif codec != 'none':
d816f61f 3568 return '.'.join(codec.split('.')[:4])
3569
3570 if field == 'vcodec' and f.get('acodec') == 'none':
3571 return 'images'
3572 elif field == 'acodec' and f.get('vcodec') == 'none':
3573 return ''
3574 return self._format_out('audio only' if field == 'vcodec' else 'video only',
3575 self.Styles.SUPPRESS)
3576
591bb9d3 3577 delim = self._format_out('\u2502', self.Styles.DELIM, '|', test_encoding=True)
8130779d 3578 table = [
3579 [
591bb9d3 3580 self._format_out(format_field(f, 'format_id'), self.Styles.ID),
8130779d 3581 format_field(f, 'ext'),
3582 format_field(f, func=self.format_resolution, ignore=('audio only', 'images')),
563e0bf8 3583 format_field(f, 'fps', '\t%d', func=round),
8130779d 3584 format_field(f, 'dynamic_range', '%s', ignore=(None, 'SDR')).replace('HDR', ''),
b8ed0f15 3585 format_field(f, 'audio_channels', '\t%s'),
8130779d 3586 delim,
3587 format_field(f, 'filesize', ' \t%s', func=format_bytes) + format_field(f, 'filesize_approx', '~\t%s', func=format_bytes),
563e0bf8 3588 format_field(f, 'tbr', '\t%dk', func=round),
8130779d 3589 shorten_protocol_name(f.get('protocol', '')),
3590 delim,
d816f61f 3591 simplified_codec(f, 'vcodec'),
563e0bf8 3592 format_field(f, 'vbr', '\t%dk', func=round),
d816f61f 3593 simplified_codec(f, 'acodec'),
563e0bf8 3594 format_field(f, 'abr', '\t%dk', func=round),
ae61d108 3595 format_field(f, 'asr', '\t%s', func=format_decimal_suffix),
8130779d 3596 join_nonempty(
591bb9d3 3597 self._format_out('UNSUPPORTED', 'light red') if f.get('ext') in ('f4f', 'f4m') else None,
8130779d 3598 format_field(f, 'language', '[%s]'),
3599 join_nonempty(format_field(f, 'format_note'),
3600 format_field(f, 'container', ignore=(None, f.get('ext'))),
3601 delim=', '),
3602 delim=' '),
3603 ] for f in formats if f.get('preference') is None or f['preference'] >= -1000]
3604 header_line = self._list_format_headers(
b8ed0f15 3605 'ID', 'EXT', 'RESOLUTION', '\tFPS', 'HDR', 'CH', delim, '\tFILESIZE', '\tTBR', 'PROTO',
8130779d 3606 delim, 'VCODEC', '\tVBR', 'ACODEC', '\tABR', '\tASR', 'MORE INFO')
3607
3608 return render_table(
3609 header_line, table, hide_empty=True,
591bb9d3 3610 delim=self._format_out('\u2500', self.Styles.DELIM, '-', test_encoding=True))
8130779d 3611
3612 def render_thumbnails_table(self, info_dict):
88f23a18 3613 thumbnails = list(info_dict.get('thumbnails') or [])
cfb56d1a 3614 if not thumbnails:
8130779d 3615 return None
3616 return render_table(
ec11a9f4 3617 self._list_format_headers('ID', 'Width', 'Height', 'URL'),
6970b600 3618 [[t.get('id'), t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails])
2412044c 3619
8130779d 3620 def render_subtitles_table(self, video_id, subtitles):
2412044c 3621 def _row(lang, formats):
49c258e1 3622 exts, names = zip(*((f['ext'], f.get('name') or 'unknown') for f in reversed(formats)))
2412044c 3623 if len(set(names)) == 1:
7aee40c1 3624 names = [] if names[0] == 'unknown' else names[:1]
2412044c 3625 return [lang, ', '.join(names), ', '.join(exts)]
3626
8130779d 3627 if not subtitles:
3628 return None
3629 return render_table(
ec11a9f4 3630 self._list_format_headers('Language', 'Name', 'Formats'),
2412044c 3631 [_row(lang, formats) for lang, formats in subtitles.items()],
8130779d 3632 hide_empty=True)
3633
3634 def __list_table(self, video_id, name, func, *args):
3635 table = func(*args)
3636 if not table:
3637 self.to_screen(f'{video_id} has no {name}')
3638 return
3639 self.to_screen(f'[info] Available {name} for {video_id}:')
3640 self.to_stdout(table)
3641
3642 def list_formats(self, info_dict):
3643 self.__list_table(info_dict['id'], 'formats', self.render_formats_table, info_dict)
3644
3645 def list_thumbnails(self, info_dict):
3646 self.__list_table(info_dict['id'], 'thumbnails', self.render_thumbnails_table, info_dict)
