]> jfr.im git - yt-dlp.git/blame - yt_dlp/YoutubeDL.py
[cleanup] Misc
[yt-dlp.git] / yt_dlp / YoutubeDL.py
CommitLineData
26e63931 1import collections
31bd3925 2import contextlib
9d2ecdbc 3import datetime
c1c9a79c 4import errno
31bd3925 5import fileinput
b5ae35ee 6import functools
8222d8de 7import io
b82f815f 8import itertools
8694c600 9import json
62fec3b2 10import locale
083c9df9 11import operator
8222d8de 12import os
f8271158 13import random
8222d8de
JMF
14import re
15import shutil
dca08720 16import subprocess
8222d8de 17import sys
21cd8fae 18import tempfile
8222d8de 19import time
67134eab 20import tokenize
8222d8de 21import traceback
524e2e4f 22import unicodedata
f9934b96 23import urllib.request
961ea474
S
24from string import ascii_letters
25
f8271158 26from .cache import Cache
14f25df2 27from .compat import compat_os_name, compat_shlex_quote
982ee69a 28from .cookies import load_cookies
f8271158 29from .downloader import FFmpegFD, get_suitable_downloader, shorten_protocol_name
30from .downloader.rtmp import rtmpdump_version
f8271158 31from .extractor import gen_extractor_classes, get_info_extractor
fe7866d0 32from .extractor.common import UnsupportedURLIE
f8271158 33from .extractor.openload import PhantomJSwrapper
34from .minicurses import format_text
35from .postprocessor import _PLUGIN_CLASSES as plugin_postprocessors
36from .postprocessor import (
37 EmbedThumbnailPP,
38 FFmpegFixupDuplicateMoovPP,
39 FFmpegFixupDurationPP,
40 FFmpegFixupM3u8PP,
41 FFmpegFixupM4aPP,
42 FFmpegFixupStretchedPP,
43 FFmpegFixupTimestampPP,
44 FFmpegMergerPP,
45 FFmpegPostProcessor,
ca9def71 46 FFmpegVideoConvertorPP,
f8271158 47 MoveFilesAfterDownloadPP,
48 get_postprocessor,
49)
ca9def71 50from .postprocessor.ffmpeg import resolve_mapping as resolve_recode_mapping
b5e7a2e6 51from .update import REPOSITORY, current_git_head, detect_variant
8c25f81b 52from .utils import (
f8271158 53 DEFAULT_OUTTMPL,
7b2c3f47 54 IDENTITY,
f8271158 55 LINK_TEMPLATES,
8dc59305 56 MEDIA_EXTENSIONS,
f8271158 57 NO_DEFAULT,
1d485a1a 58 NUMBER_RE,
f8271158 59 OUTTMPL_TYPES,
60 POSTPROCESS_WHEN,
61 STR_FORMAT_RE_TMPL,
62 STR_FORMAT_TYPES,
63 ContentTooShortError,
64 DateRange,
65 DownloadCancelled,
66 DownloadError,
67 EntryNotInPlaylist,
68 ExistingVideoReached,
69 ExtractorError,
70 GeoRestrictedError,
71 HEADRequest,
f8271158 72 ISO3166Utils,
73 LazyList,
74 MaxDownloadsReached,
19a03940 75 Namespace,
f8271158 76 PagedList,
77 PerRequestProxyHandler,
7e88d7d7 78 PlaylistEntries,
f8271158 79 Popen,
80 PostProcessingError,
81 ReExtractInfo,
82 RejectedVideoReached,
83 SameFileError,
84 UnavailableVideoError,
693f0600 85 UserNotLive,
f8271158 86 YoutubeDLCookieProcessor,
87 YoutubeDLHandler,
88 YoutubeDLRedirectHandler,
eedb7ba5
S
89 age_restricted,
90 args_to_str,
cb794ee0 91 bug_reports_message,
ce02ed60 92 date_from_str,
da4db748 93 deprecation_warning,
ce02ed60 94 determine_ext,
b5559424 95 determine_protocol,
c0384f22 96 encode_compat_str,
ce02ed60 97 encodeFilename,
a06916d9 98 error_to_compat_str,
47cdc68e 99 escapeHTML,
590bc6f6 100 expand_path,
90137ca4 101 filter_dict,
e29663c6 102 float_or_none,
02dbf93f 103 format_bytes,
e0fd9573 104 format_decimal_suffix,
f8271158 105 format_field,
525ef922 106 formatSeconds,
fc61aff4 107 get_compatible_ext,
0bb322b9 108 get_domain,
c9969434 109 int_or_none,
732044af 110 iri_to_uri,
941e881e 111 is_path_like,
34921b43 112 join_nonempty,
ce02ed60 113 locked_file,
0647d925 114 make_archive_id,
0202b52a 115 make_dir,
dca08720 116 make_HTTPS_handler,
8b7539d2 117 merge_headers,
3158150c 118 network_exceptions,
ec11a9f4 119 number_of_digits,
cd6fc19e 120 orderedSet,
5314b521 121 orderedSet_from_options,
083c9df9 122 parse_filesize,
ce02ed60 123 preferredencoding,
eedb7ba5 124 prepend_extension,
51fb4995 125 register_socks_protocols,
3efb96a6 126 remove_terminal_sequences,
cfb56d1a 127 render_table,
eedb7ba5 128 replace_extension,
ce02ed60 129 sanitize_filename,
1bb5c511 130 sanitize_path,
dcf77cf1 131 sanitize_url,
67dda517 132 sanitized_Request,
e5660ee6 133 std_headers,
1211bb6d 134 str_or_none,
e29663c6 135 strftime_or_none,
ce02ed60 136 subtitles_filename,
819e0531 137 supports_terminal_sequences,
b1f94422 138 system_identifier,
f2ebc5c7 139 timetuple_from_msec,
732044af 140 to_high_limit_path,
324ad820 141 traverse_obj,
fc61aff4 142 try_call,
6033d980 143 try_get,
29eb5174 144 url_basename,
7d1eb38a 145 variadic,
58b1f00d 146 version_tuple,
53973b4d 147 windows_enable_vt_mode,
ce02ed60
PH
148 write_json_file,
149 write_string,
4f026faf 150)
70b23409 151from .version import RELEASE_GIT_HEAD, VARIANT, __version__
8222d8de 152
e9c0cdd3
YCH
153if compat_os_name == 'nt':
154 import ctypes
155
2459b6e1 156
86e5f3ed 157class YoutubeDL:
8222d8de
JMF
158 """YoutubeDL class.
159
160 YoutubeDL objects are the ones responsible of downloading the
161 actual video file and writing it to disk if the user has requested
162 it, among some other tasks. In most cases there should be one per
163 program. As, given a video URL, the downloader doesn't know how to
164 extract all the needed information, task that InfoExtractors do, it
165 has to pass the URL to one of them.
166
167 For this, YoutubeDL objects have a method that allows
168 InfoExtractors to be registered in a given order. When it is passed
169 a URL, the YoutubeDL object handles it to the first InfoExtractor it
170 finds that reports being able to handle it. The InfoExtractor extracts
171 all the information about the video or videos the URL refers to, and
172 YoutubeDL process the extracted information, possibly using a File
173 Downloader to download the video.
174
175 YoutubeDL objects accept a lot of parameters. In order not to saturate
176 the object constructor with arguments, it receives a dictionary of
177 options instead. These options are available through the params
178 attribute for the InfoExtractors to use. The YoutubeDL also
179 registers itself as the downloader in charge for the InfoExtractors
180 that are added to it, so this is a "mutual registration".
181
182 Available options:
183
184 username: Username for authentication purposes.
185 password: Password for authentication purposes.
180940e0 186 videopassword: Password for accessing a video.
1da50aa3
S
187 ap_mso: Adobe Pass multiple-system operator identifier.
188 ap_username: Multiple-system operator account username.
189 ap_password: Multiple-system operator account password.
8222d8de
JMF
190 usenetrc: Use netrc for authentication instead.
191 verbose: Print additional info to stdout.
192 quiet: Do not print messages to stdout.
ad8915b7 193 no_warnings: Do not print out anything for warnings.
bb66c247 194 forceprint: A dict with keys WHEN mapped to a list of templates to
195 print to stdout. The allowed keys are video or any of the
196 items in utils.POSTPROCESS_WHEN.
ca30f449 197 For compatibility, a single list is also accepted
bb66c247 198 print_to_file: A dict with keys WHEN (same as forceprint) mapped to
199 a list of tuples with (template, filename)
8694c600 200 forcejson: Force printing info_dict as JSON.
63e0be34
PH
201 dump_single_json: Force printing the info_dict of the whole playlist
202 (or video) as a single JSON line.
c25228e5 203 force_write_download_archive: Force writing download archive regardless
204 of 'skip_download' or 'simulate'.
b7b04c78 205 simulate: Do not download the video files. If unset (or None),
206 simulate only if listsubtitles, listformats or list_thumbnails is used
eb8a4433 207 format: Video format code. see "FORMAT SELECTION" for more details.
093a1710 208 You can also pass a function. The function takes 'ctx' as
209 argument and returns the formats to download.
210 See "build_format_selector" for an implementation
63ad4d43 211 allow_unplayable_formats: Allow unplayable formats to be extracted and downloaded.
b7da73eb 212 ignore_no_formats_error: Ignore "No video formats" error. Usefull for
213 extracting metadata even if the video is not actually
214 available for download (experimental)
0930b11f 215 format_sort: A list of fields by which to sort the video formats.
216 See "Sorting Formats" for more details.
c25228e5 217 format_sort_force: Force the given format_sort. see "Sorting Formats"
218 for more details.
08d30158 219 prefer_free_formats: Whether to prefer video formats with free containers
220 over non-free ones of same quality.
c25228e5 221 allow_multiple_video_streams: Allow multiple video streams to be merged
222 into a single file
223 allow_multiple_audio_streams: Allow multiple audio streams to be merged
224 into a single file
0ba692ac 225 check_formats Whether to test if the formats are downloadable.
9f1a1c36 226 Can be True (check all), False (check none),
227 'selected' (check selected formats),
0ba692ac 228 or None (check only if requested by extractor)
4524baf0 229 paths: Dictionary of output paths. The allowed keys are 'home'
230 'temp' and the keys of OUTTMPL_TYPES (in utils.py)
de6000d9 231 outtmpl: Dictionary of templates for output names. Allowed keys
4524baf0 232 are 'default' and the keys of OUTTMPL_TYPES (in utils.py).
34488702 233 For compatibility with youtube-dl, a single string can also be used
a820dc72
RA
234 outtmpl_na_placeholder: Placeholder for unavailable meta fields.
235 restrictfilenames: Do not allow "&" and spaces in file names
236 trim_file_name: Limit length of filename (extension excluded)
4524baf0 237 windowsfilenames: Force the filenames to be windows compatible
b1940459 238 ignoreerrors: Do not stop on download/postprocessing errors.
239 Can be 'only_download' to ignore only download errors.
240 Default is 'only_download' for CLI, but False for API
26e2805c 241 skip_playlist_after_errors: Number of allowed failures until the rest of
242 the playlist is skipped
fe7866d0 243 allowed_extractors: List of regexes to match against extractor names that are allowed
0c3d0f51 244 overwrites: Overwrite all video and metadata files if True,
245 overwrite only non-video files if None
246 and don't overwrite any file if False
34488702 247 For compatibility with youtube-dl,
248 "nooverwrites" may also be used instead
c14e88f0 249 playlist_items: Specific indices of playlist to download.
75822ca7 250 playlistrandom: Download playlist items in random order.
7e9a6125 251 lazy_playlist: Process playlist entries as they are received.
8222d8de
JMF
252 matchtitle: Download only matching titles.
253 rejecttitle: Reject downloads for matching titles.
8bf9319e 254 logger: Log messages to a logging.Logger instance.
17ffed18 255 logtostderr: Print everything to stderr instead of stdout.
256 consoletitle: Display progress in console window's titlebar.
8222d8de
JMF
257 writedescription: Write the video description to a .description file
258 writeinfojson: Write the video description to a .info.json file
75d43ca0 259 clean_infojson: Remove private fields from the infojson
34488702 260 getcomments: Extract video comments. This will not be written to disk
06167fbb 261 unless writeinfojson is also given
1fb07d10 262 writeannotations: Write the video annotations to a .annotations.xml file
8222d8de 263 writethumbnail: Write the thumbnail image to a file
c25228e5 264 allow_playlist_files: Whether to write playlists' description, infojson etc
265 also to disk when using the 'write*' options
ec82d85a 266 write_all_thumbnails: Write all thumbnail formats to files
732044af 267 writelink: Write an internet shortcut file, depending on the
268 current platform (.url/.webloc/.desktop)
269 writeurllink: Write a Windows internet shortcut file (.url)
270 writewebloclink: Write a macOS internet shortcut file (.webloc)
271 writedesktoplink: Write a Linux internet shortcut file (.desktop)
8222d8de 272 writesubtitles: Write the video subtitles to a file
741dd8ea 273 writeautomaticsub: Write the automatically generated subtitles to a file
8222d8de 274 listsubtitles: Lists all available subtitles for the video
a504ced0 275 subtitlesformat: The format code for subtitles
c32b0aab 276 subtitleslangs: List of languages of the subtitles to download (can be regex).
277 The list may contain "all" to refer to all the available
278 subtitles. The language can be prefixed with a "-" to
62b58c09 279 exclude it from the requested languages, e.g. ['all', '-live_chat']
8222d8de
JMF
280 keepvideo: Keep the video file after post-processing
281 daterange: A DateRange object, download only if the upload_date is in the range.
282 skip_download: Skip the actual download of the video file
c35f9e72 283 cachedir: Location of the cache files in the filesystem.
a0e07d31 284 False to disable filesystem cache.
47192f92 285 noplaylist: Download single video instead of a playlist if in doubt.
8dbe9899
PH
286 age_limit: An integer representing the user's age in years.
287 Unsuitable videos for the given age are skipped.
5fe18bdb
PH
288 min_views: An integer representing the minimum view count the video
289 must have in order to not be skipped.
290 Videos without view count information are always
291 downloaded. None for no limit.
292 max_views: An integer representing the maximum view count.
293 Videos that are more popular than that are not
294 downloaded.
295 Videos without view count information are always
296 downloaded. None for no limit.
ae103564 297 download_archive: A set, or the name of a file where all downloads are recorded.
298 Videos already present in the file are not downloaded again.
8a51f564 299 break_on_existing: Stop the download process after attempting to download a
300 file that is in the archive.
301 break_on_reject: Stop the download process when encountering a video that
302 has been filtered out.
b222c271 303 break_per_url: Whether break_on_reject and break_on_existing
304 should act on each input URL as opposed to for the entire queue
d76fa1f3 305 cookiefile: File name or text stream from where cookies should be read and dumped to
f59f5ef8 306 cookiesfrombrowser: A tuple containing the name of the browser, the profile
9bd13fe5 307 name/path from where cookies are loaded, the name of the keyring,
308 and the container name, e.g. ('chrome', ) or
309 ('vivaldi', 'default', 'BASICTEXT') or ('firefox', 'default', None, 'Meta')
f81c62a6 310 legacyserverconnect: Explicitly allow HTTPS connection to servers that do not
311 support RFC 5746 secure renegotiation
f59f5ef8 312 nocheckcertificate: Do not verify SSL certificates
bb58c9ed 313 client_certificate: Path to client certificate file in PEM format. May include the private key
314 client_certificate_key: Path to private key file for client certificate
315 client_certificate_password: Password for client certificate private key, if encrypted.
316 If not provided and the key is encrypted, yt-dlp will ask interactively
7e8c0af0 317 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
c6e07cf1 318 (Only supported by some extractors)
8b7539d2 319 http_headers: A dictionary of custom headers to be used for all requests
a1ee09e8 320 proxy: URL of the proxy server to use
38cce791 321 geo_verification_proxy: URL of the proxy to use for IP address verification
504f20dd 322 on geo-restricted sites.
e344693b 323 socket_timeout: Time to wait for unresponsive hosts, in seconds
0783b09b
PH
324 bidi_workaround: Work around buggy terminals without bidirectional text
325 support, using fridibi
a0ddb8a2 326 debug_printtraffic:Print out sent and received HTTP traffic
04b4d394
PH
327 default_search: Prepend this string if an input url is not valid.
328 'auto' for elaborate guessing
62fec3b2 329 encoding: Use this encoding instead of the system-specified.
134c913c 330 extract_flat: Whether to resolve and process url_results further
331 * False: Always process (default)
332 * True: Never process
333 * 'in_playlist': Do not process inside playlist/multi_video
334 * 'discard': Always process, but don't return the result
335 from inside playlist/multi_video
336 * 'discard_in_playlist': Same as "discard", but only for
337 playlists (not multi_video)
f2ebc5c7 338 wait_for_video: If given, wait for scheduled streams to become available.
339 The value should be a tuple containing the range
340 (min_secs, max_secs) to wait between retries
4f026faf 341 postprocessors: A list of dictionaries, each with an entry
71b640cc 342 * key: The name of the postprocessor. See
7a5c1cfe 343 yt_dlp/postprocessor/__init__.py for a list.
bb66c247 344 * when: When to run the postprocessor. Allowed values are
345 the entries of utils.POSTPROCESS_WHEN
56d868db 346 Assumed to be 'post_process' if not given
71b640cc
PH
347 progress_hooks: A list of functions that get called on download
348 progress, with a dictionary with the entries
5cda4eda 349 * status: One of "downloading", "error", or "finished".
ee69b99a 350 Check this first and ignore unknown values.
3ba7740d 351 * info_dict: The extracted info_dict
71b640cc 352
5cda4eda 353 If status is one of "downloading", or "finished", the
ee69b99a
PH
354 following properties may also be present:
355 * filename: The final filename (always present)
5cda4eda 356 * tmpfilename: The filename we're currently writing to
71b640cc
PH
357 * downloaded_bytes: Bytes on disk
358 * total_bytes: Size of the whole file, None if unknown
5cda4eda
PH
359 * total_bytes_estimate: Guess of the eventual file size,
360 None if unavailable.
361 * elapsed: The number of seconds since download started.
71b640cc
PH
362 * eta: The estimated time in seconds, None if unknown
363 * speed: The download speed in bytes/second, None if
364 unknown
5cda4eda
PH
365 * fragment_index: The counter of the currently
366 downloaded video fragment.
367 * fragment_count: The number of fragments (= individual
368 files that will be merged)
71b640cc
PH
369
370 Progress hooks are guaranteed to be called at least once
371 (with status "finished") if the download is successful.
819e0531 372 postprocessor_hooks: A list of functions that get called on postprocessing
373 progress, with a dictionary with the entries
374 * status: One of "started", "processing", or "finished".
375 Check this first and ignore unknown values.
376 * postprocessor: Name of the postprocessor
377 * info_dict: The extracted info_dict
378
379 Progress hooks are guaranteed to be called at least twice
380 (with status "started" and "finished") if the processing is successful.
fc61aff4 381 merge_output_format: "/" separated list of extensions to use when merging formats.
6b591b29 382 final_ext: Expected final extension; used to detect when the file was
59a7a13e 383 already downloaded and converted
6271f1ca
PH
384 fixup: Automatically correct known faults of the file.
385 One of:
386 - "never": do nothing
387 - "warn": only emit a warning
388 - "detect_or_warn": check whether we can do anything
62cd676c 389 about it, warn otherwise (default)
504f20dd 390 source_address: Client-side IP address to bind to.
1cf376f5 391 sleep_interval_requests: Number of seconds to sleep between requests
392 during extraction
7aa589a5
S
393 sleep_interval: Number of seconds to sleep before each download when
394 used alone or a lower bound of a range for randomized
395 sleep before each download (minimum possible number
396 of seconds to sleep) when used along with
397 max_sleep_interval.
398 max_sleep_interval:Upper bound of a range for randomized sleep before each
399 download (maximum possible number of seconds to sleep).
400 Must only be used along with sleep_interval.
401 Actual sleep time will be a random float from range
402 [sleep_interval; max_sleep_interval].
1cf376f5 403 sleep_interval_subtitles: Number of seconds to sleep before each subtitle download
cfb56d1a
PH
404 listformats: Print an overview of available video formats and exit.
405 list_thumbnails: Print a table of all thumbnails and exit.
0a41f331 406 match_filter: A function that gets called for every video with the signature
407 (info_dict, *, incomplete: bool) -> Optional[str]
408 For backward compatibility with youtube-dl, the signature
409 (info_dict) -> Optional[str] is also allowed.
410 - If it returns a message, the video is ignored.
411 - If it returns None, the video is downloaded.
412 - If it returns utils.NO_DEFAULT, the user is interactively
413 asked whether to download the video.
347de493 414 match_filter_func in utils.py is one example for this.
7e5db8c9 415 no_color: Do not emit color codes in output.