3647
3648 def list_subtitles(self, video_id, subtitles, name='subtitles'):
3649 self.__list_table(video_id, name, self.render_subtitles_table, video_id, subtitles)
a504ced0 3650
dca08720
PH
3651 def urlopen(self, req):
3652 """ Start an HTTP download """
f9934b96 3653 if isinstance(req, str):
67dda517 3654 req = sanitized_Request(req)
19a41fc6 3655 return self._opener.open(req, timeout=self._socket_timeout)
dca08720
PH
3656
3657 def print_debug_header(self):
3658 if not self.params.get('verbose'):
3659 return
49a57e70 3660
560738f3 3661 # These imports can be slow. So import them only as needed
3662 from .extractor.extractors import _LAZY_LOADER
3663 from .extractor.extractors import _PLUGIN_CLASSES as plugin_extractors
3664
49a57e70 3665 def get_encoding(stream):
2a938746 3666 ret = str(getattr(stream, 'encoding', 'missing (%s)' % type(stream).__name__))
49a57e70 3667 if not supports_terminal_sequences(stream):
53973b4d 3668 from .utils import WINDOWS_VT_MODE # Must be imported locally
e3c7d495 3669 ret += ' (No VT)' if WINDOWS_VT_MODE is False else ' (No ANSI)'
49a57e70 3670 return ret
3671
591bb9d3 3672 encoding_str = 'Encodings: locale %s, fs %s, pref %s, %s' % (
49a57e70 3673 locale.getpreferredencoding(),
3674 sys.getfilesystemencoding(),
591bb9d3 3675 self.get_encoding(),
3676 ', '.join(
64fa820c 3677 f'{key} {get_encoding(stream)}' for key, stream in self._out_files.items_
591bb9d3 3678 if stream is not None and key != 'console')
3679 )
883d4b1e 3680
3681 logger = self.params.get('logger')
3682 if logger:
3683 write_debug = lambda msg: logger.debug(f'[debug] {msg}')
3684 write_debug(encoding_str)
3685 else:
96565c7e 3686 write_string(f'[debug] {encoding_str}\n', encoding=None)
49a57e70 3687 write_debug = lambda msg: self._write_string(f'[debug] {msg}\n')
734f90bb 3688
4c88ff87 3689 source = detect_variant()
70b23409 3690 if VARIANT not in (None, 'pip'):
3691 source += '*'
36eaf303 3692 write_debug(join_nonempty(
b5e7a2e6 3693 f'{"yt-dlp" if REPOSITORY == "yt-dlp/yt-dlp" else REPOSITORY} version',
3694 __version__,
36eaf303 3695 f'[{RELEASE_GIT_HEAD}]' if RELEASE_GIT_HEAD else '',
3696 '' if source == 'unknown' else f'({source})',
3697 delim=' '))
6e21fdd2 3698 if not _LAZY_LOADER:
3699 if os.environ.get('YTDLP_NO_LAZY_EXTRACTORS'):
49a57e70 3700 write_debug('Lazy loading extractors is forcibly disabled')
6e21fdd2 3701 else:
49a57e70 3702 write_debug('Lazy loading extractors is disabled')
3ae5e797 3703 if plugin_extractors or plugin_postprocessors:
49a57e70 3704 write_debug('Plugins: %s' % [
3ae5e797 3705 '%s%s' % (klass.__name__, '' if klass.__name__ == name else f' as {name}')
3706 for name, klass in itertools.chain(plugin_extractors.items(), plugin_postprocessors.items())])
8a82af35 3707 if self.params['compat_opts']:
3708 write_debug('Compatibility options: %s' % ', '.join(self.params['compat_opts']))
36eaf303 3709
b5e7a2e6 3710 if current_git_head():
3711 write_debug(f'Git HEAD: {current_git_head()}')
b1f94422 3712 write_debug(system_identifier())
d28b5171 3713
8913ef74 3714 exe_versions, ffmpeg_features = FFmpegPostProcessor.get_versions_and_features(self)
3715 ffmpeg_features = {key for key, val in ffmpeg_features.items() if val}
3716 if ffmpeg_features:
19a03940 3717 exe_versions['ffmpeg'] += ' (%s)' % ','.join(sorted(ffmpeg_features))
8913ef74 3718
4c83c967 3719 exe_versions['rtmpdump'] = rtmpdump_version()
feee8d32 3720 exe_versions['phantomjs'] = PhantomJSwrapper._version()
d28b5171 3721 exe_str = ', '.join(
2831b468 3722 f'{exe} {v}' for exe, v in sorted(exe_versions.items()) if v
3723 ) or 'none'
49a57e70 3724 write_debug('exe versions: %s' % exe_str)
dca08720 3725
1d485a1a 3726 from .compat.compat_utils import get_package_info
9b8ee23b 3727 from .dependencies import available_dependencies