0a840f58 416 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
504f20dd 417 HTTP header
0a840f58 418 geo_bypass_country:
773f291d
S
419 Two-letter ISO 3166-2 country code that will be used for
420 explicit geographic restriction bypassing via faking
504f20dd 421 X-Forwarded-For HTTP header
5f95927a
S
422 geo_bypass_ip_block:
423 IP range in CIDR notation that will be used similarly to
504f20dd 424 geo_bypass_country
52a8a1e1 425 external_downloader: A dictionary of protocol keys and the executable of the
426 external downloader to use for it. The allowed protocols
427 are default|http|ftp|m3u8|dash|rtsp|rtmp|mms.
428 Set the value to 'native' to use the native downloader
53ed7066 429 compat_opts: Compatibility options. See "Differences in default behavior".
3acf6d38 430 The following options do not work when used through the API:
b5ae35ee 431 filename, abort-on-error, multistreams, no-live-chat, format-sort
dac5df5a 432 no-clean-infojson, no-playlist-metafiles, no-keep-subs, no-attach-info-json.
e4f02757 433 Refer __init__.py for their implementation
819e0531 434 progress_template: Dictionary of templates for progress outputs.
435 Allowed keys are 'download', 'postprocess',
436 'download-title' (console title) and 'postprocess-title'.
437 The template is mapped on a dictionary with keys 'progress' and 'info'
23326151 438 retry_sleep_functions: Dictionary of functions that takes the number of attempts
439 as argument and returns the time to sleep in seconds.
440 Allowed keys are 'http', 'fragment', 'file_access'
0f446365
SW
441 download_ranges: A callback function that gets called for every video with
442 the signature (info_dict, ydl) -> Iterable[Section].
443 Only the returned sections will be downloaded.
444 Each Section is a dict with the following keys:
5ec1b6b7 445 * start_time: Start time of the section in seconds
446 * end_time: End time of the section in seconds
447 * title: Section title (Optional)
448 * index: Section number (Optional)
0f446365 449 force_keyframes_at_cuts: Re-encode the video when downloading ranges to get precise cuts
a7dc6a89 450 noprogress: Do not print the progress bar
a831c2ea 451 live_from_start: Whether to download livestreams videos from the start
fe7e0c98 452
8222d8de 453 The following parameters are not used by YoutubeDL itself, they are used by
7a5c1cfe 454 the downloader (see yt_dlp/downloader/common.py):
51d9739f 455 nopart, updatetime, buffersize, ratelimit, throttledratelimit, min_filesize,
205a0654 456 max_filesize, test, noresizebuffer, retries, file_access_retries, fragment_retries,
a7dc6a89 457 continuedl, xattr_set_filesize, hls_use_mpegts, http_chunk_size,
59a7a13e 458 external_downloader_args, concurrent_fragment_downloads.
76b1bd67
JMF
459
460 The following options are used by the post processors:
c0b7d117
S
461 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
462 to the binary or its containing directory.
43820c03 463 postprocessor_args: A dictionary of postprocessor/executable keys (in lower case)
34488702 464 and a list of additional command-line arguments for the
465 postprocessor/executable. The dict can also have "PP+EXE" keys
466 which are used when the given exe is used by the given PP.
467 Use 'default' as the name for arguments to passed to all PP
468 For compatibility with youtube-dl, a single list of args
469 can also be used
e409895f 470
471 The following options are used by the extractors:
62bff2c1 472 extractor_retries: Number of times to retry for known errors
473 dynamic_mpd: Whether to process dynamic DASH manifests (default: True)
e409895f 474 hls_split_discontinuity: Split HLS playlists to different formats at
62bff2c1 475 discontinuities such as ad breaks (default: False)
5d3a0e79 476 extractor_args: A dictionary of arguments to be passed to the extractors.
477 See "EXTRACTOR ARGUMENTS" for details.
62b58c09 478 E.g. {'youtube': {'skip': ['dash', 'hls']}}
88f23a18 479 mark_watched: Mark videos watched (even with --simulate). Only for YouTube
1890fc63 480
481 The following options are deprecated and may be removed in the future:
482
fe7866d0 483 force_generic_extractor: Force downloader to use the generic extractor
484 - Use allowed_extractors = ['generic', 'default']
7e9a6125 485 playliststart: - Use playlist_items
486 Playlist item to start at.
487 playlistend: - Use playlist_items
488 Playlist item to end at.
489 playlistreverse: - Use playlist_items
490 Download playlist items in reverse order.
1890fc63 491 forceurl: - Use forceprint
492 Force printing final URL.
493 forcetitle: - Use forceprint
494 Force printing title.
495 forceid: - Use forceprint
496 Force printing ID.
497 forcethumbnail: - Use forceprint
498 Force printing thumbnail URL.
499 forcedescription: - Use forceprint
500 Force printing description.
501 forcefilename: - Use forceprint
502 Force printing final filename.
503 forceduration: - Use forceprint
504 Force printing duration.
505 allsubtitles: - Use subtitleslangs = ['all']
506 Downloads all the subtitles of the video
507 (requires writesubtitles or writeautomaticsub)
508 include_ads: - Doesn't work
509 Download ads as well
510 call_home: - Not implemented
511 Boolean, true iff we are allowed to contact the
512 yt-dlp servers for debugging.
513 post_hooks: - Register a custom postprocessor
514 A list of functions that get called as the final step
515 for each video file, after all postprocessors have been
516 called. The filename will be passed as the only argument.
517 hls_prefer_native: - Use external_downloader = {'m3u8': 'native'} or {'m3u8': 'ffmpeg'}.
518 Use the native HLS downloader instead of ffmpeg/avconv
519 if True, otherwise use ffmpeg/avconv if False, otherwise
520 use downloader suggested by extractor if None.
521 prefer_ffmpeg: - avconv support is deprecated
522 If False, use avconv instead of ffmpeg if both are available,
523 otherwise prefer ffmpeg.
524 youtube_include_dash_manifest: - Use extractor_args
5d3a0e79 525 If True (default), DASH manifests and related
62bff2c1 526 data will be downloaded and processed by extractor.
527 You can reduce network I/O by disabling it if you don't
528 care about DASH. (only for youtube)
1890fc63 529 youtube_include_hls_manifest: - Use extractor_args
5d3a0e79 530 If True (default), HLS manifests and related
62bff2c1 531 data will be downloaded and processed by extractor.
532 You can reduce network I/O by disabling it if you don't
533 care about HLS. (only for youtube)
8222d8de
JMF
534 """
535
86e5f3ed 536 _NUMERIC_FIELDS = {
b8ed0f15 537 'width', 'height', 'asr', 'audio_channels', 'fps',
538 'tbr', 'abr', 'vbr', 'filesize', 'filesize_approx',
e6f21b3d 539 'timestamp', 'release_timestamp',
c9969434
S
540 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
541 'average_rating', 'comment_count', 'age_limit',
542 'start_time', 'end_time',
543 'chapter_number', 'season_number', 'episode_number',
544 'track_number', 'disc_number', 'release_year',
86e5f3ed 545 }
c9969434 546
6db9c4d5 547 _format_fields = {
548 # NB: Keep in sync with the docstring of extractor/common.py
a44ca5a4 549 'url', 'manifest_url', 'manifest_stream_number', 'ext', 'format', 'format_id', 'format_note',
b8ed0f15 550 'width', 'height', 'resolution', 'dynamic_range', 'tbr', 'abr', 'acodec', 'asr', 'audio_channels',
d5d1df8a 551 'vbr', 'fps', 'vcodec', 'container', 'filesize', 'filesize_approx', 'rows', 'columns',
6db9c4d5 552 'player_url', 'protocol', 'fragment_base_url', 'fragments', 'is_from_start',
553 'preference', 'language', 'language_preference', 'quality', 'source_preference',
554 'http_headers', 'stretched_ratio', 'no_resume', 'has_drm', 'downloader_options',
555 'page_url', 'app', 'play_path', 'tc_url', 'flash_version', 'rtmp_live', 'rtmp_conn', 'rtmp_protocol', 'rtmp_real_time'
556 }
48ee10ee 557 _format_selection_exts = {
8dc59305 558 'audio': set(MEDIA_EXTENSIONS.common_audio),
559 'video': set(MEDIA_EXTENSIONS.common_video + ('3gp', )),
560 'storyboards': set(MEDIA_EXTENSIONS.storyboards),
48ee10ee 561 }
562
3511266b 563 def __init__(self, params=None, auto_init=True):
883d4b1e 564 """Create a FileDownloader object with the given options.
565 @param auto_init Whether to load the default extractors and print header (if verbose).
49a57e70 566 Set to 'no_verbose_header' to not print the header
883d4b1e 567 """
e9f9a10f
JMF
568 if params is None:
569 params = {}
592b7485 570 self.params = params
8b7491c8 571 self._ies = {}
56c73665 572 self._ies_instances = {}
1e43a6f7 573 self._pps = {k: [] for k in POSTPROCESS_WHEN}
b35496d8 574 self._printed_messages = set()
1cf376f5 575 self._first_webpage_request = True
ab8e5e51 576 self._post_hooks = []
933605d7 577 self._progress_hooks = []
819e0531 578 self._postprocessor_hooks = []
8222d8de
JMF
579 self._download_retcode = 0
580 self._num_downloads = 0
9c906919 581 self._num_videos = 0
592b7485 582 self._playlist_level = 0
583 self._playlist_urls = set()
a0e07d31 584 self.cache = Cache(self)
34308b30 585
819e0531 586 windows_enable_vt_mode()
591bb9d3 587 stdout = sys.stderr if self.params.get('logtostderr') else sys.stdout
588 self._out_files = Namespace(
589 out=stdout,
590 error=sys.stderr,
591 screen=sys.stderr if self.params.get('quiet') else stdout,
592 console=None if compat_os_name == 'nt' else next(
cf4f42cb 593 filter(supports_terminal_sequences, (sys.stderr, sys.stdout)), None)
591bb9d3 594 )
595 self._allow_colors = Namespace(**{
596 type_: not self.params.get('no_color') and supports_terminal_sequences(stream)
64fa820c 597 for type_, stream in self._out_files.items_ if type_ != 'console'
591bb9d3 598 })
819e0531 599
6929b41a 600 # The code is left like this to be reused for future deprecations
601 MIN_SUPPORTED, MIN_RECOMMENDED = (3, 7), (3, 7)
eff42759 602 current_version = sys.version_info[:2]
603 if current_version < MIN_RECOMMENDED:
9d339c41 604 msg = ('Support for Python version %d.%d has been deprecated. '
24093d52 605 'See https://github.com/yt-dlp/yt-dlp/issues/3764 for more details.'
c6e07cf1 606 '\n You will no longer receive updates on this version')
eff42759 607 if current_version < MIN_SUPPORTED:
608 msg = 'Python version %d.%d is no longer supported'
609 self.deprecation_warning(
610 f'{msg}! Please update to Python %d.%d or above' % (*current_version, *MIN_RECOMMENDED))
a61f4b28 611
88acdbc2 612 if self.params.get('allow_unplayable_formats'):
613 self.report_warning(
ec11a9f4 614 f'You have asked for {self._format_err("UNPLAYABLE", self.Styles.EMPHASIS)} formats to be listed/downloaded. '
819e0531 615 'This is a developer option intended for debugging. \n'
616 ' If you experience any issues while using this option, '
ec11a9f4 617 f'{self._format_err("DO NOT", self.Styles.ERROR)} open a bug report')
88acdbc2 618
497074f0 619 if self.params.get('bidi_workaround', False):
620 try:
621 import pty
622 master, slave = pty.openpty()
623 width = shutil.get_terminal_size().columns
624 width_args = [] if width is None else ['-w', str(width)]
625 sp_kwargs = {'stdin': subprocess.PIPE, 'stdout': slave, 'stderr': self._out_files.error}
626 try:
627 self._output_process = Popen(['bidiv'] + width_args, **sp_kwargs)
628 except OSError:
629 self._output_process = Popen(['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
630 self._output_channel = os.fdopen(master, 'rb')
631 except OSError as ose:
632 if ose.errno == errno.ENOENT:
633 self.report_warning(
634 'Could not find fribidi executable, ignoring --bidi-workaround. '
635 'Make sure that fribidi is an executable file in one of the directories in your $PATH.')
636 else:
637 raise
638
639 self.params['compat_opts'] = set(self.params.get('compat_opts', ()))
640 if auto_init and auto_init != 'no_verbose_header':
641 self.print_debug_header()
642
be5df5ee
S
643 def check_deprecated(param, option, suggestion):
644 if self.params.get(param) is not None:
86e5f3ed 645 self.report_warning(f'{option} is deprecated. Use {suggestion} instead')
be5df5ee
S
646 return True
647 return False
648
649 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
38cce791
YCH
650 if self.params.get('geo_verification_proxy') is None:
651 self.params['geo_verification_proxy'] = self.params['cn_verification_proxy']
652
0d1bb027 653 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
654 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
53ed7066 655 check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"')
0d1bb027 656
49a57e70 657 for msg in self.params.get('_warnings', []):
0d1bb027 658 self.report_warning(msg)
ee8dd27a 659 for msg in self.params.get('_deprecation_warnings', []):
da4db748 660 self.deprecated_feature(msg)
0d1bb027 661
8a82af35 662 if 'list-formats' in self.params['compat_opts']:
ec11a9f4 663 self.params['listformats_table'] = False
664
b5ae35ee 665 if 'overwrites' not in self.params and self.params.get('nooverwrites') is not None:
b868936c 666 # nooverwrites was unnecessarily changed to overwrites
667 # in 0c3d0f51778b153f65c21906031c2e091fcfb641
668 # This ensures compatibility with both keys
669 self.params['overwrites'] = not self.params['nooverwrites']
b5ae35ee 670 elif self.params.get('overwrites') is None:
671 self.params.pop('overwrites', None)
b868936c 672 else:
673 self.params['nooverwrites'] = not self.params['overwrites']
b9d973be 674
e4221b70 675 if self.params.get('simulate') is None and any((
676 self.params.get('list_thumbnails'),
677 self.params.get('listformats'),
678 self.params.get('listsubtitles'),
679 )):
680 self.params['simulate'] = 'list_only'
681
455a15e2 682 self.params.setdefault('forceprint', {})
683 self.params.setdefault('print_to_file', {})
bb66c247 684
685 # Compatibility with older syntax
ca30f449 686 if not isinstance(params['forceprint'], dict):
455a15e2 687 self.params['forceprint'] = {'video': params['forceprint']}
ca30f449 688
97ec5bc5 689 if auto_init:
97ec5bc5 690 self.add_default_info_extractors()
691
3089bc74
S
692 if (sys.platform != 'win32'
693 and sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
455a15e2 694 and not self.params.get('restrictfilenames', False)):
e9137224 695 # Unicode filesystem API will throw errors (#1474, #13027)
34308b30 696 self.report_warning(
6febd1c1 697 'Assuming --restrict-filenames since file system encoding '
1b725173 698 'cannot encode all characters. '
6febd1c1 699 'Set the LC_ALL environment variable to fix this.')
4a98cdbf 700 self.params['restrictfilenames'] = True
34308b30 701
bf1824b3 702 self._parse_outtmpl()
486dd09e 703
187986a8 704 # Creating format selector here allows us to catch syntax errors before the extraction
705 self.format_selector = (
fa9f30b8 706 self.params.get('format') if self.params.get('format') in (None, '-')
093a1710 707 else self.params['format'] if callable(self.params['format'])
187986a8 708 else self.build_format_selector(self.params['format']))
709
8b7539d2 710 # Set http_headers defaults according to std_headers
711 self.params['http_headers'] = merge_headers(std_headers, self.params.get('http_headers', {}))
712
013b50b7 713 hooks = {
714 'post_hooks': self.add_post_hook,
715 'progress_hooks': self.add_progress_hook,
716 'postprocessor_hooks': self.add_postprocessor_hook,
717 }
718 for opt, fn in hooks.items():
719 for ph in self.params.get(opt, []):
720 fn(ph)
71b640cc 721
5bfc8bee 722 for pp_def_raw in self.params.get('postprocessors', []):
723 pp_def = dict(pp_def_raw)
724 when = pp_def.pop('when', 'post_process')
725 self.add_post_processor(
f9934b96 726 get_postprocessor(pp_def.pop('key'))(self, **pp_def),
5bfc8bee 727 when=when)
728
97ec5bc5 729 self._setup_opener()
51fb4995
YCH
730 register_socks_protocols()
731
ed39cac5 732 def preload_download_archive(fn):
733 """Preload the archive, if any is specified"""
ae103564 734 archive = set()
ed39cac5 735 if fn is None:
ae103564 736 return archive
941e881e 737 elif not is_path_like(fn):
ae103564 738 return fn
739
49a57e70 740 self.write_debug(f'Loading archive file {fn!r}')
ed39cac5 741 try:
742 with locked_file(fn, 'r', encoding='utf-8') as archive_file:
743 for line in archive_file:
ae103564 744 archive.add(line.strip())
86e5f3ed 745 except OSError as ioe:
ed39cac5 746 if ioe.errno != errno.ENOENT:
747 raise
ae103564 748 return archive
ed39cac5 749
ae103564 750 self.archive = preload_download_archive(self.params.get('download_archive'))
ed39cac5 751
7d4111ed
PH
752 def warn_if_short_id(self, argv):
753 # short YouTube ID starting with dash?
754 idxs = [
755 i for i, a in enumerate(argv)
756 if re.match(r'^-[0-9A-Za-z_-]{10}$', a)]
757 if idxs:
758 correct_argv = (
7a5c1cfe 759 ['yt-dlp']
3089bc74
S
760 + [a for i, a in enumerate(argv) if i not in idxs]
761 + ['--'] + [argv[i] for i in idxs]
7d4111ed
PH
762 )
763 self.report_warning(
764 'Long argument string detected. '
49a57e70 765 'Use -- to separate parameters and URLs, like this:\n%s' %
7d4111ed
PH
766 args_to_str(correct_argv))
767
8222d8de
JMF
768 def add_info_extractor(self, ie):
769 """Add an InfoExtractor object to the end of the list."""
8b7491c8 770 ie_key = ie.ie_key()
771 self._ies[ie_key] = ie
e52d7f85 772 if not isinstance(ie, type):
8b7491c8 773 self._ies_instances[ie_key] = ie
e52d7f85 774 ie.set_downloader(self)
8222d8de 775
56c73665
JMF
776 def get_info_extractor(self, ie_key):
777 """
778 Get an instance of an IE with name ie_key, it will try to get one from
779 the _ies list, if there's no instance it will create a new one and add
780 it to the extractor list.
781 """
782 ie = self._ies_instances.get(ie_key)
783 if ie is None:
784 ie = get_info_extractor(ie_key)()
785 self.add_info_extractor(ie)
786 return ie
787
023fa8c4
JMF
788 def add_default_info_extractors(self):
789 """
790 Add the InfoExtractors returned by gen_extractors to the end of the list
791 """
fe7866d0 792 all_ies = {ie.IE_NAME.lower(): ie for ie in gen_extractor_classes()}
793 all_ies['end'] = UnsupportedURLIE()
794 try:
795 ie_names = orderedSet_from_options(
796 self.params.get('allowed_extractors', ['default']), {
797 'all': list(all_ies),
798 'default': [name for name, ie in all_ies.items() if ie._ENABLED],
799 }, use_regex=True)
800 except re.error as e:
801 raise ValueError(f'Wrong regex for allowed_extractors: {e.pattern}')
802 for name in ie_names:
803 self.add_info_extractor(all_ies[name])
804 self.write_debug(f'Loaded {len(ie_names)} extractors')
023fa8c4 805
56d868db 806 def add_post_processor(self, pp, when='post_process'):
8222d8de 807 """Add a PostProcessor object to the end of the chain."""