3728
3729 write_debug('Optional libraries: %s' % (', '.join(sorted({
1d485a1a 3730 join_nonempty(*get_package_info(m)) for m in available_dependencies.values()
9b8ee23b 3731 })) or 'none'))
2831b468 3732
97ec5bc5 3733 self._setup_opener()
dca08720
PH
3734 proxy_map = {}
3735 for handler in self._opener.handlers:
3736 if hasattr(handler, 'proxies'):
3737 proxy_map.update(handler.proxies)
49a57e70 3738 write_debug(f'Proxy map: {proxy_map}')
dca08720 3739
49a57e70 3740 # Not implemented
3741 if False and self.params.get('call_home'):
0f06bcd7 3742 ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode()
49a57e70 3743 write_debug('Public IP address: %s' % ipaddr)
58b1f00d 3744 latest_version = self.urlopen(
0f06bcd7 3745 'https://yt-dl.org/latest/version').read().decode()
58b1f00d
PH
3746 if version_tuple(latest_version) > version_tuple(__version__):
3747 self.report_warning(
3748 'You are using an outdated version (newest version: %s)! '
3749 'See https://yt-dl.org/update if you need help updating.' %
3750 latest_version)
3751
e344693b 3752 def _setup_opener(self):
97ec5bc5 3753 if hasattr(self, '_opener'):
3754 return
6ad14cab 3755 timeout_val = self.params.get('socket_timeout')
17bddf3e 3756 self._socket_timeout = 20 if timeout_val is None else float(timeout_val)
6ad14cab 3757
982ee69a 3758 opts_cookiesfrombrowser = self.params.get('cookiesfrombrowser')
dca08720
PH
3759 opts_cookiefile = self.params.get('cookiefile')
3760 opts_proxy = self.params.get('proxy')
3761
982ee69a 3762 self.cookiejar = load_cookies(opts_cookiefile, opts_cookiesfrombrowser, self)
dca08720 3763
6a3f4c3f 3764 cookie_processor = YoutubeDLCookieProcessor(self.cookiejar)
dca08720
PH
3765 if opts_proxy is not None:
3766 if opts_proxy == '':
3767 proxies = {}
3768 else:
3769 proxies = {'http': opts_proxy, 'https': opts_proxy}
3770 else:
ac668111 3771 proxies = urllib.request.getproxies()
067aa17e 3772 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
dca08720
PH
3773 if 'http' in proxies and 'https' not in proxies:
3774 proxies['https'] = proxies['http']
91410c9b 3775 proxy_handler = PerRequestProxyHandler(proxies)
a0ddb8a2
PH
3776
3777 debuglevel = 1 if self.params.get('debug_printtraffic') else 0
be4a824d
PH
3778 https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel)
3779 ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel)
fca6dba8 3780 redirect_handler = YoutubeDLRedirectHandler()
f9934b96 3781 data_handler = urllib.request.DataHandler()
6240b0a2
JMF
3782
3783 # When passing our own FileHandler instance, build_opener won't add the
3784 # default FileHandler and allows us to disable the file protocol, which
3785 # can be used for malicious purposes (see
067aa17e 3786 # https://github.com/ytdl-org/youtube-dl/issues/8227)
ac668111 3787 file_handler = urllib.request.FileHandler()
6240b0a2
JMF
3788
3789 def file_open(*args, **kwargs):
ac668111 3790 raise urllib.error.URLError('file:// scheme is explicitly disabled in yt-dlp for security reasons')
6240b0a2
JMF
3791 file_handler.file_open = file_open
3792
ac668111 3793 opener = urllib.request.build_opener(
fca6dba8 3794 proxy_handler, https_handler, cookie_processor, ydlh, redirect_handler, data_handler, file_handler)
2461f79d 3795
dca08720
PH
3796 # Delete the default user-agent header, which would otherwise apply in
3797 # cases where our custom HTTP handler doesn't come into play
067aa17e 3798 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
dca08720
PH
3799 opener.addheaders = []
3800 self._opener = opener
62fec3b2
PH
3801
3802 def encode(self, s):
3803 if isinstance(s, bytes):
3804 return s # Already encoded
3805
3806 try:
3807 return s.encode(self.get_encoding())
3808 except UnicodeEncodeError as err:
3809 err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.'