8aa0e7cd 808 assert when in POSTPROCESS_WHEN, f'Invalid when={when}'
5bfa4862 809 self._pps[when].append(pp)
8222d8de
JMF
810 pp.set_downloader(self)
811
ab8e5e51
AM
812 def add_post_hook(self, ph):
813 """Add the post hook"""
814 self._post_hooks.append(ph)
815
933605d7 816 def add_progress_hook(self, ph):
819e0531 817 """Add the download progress hook"""
933605d7 818 self._progress_hooks.append(ph)
8ab470f1 819
819e0531 820 def add_postprocessor_hook(self, ph):
821 """Add the postprocessing progress hook"""
822 self._postprocessor_hooks.append(ph)
5bfc8bee 823 for pps in self._pps.values():
824 for pp in pps:
825 pp.add_progress_hook(ph)
819e0531 826
1c088fa8 827 def _bidi_workaround(self, message):
5d681e96 828 if not hasattr(self, '_output_channel'):
1c088fa8
PH
829 return message
830
5d681e96 831 assert hasattr(self, '_output_process')
14f25df2 832 assert isinstance(message, str)
6febd1c1 833 line_count = message.count('\n') + 1
0f06bcd7 834 self._output_process.stdin.write((message + '\n').encode())
5d681e96 835 self._output_process.stdin.flush()
0f06bcd7 836 res = ''.join(self._output_channel.readline().decode()
9e1a5b84 837 for _ in range(line_count))
6febd1c1 838 return res[:-len('\n')]
1c088fa8 839
b35496d8 840 def _write_string(self, message, out=None, only_once=False):
841 if only_once:
842 if message in self._printed_messages:
843 return
844 self._printed_messages.add(message)
845 write_string(message, out=out, encoding=self.params.get('encoding'))
734f90bb 846
cf4f42cb 847 def to_stdout(self, message, skip_eol=False, quiet=None):
0760b0a7 848 """Print message to stdout"""
cf4f42cb 849 if quiet is not None:
da4db748 850 self.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument quiet. '
851 'Use "YoutubeDL.to_screen" instead')
8a82af35 852 if skip_eol is not False:
da4db748 853 self.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument skip_eol. '
854 'Use "YoutubeDL.to_screen" instead')
0bf9dc1e 855 self._write_string(f'{self._bidi_workaround(message)}\n', self._out_files.out)
cf4f42cb 856
dfea94f8 857 def to_screen(self, message, skip_eol=False, quiet=None, only_once=False):
cf4f42cb 858 """Print message to screen if not in quiet mode"""
8bf9319e 859 if self.params.get('logger'):
43afe285 860 self.params['logger'].debug(message)
cf4f42cb 861 return
862 if (self.params.get('quiet') if quiet is None else quiet) and not self.params.get('verbose'):
863 return
864 self._write_string(
865 '%s%s' % (self._bidi_workaround(message), ('' if skip_eol else '\n')),
dfea94f8 866 self._out_files.screen, only_once=only_once)
8222d8de 867
b35496d8 868 def to_stderr(self, message, only_once=False):
0760b0a7 869 """Print message to stderr"""
14f25df2 870 assert isinstance(message, str)
8bf9319e 871 if self.params.get('logger'):
43afe285
IB
872 self.params['logger'].error(message)
873 else:
5792c950 874 self._write_string(f'{self._bidi_workaround(message)}\n', self._out_files.error, only_once=only_once)
cf4f42cb 875
876 def _send_console_code(self, code):
591bb9d3 877 if compat_os_name == 'nt' or not self._out_files.console:
cf4f42cb 878 return
591bb9d3 879 self._write_string(code, self._out_files.console)
8222d8de 880
1e5b9a95
PH
881 def to_console_title(self, message):
882 if not self.params.get('consoletitle', False):
883 return
3efb96a6 884 message = remove_terminal_sequences(message)
4bede0d8
C
885 if compat_os_name == 'nt':
886 if ctypes.windll.kernel32.GetConsoleWindow():
887 # c_wchar_p() might not be necessary if `message` is
888 # already of type unicode()
889 ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
cf4f42cb 890 else:
891 self._send_console_code(f'\033]0;{message}\007')
1e5b9a95 892
bdde425c 893 def save_console_title(self):
cf4f42cb 894 if not self.params.get('consoletitle') or self.params.get('simulate'):
bdde425c 895 return
592b7485 896 self._send_console_code('\033[22;0t') # Save the title on stack
bdde425c
PH
897
898 def restore_console_title(self):
cf4f42cb 899 if not self.params.get('consoletitle') or self.params.get('simulate'):
bdde425c 900 return
592b7485 901 self._send_console_code('\033[23;0t') # Restore the title from stack
bdde425c
PH
902
903 def __enter__(self):
904 self.save_console_title()
905 return self
906
907 def __exit__(self, *args):
908 self.restore_console_title()
f89197d7 909
dca08720 910 if self.params.get('cookiefile') is not None:
1bab3437 911 self.cookiejar.save(ignore_discard=True, ignore_expires=True)
bdde425c 912
fa9f30b8 913 def trouble(self, message=None, tb=None, is_error=True):
8222d8de
JMF
914 """Determine action to take when a download problem appears.
915
916 Depending on if the downloader has been configured to ignore
917 download errors or not, this method may throw an exception or
918 not when errors are found, after printing the message.
919
fa9f30b8 920 @param tb If given, is additional traceback information
921 @param is_error Whether to raise error according to ignorerrors
8222d8de
JMF
922 """
923 if message is not None:
924 self.to_stderr(message)
925 if self.params.get('verbose'):
926 if tb is None:
927 if sys.exc_info()[0]: # if .trouble has been called from an except block
6febd1c1 928 tb = ''
8222d8de 929 if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
6febd1c1 930 tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
c0384f22 931 tb += encode_compat_str(traceback.format_exc())
8222d8de
JMF
932 else:
933 tb_data = traceback.format_list(traceback.extract_stack())
6febd1c1 934 tb = ''.join(tb_data)
c19bc311 935 if tb:
936 self.to_stderr(tb)
fa9f30b8 937 if not is_error:
938 return
b1940459 939 if not self.params.get('ignoreerrors'):
8222d8de
JMF
940 if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
941 exc_info = sys.exc_info()[1].exc_info
942 else:
943 exc_info = sys.exc_info()
944 raise DownloadError(message, exc_info)
945 self._download_retcode = 1
946
19a03940 947 Styles = Namespace(
948 HEADERS='yellow',
949 EMPHASIS='light blue',
492272fe 950 FILENAME='green',
19a03940 951 ID='green',
952 DELIM='blue',
953 ERROR='red',
954 WARNING='yellow',
955 SUPPRESS='light black',
956 )
ec11a9f4 957
7578d77d 958 def _format_text(self, handle, allow_colors, text, f, fallback=None, *, test_encoding=False):
e5a998f3 959 text = str(text)
ec11a9f4 960 if test_encoding:
961 original_text = text
5c104538 962 # handle.encoding can be None. See https://github.com/yt-dlp/yt-dlp/issues/2711
963 encoding = self.params.get('encoding') or getattr(handle, 'encoding', None) or 'ascii'
ec11a9f4 964 text = text.encode(encoding, 'ignore').decode(encoding)
965 if fallback is not None and text != original_text:
966 text = fallback
7578d77d 967 return format_text(text, f) if allow_colors else text if fallback is None else fallback
ec11a9f4 968
591bb9d3 969 def _format_out(self, *args, **kwargs):
970 return self._format_text(self._out_files.out, self._allow_colors.out, *args, **kwargs)
971
ec11a9f4 972 def _format_screen(self, *args, **kwargs):
591bb9d3 973 return self._format_text(self._out_files.screen, self._allow_colors.screen, *args, **kwargs)
ec11a9f4 974
975 def _format_err(self, *args, **kwargs):
591bb9d3 976 return self._format_text(self._out_files.error, self._allow_colors.error, *args, **kwargs)
819e0531 977
c84aeac6 978 def report_warning(self, message, only_once=False):
8222d8de
JMF
979 '''
980 Print the message to stderr, it will be prefixed with 'WARNING:'
981 If stderr is a tty file the 'WARNING:' will be colored
982 '''
6d07ce01
JMF
983 if self.params.get('logger') is not None:
984 self.params['logger'].warning(message)
8222d8de 985 else:
ad8915b7
PH
986 if self.params.get('no_warnings'):
987 return
ec11a9f4 988 self.to_stderr(f'{self._format_err("WARNING:", self.Styles.WARNING)} {message}', only_once)
8222d8de 989
da4db748 990 def deprecation_warning(self, message, *, stacklevel=0):
991 deprecation_warning(
992 message, stacklevel=stacklevel + 1, printer=self.report_error, is_error=False)
993
994 def deprecated_feature(self, message):
ee8dd27a 995 if self.params.get('logger') is not None:
da4db748 996 self.params['logger'].warning(f'Deprecated Feature: {message}')
997 self.to_stderr(f'{self._format_err("Deprecated Feature:", self.Styles.ERROR)} {message}', True)
ee8dd27a 998
fa9f30b8 999 def report_error(self, message, *args, **kwargs):
8222d8de
JMF
1000 '''
1001 Do the same as trouble, but prefixes the message with 'ERROR:', colored
1002 in red if stderr is a tty file.
1003 '''
fa9f30b8 1004 self.trouble(f'{self._format_err("ERROR:", self.Styles.ERROR)} {message}', *args, **kwargs)
8222d8de 1005
b35496d8 1006 def write_debug(self, message, only_once=False):
0760b0a7 1007 '''Log debug message or Print message to stderr'''
1008 if not self.params.get('verbose', False):
1009 return
8a82af35 1010 message = f'[debug] {message}'
0760b0a7 1011 if self.params.get('logger'):
1012 self.params['logger'].debug(message)
1013 else:
b35496d8 1014 self.to_stderr(message, only_once)
0760b0a7 1015
8222d8de
JMF
1016 def report_file_already_downloaded(self, file_name):
1017 """Report file has already been fully downloaded."""
1018 try:
6febd1c1 1019 self.to_screen('[download] %s has already been downloaded' % file_name)
ce02ed60 1020 except UnicodeEncodeError:
6febd1c1 1021 self.to_screen('[download] The file has already been downloaded')
8222d8de 1022
0c3d0f51 1023 def report_file_delete(self, file_name):
1024 """Report that existing file will be deleted."""
1025 try:
c25228e5 1026 self.to_screen('Deleting existing file %s' % file_name)
0c3d0f51 1027 except UnicodeEncodeError:
c25228e5 1028 self.to_screen('Deleting existing file')
0c3d0f51 1029
319b6059 1030 def raise_no_formats(self, info, forced=False, *, msg=None):
0a5a191a 1031 has_drm = info.get('_has_drm')
319b6059 1032 ignored, expected = self.params.get('ignore_no_formats_error'), bool(msg)
1033 msg = msg or has_drm and 'This video is DRM protected' or 'No video formats found!'
1034 if forced or not ignored:
1151c407 1035 raise ExtractorError(msg, video_id=info['id'], ie=info['extractor'],
319b6059 1036 expected=has_drm or ignored or expected)
88acdbc2 1037 else:
1038 self.report_warning(msg)
1039
de6000d9 1040 def parse_outtmpl(self):
bf1824b3 1041 self.deprecation_warning('"YoutubeDL.parse_outtmpl" is deprecated and may be removed in a future version')
1042 self._parse_outtmpl()
1043 return self.params['outtmpl']
1044
1045 def _parse_outtmpl(self):
7b2c3f47 1046 sanitize = IDENTITY
bf1824b3 1047 if self.params.get('restrictfilenames'): # Remove spaces in the default template
71ce444a 1048 sanitize = lambda x: x.replace(' - ', ' ').replace(' ', '-')
bf1824b3 1049
1050 outtmpl = self.params.setdefault('outtmpl', {})
1051 if not isinstance(outtmpl, dict):
1052 self.params['outtmpl'] = outtmpl = {'default': outtmpl}
1053 outtmpl.update({k: sanitize(v) for k, v in DEFAULT_OUTTMPL.items() if outtmpl.get(k) is None})
de6000d9 1054
21cd8fae 1055 def get_output_path(self, dir_type='', filename=None):
1056 paths = self.params.get('paths', {})
d2c8aadf 1057 assert isinstance(paths, dict), '"paths" parameter must be a dictionary'
21cd8fae 1058 path = os.path.join(
1059 expand_path(paths.get('home', '').strip()),
1060 expand_path(paths.get(dir_type, '').strip()) if dir_type else '',
1061 filename or '')
21cd8fae 1062 return sanitize_path(path, force=self.params.get('windowsfilenames'))
1063
76a264ac 1064 @staticmethod
901130bb 1065 def _outtmpl_expandpath(outtmpl):
1066 # expand_path translates '%%' into '%' and '$$' into '$'
1067 # correspondingly that is not what we want since we need to keep
1068 # '%%' intact for template dict substitution step. Working around
1069 # with boundary-alike separator hack.
1070 sep = ''.join([random.choice(ascii_letters) for _ in range(32)])
86e5f3ed 1071 outtmpl = outtmpl.replace('%%', f'%{sep}%').replace('$$', f'${sep}$')
901130bb 1072
1073 # outtmpl should be expand_path'ed before template dict substitution
1074 # because meta fields may contain env variables we don't want to
62b58c09 1075 # be expanded. E.g. for outtmpl "%(title)s.%(ext)s" and
901130bb 1076 # title "Hello $PATH", we don't want `$PATH` to be expanded.
1077 return expand_path(outtmpl).replace(sep, '')
1078
1079 @staticmethod
1080 def escape_outtmpl(outtmpl):
1081 ''' Escape any remaining strings like %s, %abc% etc. '''
1082 return re.sub(
1083 STR_FORMAT_RE_TMPL.format('', '(?![%(\0])'),
1084 lambda mobj: ('' if mobj.group('has_key') else '%') + mobj.group(0),
1085 outtmpl)
1086
1087 @classmethod
1088 def validate_outtmpl(cls, outtmpl):
76a264ac 1089 ''' @return None or Exception object '''
7d1eb38a 1090 outtmpl = re.sub(
47cdc68e 1091 STR_FORMAT_RE_TMPL.format('[^)]*', '[ljhqBUDS]'),
7d1eb38a 1092 lambda mobj: f'{mobj.group(0)[:-1]}s',
1093 cls._outtmpl_expandpath(outtmpl))
76a264ac 1094 try:
7d1eb38a 1095 cls.escape_outtmpl(outtmpl) % collections.defaultdict(int)
76a264ac 1096 return None
1097 except ValueError as err:
1098 return err
1099
03b4de72 1100 @staticmethod
1101 def _copy_infodict(info_dict):
1102 info_dict = dict(info_dict)
09b49e1f 1103 info_dict.pop('__postprocessors', None)
415f8d51 1104 info_dict.pop('__pending_error', None)
03b4de72 1105 return info_dict
1106
e0fd9573 1107 def prepare_outtmpl(self, outtmpl, info_dict, sanitize=False):
1108 """ Make the outtmpl and info_dict suitable for substitution: ydl.escape_outtmpl(outtmpl) % info_dict
1109 @param sanitize Whether to sanitize the output as a filename.
1110 For backward compatibility, a function can also be passed
1111 """
1112
6e84b215 1113 info_dict.setdefault('epoch', int(time.time())) # keep epoch consistent once set
143db31d 1114
03b4de72 1115 info_dict = self._copy_infodict(info_dict)
752cda38 1116 info_dict['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
53c18592 1117 formatSeconds(info_dict['duration'], '-' if sanitize else ':')
143db31d 1118 if info_dict.get('duration', None) is not None
1119 else None)
1d485a1a 1120 info_dict['autonumber'] = int(self.params.get('autonumber_start', 1) - 1 + self._num_downloads)
9c906919 1121 info_dict['video_autonumber'] = self._num_videos
752cda38 1122 if info_dict.get('resolution') is None:
1123 info_dict['resolution'] = self.format_resolution(info_dict, default=None)
143db31d 1124
e6f21b3d 1125 # For fields playlist_index, playlist_autonumber and autonumber convert all occurrences
143db31d 1126 # of %(field)s to %(field)0Nd for backward compatibility
1127 field_size_compat_map = {
0a5a191a 1128 'playlist_index': number_of_digits(info_dict.get('__last_playlist_index') or 0),
ec11a9f4 1129 'playlist_autonumber': number_of_digits(info_dict.get('n_entries') or 0),
752cda38 1130 'autonumber': self.params.get('autonumber_size') or 5,
143db31d 1131 }
752cda38 1132
385a27fa 1133 TMPL_DICT = {}
47cdc68e 1134 EXTERNAL_FORMAT_RE = re.compile(STR_FORMAT_RE_TMPL.format('[^)]*', f'[{STR_FORMAT_TYPES}ljhqBUDS]'))
385a27fa 1135 MATH_FUNCTIONS = {
1136 '+': float.__add__,
1137 '-': float.__sub__,
1138 }
e625be0d 1139 # Field is of the form key1.key2...
07a1250e 1140 # where keys (except first) can be string, int, slice or "{field, ...}"
1141 FIELD_INNER_RE = r'(?:\w+|%(num)s|%(num)s?(?::%(num)s?){1,2})' % {'num': r'(?:-?\d+)'}
1142 FIELD_RE = r'\w*(?:\.(?:%(inner)s|{%(field)s(?:,%(field)s)*}))*' % {
1143 'inner': FIELD_INNER_RE,
1144 'field': rf'\w*(?:\.{FIELD_INNER_RE})*'
1145 }
1d485a1a 1146 MATH_FIELD_RE = rf'(?:{FIELD_RE}|-?{NUMBER_RE})'
385a27fa 1147 MATH_OPERATORS_RE = r'(?:%s)' % '|'.join(map(re.escape, MATH_FUNCTIONS.keys()))
1d485a1a 1148 INTERNAL_FORMAT_RE = re.compile(rf'''(?x)
e625be0d 1149 (?P<negate>-)?
1d485a1a 1150 (?P<fields>{FIELD_RE})
1151 (?P<maths>(?:{MATH_OPERATORS_RE}{MATH_FIELD_RE})*)
e625be0d 1152 (?:>(?P<strf_format>.+?))?
34baa9fd 1153 (?P<remaining>
1154 (?P<alternate>(?<!\\),[^|&)]+)?
1155 (?:&(?P<replacement>.*?))?
1156 (?:\|(?P<default>.*?))?
1d485a1a 1157 )$''')
752cda38 1158
07a1250e 1159 def _traverse_infodict(fields):
1160 fields = [f for x in re.split(r'\.({.+?})\.?', fields)
1161 for f in ([x] if x.startswith('{') else x.split('.'))]
1162 for i in (0, -1):
1163 if fields and not fields[i]:
1164 fields.pop(i)
1165
1166 for i, f in enumerate(fields):
1167 if not f.startswith('{'):
1168 continue
1169 assert f.endswith('}'), f'No closing brace for {f} in {fields}'
1170 fields[i] = {k: k.split('.') for k in f[1:-1].split(',')}
1171
1172 return traverse_obj(info_dict, fields, is_user_input=True, traverse_string=True)
76a264ac 1173
752cda38 1174 def get_value(mdict):
1175 # Object traversal
2b8a2973 1176 value = _traverse_infodict(mdict['fields'])
752cda38 1177 # Negative
1178 if mdict['negate']:
1179 value = float_or_none(value)
1180 if value is not None:
1181 value *= -1
1182 # Do maths
385a27fa 1183 offset_key = mdict['maths']
1184 if offset_key:
752cda38 1185 value = float_or_none(value)
1186 operator = None
385a27fa 1187 while offset_key:
1188 item = re.match(
1189 MATH_FIELD_RE if operator else MATH_OPERATORS_RE,
1190 offset_key).group(0)
1191 offset_key = offset_key[len(item):]
1192 if operator is None:
752cda38 1193 operator = MATH_FUNCTIONS[item]
385a27fa 1194 continue
1195 item, multiplier = (item[1:], -1) if item[0] == '-' else (item, 1)
1196 offset = float_or_none(item)
1197 if offset is None:
2b8a2973 1198 offset = float_or_none(_traverse_infodict(item))
385a27fa 1199 try:
1200 value = operator(value, multiplier * offset)
1201 except (TypeError, ZeroDivisionError):
1202 return None
1203 operator = None
752cda38 1204 # Datetime formatting
1205 if mdict['strf_format']:
7c37ff97 1206 value = strftime_or_none(value, mdict['strf_format'].replace('\\,', ','))
752cda38 1207
a6bcaf71 1208 # XXX: Workaround for https://github.com/yt-dlp/yt-dlp/issues/4485
1209 if sanitize and value == '':
1210 value = None
752cda38 1211 return value
1212
b868936c 1213 na = self.params.get('outtmpl_na_placeholder', 'NA')
1214
e0fd9573 1215 def filename_sanitizer(key, value, restricted=self.params.get('restrictfilenames')):
5c3895ff 1216 return sanitize_filename(str(value), restricted=restricted, is_id=(
1217 bool(re.search(r'(^|[_.])id(\.|$)', key))
8a82af35 1218 if 'filename-sanitization' in self.params['compat_opts']
5c3895ff 1219 else NO_DEFAULT))
e0fd9573 1220
1221 sanitizer = sanitize if callable(sanitize) else filename_sanitizer
1222 sanitize = bool(sanitize)
1223
6e84b215 1224 def _dumpjson_default(obj):
1225 if isinstance(obj, (set, LazyList)):
1226 return list(obj)
adbc4ec4 1227 return repr(obj)
6e84b215 1228
752cda38 1229 def create_key(outer_mobj):
1230 if not outer_mobj.group('has_key'):
b836dc94 1231 return outer_mobj.group(0)
752cda38 1232 key = outer_mobj.group('key')
752cda38 1233 mobj = re.match(INTERNAL_FORMAT_RE, key)
e0fd9573 1234 initial_field = mobj.group('fields') if mobj else ''
e978789f 1235 value, replacement, default = None, None, na
7c37ff97 1236 while mobj:
e625be0d 1237 mobj = mobj.groupdict()
7c37ff97 1238 default = mobj['default'] if mobj['default'] is not None else default
752cda38 1239 value = get_value(mobj)
e978789f 1240 replacement = mobj['replacement']
7c37ff97 1241 if value is None and mobj['alternate']:
34baa9fd 1242 mobj = re.match(INTERNAL_FORMAT_RE, mobj['remaining'][1:])
7c37ff97 1243 else:
1244 break
752cda38 1245
b868936c 1246 fmt = outer_mobj.group('format')
752cda38 1247 if fmt == 's' and value is not None and key in field_size_compat_map.keys():
86e5f3ed 1248 fmt = f'0{field_size_compat_map[key]:d}d'
752cda38 1249
e978789f 1250 value = default if value is None else value if replacement is None else replacement
752cda38 1251
4476d2c7 1252 flags = outer_mobj.group('conversion') or ''
7d1eb38a 1253 str_fmt = f'{fmt[:-1]}s'
524e2e4f 1254 if fmt[-1] == 'l': # list
4476d2c7 1255 delim = '\n' if '#' in flags else ', '
9e907ebd 1256 value, fmt = delim.join(map(str, variadic(value, allowed_types=(str, bytes)))), str_fmt
524e2e4f 1257 elif fmt[-1] == 'j': # json
deae7c17 1258 value, fmt = json.dumps(
1259 value, default=_dumpjson_default,
9b9dad11 1260 indent=4 if '#' in flags else None, ensure_ascii='+' not in flags), str_fmt
47cdc68e 1261 elif fmt[-1] == 'h': # html
deae7c17 1262 value, fmt = escapeHTML(str(value)), str_fmt
524e2e4f 1263 elif fmt[-1] == 'q': # quoted
4476d2c7 1264 value = map(str, variadic(value) if '#' in flags else [value])
1265 value, fmt = ' '.join(map(compat_shlex_quote, value)), str_fmt
524e2e4f 1266 elif fmt[-1] == 'B': # bytes
0f06bcd7 1267 value = f'%{str_fmt}'.encode() % str(value).encode()
f5aa5cfb 1268 value, fmt = value.decode('utf-8', 'ignore'), 's'
524e2e4f 1269 elif fmt[-1] == 'U': # unicode normalized
524e2e4f 1270 value, fmt = unicodedata.normalize(
1271 # "+" = compatibility equivalence, "#" = NFD
4476d2c7 1272 'NF%s%s' % ('K' if '+' in flags else '', 'D' if '#' in flags else 'C'),
524e2e4f 1273 value), str_fmt
e0fd9573 1274 elif fmt[-1] == 'D': # decimal suffix
abbeeebc 1275 num_fmt, fmt = fmt[:-1].replace('#', ''), 's'
1276 value = format_decimal_suffix(value, f'%{num_fmt}f%s' if num_fmt else '%d%s',
1277 factor=1024 if '#' in flags else 1000)
37893bb0 1278 elif fmt[-1] == 'S': # filename sanitization
e0fd9573 1279 value, fmt = filename_sanitizer(initial_field, value, restricted='#' in flags), str_fmt
7d1eb38a 1280 elif fmt[-1] == 'c':
524e2e4f 1281 if value:
1282 value = str(value)[0]
76a264ac 1283 else:
524e2e4f 1284 fmt = str_fmt
76a264ac 1285 elif fmt[-1] not in 'rs': # numeric
a439a3a4 1286 value = float_or_none(value)
752cda38 1287 if value is None:
1288 value, fmt = default, 's'
901130bb 1289
752cda38 1290 if sanitize:
1291 if fmt[-1] == 'r':
1292 # If value is an object, sanitize might convert it to a string
1293 # So we convert it to repr first
7d1eb38a 1294 value, fmt = repr(value), str_fmt
639f1cea 1295 if fmt[-1] in 'csr':
e0fd9573 1296 value = sanitizer(initial_field, value)
901130bb 1297
b868936c 1298 key = '%s\0%s' % (key.replace('%', '%\0'), outer_mobj.group('format'))
385a27fa 1299 TMPL_DICT[key] = value
b868936c 1300 return '{prefix}%({key}){fmt}'.format(key=key, fmt=fmt, prefix=outer_mobj.group('prefix'))
752cda38 1301
385a27fa 1302 return EXTERNAL_FORMAT_RE.sub(create_key, outtmpl), TMPL_DICT
143db31d 1303
819e0531 1304 def evaluate_outtmpl(self, outtmpl, info_dict, *args, **kwargs):
1305 outtmpl, info_dict = self.prepare_outtmpl(outtmpl, info_dict, *args, **kwargs)
1306 return self.escape_outtmpl(outtmpl) % info_dict
1307
5127e92a 1308 def _prepare_filename(self, info_dict, *, outtmpl=None, tmpl_type=None):
1309 assert None in (outtmpl, tmpl_type), 'outtmpl and tmpl_type are mutually exclusive'
1310 if outtmpl is None:
bf1824b3 1311 outtmpl = self.params['outtmpl'].get(tmpl_type or 'default', self.params['outtmpl']['default'])
8222d8de 1312 try:
5127e92a 1313 outtmpl = self._outtmpl_expandpath(outtmpl)
e0fd9573 1314 filename = self.evaluate_outtmpl(outtmpl, info_dict, True)
6a0546e3 1315 if not filename:
1316 return None
15da37c7 1317
5127e92a 1318 if tmpl_type in ('', 'temp'):
6a0546e3 1319 final_ext, ext = self.params.get('final_ext'), info_dict.get('ext')
1320 if final_ext and ext and final_ext != ext and filename.endswith(f'.{final_ext}'):
1321 filename = replace_extension(filename, ext, final_ext)
5127e92a 1322 elif tmpl_type:
6a0546e3 1323 force_ext = OUTTMPL_TYPES[tmpl_type]
1324 if force_ext:
1325 filename = replace_extension(filename, force_ext, info_dict.get('ext'))
de6000d9 1326
bdc3fd2f
U
1327 # https://github.com/blackjack4494/youtube-dlc/issues/85
1328 trim_file_name = self.params.get('trim_file_name', False)
1329 if trim_file_name:
5c22c63d 1330 no_ext, *ext = filename.rsplit('.', 2)
1331 filename = join_nonempty(no_ext[:trim_file_name], *ext, delim='.')
bdc3fd2f 1332
0202b52a 1333 return filename
8222d8de 1334 except ValueError as err:
6febd1c1 1335 self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')')
8222d8de
JMF
1336 return None
1337
5127e92a 1338 def prepare_filename(self, info_dict, dir_type='', *, outtmpl=None, warn=False):
1339 """Generate the output filename"""
1340 if outtmpl:
1341 assert not dir_type, 'outtmpl and dir_type are mutually exclusive'
1342 dir_type = None
1343 filename = self._prepare_filename(info_dict, tmpl_type=dir_type, outtmpl=outtmpl)
80c03fa9 1344 if not filename and dir_type not in ('', 'temp'):
1345 return ''
de6000d9 1346
c84aeac6 1347 if warn:
21cd8fae 1348 if not self.params.get('paths'):
de6000d9 1349 pass
1350 elif filename == '-':
c84aeac6 1351 self.report_warning('--paths is ignored when an outputting to stdout', only_once=True)
de6000d9 1352 elif os.path.isabs(filename):
c84aeac6 1353 self.report_warning('--paths is ignored since an absolute path is given in output template', only_once=True)
de6000d9 1354 if filename == '-' or not filename:
1355 return filename
1356
21cd8fae 1357 return self.get_output_path(dir_type, filename)
0202b52a 1358
120fe513 1359 def _match_entry(self, info_dict, incomplete=False, silent=False):
6368e2e6 1360 """Returns None if the file should be downloaded"""
d7b460d0 1361 _type = info_dict.get('_type', 'video')
1362 assert incomplete or _type == 'video', 'Only video result can be considered complete'
8222d8de 1363
3bec830a 1364 video_title = info_dict.get('title', info_dict.get('id', 'entry'))
c77495e3 1365
8b0d7497 1366 def check_filter():
d7b460d0 1367 if _type in ('playlist', 'multi_video'):
1368 return
1369 elif _type in ('url', 'url_transparent') and not try_call(
1370 lambda: self.get_info_extractor(info_dict['ie_key']).is_single_video(info_dict['url'])):
1371 return
1372
8b0d7497 1373 if 'title' in info_dict:
1374 # This can happen when we're just evaluating the playlist
1375 title = info_dict['title']
1376 matchtitle = self.params.get('matchtitle', False)
1377 if matchtitle:
1378 if not re.search(matchtitle, title, re.IGNORECASE):
1379 return '"' + title + '" title did not match pattern "' + matchtitle + '"'
1380 rejecttitle = self.params.get('rejecttitle', False)
1381 if rejecttitle:
1382 if re.search(rejecttitle, title, re.IGNORECASE):
1383 return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
6368e2e6 1384
8b0d7497 1385 date = info_dict.get('upload_date')
1386 if date is not None:
1387 dateRange = self.params.get('daterange', DateRange())
1388 if date not in dateRange:
86e5f3ed 1389 return f'{date_from_str(date).isoformat()} upload date is not in range {dateRange}'
8b0d7497 1390 view_count = info_dict.get('view_count')
1391 if view_count is not None:
1392 min_views = self.params.get('min_views')
1393 if min_views is not None and view_count < min_views:
1394 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
1395 max_views = self.params.get('max_views')
1396 if max_views is not None and view_count > max_views:
1397 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
1398 if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')):
1399 return 'Skipping "%s" because it is age restricted' % video_title
8b0d7497 1400
8f18aca8 1401 match_filter = self.params.get('match_filter')
1402 if match_filter is not None:
1403 try:
1404 ret = match_filter(info_dict, incomplete=incomplete)
1405 except TypeError:
1406 # For backward compatibility
1407 ret = None if incomplete else match_filter(info_dict)
492272fe 1408 if ret is NO_DEFAULT:
1409 while True:
1410 filename = self._format_screen(self.prepare_filename(info_dict), self.Styles.FILENAME)
1411 reply = input(self._format_screen(
1412 f'Download "{filename}"? (Y/n): ', self.Styles.EMPHASIS)).lower().strip()
1413 if reply in {'y', ''}:
1414 return None
1415 elif reply == 'n':
1416 return f'Skipping {video_title}'
492272fe 1417 elif ret is not None:
8f18aca8 1418 return ret
8b0d7497 1419 return None
1420
c77495e3 1421 if self.in_download_archive(info_dict):
1422 reason = '%s has already been recorded in the archive' % video_title
1423 break_opt, break_err = 'break_on_existing', ExistingVideoReached
1424 else:
1425 reason = check_filter()
1426 break_opt, break_err = 'break_on_reject', RejectedVideoReached
8b0d7497 1427 if reason is not None:
120fe513 1428 if not silent:
1429 self.to_screen('[download] ' + reason)
c77495e3 1430 if self.params.get(break_opt, False):
1431 raise break_err()
8b0d7497 1432 return reason
fe7e0c98 1433
b6c45014
JMF
1434 @staticmethod
1435 def add_extra_info(info_dict, extra_info):
1436 '''Set the keys from extra_info in info dict if they are missing'''
1437 for key, value in extra_info.items():
1438 info_dict.setdefault(key, value)
1439
409e1828 1440 def extract_info(self, url, download=True, ie_key=None, extra_info=None,
61aa5ba3 1441 process=True, force_generic_extractor=False):
41d1cca3 1442 """
17ffed18 1443 Extract and return the information dictionary of the URL
41d1cca3 1444
1445 Arguments:
17ffed18 1446 @param url URL to extract
41d1cca3 1447
1448 Keyword arguments:
17ffed18 1449 @param download Whether to download videos
1450 @param process Whether to resolve all unresolved references (URLs, playlist items).
1451 Must be True for download to work
1452 @param ie_key Use only the extractor with this key
1453
1454 @param extra_info Dictionary containing the extra values to add to the info (For internal use only)
1455 @force_generic_extractor Force using the generic extractor (Deprecated; use ie_key='Generic')
41d1cca3 1456 """
fe7e0c98 1457
409e1828 1458 if extra_info is None:
1459 extra_info = {}
1460
61aa5ba3 1461 if not ie_key and force_generic_extractor:
d22dec74
S
1462 ie_key = 'Generic'
1463
8222d8de 1464 if ie_key:
fe7866d0 1465 ies = {ie_key: self._ies[ie_key]} if ie_key in self._ies else {}
8222d8de
JMF
1466 else:
1467 ies = self._ies
1468
fe7866d0 1469 for key, ie in ies.items():
8222d8de
JMF
1470 if not ie.suitable(url):
1471 continue
1472
1473 if not ie.working():
6febd1c1
PH
1474 self.report_warning('The program functionality for this site has been marked as broken, '
1475 'and will probably not work.')
8222d8de 1476
1151c407 1477 temp_id = ie.get_temp_id(url)
fe7866d0 1478 if temp_id is not None and self.in_download_archive({'id': temp_id, 'ie_key': key}):
1479 self.to_screen(f'[{key}] {temp_id}: has already been recorded in the archive')
5e5be0c0 1480 if self.params.get('break_on_existing', False):
1481 raise ExistingVideoReached()
a0566bbf 1482 break
fe7866d0 1483 return self.__extract_info(url, self.get_info_extractor(key), download, extra_info, process)
a0566bbf 1484 else:
fe7866d0 1485 extractors_restricted = self.params.get('allowed_extractors') not in (None, ['default'])
1486 self.report_error(f'No suitable extractor{format_field(ie_key, None, " (%s)")} found for URL {url}',
1487 tb=False if extractors_restricted else None)
a0566bbf 1488
7e88d7d7 1489 def _handle_extraction_exceptions(func):
b5ae35ee 1490 @functools.wraps(func)
a0566bbf 1491 def wrapper(self, *args, **kwargs):
6da22e7d 1492 while True:
1493 try:
1494 return func(self, *args, **kwargs)
1495 except (DownloadCancelled, LazyList.IndexError, PagedList.IndexError):
8222d8de 1496 raise
6da22e7d 1497 except ReExtractInfo as e:
1498 if e.expected:
1499 self.to_screen(f'{e}; Re-extracting data')
1500 else:
1501 self.to_stderr('\r')
1502 self.report_warning(f'{e}; Re-extracting data')
1503 continue
1504 except GeoRestrictedError as e:
1505 msg = e.msg
1506 if e.countries:
1507 msg += '\nThis video is available in %s.' % ', '.join(
1508 map(ISO3166Utils.short2full, e.countries))
1509 msg += '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1510 self.report_error(msg)
1511 except ExtractorError as e: # An error we somewhat expected
1512 self.report_error(str(e), e.format_traceback())
1513 except Exception as e:
1514 if self.params.get('ignoreerrors'):
1515 self.report_error(str(e), tb=encode_compat_str(traceback.format_exc()))
1516 else:
1517 raise
1518 break
a0566bbf 1519 return wrapper
1520
693f0600 1521 def _wait_for_video(self, ie_result={}):
f2ebc5c7 1522 if (not self.params.get('wait_for_video')
1523 or ie_result.get('_type', 'video') != 'video'
1524 or ie_result.get('formats') or ie_result.get('url')):
1525 return
1526
1527 format_dur = lambda dur: '%02d:%02d:%02d' % timetuple_from_msec(dur * 1000)[:-1]
1528 last_msg = ''
1529
1530 def progress(msg):
1531 nonlocal last_msg
a7dc6a89 1532 full_msg = f'{msg}\n'
1533 if not self.params.get('noprogress'):
1534 full_msg = msg + ' ' * (len(last_msg) - len(msg)) + '\r'
1535 elif last_msg:
1536 return
1537 self.to_screen(full_msg, skip_eol=True)
f2ebc5c7 1538 last_msg = msg
1539
1540 min_wait, max_wait = self.params.get('wait_for_video')
1541 diff = try_get(ie_result, lambda x: x['release_timestamp'] - time.time())
1542 if diff is None and ie_result.get('live_status') == 'is_upcoming':
16c620bc 1543 diff = round(random.uniform(min_wait, max_wait) if (max_wait and min_wait) else (max_wait or min_wait), 0)
f2ebc5c7 1544 self.report_warning('Release time of video is not known')
693f0600 1545 elif ie_result and (diff or 0) <= 0:
f2ebc5c7 1546 self.report_warning('Video should already be available according to extracted info')
38d79fd1 1547 diff = min(max(diff or 0, min_wait or 0), max_wait or float('inf'))
f2ebc5c7 1548 self.to_screen(f'[wait] Waiting for {format_dur(diff)} - Press Ctrl+C to try now')
1549
1550 wait_till = time.time() + diff
1551 try:
1552 while True:
1553 diff = wait_till - time.time()
1554 if diff <= 0:
1555 progress('')
1556 raise ReExtractInfo('[wait] Wait period ended', expected=True)
1557 progress(f'[wait] Remaining time until next attempt: {self._format_screen(format_dur(diff), self.Styles.EMPHASIS)}')
1558 time.sleep(1)
1559 except KeyboardInterrupt:
1560 progress('')
1561 raise ReExtractInfo('[wait] Interrupted by user', expected=True)
1562 except BaseException as e:
1563 if not isinstance(e, ReExtractInfo):
1564 self.to_screen('')
1565 raise
1566
7e88d7d7 1567 @_handle_extraction_exceptions
58f197b7 1568 def __extract_info(self, url, ie, download, extra_info, process):
693f0600 1569 try:
1570 ie_result = ie.extract(url)
1571 except UserNotLive as e:
1572 if process:
1573 if self.params.get('wait_for_video'):
1574 self.report_warning(e)
1575 self._wait_for_video()
1576 raise
a0566bbf 1577 if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
cb794ee0 1578 self.report_warning(f'Extractor {ie.IE_NAME} returned nothing{bug_reports_message()}')
a0566bbf 1579 return
1580 if isinstance(ie_result, list):
1581 # Backwards compatibility: old IE result format
1582 ie_result = {
1583 '_type': 'compat_list',
1584 'entries': ie_result,
1585 }
e37d0efb 1586 if extra_info.get('original_url'):
1587 ie_result.setdefault('original_url', extra_info['original_url'])
a0566bbf 1588 self.add_default_extra_info(ie_result, ie, url)
1589 if process:
f2ebc5c7 1590 self._wait_for_video(ie_result)
a0566bbf 1591 return self.process_ie_result(ie_result, download, extra_info)
8222d8de 1592 else:
a0566bbf 1593 return ie_result
fe7e0c98 1594
ea38e55f 1595 def add_default_extra_info(self, ie_result, ie, url):
6033d980 1596 if url is not None:
1597 self.add_extra_info(ie_result, {
1598 'webpage_url': url,
1599 'original_url': url,
57ebfca3 1600 })
1601 webpage_url = ie_result.get('webpage_url')
1602 if webpage_url:
1603 self.add_extra_info(ie_result, {
1604 'webpage_url_basename': url_basename(webpage_url),
1605 'webpage_url_domain': get_domain(webpage_url),
6033d980 1606 })
1607 if ie is not None:
1608 self.add_extra_info(ie_result, {
1609 'extractor': ie.IE_NAME,
1610 'extractor_key': ie.ie_key(),
1611 })
ea38e55f 1612
58adec46 1613 def process_ie_result(self, ie_result, download=True, extra_info=None):
8222d8de
JMF
1614 """
1615 Take the result of the ie(may be modified) and resolve all unresolved
1616 references (URLs, playlist items).
1617
1618 It will also download the videos if 'download'.
1619 Returns the resolved ie_result.
1620 """
58adec46 1621 if extra_info is None:
1622 extra_info = {}
e8ee972c
PH
1623 result_type = ie_result.get('_type', 'video')
1624
057a5206 1625 if result_type in ('url', 'url_transparent'):
8f97a15d 1626 ie_result['url'] = sanitize_url(
1627 ie_result['url'], scheme='http' if self.params.get('prefer_insecure') else 'https')
e37d0efb 1628 if ie_result.get('original_url'):
1629 extra_info.setdefault('original_url', ie_result['original_url'])
1630
057a5206 1631 extract_flat = self.params.get('extract_flat', False)
3089bc74
S
1632 if ((extract_flat == 'in_playlist' and 'playlist' in extra_info)
1633 or extract_flat is True):
ecb54191 1634 info_copy = ie_result.copy()
6033d980 1635 ie = try_get(ie_result.get('ie_key'), self.get_info_extractor)
360167b9 1636 if ie and not ie_result.get('id'):
4614bc22 1637 info_copy['id'] = ie.get_temp_id(ie_result['url'])
6033d980 1638 self.add_default_extra_info(info_copy, ie, ie_result['url'])
4614bc22 1639 self.add_extra_info(info_copy, extra_info)
b5475f11 1640 info_copy, _ = self.pre_process(info_copy)
94dc8604 1641 self._fill_common_fields(info_copy, False)
ecb54191 1642 self.__forced_printings(info_copy, self.prepare_filename(info_copy), incomplete=True)
415f8d51 1643 self._raise_pending_errors(info_copy)
4614bc22 1644 if self.params.get('force_write_download_archive', False):
1645 self.record_download_archive(info_copy)
e8ee972c
PH
1646 return ie_result
1647
8222d8de 1648 if result_type == 'video':
b6c45014 1649 self.add_extra_info(ie_result, extra_info)
9c2b75b5 1650 ie_result = self.process_video_result(ie_result, download=download)
415f8d51 1651 self._raise_pending_errors(ie_result)
28b0eb0f 1652 additional_urls = (ie_result or {}).get('additional_urls')
9c2b75b5 1653 if additional_urls:
e9f4ccd1 1654 # TODO: Improve MetadataParserPP to allow setting a list
14f25df2 1655 if isinstance(additional_urls, str):
9c2b75b5 1656 additional_urls = [additional_urls]
1657 self.to_screen(
1658 '[info] %s: %d additional URL(s) requested' % (ie_result['id'], len(additional_urls)))
1659 self.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls))
1660 ie_result['additional_entries'] = [
1661 self.extract_info(
b69fd25c 1662 url, download, extra_info=extra_info,
9c2b75b5 1663 force_generic_extractor=self.params.get('force_generic_extractor'))
1664 for url in additional_urls
1665 ]
1666 return ie_result
8222d8de
JMF
1667 elif result_type == 'url':
1668 # We have to add extra_info to the results because it may be
1669 # contained in a playlist
07cce701 1670 return self.extract_info(
1671 ie_result['url'], download,
1672 ie_key=ie_result.get('ie_key'),
1673 extra_info=extra_info)
7fc3fa05
PH
1674 elif result_type == 'url_transparent':
1675 # Use the information from the embedding page
1676 info = self.extract_info(
1677 ie_result['url'], ie_key=ie_result.get('ie_key'),
1678 extra_info=extra_info, download=False, process=False)
1679
1640eb09
S
1680 # extract_info may return None when ignoreerrors is enabled and
1681 # extraction failed with an error, don't crash and return early
1682 # in this case
1683 if not info:
1684 return info
1685
3975b4d2 1686 exempted_fields = {'_type', 'url', 'ie_key'}
1687 if not ie_result.get('section_end') and ie_result.get('section_start') is None:
1688 # For video clips, the id etc of the clip extractor should be used
1689 exempted_fields |= {'id', 'extractor', 'extractor_key'}
1690
412c617d 1691 new_result = info.copy()
3975b4d2 1692 new_result.update(filter_dict(ie_result, lambda k, v: v is not None and k not in exempted_fields))
7fc3fa05 1693
0563f7ac
S
1694 # Extracted info may not be a video result (i.e.