3810 raise
3811
3812 def get_encoding(self):
3813 encoding = self.params.get('encoding')
3814 if encoding is None:
3815 encoding = preferredencoding()
3816 return encoding
ec82d85a 3817
e08a85d8 3818 def _write_info_json(self, label, ie_result, infofn, overwrite=None):
cb96c5be 3819 ''' Write infojson and returns True = written, 'exists' = Already exists, False = skip, None = error '''
e08a85d8 3820 if overwrite is None:
3821 overwrite = self.params.get('overwrites', True)
80c03fa9 3822 if not self.params.get('writeinfojson'):
3823 return False
3824 elif not infofn:
3825 self.write_debug(f'Skipping writing {label} infojson')
3826 return False
3827 elif not self._ensure_dir_exists(infofn):
3828 return None
e08a85d8 3829 elif not overwrite and os.path.exists(infofn):
80c03fa9 3830 self.to_screen(f'[info] {label.title()} metadata is already present')
cb96c5be 3831 return 'exists'
3832
3833 self.to_screen(f'[info] Writing {label} metadata as JSON to: {infofn}')
3834 try:
3835 write_json_file(self.sanitize_info(ie_result, self.params.get('clean_infojson', True)), infofn)
3836 return True
86e5f3ed 3837 except OSError:
cb96c5be 3838 self.report_error(f'Cannot write {label} metadata to JSON file {infofn}')
3839 return None
80c03fa9 3840
3841 def _write_description(self, label, ie_result, descfn):
3842 ''' Write description and returns True = written, False = skip, None = error '''
3843 if not self.params.get('writedescription'):
3844 return False
3845 elif not descfn:
3846 self.write_debug(f'Skipping writing {label} description')
3847 return False
3848 elif not self._ensure_dir_exists(descfn):
3849 return None
3850 elif not self.params.get('overwrites', True) and os.path.exists(descfn):
3851 self.to_screen(f'[info] {label.title()} description is already present')
3852 elif ie_result.get('description') is None:
3853 self.report_warning(f'There\'s no {label} description to write')
3854 return False
3855 else:
3856 try:
3857 self.to_screen(f'[info] Writing {label} description to: {descfn}')
86e5f3ed 3858 with open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
80c03fa9 3859 descfile.write(ie_result['description'])
86e5f3ed 3860 except OSError:
80c03fa9 3861 self.report_error(f'Cannot write {label} description file {descfn}')
3862 return None
3863 return True
3864
3865 def _write_subtitles(self, info_dict, filename):
3866 ''' Write subtitles to file and return list of (sub_filename, final_sub_filename); or None if error'''
3867 ret = []
3868 subtitles = info_dict.get('requested_subtitles')
3869 if not subtitles or not (self.params.get('writesubtitles') or self.params.get('writeautomaticsub')):
3870 # subtitles download errors are already managed as troubles in relevant IE
3871 # that way it will silently go on when used with unsupporting IE
3872 return ret
3873
3874 sub_filename_base = self.prepare_filename(info_dict, 'subtitle')
3875 if not sub_filename_base:
3876 self.to_screen('[info] Skipping writing video subtitles')
3877 return ret
3878 for sub_lang, sub_info in subtitles.items():
3879 sub_format = sub_info['ext']
3880 sub_filename = subtitles_filename(filename, sub_lang, sub_format, info_dict.get('ext'))
3881 sub_filename_final = subtitles_filename(sub_filename_base, sub_lang, sub_format, info_dict.get('ext'))
e04938ab 3882 existing_sub = self.existing_file((sub_filename_final, sub_filename))
3883 if existing_sub:
80c03fa9 3884 self.to_screen(f'[info] Video subtitle {sub_lang}.{sub_format} is already present')
e04938ab 3885 sub_info['filepath'] = existing_sub
3886 ret.append((existing_sub, sub_filename_final))
80c03fa9 3887 continue
3888