1695 # info.get('_type', 'video') != video) but rather an url or
1696 # url_transparent. In such cases outer metadata (from ie_result)
1697 # should be propagated to inner one (info). For this to happen
1698 # _type of info should be overridden with url_transparent. This
067aa17e 1699 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
0563f7ac
S
1700 if new_result.get('_type') == 'url':
1701 new_result['_type'] = 'url_transparent'
7fc3fa05
PH
1702
1703 return self.process_ie_result(
1704 new_result, download=download, extra_info=extra_info)
40fcba5e 1705 elif result_type in ('playlist', 'multi_video'):
30a074c2 1706 # Protect from infinite recursion due to recursively nested playlists
1707 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
0bd5a039 1708 webpage_url = ie_result.get('webpage_url') # Playlists maynot have webpage_url
1709 if webpage_url and webpage_url in self._playlist_urls:
7e85e872 1710 self.to_screen(
30a074c2 1711 '[download] Skipping already downloaded playlist: %s'
1712 % ie_result.get('title') or ie_result.get('id'))
1713 return
7e85e872 1714
30a074c2 1715 self._playlist_level += 1
1716 self._playlist_urls.add(webpage_url)
03f83004 1717 self._fill_common_fields(ie_result, False)
bc516a3f 1718 self._sanitize_thumbnails(ie_result)
30a074c2 1719 try:
1720 return self.__process_playlist(ie_result, download)
1721 finally:
1722 self._playlist_level -= 1
1723 if not self._playlist_level:
1724 self._playlist_urls.clear()
8222d8de 1725 elif result_type == 'compat_list':
c9bf4114
PH
1726 self.report_warning(
1727 'Extractor %s returned a compat_list result. '
1728 'It needs to be updated.' % ie_result.get('extractor'))
5f6a1245 1729
8222d8de 1730 def _fixup(r):
b868936c 1731 self.add_extra_info(r, {
1732 'extractor': ie_result['extractor'],
1733 'webpage_url': ie_result['webpage_url'],
1734 'webpage_url_basename': url_basename(ie_result['webpage_url']),
0bb322b9 1735 'webpage_url_domain': get_domain(ie_result['webpage_url']),
b868936c 1736 'extractor_key': ie_result['extractor_key'],
1737 })
8222d8de
JMF
1738 return r
1739 ie_result['entries'] = [
b6c45014 1740 self.process_ie_result(_fixup(r), download, extra_info)
8222d8de
JMF
1741 for r in ie_result['entries']
1742 ]
1743 return ie_result
1744 else:
1745 raise Exception('Invalid result type: %s' % result_type)
1746
e92caff5 1747 def _ensure_dir_exists(self, path):
1748 return make_dir(path, self.report_error)
1749
3b603dbd 1750 @staticmethod
3bec830a 1751 def _playlist_infodict(ie_result, strict=False, **kwargs):
1752 info = {
1753 'playlist_count': ie_result.get('playlist_count'),
3b603dbd 1754 'playlist': ie_result.get('title') or ie_result.get('id'),
1755 'playlist_id': ie_result.get('id'),
1756 'playlist_title': ie_result.get('title'),
1757 'playlist_uploader': ie_result.get('uploader'),
1758 'playlist_uploader_id': ie_result.get('uploader_id'),
3b603dbd 1759 **kwargs,
1760 }
3bec830a 1761 if strict:
1762 return info
0bd5a039 1763 if ie_result.get('webpage_url'):
1764 info.update({
1765 'webpage_url': ie_result['webpage_url'],
1766 'webpage_url_basename': url_basename(ie_result['webpage_url']),
1767 'webpage_url_domain': get_domain(ie_result['webpage_url']),
1768 })
3bec830a 1769 return {
1770 **info,
1771 'playlist_index': 0,
1772 '__last_playlist_index': max(ie_result['requested_entries'] or (0, 0)),
1773 'extractor': ie_result['extractor'],
3bec830a 1774 'extractor_key': ie_result['extractor_key'],
1775 }
3b603dbd 1776
30a074c2 1777 def __process_playlist(self, ie_result, download):
7e88d7d7 1778 """Process each entry in the playlist"""
f5ea4748 1779 assert ie_result['_type'] in ('playlist', 'multi_video')
1780
3bec830a 1781 common_info = self._playlist_infodict(ie_result, strict=True)
3955b207 1782 title = common_info.get('playlist') or '<Untitled>'
3bec830a 1783 if self._match_entry(common_info, incomplete=True) is not None:
1784 return
c6e07cf1 1785 self.to_screen(f'[download] Downloading {ie_result["_type"]}: {title}')
f0d785d3 1786
7e88d7d7 1787 all_entries = PlaylistEntries(self, ie_result)
7e9a6125 1788 entries = orderedSet(all_entries.get_requested_items(), lazy=True)
1789
1790 lazy = self.params.get('lazy_playlist')
1791 if lazy:
1792 resolved_entries, n_entries = [], 'N/A'
1793 ie_result['requested_entries'], ie_result['entries'] = None, None
1794 else:
1795 entries = resolved_entries = list(entries)
1796 n_entries = len(resolved_entries)
1797 ie_result['requested_entries'], ie_result['entries'] = tuple(zip(*resolved_entries)) or ([], [])
1798 if not ie_result.get('playlist_count'):
1799 # Better to do this after potentially exhausting entries
1800 ie_result['playlist_count'] = all_entries.get_full_count()
498f5606 1801
0647d925 1802 extra = self._playlist_infodict(ie_result, n_entries=int_or_none(n_entries))
1803 ie_copy = collections.ChainMap(ie_result, extra)
3bec830a 1804
e08a85d8 1805 _infojson_written = False
0bfc53d0 1806 write_playlist_files = self.params.get('allow_playlist_files', True)
1807 if write_playlist_files and self.params.get('list_thumbnails'):
1808 self.list_thumbnails(ie_result)
1809 if write_playlist_files and not self.params.get('simulate'):
e08a85d8 1810 _infojson_written = self._write_info_json(
1811 'playlist', ie_result, self.prepare_filename(ie_copy, 'pl_infojson'))
1812 if _infojson_written is None:
80c03fa9 1813 return
1814 if self._write_description('playlist', ie_result,
1815 self.prepare_filename(ie_copy, 'pl_description')) is None:
1816 return
681de68e 1817 # TODO: This should be passed to ThumbnailsConvertor if necessary
3bec830a 1818 self._write_thumbnails('playlist', ie_result, self.prepare_filename(ie_copy, 'pl_thumbnail'))
30a074c2 1819
7e9a6125 1820 if lazy:
1821 if self.params.get('playlistreverse') or self.params.get('playlistrandom'):
1822 self.report_warning('playlistreverse and playlistrandom are not supported with lazy_playlist', only_once=True)
1823 elif self.params.get('playlistreverse'):
1824 entries.reverse()
1825 elif self.params.get('playlistrandom'):
30a074c2 1826 random.shuffle(entries)
1827
bc5c2f8a 1828 self.to_screen(f'[{ie_result["extractor"]}] Playlist {title}: Downloading {n_entries} items'
7e88d7d7 1829 f'{format_field(ie_result, "playlist_count", " of %s")}')
30a074c2 1830
134c913c 1831 keep_resolved_entries = self.params.get('extract_flat') != 'discard'
1832 if self.params.get('extract_flat') == 'discard_in_playlist':
1833 keep_resolved_entries = ie_result['_type'] != 'playlist'
1834 if keep_resolved_entries:
1835 self.write_debug('The information of all playlist entries will be held in memory')
1836
26e2805c 1837 failures = 0
1838 max_failures = self.params.get('skip_playlist_after_errors') or float('inf')
7e9a6125 1839 for i, (playlist_index, entry) in enumerate(entries):
1840 if lazy:
1841 resolved_entries.append((playlist_index, entry))
3bec830a 1842 if not entry:
7e88d7d7 1843 continue
1844
7e88d7d7 1845 entry['__x_forwarded_for_ip'] = ie_result.get('__x_forwarded_for_ip')
7e9a6125 1846 if not lazy and 'playlist-index' in self.params.get('compat_opts', []):
1847 playlist_index = ie_result['requested_entries'][i]
1848
0647d925 1849 entry_copy = collections.ChainMap(entry, {
3bec830a 1850 **common_info,
3955b207 1851 'n_entries': int_or_none(n_entries),
71729754 1852 'playlist_index': playlist_index,
7e9a6125 1853 'playlist_autonumber': i + 1,
0647d925 1854 })
3bec830a 1855
0647d925 1856 if self._match_entry(entry_copy, incomplete=True) is not None:
f0ad6f8c 1857 # For compatabilty with youtube-dl. See https://github.com/yt-dlp/yt-dlp/issues/4369
1858 resolved_entries[i] = (playlist_index, NO_DEFAULT)
3bec830a 1859 continue
1860
bc5c2f8a 1861 self.to_screen('[download] Downloading item %s of %s' % (
3bec830a 1862 self._format_screen(i + 1, self.Styles.ID), self._format_screen(n_entries, self.Styles.EMPHASIS)))
1863
a6ca61d4 1864 extra.update({
1865 'playlist_index': playlist_index,
1866 'playlist_autonumber': i + 1,
1867 })
3bec830a 1868 entry_result = self.__process_iterable_entry(entry, download, extra)
26e2805c 1869 if not entry_result:
1870 failures += 1
1871 if failures >= max_failures:
1872 self.report_error(
7e88d7d7 1873 f'Skipping the remaining entries in playlist "{title}" since {failures} items failed extraction')
26e2805c 1874 break
134c913c 1875 if keep_resolved_entries:
1876 resolved_entries[i] = (playlist_index, entry_result)
7e88d7d7 1877
1878 # Update with processed data
f0ad6f8c 1879 ie_result['entries'] = [e for _, e in resolved_entries if e is not NO_DEFAULT]
bc5c2f8a 1880 ie_result['requested_entries'] = [i for i, e in resolved_entries if e is not NO_DEFAULT]
1881 if ie_result['requested_entries'] == try_call(lambda: list(range(1, ie_result['playlist_count'] + 1))):
1882 # Do not set for full playlist
1883 ie_result.pop('requested_entries')
e08a85d8 1884
1885 # Write the updated info to json
cb96c5be 1886 if _infojson_written is True and self._write_info_json(
e08a85d8 1887 'updated playlist', ie_result,
1888 self.prepare_filename(ie_copy, 'pl_infojson'), overwrite=True) is None:
1889 return
ca30f449 1890
ed5835b4 1891 ie_result = self.run_all_pps('playlist', ie_result)
7e88d7d7 1892 self.to_screen(f'[download] Finished downloading playlist: {title}')
30a074c2 1893 return ie_result
1894
7e88d7d7 1895 @_handle_extraction_exceptions
a0566bbf 1896 def __process_iterable_entry(self, entry, download, extra_info):
1897 return self.process_ie_result(
1898 entry, download=download, extra_info=extra_info)
1899
67134eab
JMF
1900 def _build_format_filter(self, filter_spec):
1901 " Returns a function to filter the formats according to the filter_spec "
083c9df9
PH
1902
1903 OPERATORS = {
1904 '<': operator.lt,
1905 '<=': operator.le,
1906 '>': operator.gt,
1907 '>=': operator.ge,
1908 '=': operator.eq,
1909 '!=': operator.ne,
1910 }
67134eab 1911 operator_rex = re.compile(r'''(?x)\s*
187986a8 1912 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)\s*
1913 (?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1914 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)\s*
083c9df9 1915 ''' % '|'.join(map(re.escape, OPERATORS.keys())))
187986a8 1916 m = operator_rex.fullmatch(filter_spec)
9ddb6925
S
1917 if m:
1918 try:
1919 comparison_value = int(m.group('value'))
1920 except ValueError:
1921 comparison_value = parse_filesize(m.group('value'))
1922 if comparison_value is None:
1923 comparison_value = parse_filesize(m.group('value') + 'B')
1924 if comparison_value is None:
1925 raise ValueError(
1926 'Invalid value %r in format specification %r' % (
67134eab 1927 m.group('value'), filter_spec))
9ddb6925
S
1928 op = OPERATORS[m.group('op')]
1929
083c9df9 1930 if not m:
9ddb6925
S
1931 STR_OPERATORS = {
1932 '=': operator.eq,
10d33b34
YCH
1933 '^=': lambda attr, value: attr.startswith(value),
1934 '$=': lambda attr, value: attr.endswith(value),
1935 '*=': lambda attr, value: value in attr,
1ce9a3cb 1936 '~=': lambda attr, value: value.search(attr) is not None
9ddb6925 1937 }
187986a8 1938 str_operator_rex = re.compile(r'''(?x)\s*
1939 (?P<key>[a-zA-Z0-9._-]+)\s*
1ce9a3cb
LF
1940 (?P<negation>!\s*)?(?P<op>%s)\s*(?P<none_inclusive>\?\s*)?
1941 (?P<quote>["'])?
1942 (?P<value>(?(quote)(?:(?!(?P=quote))[^\\]|\\.)+|[\w.-]+))
1943 (?(quote)(?P=quote))\s*
9ddb6925 1944 ''' % '|'.join(map(re.escape, STR_OPERATORS.keys())))
187986a8 1945 m = str_operator_rex.fullmatch(filter_spec)
9ddb6925 1946 if m:
1ce9a3cb
LF
1947 if m.group('op') == '~=':
1948 comparison_value = re.compile(m.group('value'))
1949 else:
1950 comparison_value = re.sub(r'''\\([\\"'])''', r'\1', m.group('value'))
2cc779f4
S
1951 str_op = STR_OPERATORS[m.group('op')]
1952 if m.group('negation'):
e118a879 1953 op = lambda attr, value: not str_op(attr, value)
2cc779f4
S
1954 else:
1955 op = str_op
083c9df9 1956
9ddb6925 1957 if not m:
187986a8 1958 raise SyntaxError('Invalid filter specification %r' % filter_spec)
083c9df9
PH
1959
1960 def _filter(f):
1961 actual_value = f.get(m.group('key'))
1962 if actual_value is None:
1963 return m.group('none_inclusive')
1964 return op(actual_value, comparison_value)
67134eab
JMF
1965 return _filter
1966
9f1a1c36 1967 def _check_formats(self, formats):
1968 for f in formats:
1969 self.to_screen('[info] Testing format %s' % f['format_id'])
75689fe5 1970 path = self.get_output_path('temp')
1971 if not self._ensure_dir_exists(f'{path}/'):
1972 continue
1973 temp_file = tempfile.NamedTemporaryFile(suffix='.tmp', delete=False, dir=path or None)
9f1a1c36 1974 temp_file.close()
1975 try:
1976 success, _ = self.dl(temp_file.name, f, test=True)
8a82af35 1977 except (DownloadError, OSError, ValueError) + network_exceptions:
9f1a1c36 1978 success = False
1979 finally:
1980 if os.path.exists(temp_file.name):
1981 try:
1982 os.remove(temp_file.name)
1983 except OSError:
1984 self.report_warning('Unable to delete temporary file "%s"' % temp_file.name)
1985 if success:
1986 yield f
1987 else:
1988 self.to_screen('[info] Unable to download format %s. Skipping...' % f['format_id'])
1989
0017d9ad 1990 def _default_format_spec(self, info_dict, download=True):
0017d9ad 1991
af0f7428
S
1992 def can_merge():
1993 merger = FFmpegMergerPP(self)
1994 return merger.available and merger.can_merge()
1995
91ebc640 1996 prefer_best = (
b7b04c78 1997 not self.params.get('simulate')
91ebc640 1998 and download
1999 and (
2000 not can_merge()
21633673 2001 or info_dict.get('is_live') and not self.params.get('live_from_start')
bf1824b3 2002 or self.params['outtmpl']['default'] == '-'))
53ed7066 2003 compat = (
2004 prefer_best
2005 or self.params.get('allow_multiple_audio_streams', False)
8a82af35 2006 or 'format-spec' in self.params['compat_opts'])
91ebc640 2007
2008 return (
53ed7066 2009 'best/bestvideo+bestaudio' if prefer_best
2010 else 'bestvideo*+bestaudio/best' if not compat
91ebc640 2011 else 'bestvideo+bestaudio/best')
0017d9ad 2012
67134eab
JMF
2013 def build_format_selector(self, format_spec):
2014 def syntax_error(note, start):
2015 message = (
2016 'Invalid format specification: '
86e5f3ed 2017 '{}\n\t{}\n\t{}^'.format(note, format_spec, ' ' * start[1]))
67134eab
JMF
2018 return SyntaxError(message)
2019
2020 PICKFIRST = 'PICKFIRST'
2021 MERGE = 'MERGE'
2022 SINGLE = 'SINGLE'
0130afb7 2023 GROUP = 'GROUP'
67134eab
JMF
2024 FormatSelector = collections.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
2025
91ebc640 2026 allow_multiple_streams = {'audio': self.params.get('allow_multiple_audio_streams', False),
2027 'video': self.params.get('allow_multiple_video_streams', False)}
909d24dd 2028
9f1a1c36 2029 check_formats = self.params.get('check_formats') == 'selected'
e8e73840 2030
67134eab
JMF
2031 def _parse_filter(tokens):
2032 filter_parts = []
2033 for type, string, start, _, _ in tokens:
2034 if type == tokenize.OP and string == ']':
2035 return ''.join(filter_parts)
2036 else:
2037 filter_parts.append(string)
2038
232541df 2039 def _remove_unused_ops(tokens):
62b58c09
L
2040 # Remove operators that we don't use and join them with the surrounding strings.
2041 # E.g. 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
232541df
JMF
2042 ALLOWED_OPS = ('/', '+', ',', '(', ')')
2043 last_string, last_start, last_end, last_line = None, None, None, None
2044 for type, string, start, end, line in tokens:
2045 if type == tokenize.OP and string == '[':
2046 if last_string:
2047 yield tokenize.NAME, last_string, last_start, last_end, last_line
2048 last_string = None
2049 yield type, string, start, end, line
2050 # everything inside brackets will be handled by _parse_filter
2051 for type, string, start, end, line in tokens:
2052 yield type, string, start, end, line
2053 if type == tokenize.OP and string == ']':
2054 break
2055 elif type == tokenize.OP and string in ALLOWED_OPS:
2056 if last_string:
2057 yield tokenize.NAME, last_string, last_start, last_end, last_line
2058 last_string = None
2059 yield type, string, start, end, line
2060 elif type in [tokenize.NAME, tokenize.NUMBER, tokenize.OP]:
2061 if not last_string:
2062 last_string = string
2063 last_start = start
2064 last_end = end
2065 else:
2066 last_string += string
2067 if last_string:
2068 yield tokenize.NAME, last_string, last_start, last_end, last_line
2069
cf2ac6df 2070 def _parse_format_selection(tokens, inside_merge=False, inside_choice=False, inside_group=False):
67134eab
JMF
2071 selectors = []
2072 current_selector = None
2073 for type, string, start, _, _ in tokens:
2074 # ENCODING is only defined in python 3.x
2075 if type == getattr(tokenize, 'ENCODING', None):
2076 continue
2077 elif type in [tokenize.NAME, tokenize.NUMBER]:
2078 current_selector = FormatSelector(SINGLE, string, [])
2079 elif type == tokenize.OP:
cf2ac6df
JMF
2080 if string == ')':
2081 if not inside_group:
2082 # ')' will be handled by the parentheses group
2083 tokens.restore_last_token()
67134eab 2084 break
cf2ac6df 2085 elif inside_merge and string in ['/', ',']:
0130afb7
JMF
2086 tokens.restore_last_token()
2087 break
cf2ac6df
JMF
2088 elif inside_choice and string == ',':
2089 tokens.restore_last_token()
2090 break
2091 elif string == ',':
0a31a350
JMF
2092 if not current_selector:
2093 raise syntax_error('"," must follow a format selector', start)
67134eab
JMF
2094 selectors.append(current_selector)
2095 current_selector = None
2096 elif string == '/':
d96d604e
JMF
2097 if not current_selector:
2098 raise syntax_error('"/" must follow a format selector', start)
67134eab 2099 first_choice = current_selector
cf2ac6df 2100 second_choice = _parse_format_selection(tokens, inside_choice=True)
f5f4a27a 2101 current_selector = FormatSelector(PICKFIRST, (first_choice, second_choice), [])
67134eab
JMF
2102 elif string == '[':
2103 if not current_selector:
2104 current_selector = FormatSelector(SINGLE, 'best', [])
2105 format_filter = _parse_filter(tokens)
2106 current_selector.filters.append(format_filter)
0130afb7
JMF
2107 elif string == '(':
2108 if current_selector:
2109 raise syntax_error('Unexpected "("', start)
cf2ac6df
JMF
2110 group = _parse_format_selection(tokens, inside_group=True)
2111 current_selector = FormatSelector(GROUP, group, [])
67134eab 2112 elif string == '+':
d03cfdce 2113 if not current_selector:
2114 raise syntax_error('Unexpected "+"', start)
2115 selector_1 = current_selector
2116 selector_2 = _parse_format_selection(tokens, inside_merge=True)
2117 if not selector_2:
2118 raise syntax_error('Expected a selector', start)
2119 current_selector = FormatSelector(MERGE, (selector_1, selector_2), [])
67134eab 2120 else:
86e5f3ed 2121 raise syntax_error(f'Operator not recognized: "{string}"', start)
67134eab
JMF
2122 elif type == tokenize.ENDMARKER:
2123 break
2124 if current_selector:
2125 selectors.append(current_selector)
2126 return selectors
2127
f8d4ad9a 2128 def _merge(formats_pair):
2129 format_1, format_2 = formats_pair
2130
2131 formats_info = []
2132 formats_info.extend(format_1.get('requested_formats', (format_1,)))
2133 formats_info.extend(format_2.get('requested_formats', (format_2,)))
2134
2135 if not allow_multiple_streams['video'] or not allow_multiple_streams['audio']:
551f9388 2136 get_no_more = {'video': False, 'audio': False}
f8d4ad9a 2137 for (i, fmt_info) in enumerate(formats_info):
551f9388 2138 if fmt_info.get('acodec') == fmt_info.get('vcodec') == 'none':
2139 formats_info.pop(i)
2140 continue
2141 for aud_vid in ['audio', 'video']:
f8d4ad9a 2142 if not allow_multiple_streams[aud_vid] and fmt_info.get(aud_vid[0] + 'codec') != 'none':
2143 if get_no_more[aud_vid]:
2144 formats_info.pop(i)
f5510afe 2145 break
f8d4ad9a 2146 get_no_more[aud_vid] = True
2147
2148 if len(formats_info) == 1:
2149 return formats_info[0]
2150
2151 video_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('vcodec') != 'none']
2152 audio_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('acodec') != 'none']
2153
2154 the_only_video = video_fmts[0] if len(video_fmts) == 1 else None
2155 the_only_audio = audio_fmts[0] if len(audio_fmts) == 1 else None
2156
fc61aff4
LL
2157 output_ext = get_compatible_ext(
2158 vcodecs=[f.get('vcodec') for f in video_fmts],
2159 acodecs=[f.get('acodec') for f in audio_fmts],
2160 vexts=[f['ext'] for f in video_fmts],
2161 aexts=[f['ext'] for f in audio_fmts],
2162 preferences=(try_call(lambda: self.params['merge_output_format'].split('/'))
2163 or self.params.get('prefer_free_formats') and ('webm', 'mkv')))
f8d4ad9a 2164
975a0d0d 2165 filtered = lambda *keys: filter(None, (traverse_obj(fmt, *keys) for fmt in formats_info))
2166
f8d4ad9a 2167 new_dict = {
2168 'requested_formats': formats_info,
975a0d0d 2169 'format': '+'.join(filtered('format')),
2170 'format_id': '+'.join(filtered('format_id')),
f8d4ad9a 2171 'ext': output_ext,
975a0d0d 2172 'protocol': '+'.join(map(determine_protocol, formats_info)),
093a1710 2173 'language': '+'.join(orderedSet(filtered('language'))) or None,
2174 'format_note': '+'.join(orderedSet(filtered('format_note'))) or None,
2175 'filesize_approx': sum(filtered('filesize', 'filesize_approx')) or None,
975a0d0d 2176 'tbr': sum(filtered('tbr', 'vbr', 'abr')),
f8d4ad9a 2177 }
2178
2179 if the_only_video:
2180 new_dict.update({
2181 'width': the_only_video.get('width'),
2182 'height': the_only_video.get('height'),
2183 'resolution': the_only_video.get('resolution') or self.format_resolution(the_only_video),
2184 'fps': the_only_video.get('fps'),
49a57e70 2185 'dynamic_range': the_only_video.get('dynamic_range'),
f8d4ad9a 2186 'vcodec': the_only_video.get('vcodec'),
2187 'vbr': the_only_video.get('vbr'),
2188 'stretched_ratio': the_only_video.get('stretched_ratio'),
2189 })
2190
2191 if the_only_audio:
2192 new_dict.update({
2193 'acodec': the_only_audio.get('acodec'),
2194 'abr': the_only_audio.get('abr'),
975a0d0d 2195 'asr': the_only_audio.get('asr'),
b8ed0f15 2196 'audio_channels': the_only_audio.get('audio_channels')
f8d4ad9a 2197 })
2198
2199 return new_dict
2200
e8e73840 2201 def _check_formats(formats):
981052c9 2202 if not check_formats:
2203 yield from formats
b5ac45b1 2204 return
9f1a1c36 2205 yield from self._check_formats(formats)
e8e73840 2206
67134eab 2207 def _build_selector_function(selector):
909d24dd 2208 if isinstance(selector, list): # ,
67134eab
JMF
2209 fs = [_build_selector_function(s) for s in selector]
2210
317f7ab6 2211 def selector_function(ctx):
67134eab 2212 for f in fs:
981052c9 2213 yield from f(ctx)
67134eab 2214 return selector_function
909d24dd 2215
2216 elif selector.type == GROUP: # ()
0130afb7 2217 selector_function = _build_selector_function(selector.selector)
909d24dd 2218
2219 elif selector.type == PICKFIRST: # /
67134eab
JMF
2220 fs = [_build_selector_function(s) for s in selector.selector]
2221
317f7ab6 2222 def selector_function(ctx):
67134eab 2223 for f in fs:
317f7ab6 2224 picked_formats = list(f(ctx))
67134eab
JMF
2225 if picked_formats:
2226 return picked_formats
2227 return []
67134eab 2228
981052c9 2229 elif selector.type == MERGE: # +
2230 selector_1, selector_2 = map(_build_selector_function, selector.selector)
2231
2232 def selector_function(ctx):
adbc4ec4 2233 for pair in itertools.product(selector_1(ctx), selector_2(ctx)):
981052c9 2234 yield _merge(pair)
2235
909d24dd 2236 elif selector.type == SINGLE: # atom
598d185d 2237 format_spec = selector.selector or 'best'
909d24dd 2238
f8d4ad9a 2239 # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
909d24dd 2240 if format_spec == 'all':
2241 def selector_function(ctx):
9222c381 2242 yield from _check_formats(ctx['formats'][::-1])
f8d4ad9a 2243 elif format_spec == 'mergeall':
2244 def selector_function(ctx):
316f2650 2245 formats = list(_check_formats(
2246 f for f in ctx['formats'] if f.get('vcodec') != 'none' or f.get('acodec') != 'none'))
e01d6aa4 2247 if not formats:
2248 return
921b76ca 2249 merged_format = formats[-1]
2250 for f in formats[-2::-1]:
f8d4ad9a 2251 merged_format = _merge((merged_format, f))
2252 yield merged_format
909d24dd 2253
2254 else:
85e801a9 2255 format_fallback, seperate_fallback, format_reverse, format_idx = False, None, True, 1
eff63539 2256 mobj = re.match(
2257 r'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
2258 format_spec)
2259 if mobj is not None:
2260 format_idx = int_or_none(mobj.group('n'), default=1)
e8e73840 2261 format_reverse = mobj.group('bw')[0] == 'b'
eff63539 2262 format_type = (mobj.group('type') or [None])[0]
2263 not_format_type = {'v': 'a', 'a': 'v'}.get(format_type)
2264 format_modified = mobj.group('mod') is not None
909d24dd 2265
2266 format_fallback = not format_type and not format_modified # for b, w
8326b00a 2267 _filter_f = (
eff63539 2268 (lambda f: f.get('%scodec' % format_type) != 'none')
2269 if format_type and format_modified # bv*, ba*, wv*, wa*
2270 else (lambda f: f.get('%scodec' % not_format_type) == 'none')
2271 if format_type # bv, ba, wv, wa
2272 else (lambda f: f.get('vcodec') != 'none' and f.get('acodec') != 'none')
2273 if not format_modified # b, w
8326b00a 2274 else lambda f: True) # b*, w*
2275 filter_f = lambda f: _filter_f(f) and (
2276 f.get('vcodec') != 'none' or f.get('acodec') != 'none')
67134eab 2277 else:
48ee10ee 2278 if format_spec in self._format_selection_exts['audio']:
b11c04a8 2279 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none'
48ee10ee 2280 elif format_spec in self._format_selection_exts['video']:
b11c04a8 2281 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none' and f.get('vcodec') != 'none'
85e801a9 2282 seperate_fallback = lambda f: f.get('ext') == format_spec and f.get('vcodec') != 'none'
48ee10ee 2283 elif format_spec in self._format_selection_exts['storyboards']:
b11c04a8 2284 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') == 'none' and f.get('vcodec') == 'none'
2285 else:
b5ae35ee 2286 filter_f = lambda f: f.get('format_id') == format_spec # id
909d24dd 2287
2288 def selector_function(ctx):
2289 formats = list(ctx['formats'])
909d24dd 2290 matches = list(filter(filter_f, formats)) if filter_f is not None else formats
85e801a9 2291 if not matches:
2292 if format_fallback and ctx['incomplete_formats']:
2293 # for extractors with incomplete formats (audio only (soundcloud)
2294 # or video only (imgur)) best/worst will fallback to
2295 # best/worst {video,audio}-only format
2296 matches = formats
2297 elif seperate_fallback and not ctx['has_merged_format']:
2298 # for compatibility with youtube-dl when there is no pre-merged format
2299 matches = list(filter(seperate_fallback, formats))
981052c9 2300 matches = LazyList(_check_formats(matches[::-1 if format_reverse else 1]))
2301 try:
e8e73840 2302 yield matches[format_idx - 1]
4abea8ca 2303 except LazyList.IndexError:
981052c9 2304 return
083c9df9 2305
67134eab 2306 filters = [self._build_format_filter(f) for f in selector.filters]
083c9df9 2307
317f7ab6 2308 def final_selector(ctx):
adbc4ec4 2309 ctx_copy = dict(ctx)
67134eab 2310 for _filter in filters:
317f7ab6
S
2311 ctx_copy['formats'] = list(filter(_filter, ctx_copy['formats']))
2312 return selector_function(ctx_copy)
67134eab 2313 return final_selector
083c9df9 2314
0f06bcd7 2315 stream = io.BytesIO(format_spec.encode())
0130afb7 2316 try:
f9934b96 2317 tokens = list(_remove_unused_ops(tokenize.tokenize(stream.readline)))
0130afb7
JMF
2318 except tokenize.TokenError:
2319 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec)))
2320
86e5f3ed 2321 class TokenIterator:
0130afb7
JMF
2322 def __init__(self, tokens):
2323 self.tokens = tokens
2324 self.counter = 0
2325
2326 def __iter__(self):
2327 return self
2328
2329 def __next__(self):
2330 if self.counter >= len(self.tokens):
2331 raise StopIteration()
2332 value = self.tokens[self.counter]
2333 self.counter += 1
2334 return value
2335
2336 next = __next__
2337
2338 def restore_last_token(self):
2339 self.counter -= 1
2340
2341 parsed_selector = _parse_format_selection(iter(TokenIterator(tokens)))
67134eab 2342 return _build_selector_function(parsed_selector)
a9c58ad9 2343
e5660ee6 2344 def _calc_headers(self, info_dict):
8b7539d2 2345 res = merge_headers(self.params['http_headers'], info_dict.get('http_headers') or {})
e5660ee6 2346
c487cf00 2347 cookies = self._calc_cookies(info_dict['url'])
e5660ee6
JMF
2348 if cookies:
2349 res['Cookie'] = cookies
2350
0016b84e
S
2351 if 'X-Forwarded-For' not in res:
2352 x_forwarded_for_ip = info_dict.get('__x_forwarded_for_ip')
2353 if x_forwarded_for_ip:
2354 res['X-Forwarded-For'] = x_forwarded_for_ip
2355
e5660ee6
JMF
2356 return res
2357
c487cf00 2358 def _calc_cookies(self, url):
2359 pr = sanitized_Request(url)
e5660ee6 2360 self.cookiejar.add_cookie_header(pr)
662435f7 2361 return pr.get_header('Cookie')
e5660ee6 2362
9f1a1c36 2363 def _sort_thumbnails(self, thumbnails):
2364 thumbnails.sort(key=lambda t: (
2365 t.get('preference') if t.get('preference') is not None else -1,
2366 t.get('width') if t.get('width') is not None else -1,
2367 t.get('height') if t.get('height') is not None else -1,
2368 t.get('id') if t.get('id') is not None else '',
2369 t.get('url')))
2370
b0249bca 2371 def _sanitize_thumbnails(self, info_dict):
bc516a3f 2372 thumbnails = info_dict.get('thumbnails')
2373 if thumbnails is None:
2374 thumbnail = info_dict.get('thumbnail')
2375 if thumbnail:
2376 info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}]
9f1a1c36 2377 if not thumbnails:
2378 return
2379
2380 def check_thumbnails(thumbnails):
2381 for t in thumbnails:
2382 self.to_screen(f'[info] Testing thumbnail {t["id"]}')
2383 try:
2384 self.urlopen(HEADRequest(t['url']))
2385 except network_exceptions as err:
2386 self.to_screen(f'[info] Unable to connect to thumbnail {t["id"]} URL {t["url"]!r} - {err}. Skipping...')
2387 continue
2388 yield t
2389
2390 self._sort_thumbnails(thumbnails)
2391 for i, t in enumerate(thumbnails):
2392 if t.get('id') is None:
2393 t['id'] = '%d' % i
2394 if t.get('width') and t.get('height'):
2395 t['resolution'] = '%dx%d' % (t['width'], t['height'])
2396 t['url'] = sanitize_url(t['url'])
2397
2398 if self.params.get('check_formats') is True:
282f5709 2399 info_dict['thumbnails'] = LazyList(check_thumbnails(thumbnails[::-1]), reverse=True)
9f1a1c36 2400 else:
2401 info_dict['thumbnails'] = thumbnails
bc516a3f 2402
94dc8604 2403 def _fill_common_fields(self, info_dict, final=True):
03f83004 2404 # TODO: move sanitization here
94dc8604 2405 if final:
d4736fdb 2406 title = info_dict.get('title', NO_DEFAULT)
2407 if title is NO_DEFAULT:
03f83004
LNO
2408 raise ExtractorError('Missing "title" field in extractor result',
2409 video_id=info_dict['id'], ie=info_dict['extractor'])
d4736fdb 2410 info_dict['fulltitle'] = title
2411 if not title:
2412 if title == '':
2413 self.write_debug('Extractor gave empty title. Creating a generic title')
2414 else:
2415 self.report_warning('Extractor failed to obtain "title". Creating a generic title instead')
1d485a1a 2416 info_dict['title'] = f'{info_dict["extractor"].replace(":", "-")} video #{info_dict["id"]}'
03f83004
LNO
2417
2418 if info_dict.get('duration') is not None:
2419 info_dict['duration_string'] = formatSeconds(info_dict['duration'])
2420
2421 for ts_key, date_key in (
2422 ('timestamp', 'upload_date'),
2423 ('release_timestamp', 'release_date'),
2424 ('modified_timestamp', 'modified_date'),
2425 ):
2426 if info_dict.get(date_key) is None and info_dict.get(ts_key) is not None:
2427 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
2428 # see http://bugs.python.org/issue1646728)
19a03940 2429 with contextlib.suppress(ValueError, OverflowError, OSError):
03f83004
LNO
2430 upload_date = datetime.datetime.utcfromtimestamp(info_dict[ts_key])
2431 info_dict[date_key] = upload_date.strftime('%Y%m%d')
03f83004
LNO
2432
2433 live_keys = ('is_live', 'was_live')
2434 live_status = info_dict.get('live_status')
2435 if live_status is None:
2436 for key in live_keys:
2437 if info_dict.get(key) is False:
2438 continue
2439 if info_dict.get(key):
2440 live_status = key
2441 break
2442 if all(info_dict.get(key) is False for key in live_keys):
2443 live_status = 'not_live'
2444 if live_status:
2445 info_dict['live_status'] = live_status
2446 for key in live_keys:
2447 if info_dict.get(key) is None:
2448 info_dict[key] = (live_status == key)
a057779d 2449 if live_status == 'post_live':
2450 info_dict['was_live'] = True
03f83004
LNO
2451
2452 # Auto generate title fields corresponding to the *_number fields when missing
2453 # in order to always have clean titles. This is very common for TV series.