3889 self.to_screen(f'[info] Writing video subtitles to: {sub_filename}')
3890 if sub_info.get('data') is not None:
3891 try:
3892 # Use newline='' to prevent conversion of newline characters
3893 # See https://github.com/ytdl-org/youtube-dl/issues/10268
86e5f3ed 3894 with open(sub_filename, 'w', encoding='utf-8', newline='') as subfile:
80c03fa9 3895 subfile.write(sub_info['data'])
3896 sub_info['filepath'] = sub_filename
3897 ret.append((sub_filename, sub_filename_final))
3898 continue
86e5f3ed 3899 except OSError:
80c03fa9 3900 self.report_error(f'Cannot write video subtitles file {sub_filename}')
3901 return None
3902
3903 try:
3904 sub_copy = sub_info.copy()
3905 sub_copy.setdefault('http_headers', info_dict.get('http_headers'))
3906 self.dl(sub_filename, sub_copy, subtitle=True)
3907 sub_info['filepath'] = sub_filename
3908 ret.append((sub_filename, sub_filename_final))
6020e05d 3909 except (DownloadError, ExtractorError, IOError, OSError, ValueError) + network_exceptions as err:
c70c418d 3910 msg = f'Unable to download video subtitles for {sub_lang!r}: {err}'
6020e05d 3911 if self.params.get('ignoreerrors') is not True: # False or 'only_download'
c70c418d 3912 if not self.params.get('ignoreerrors'):
3913 self.report_error(msg)
3914 raise DownloadError(msg)
3915 self.report_warning(msg)
519804a9 3916 return ret
80c03fa9 3917
3918 def _write_thumbnails(self, label, info_dict, filename, thumb_filename_base=None):
3919 ''' Write thumbnails to file and return list of (thumb_filename, final_thumb_filename) '''
6c4fd172 3920 write_all = self.params.get('write_all_thumbnails', False)
80c03fa9 3921 thumbnails, ret = [], []
6c4fd172 3922 if write_all or self.params.get('writethumbnail', False):
0202b52a 3923 thumbnails = info_dict.get('thumbnails') or []
6c4fd172 3924 multiple = write_all and len(thumbnails) > 1
ec82d85a 3925
80c03fa9 3926 if thumb_filename_base is None:
3927 thumb_filename_base = filename
3928 if thumbnails and not thumb_filename_base:
3929 self.write_debug(f'Skipping writing {label} thumbnail')
3930 return ret
3931
dd0228ce 3932 for idx, t in list(enumerate(thumbnails))[::-1]:
80c03fa9 3933 thumb_ext = (f'{t["id"]}.' if multiple else '') + determine_ext(t['url'], 'jpg')
aa9369a2 3934 thumb_display_id = f'{label} thumbnail {t["id"]}'
80c03fa9 3935 thumb_filename = replace_extension(filename, thumb_ext, info_dict.get('ext'))
3936 thumb_filename_final = replace_extension(thumb_filename_base, thumb_ext, info_dict.get('ext'))
ec82d85a 3937
e04938ab 3938 existing_thumb = self.existing_file((thumb_filename_final, thumb_filename))
3939 if existing_thumb:
aa9369a2 3940 self.to_screen('[info] %s is already present' % (
3941 thumb_display_id if multiple else f'{label} thumbnail').capitalize())
e04938ab 3942 t['filepath'] = existing_thumb
3943 ret.append((existing_thumb, thumb_filename_final))
ec82d85a 3944 else:
80c03fa9 3945 self.to_screen(f'[info] Downloading {thumb_display_id} ...')
ec82d85a 3946 try:
297e9952 3947 uf = self.urlopen(sanitized_Request(t['url'], headers=t.get('http_headers', {})))
80c03fa9 3948 self.to_screen(f'[info] Writing {thumb_display_id} to: {thumb_filename}')
d3d89c32 3949 with open(encodeFilename(thumb_filename), 'wb') as thumbf:
ec82d85a 3950 shutil.copyfileobj(uf, thumbf)
80c03fa9 3951 ret.append((thumb_filename, thumb_filename_final))
885cc0b7 3952 t['filepath'] = thumb_filename
3158150c 3953 except network_exceptions as err:
dd0228ce 3954 thumbnails.pop(idx)
80c03fa9 3955 self.report_warning(f'Unable to download {thumb_display_id}: {err}')
6c4fd172 3956 if ret and not write_all:
3957 break
0202b52a 3958 return ret