2454 for field in ('chapter', 'season', 'episode'):
94dc8604 2455 if final and info_dict.get('%s_number' % field) is not None and not info_dict.get(field):
03f83004
LNO
2456 info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field])
2457
415f8d51 2458 def _raise_pending_errors(self, info):
2459 err = info.pop('__pending_error', None)
2460 if err:
2461 self.report_error(err, tb=False)
2462
dd82ffea
JMF
2463 def process_video_result(self, info_dict, download=True):
2464 assert info_dict.get('_type', 'video') == 'video'
9c906919 2465 self._num_videos += 1
dd82ffea 2466
bec1fad2 2467 if 'id' not in info_dict:
fc08bdd6 2468 raise ExtractorError('Missing "id" field in extractor result', ie=info_dict['extractor'])
2469 elif not info_dict.get('id'):
2470 raise ExtractorError('Extractor failed to obtain "id"', ie=info_dict['extractor'])
455a15e2 2471
c9969434
S
2472 def report_force_conversion(field, field_not, conversion):
2473 self.report_warning(
2474 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
2475 % (field, field_not, conversion))
2476
2477 def sanitize_string_field(info, string_field):
2478 field = info.get(string_field)
14f25df2 2479 if field is None or isinstance(field, str):
c9969434
S
2480 return
2481 report_force_conversion(string_field, 'a string', 'string')
14f25df2 2482 info[string_field] = str(field)
c9969434
S
2483
2484 def sanitize_numeric_fields(info):
2485 for numeric_field in self._NUMERIC_FIELDS:
2486 field = info.get(numeric_field)
f9934b96 2487 if field is None or isinstance(field, (int, float)):
c9969434
S
2488 continue
2489 report_force_conversion(numeric_field, 'numeric', 'int')
2490 info[numeric_field] = int_or_none(field)
2491
2492 sanitize_string_field(info_dict, 'id')
2493 sanitize_numeric_fields(info_dict)
3975b4d2 2494 if info_dict.get('section_end') and info_dict.get('section_start') is not None:
2495 info_dict['duration'] = round(info_dict['section_end'] - info_dict['section_start'], 3)
4c3f8c3f 2496 if (info_dict.get('duration') or 0) <= 0 and info_dict.pop('duration', None):
50e93e03 2497 self.report_warning('"duration" field is negative, there is an error in extractor')
be6217b2 2498
9eef7c4e 2499 chapters = info_dict.get('chapters') or []
a3976e07 2500 if chapters and chapters[0].get('start_time'):
2501 chapters.insert(0, {'start_time': 0})
2502
9eef7c4e 2503 dummy_chapter = {'end_time': 0, 'start_time': info_dict.get('duration')}
a3976e07 2504 for idx, (prev, current, next_) in enumerate(zip(
2505 (dummy_chapter, *chapters), chapters, (*chapters[1:], dummy_chapter)), 1):
9eef7c4e 2506 if current.get('start_time') is None:
2507 current['start_time'] = prev.get('end_time')
2508 if not current.get('end_time'):
2509 current['end_time'] = next_.get('start_time')
a3976e07 2510 if not current.get('title'):
2511 current['title'] = f'<Untitled Chapter {idx}>'
9eef7c4e 2512
dd82ffea
JMF
2513 if 'playlist' not in info_dict:
2514 # It isn't part of a playlist
2515 info_dict['playlist'] = None
2516 info_dict['playlist_index'] = None
2517
bc516a3f 2518 self._sanitize_thumbnails(info_dict)
d5519808 2519
536a55da 2520 thumbnail = info_dict.get('thumbnail')
bc516a3f 2521 thumbnails = info_dict.get('thumbnails')
536a55da
S
2522 if thumbnail:
2523 info_dict['thumbnail'] = sanitize_url(thumbnail)
2524 elif thumbnails:
d5519808
PH
2525 info_dict['thumbnail'] = thumbnails[-1]['url']
2526
ae30b840 2527 if info_dict.get('display_id') is None and 'id' in info_dict:
0afef30b
PH
2528 info_dict['display_id'] = info_dict['id']
2529
03f83004 2530 self._fill_common_fields(info_dict)
33d2fc2f 2531
05108a49
S
2532 for cc_kind in ('subtitles', 'automatic_captions'):
2533 cc = info_dict.get(cc_kind)
2534 if cc:
2535 for _, subtitle in cc.items():
2536 for subtitle_format in subtitle:
2537 if subtitle_format.get('url'):
2538 subtitle_format['url'] = sanitize_url(subtitle_format['url'])
2539 if subtitle_format.get('ext') is None:
2540 subtitle_format['ext'] = determine_ext(subtitle_format['url']).lower()
2541
2542 automatic_captions = info_dict.get('automatic_captions')
4bba3716 2543 subtitles = info_dict.get('subtitles')
4bba3716 2544
360e1ca5 2545 info_dict['requested_subtitles'] = self.process_subtitles(
05108a49 2546 info_dict['id'], subtitles, automatic_captions)
a504ced0 2547
aebb4f4b 2548 formats = self._get_formats(info_dict)
dd82ffea 2549
0a5a191a 2550 # or None ensures --clean-infojson removes it
2551 info_dict['_has_drm'] = any(f.get('has_drm') for f in formats) or None
88acdbc2 2552 if not self.params.get('allow_unplayable_formats'):
2553 formats = [f for f in formats if not f.get('has_drm')]
17ffed18 2554
2555 if formats and all(f.get('acodec') == f.get('vcodec') == 'none' for f in formats):
2556 self.report_warning(
2557 f'{"This video is DRM protected and " if info_dict["_has_drm"] else ""}'
2558 'only images are available for download. Use --list-formats to see them'.capitalize())
88acdbc2 2559
319b6059 2560 get_from_start = not info_dict.get('is_live') or bool(self.params.get('live_from_start'))
2561 if not get_from_start:
2562 info_dict['title'] += ' ' + datetime.datetime.now().strftime('%Y-%m-%d %H:%M')
2563 if info_dict.get('is_live') and formats:
adbc4ec4 2564 formats = [f for f in formats if bool(f.get('is_from_start')) == get_from_start]
319b6059 2565 if get_from_start and not formats:
a44ca5a4 2566 self.raise_no_formats(info_dict, msg=(
2567 '--live-from-start is passed, but there are no formats that can be downloaded from the start. '
2568 'If you want to download from the current time, use --no-live-from-start'))
adbc4ec4 2569
73af5cc8
S
2570 def is_wellformed(f):
2571 url = f.get('url')
a5ac0c47 2572 if not url:
73af5cc8
S
2573 self.report_warning(
2574 '"url" field is missing or empty - skipping format, '
2575 'there is an error in extractor')
a5ac0c47
S
2576 return False
2577 if isinstance(url, bytes):
2578 sanitize_string_field(f, 'url')
2579 return True
73af5cc8
S
2580
2581 # Filter out malformed formats for better extraction robustness
1ac7f461 2582 formats = list(filter(is_wellformed, formats or []))
2583
2584 if not formats:
2585 self.raise_no_formats(info_dict)
73af5cc8 2586
181c7053
S
2587 formats_dict = {}
2588
dd82ffea 2589 # We check that all the formats have the format and format_id fields
db95dc13 2590 for i, format in enumerate(formats):
c9969434
S
2591 sanitize_string_field(format, 'format_id')
2592 sanitize_numeric_fields(format)
dcf77cf1 2593 format['url'] = sanitize_url(format['url'])
e74e3b63 2594 if not format.get('format_id'):
14f25df2 2595 format['format_id'] = str(i)
e2effb08
S
2596 else:
2597 # Sanitize format_id from characters used in format selector expression
ec85ded8 2598 format['format_id'] = re.sub(r'[\s,/+\[\]()]', '_', format['format_id'])
181c7053
S
2599 format_id = format['format_id']
2600 if format_id not in formats_dict:
2601 formats_dict[format_id] = []
2602 formats_dict[format_id].append(format)
2603
2604 # Make sure all formats have unique format_id
03b4de72 2605 common_exts = set(itertools.chain(*self._format_selection_exts.values()))
181c7053 2606 for format_id, ambiguous_formats in formats_dict.items():
48ee10ee 2607 ambigious_id = len(ambiguous_formats) > 1
2608 for i, format in enumerate(ambiguous_formats):
2609 if ambigious_id:
181c7053 2610 format['format_id'] = '%s-%d' % (format_id, i)
48ee10ee 2611 if format.get('ext') is None:
2612 format['ext'] = determine_ext(format['url']).lower()
2613 # Ensure there is no conflict between id and ext in format selection
2614 # See https://github.com/yt-dlp/yt-dlp/issues/1282
2615 if format['format_id'] != format['ext'] and format['format_id'] in common_exts:
2616 format['format_id'] = 'f%s' % format['format_id']
181c7053
S
2617
2618 for i, format in enumerate(formats):
8c51aa65 2619 if format.get('format') is None:
6febd1c1 2620 format['format'] = '{id} - {res}{note}'.format(
8c51aa65
JMF
2621 id=format['format_id'],
2622 res=self.format_resolution(format),
b868936c 2623 note=format_field(format, 'format_note', ' (%s)'),
8c51aa65 2624 )
6f0be937 2625 if format.get('protocol') is None:
b5559424 2626 format['protocol'] = determine_protocol(format)
239df021 2627 if format.get('resolution') is None:
2628 format['resolution'] = self.format_resolution(format, default=None)
176f1866 2629 if format.get('dynamic_range') is None and format.get('vcodec') != 'none':
2630 format['dynamic_range'] = 'SDR'
f2fe69c7 2631 if (info_dict.get('duration') and format.get('tbr')
2632 and not format.get('filesize') and not format.get('filesize_approx')):
56ba69e4 2633 format['filesize_approx'] = int(info_dict['duration'] * format['tbr'] * (1024 / 8))
f2fe69c7 2634
e5660ee6
JMF
2635 # Add HTTP headers, so that external programs can use them from the
2636 # json output
2637 full_format_info = info_dict.copy()
2638 full_format_info.update(format)
2639 format['http_headers'] = self._calc_headers(full_format_info)
0016b84e
S
2640 # Remove private housekeeping stuff
2641 if '__x_forwarded_for_ip' in info_dict:
2642 del info_dict['__x_forwarded_for_ip']
dd82ffea 2643
9f1a1c36 2644 if self.params.get('check_formats') is True:
282f5709 2645 formats = LazyList(self._check_formats(formats[::-1]), reverse=True)
9f1a1c36 2646
88acdbc2 2647 if not formats or formats[0] is not info_dict:
b3d9ef88
JMF
2648 # only set the 'formats' fields if the original info_dict list them
2649 # otherwise we end up with a circular reference, the first (and unique)
f89197d7 2650 # element in the 'formats' field in info_dict is info_dict itself,
dfb1b146 2651 # which can't be exported to json
b3d9ef88 2652 info_dict['formats'] = formats
4ec82a72 2653
2654 info_dict, _ = self.pre_process(info_dict)
2655
6db9c4d5 2656 if self._match_entry(info_dict, incomplete=self._format_fields) is not None:
09b49e1f 2657 return info_dict
2658
2659 self.post_extract(info_dict)
2660 info_dict, _ = self.pre_process(info_dict, 'after_filter')
2661
093a1710 2662 # The pre-processors may have modified the formats
aebb4f4b 2663 formats = self._get_formats(info_dict)
093a1710 2664
e4221b70 2665 list_only = self.params.get('simulate') == 'list_only'
fa9f30b8 2666 interactive_format_selection = not list_only and self.format_selector == '-'
b7b04c78 2667 if self.params.get('list_thumbnails'):
2668 self.list_thumbnails(info_dict)
b7b04c78 2669 if self.params.get('listsubtitles'):
2670 if 'automatic_captions' in info_dict:
2671 self.list_subtitles(
2672 info_dict['id'], automatic_captions, 'automatic captions')
2673 self.list_subtitles(info_dict['id'], subtitles, 'subtitles')
fa9f30b8 2674 if self.params.get('listformats') or interactive_format_selection:
b69fd25c 2675 self.list_formats(info_dict)
169dbde9 2676 if list_only:
b7b04c78 2677 # Without this printing, -F --print-json will not work
169dbde9 2678 self.__forced_printings(info_dict, self.prepare_filename(info_dict), incomplete=True)
c487cf00 2679 return info_dict
bfaae0a7 2680
187986a8 2681 format_selector = self.format_selector
2682 if format_selector is None:
0017d9ad 2683 req_format = self._default_format_spec(info_dict, download=download)
0760b0a7 2684 self.write_debug('Default format spec: %s' % req_format)
187986a8 2685 format_selector = self.build_format_selector(req_format)
317f7ab6 2686
fa9f30b8 2687 while True:
2688 if interactive_format_selection:
2689 req_format = input(
2690 self._format_screen('\nEnter format selector: ', self.Styles.EMPHASIS))
2691 try:
2692 format_selector = self.build_format_selector(req_format)
2693 except SyntaxError as err:
2694 self.report_error(err, tb=False, is_error=False)
2695 continue
2696
85e801a9 2697 formats_to_download = list(format_selector({
fa9f30b8 2698 'formats': formats,
85e801a9 2699 'has_merged_format': any('none' not in (f.get('acodec'), f.get('vcodec')) for f in formats),
2700 'incomplete_formats': (
2701 # All formats are video-only or
2702 all(f.get('vcodec') != 'none' and f.get('acodec') == 'none' for f in formats)
2703 # all formats are audio-only
2704 or all(f.get('vcodec') == 'none' and f.get('acodec') != 'none' for f in formats)),
2705 }))
fa9f30b8 2706 if interactive_format_selection and not formats_to_download:
2707 self.report_error('Requested format is not available', tb=False, is_error=False)
2708 continue
2709 break
317f7ab6 2710
dd82ffea 2711 if not formats_to_download:
b7da73eb 2712 if not self.params.get('ignore_no_formats_error'):
c0b6e5c7 2713 raise ExtractorError(
2714 'Requested format is not available. Use --list-formats for a list of available formats',
2715 expected=True, video_id=info_dict['id'], ie=info_dict['extractor'])
b62fa6d7 2716 self.report_warning('Requested format is not available')
2717 # Process what we can, even without any available formats.
2718 formats_to_download = [{}]
a13e6848 2719
0500ee3d 2720 requested_ranges = tuple(self.params.get('download_ranges', lambda *_: [{}])(info_dict, self))
5ec1b6b7 2721 best_format, downloaded_formats = formats_to_download[-1], []
b62fa6d7 2722 if download:
0500ee3d 2723 if best_format and requested_ranges:
5ec1b6b7 2724 def to_screen(*msg):
2725 self.to_screen(f'[info] {info_dict["id"]}: {" ".join(", ".join(variadic(m)) for m in msg)}')
2726
2727 to_screen(f'Downloading {len(formats_to_download)} format(s):',
2728 (f['format_id'] for f in formats_to_download))
0500ee3d 2729 if requested_ranges != ({}, ):
5ec1b6b7 2730 to_screen(f'Downloading {len(requested_ranges)} time ranges:',
fc2ba496 2731 (f'{c["start_time"]:.1f}-{c["end_time"]:.1f}' for c in requested_ranges))
a13e6848 2732 max_downloads_reached = False
5ec1b6b7 2733
0500ee3d 2734 for fmt, chapter in itertools.product(formats_to_download, requested_ranges):
5ec1b6b7 2735 new_info = self._copy_infodict(info_dict)
b7da73eb 2736 new_info.update(fmt)
3975b4d2 2737 offset, duration = info_dict.get('section_start') or 0, info_dict.get('duration') or float('inf')
fc2ba496 2738 end_time = offset + min(chapter.get('end_time', duration), duration)
3975b4d2 2739 if chapter or offset:
5ec1b6b7 2740 new_info.update({
3975b4d2 2741 'section_start': offset + chapter.get('start_time', 0),
2576d53a 2742 # duration may not be accurate. So allow deviations <1sec
2743 'section_end': end_time if end_time <= offset + duration + 1 else None,
5ec1b6b7 2744 'section_title': chapter.get('title'),
2745 'section_number': chapter.get('index'),
2746 })
2747 downloaded_formats.append(new_info)
a13e6848 2748 try:
2749 self.process_info(new_info)
2750 except MaxDownloadsReached:
2751 max_downloads_reached = True
415f8d51 2752 self._raise_pending_errors(new_info)
f46e2f9d 2753 # Remove copied info
2754 for key, val in tuple(new_info.items()):
2755 if info_dict.get(key) == val:
2756 new_info.pop(key)
a13e6848 2757 if max_downloads_reached:
2758 break
ebed8b37 2759
5ec1b6b7 2760 write_archive = {f.get('__write_download_archive', False) for f in downloaded_formats}
a13e6848 2761 assert write_archive.issubset({True, False, 'ignore'})
2762 if True in write_archive and False not in write_archive:
2763 self.record_download_archive(info_dict)
be72c624 2764
5ec1b6b7 2765 info_dict['requested_downloads'] = downloaded_formats
ed5835b4 2766 info_dict = self.run_all_pps('after_video', info_dict)
a13e6848 2767 if max_downloads_reached:
2768 raise MaxDownloadsReached()
ebed8b37 2769
49a57e70 2770 # We update the info dict with the selected best quality format (backwards compatibility)
be72c624 2771 info_dict.update(best_format)
dd82ffea
JMF
2772 return info_dict
2773
98c70d6f 2774 def process_subtitles(self, video_id, normal_subtitles, automatic_captions):
a504ced0 2775 """Select the requested subtitles and their format"""
d8a58ddc 2776 available_subs, normal_sub_langs = {}, []
98c70d6f
JMF
2777 if normal_subtitles and self.params.get('writesubtitles'):
2778 available_subs.update(normal_subtitles)
d8a58ddc 2779 normal_sub_langs = tuple(normal_subtitles.keys())
98c70d6f
JMF
2780 if automatic_captions and self.params.get('writeautomaticsub'):
2781 for lang, cap_info in automatic_captions.items():
360e1ca5
JMF
2782 if lang not in available_subs:
2783 available_subs[lang] = cap_info
2784
d2c8aadf 2785 if not available_subs or (
2786 not self.params.get('writesubtitles')
2787 and not self.params.get('writeautomaticsub')):
4d171848 2788 return None
a504ced0 2789
d8a58ddc 2790 all_sub_langs = tuple(available_subs.keys())
a504ced0 2791 if self.params.get('allsubtitles', False):
c32b0aab 2792 requested_langs = all_sub_langs
2793 elif self.params.get('subtitleslangs', False):
5314b521 2794 try:
2795 requested_langs = orderedSet_from_options(
2796 self.params.get('subtitleslangs'), {'all': all_sub_langs}, use_regex=True)
2797 except re.error as e:
2798 raise ValueError(f'Wrong regex for subtitlelangs: {e.pattern}')
d8a58ddc 2799 elif normal_sub_langs:
2800 requested_langs = ['en'] if 'en' in normal_sub_langs else normal_sub_langs[:1]
a504ced0 2801 else:
d8a58ddc 2802 requested_langs = ['en'] if 'en' in all_sub_langs else all_sub_langs[:1]
ad3dc496 2803 if requested_langs:
d2c8aadf 2804 self.to_screen(f'[info] {video_id}: Downloading subtitles: {", ".join(requested_langs)}')
a504ced0
JMF
2805
2806 formats_query = self.params.get('subtitlesformat', 'best')
2807 formats_preference = formats_query.split('/') if formats_query else []
2808 subs = {}
2809 for lang in requested_langs:
2810 formats = available_subs.get(lang)
2811 if formats is None:
86e5f3ed 2812 self.report_warning(f'{lang} subtitles not available for {video_id}')
a504ced0 2813 continue
a504ced0
JMF
2814 for ext in formats_preference:
2815 if ext == 'best':
2816 f = formats[-1]
2817 break
2818 matches = list(filter(lambda f: f['ext'] == ext, formats))
2819 if matches:
2820 f = matches[-1]
2821 break
2822 else:
2823 f = formats[-1]
2824 self.report_warning(
2825 'No subtitle format found matching "%s" for language %s, '
2826 'using %s' % (formats_query, lang, f['ext']))
2827 subs[lang] = f
2828 return subs
2829
bb66c247 2830 def _forceprint(self, key, info_dict):
2831 if info_dict is None:
2832 return
2833 info_copy = info_dict.copy()
2834 info_copy['formats_table'] = self.render_formats_table(info_dict)
2835 info_copy['thumbnails_table'] = self.render_thumbnails_table(info_dict)
2836 info_copy['subtitles_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('subtitles'))
2837 info_copy['automatic_captions_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('automatic_captions'))
2838
2839 def format_tmpl(tmpl):
48c8424b 2840 mobj = re.fullmatch(r'([\w.:,]|-\d|(?P<dict>{([\w.:,]|-\d)+}))+=?', tmpl)
07a1250e 2841 if not mobj:
2842 return tmpl
48c8424b 2843
2844 fmt = '%({})s'
2845 if tmpl.startswith('{'):
2846 tmpl = f'.{tmpl}'
2847 if tmpl.endswith('='):
2848 tmpl, fmt = tmpl[:-1], '{0} = %({0})#j'
2849 return '\n'.join(map(fmt.format, [tmpl] if mobj.group('dict') else tmpl.split(',')))
8130779d 2850
bb66c247 2851 for tmpl in self.params['forceprint'].get(key, []):
2852 self.to_stdout(self.evaluate_outtmpl(format_tmpl(tmpl), info_copy))
2853
2854 for tmpl, file_tmpl in self.params['print_to_file'].get(key, []):
5127e92a 2855 filename = self.prepare_filename(info_dict, outtmpl=file_tmpl)
bb66c247 2856 tmpl = format_tmpl(tmpl)
2857 self.to_screen(f'[info] Writing {tmpl!r} to: {filename}')
8d93e69d 2858 if self._ensure_dir_exists(filename):
86e5f3ed 2859 with open(filename, 'a', encoding='utf-8') as f:
8d93e69d 2860 f.write(self.evaluate_outtmpl(tmpl, info_copy) + '\n')
ca30f449 2861
d06daf23 2862 def __forced_printings(self, info_dict, filename, incomplete):
53c18592 2863 def print_mandatory(field, actual_field=None):
2864 if actual_field is None:
2865 actual_field = field
d06daf23 2866 if (self.params.get('force%s' % field, False)
53c18592 2867 and (not incomplete or info_dict.get(actual_field) is not None)):
2868 self.to_stdout(info_dict[actual_field])
d06daf23
S
2869
2870 def print_optional(field):
2871 if (self.params.get('force%s' % field, False)
2872 and info_dict.get(field) is not None):
2873 self.to_stdout(info_dict[field])
2874
53c18592 2875 info_dict = info_dict.copy()
2876 if filename is not None:
2877 info_dict['filename'] = filename
2878 if info_dict.get('requested_formats') is not None:
2879 # For RTMP URLs, also include the playpath
2880 info_dict['urls'] = '\n'.join(f['url'] + f.get('play_path', '') for f in info_dict['requested_formats'])
10331a26 2881 elif info_dict.get('url'):
53c18592 2882 info_dict['urls'] = info_dict['url'] + info_dict.get('play_path', '')
2883
bb66c247 2884 if (self.params.get('forcejson')
2885 or self.params['forceprint'].get('video')
2886 or self.params['print_to_file'].get('video')):
2b8a2973 2887 self.post_extract(info_dict)
bb66c247 2888 self._forceprint('video', info_dict)
53c18592 2889
d06daf23
S
2890 print_mandatory('title')
2891 print_mandatory('id')
53c18592 2892 print_mandatory('url', 'urls')
d06daf23
S
2893 print_optional('thumbnail')
2894 print_optional('description')
53c18592 2895 print_optional('filename')
b868936c 2896 if self.params.get('forceduration') and info_dict.get('duration') is not None:
d06daf23
S
2897 self.to_stdout(formatSeconds(info_dict['duration']))
2898 print_mandatory('format')
53c18592 2899
2b8a2973 2900 if self.params.get('forcejson'):
6e84b215 2901 self.to_stdout(json.dumps(self.sanitize_info(info_dict)))
d06daf23 2902
e8e73840 2903 def dl(self, name, info, subtitle=False, test=False):
88acdbc2 2904 if not info.get('url'):
1151c407 2905 self.raise_no_formats(info, True)
e8e73840 2906
2907 if test:
2908 verbose = self.params.get('verbose')
2909 params = {
2910 'test': True,
a169858f 2911 'quiet': self.params.get('quiet') or not verbose,
e8e73840 2912 'verbose': verbose,
2913 'noprogress': not verbose,
2914 'nopart': True,
2915 'skip_unavailable_fragments': False,
2916 'keep_fragments': False,
2917 'overwrites': True,
2918 '_no_ytdl_file': True,
2919 }
2920 else:
2921 params = self.params
96fccc10 2922 fd = get_suitable_downloader(info, params, to_stdout=(name == '-'))(self, params)
e8e73840 2923 if not test:
2924 for ph in self._progress_hooks:
2925 fd.add_progress_hook(ph)
42676437
M
2926 urls = '", "'.join(
2927 (f['url'].split(',')[0] + ',<data>' if f['url'].startswith('data:') else f['url'])
2928 for f in info.get('requested_formats', []) or [info])
3a408f9d 2929 self.write_debug(f'Invoking {fd.FD_NAME} downloader on "{urls}"')
03b4de72 2930
adbc4ec4
THD
2931 # Note: Ideally info should be a deep-copied so that hooks cannot modify it.
2932 # But it may contain objects that are not deep-copyable
2933 new_info = self._copy_infodict(info)
e8e73840 2934 if new_info.get('http_headers') is None:
2935 new_info['http_headers'] = self._calc_headers(new_info)
2936 return fd.download(name, new_info, subtitle)
2937
e04938ab 2938 def existing_file(self, filepaths, *, default_overwrite=True):
2939 existing_files = list(filter(os.path.exists, orderedSet(filepaths)))
2940 if existing_files and not self.params.get('overwrites', default_overwrite):
2941 return existing_files[0]
2942
2943 for file in existing_files:
2944 self.report_file_delete(file)
2945 os.remove(file)
2946 return None
2947
8222d8de 2948 def process_info(self, info_dict):
09b49e1f 2949 """Process a single resolved IE result. (Modifies it in-place)"""
8222d8de
JMF
2950
2951 assert info_dict.get('_type', 'video') == 'video'
f46e2f9d 2952 original_infodict = info_dict
fd288278 2953
4513a41a 2954 if 'format' not in info_dict and 'ext' in info_dict:
8222d8de
JMF
2955 info_dict['format'] = info_dict['ext']
2956
c77495e3 2957 if self._match_entry(info_dict) is not None:
9e907ebd 2958 info_dict['__write_download_archive'] = 'ignore'
8222d8de
JMF
2959 return
2960
09b49e1f 2961 # Does nothing under normal operation - for backward compatibility of process_info
277d6ff5 2962 self.post_extract(info_dict)
0c14d66a 2963 self._num_downloads += 1
8222d8de 2964
dcf64d43 2965 # info_dict['_filename'] needs to be set for backward compatibility
de6000d9 2966 info_dict['_filename'] = full_filename = self.prepare_filename(info_dict, warn=True)
2967 temp_filename = self.prepare_filename(info_dict, 'temp')
0202b52a 2968 files_to_move = {}
8222d8de
JMF
2969
2970 # Forced printings
4513a41a 2971 self.__forced_printings(info_dict, full_filename, incomplete=('format' not in info_dict))
8222d8de 2972
ca6d59d2 2973 def check_max_downloads():
2974 if self._num_downloads >= float(self.params.get('max_downloads') or 'inf'):
2975 raise MaxDownloadsReached()
2976
b7b04c78 2977 if self.params.get('simulate'):
9e907ebd 2978 info_dict['__write_download_archive'] = self.params.get('force_write_download_archive')
ca6d59d2 2979 check_max_downloads()
8222d8de
JMF
2980 return
2981
de6000d9 2982 if full_filename is None:
8222d8de 2983 return
e92caff5 2984 if not self._ensure_dir_exists(encodeFilename(full_filename)):
0202b52a 2985 return
e92caff5 2986 if not self._ensure_dir_exists(encodeFilename(temp_filename)):
8222d8de
JMF
2987 return
2988
80c03fa9 2989 if self._write_description('video', info_dict,
2990 self.prepare_filename(info_dict, 'description')) is None:
2991 return
2992
2993 sub_files = self._write_subtitles(info_dict, temp_filename)
2994 if sub_files is None:
2995 return
2996 files_to_move.update(dict(sub_files))
2997
2998 thumb_files = self._write_thumbnails(
2999 'video', info_dict, temp_filename, self.prepare_filename(info_dict, 'thumbnail'))
3000 if thumb_files is None:
3001 return
3002 files_to_move.update(dict(thumb_files))
8222d8de 3003
80c03fa9 3004 infofn = self.prepare_filename(info_dict, 'infojson')
3005 _infojson_written = self._write_info_json('video', info_dict, infofn)
3006 if _infojson_written:
dac5df5a 3007 info_dict['infojson_filename'] = infofn
e75bb0d6 3008 # For backward compatibility, even though it was a private field
80c03fa9 3009 info_dict['__infojson_filename'] = infofn
3010 elif _infojson_written is None:
3011 return
3012
3013 # Note: Annotations are deprecated
3014 annofn = None
1fb07d10 3015 if self.params.get('writeannotations', False):
de6000d9 3016 annofn = self.prepare_filename(info_dict, 'annotation')
80c03fa9 3017 if annofn:
e92caff5 3018 if not self._ensure_dir_exists(encodeFilename(annofn)):
0202b52a 3019 return
0c3d0f51 3020 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(annofn)):
6febd1c1 3021 self.to_screen('[info] Video annotations are already present')
ffddb112
RA
3022 elif not info_dict.get('annotations'):
3023 self.report_warning('There are no annotations to write.')
7b6fefc9
PH
3024 else:
3025 try:
6febd1c1 3026 self.to_screen('[info] Writing video annotations to: ' + annofn)
86e5f3ed 3027 with open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
7b6fefc9
PH
3028 annofile.write(info_dict['annotations'])
3029 except (KeyError, TypeError):
6febd1c1 3030 self.report_warning('There are no annotations to write.')
86e5f3ed 3031 except OSError:
6febd1c1 3032 self.report_error('Cannot write annotations file: ' + annofn)
7b6fefc9 3033 return
1fb07d10 3034
732044af 3035 # Write internet shortcut files
08438d2c 3036 def _write_link_file(link_type):
60f3e995 3037 url = try_get(info_dict['webpage_url'], iri_to_uri)
3038 if not url:
3039 self.report_warning(
3040 f'Cannot write internet shortcut file because the actual URL of "{info_dict["webpage_url"]}" is unknown')
3041 return True
08438d2c 3042 linkfn = replace_extension(self.prepare_filename(info_dict, 'link'), link_type, info_dict.get('ext'))
0e6b018a
Z
3043 if not self._ensure_dir_exists(encodeFilename(linkfn)):
3044 return False
10e3742e 3045 if self.params.get('overwrites', True) and os.path.exists(encodeFilename(linkfn)):
08438d2c 3046 self.to_screen(f'[info] Internet shortcut (.{link_type}) is already present')
3047 return True
3048 try:
3049 self.to_screen(f'[info] Writing internet shortcut (.{link_type}) to: {linkfn}')
86e5f3ed 3050 with open(encodeFilename(to_high_limit_path(linkfn)), 'w', encoding='utf-8',
3051 newline='\r\n' if link_type == 'url' else '\n') as linkfile:
60f3e995 3052 template_vars = {'url': url}
08438d2c 3053 if link_type == 'desktop':
3054 template_vars['filename'] = linkfn[:-(len(link_type) + 1)]
3055 linkfile.write(LINK_TEMPLATES[link_type] % template_vars)
86e5f3ed 3056 except OSError:
08438d2c 3057 self.report_error(f'Cannot write internet shortcut {linkfn}')
3058 return False
732044af 3059 return True
3060
08438d2c 3061 write_links = {
3062 'url': self.params.get('writeurllink'),
3063 'webloc': self.params.get('writewebloclink'),
3064 'desktop': self.params.get('writedesktoplink'),
3065 }
3066 if self.params.get('writelink'):
3067 link_type = ('webloc' if sys.platform == 'darwin'
3068 else 'desktop' if sys.platform.startswith('linux')
3069 else 'url')
3070 write_links[link_type] = True
3071
3072 if any(should_write and not _write_link_file(link_type)
3073 for link_type, should_write in write_links.items()):
3074 return
732044af 3075
f46e2f9d 3076 def replace_info_dict(new_info):
3077 nonlocal info_dict
3078 if new_info == info_dict:
3079 return
3080 info_dict.clear()
3081 info_dict.update(new_info)
3082
415f8d51 3083 new_info, files_to_move = self.pre_process(info_dict, 'before_dl', files_to_move)
3084 replace_info_dict(new_info)
56d868db 3085
a13e6848 3086 if self.params.get('skip_download'):
56d868db 3087 info_dict['filepath'] = temp_filename
3088 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
3089 info_dict['__files_to_move'] = files_to_move
f46e2f9d 3090 replace_info_dict(self.run_pp(MoveFilesAfterDownloadPP(self, False), info_dict))
9e907ebd 3091 info_dict['__write_download_archive'] = self.params.get('force_write_download_archive')
56d868db 3092 else:
3093 # Download
b868936c 3094 info_dict.setdefault('__postprocessors', [])
4340deca 3095 try:
0202b52a 3096
e04938ab 3097 def existing_video_file(*filepaths):
6b591b29 3098 ext = info_dict.get('ext')
e04938ab 3099 converted = lambda file: replace_extension(file, self.params.get('final_ext') or ext, ext)
3100 file = self.existing_file(itertools.chain(*zip(map(converted, filepaths), filepaths)),
3101 default_overwrite=False)
3102 if file:
3103 info_dict['ext'] = os.path.splitext(file)[1][1:]
3104 return file
0202b52a 3105
7b2c3f47 3106 fd, success = None, True
fccf90e7 3107 if info_dict.get('protocol') or info_dict.get('url'):
56ba69e4 3108 fd = get_suitable_downloader(info_dict, self.params, to_stdout=temp_filename == '-')
3109 if fd is not FFmpegFD and (
3110 info_dict.get('section_start') or info_dict.get('section_end')):
7b2c3f47 3111 msg = ('This format cannot be partially downloaded' if FFmpegFD.available()
56ba69e4 3112 else 'You have requested downloading the video partially, but ffmpeg is not installed')
3113 self.report_error(f'{msg}. Aborting')
5ec1b6b7 3114 return
5ec1b6b7 3115
4340deca 3116 if info_dict.get('requested_formats') is not None:
81cd954a 3117 requested_formats = info_dict['requested_formats']
0202b52a 3118 old_ext = info_dict['ext']
4e3b637d 3119 if self.params.get('merge_output_format') is None:
4e3b637d 3120 if (info_dict['ext'] == 'webm'
3121 and info_dict.get('thumbnails')
3122 # check with type instead of pp_key, __name__, or isinstance
3123 # since we dont want any custom PPs to trigger this
c487cf00 3124 and any(type(pp) == EmbedThumbnailPP for pp in self._pps['post_process'])): # noqa: E721
4e3b637d 3125 info_dict['ext'] = 'mkv'
3126 self.report_warning(
3127 'webm doesn\'t support embedding a thumbnail, mkv will be used')
124bc071 3128 new_ext = info_dict['ext']
0202b52a 3129
124bc071 3130 def correct_ext(filename, ext=new_ext):
96fccc10 3131 if filename == '-':
3132 return filename
0202b52a 3133 filename_real_ext = os.path.splitext(filename)[1][1:]
3134 filename_wo_ext = (
3135 os.path.splitext(filename)[0]
124bc071 3136 if filename_real_ext in (old_ext, new_ext)
0202b52a 3137 else filename)
86e5f3ed 3138 return f'{filename_wo_ext}.{ext}'
0202b52a 3139
38c6902b 3140 # Ensure filename always has a correct extension for successful merge
0202b52a 3141 full_filename = correct_ext(full_filename)
3142 temp_filename = correct_ext(temp_filename)
e04938ab 3143 dl_filename = existing_video_file(full_filename, temp_filename)
1ea24129 3144 info_dict['__real_download'] = False
18e674b4 3145
7b2c3f47 3146 merger = FFmpegMergerPP(self)
adbc4ec4 3147 downloaded = []
dbf5416a 3148 if dl_filename is not None:
6c7274ec 3149 self.report_file_already_downloaded(dl_filename)
adbc4ec4
THD
3150 elif fd:
3151 for f in requested_formats if fd != FFmpegFD else []:
3152 f['filepath'] = fname = prepend_extension(
3153 correct_ext(temp_filename, info_dict['ext']),
3154 'f%s' % f['format_id'], info_dict['ext'])
3155 downloaded.append(fname)
dbf5416a 3156 info_dict['url'] = '\n'.join(f['url'] for f in requested_formats)
3157 success, real_download = self.dl(temp_filename, info_dict)
3158 info_dict['__real_download'] = real_download
18e674b4 3159 else:
18e674b4 3160 if self.params.get('allow_unplayable_formats'):
3161 self.report_warning(
3162 'You have requested merging of multiple formats '
3163 'while also allowing unplayable formats to be downloaded. '
3164 'The formats won\'t be merged to prevent data corruption.')
3165 elif not merger.available:
e8969bda 3166 msg = 'You have requested merging of multiple formats but ffmpeg is not installed'
3167 if not self.params.get('ignoreerrors'):
3168 self.report_error(f'{msg}. Aborting due to --abort-on-error')
3169 return
3170 self.report_warning(f'{msg}. The formats won\'t be merged')
18e674b4 3171
96fccc10 3172 if temp_filename == '-':
adbc4ec4 3173 reason = ('using a downloader other than ffmpeg' if FFmpegFD.can_merge_formats(info_dict, self.params)
96fccc10 3174 else 'but the formats are incompatible for simultaneous download' if merger.available
3175 else 'but ffmpeg is not installed')
3176 self.report_warning(
3177 f'You have requested downloading multiple formats to stdout {reason}. '
3178 'The formats will be streamed one after the other')
3179 fname = temp_filename
dbf5416a 3180 for f in requested_formats:
3181 new_info = dict(info_dict)
3182 del new_info['requested_formats']
3183 new_info.update(f)
96fccc10 3184 if temp_filename != '-':
124bc071 3185 fname = prepend_extension(
3186 correct_ext(temp_filename, new_info['ext']),
3187 'f%s' % f['format_id'], new_info['ext'])
96fccc10 3188 if not self._ensure_dir_exists(fname):
3189 return
a21e0ab1 3190 f['filepath'] = fname
96fccc10 3191 downloaded.append(fname)
dbf5416a 3192 partial_success, real_download = self.dl(fname, new_info)
3193 info_dict['__real_download'] = info_dict['__real_download'] or real_download
3194 success = success and partial_success
adbc4ec4
THD
3195
3196 if downloaded and merger.available and not self.params.get('allow_unplayable_formats'):
3197 info_dict['__postprocessors'].append(merger)
3198 info_dict['__files_to_merge'] = downloaded
3199 # Even if there were no downloads, it is being merged only now
3200 info_dict['__real_download'] = True
3201 else:
3202 for file in downloaded:
3203 files_to_move[file] = None
4340deca
P
3204 else:
3205 # Just a single file
e04938ab 3206 dl_filename = existing_video_file(full_filename, temp_filename)
6c7274ec 3207 if dl_filename is None or dl_filename == temp_filename:
3208 # dl_filename == temp_filename could mean that the file was partially downloaded with --no-part.
3209 # So we should try to resume the download
e8e73840 3210 success, real_download = self.dl(temp_filename, info_dict)
0202b52a 3211 info_dict['__real_download'] = real_download
6c7274ec 3212 else:
3213 self.report_file_already_downloaded(dl_filename)
0202b52a 3214
0202b52a 3215 dl_filename = dl_filename or temp_filename
c571435f 3216 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
0202b52a 3217
3158150c 3218 except network_exceptions as err:
7960b056 3219 self.report_error('unable to download video data: %s' % error_to_compat_str(err))
4340deca 3220 return
86e5f3ed 3221 except OSError as err:
4340deca
P
3222 raise UnavailableVideoError(err)
3223 except (ContentTooShortError, ) as err:
86e5f3ed 3224 self.report_error(f'content too short (expected {err.expected} bytes and served {err.downloaded})')
4340deca 3225 return
8222d8de 3226
415f8d51 3227 self._raise_pending_errors(info_dict)
de6000d9 3228 if success and full_filename != '-':
f17f8651 3229
fd7cfb64 3230 def fixup():
3231 do_fixup = True
3232 fixup_policy = self.params.get('fixup')
3233 vid = info_dict['id']
3234
3235 if fixup_policy in ('ignore', 'never'):
3236 return
3237 elif fixup_policy == 'warn':
3fe75fdc 3238 do_fixup = 'warn'
f89b3e2d 3239 elif fixup_policy != 'force':
3240 assert fixup_policy in ('detect_or_warn', None)
3241 if not info_dict.get('__real_download'):
3242 do_fixup = False
fd7cfb64 3243
3244 def ffmpeg_fixup(cndn, msg, cls):
3fe75fdc 3245 if not (do_fixup and cndn):
fd7cfb64 3246 return
3fe75fdc 3247 elif do_fixup == 'warn':
fd7cfb64 3248 self.report_warning(f'{vid}: {msg}')
3249 return
3250 pp = cls(self)
3251 if pp.available:
3252 info_dict['__postprocessors'].append(pp)
3253 else:
3254 self.report_warning(f'{vid}: {msg}. Install ffmpeg to fix this automatically')
3255
3256 stretched_ratio = info_dict.get('stretched_ratio')
ca9def71
LNO
3257 ffmpeg_fixup(stretched_ratio not in (1, None),
3258 f'Non-uniform pixel ratio {stretched_ratio}',
3259 FFmpegFixupStretchedPP)
fd7cfb64 3260
993191c0 3261 downloader = get_suitable_downloader(info_dict, self.params) if 'protocol' in info_dict else None
24146491 3262 downloader = downloader.FD_NAME if downloader else None
adbc4ec4 3263
ca9def71
LNO
3264 ext = info_dict.get('ext')
3265 postprocessed_by_ffmpeg = info_dict.get('requested_formats') or any((
3266 isinstance(pp, FFmpegVideoConvertorPP)
3267 and resolve_recode_mapping(ext, pp.mapping)[0] not in (ext, None)
3268 ) for pp in self._pps['post_process'])
3269
3270 if not postprocessed_by_ffmpeg:
3271 ffmpeg_fixup(ext == 'm4a' and info_dict.get('container') == 'm4a_dash',
f2df4071 3272 'writing DASH m4a. Only some players support this container',
3273 FFmpegFixupM4aPP)
24146491 3274 ffmpeg_fixup(downloader == 'hlsnative' and not self.params.get('hls_use_mpegts')
494f5230 3275 or info_dict.get('is_live') and self.params.get('hls_use_mpegts') is None,
adbc4ec4
THD
3276 'Possible MPEG-TS in MP4 container or malformed AAC timestamps',
3277 FFmpegFixupM3u8PP)
3278 ffmpeg_fixup(info_dict.get('is_live') and downloader == 'DashSegmentsFD',
3279 'Possible duplicate MOOV atoms', FFmpegFixupDuplicateMoovPP)
3280
24146491 3281 ffmpeg_fixup(downloader == 'web_socket_fragment', 'Malformed timestamps detected', FFmpegFixupTimestampPP)
3282 ffmpeg_fixup(downloader == 'web_socket_fragment', 'Malformed duration detected', FFmpegFixupDurationPP)
fd7cfb64 3283
3284 fixup()
8222d8de 3285 try:
f46e2f9d 3286 replace_info_dict(self.post_process(dl_filename, info_dict, files_to_move))
af819c21 3287 except PostProcessingError as err:
3288 self.report_error('Postprocessing: %s' % str(err))
8222d8de 3289 return
ab8e5e51
AM
3290 try:
3291 for ph in self._post_hooks:
23c1a667 3292 ph(info_dict['filepath'])
ab8e5e51
AM
3293 except Exception as err:
3294 self.report_error('post hooks: %s' % str(err))
3295 return
9e907ebd 3296 info_dict['__write_download_archive'] = True
2d30509f 3297
c487cf00 3298 assert info_dict is original_infodict # Make sure the info_dict was modified in-place
a13e6848 3299 if self.params.get('force_write_download_archive'):
9e907ebd 3300 info_dict['__write_download_archive'] = True
ca6d59d2 3301 check_max_downloads()
8222d8de 3302
aa9369a2 3303 def __download_wrapper(self, func):
3304 @functools.wraps(func)
3305 def wrapper(*args, **kwargs):
3306 try:
3307 res = func(*args, **kwargs)
3308 except UnavailableVideoError as e:
3309 self.report_error(e)
b222c271 3310 except DownloadCancelled as e:
3311 self.to_screen(f'[info] {e}')
3312 if not self.params.get('break_per_url'):
3313 raise
fd404bec 3314 self._num_downloads = 0
aa9369a2 3315 else:
3316 if self.params.get('dump_single_json', False):
3317 self.post_extract(res)
3318 self.to_stdout(json.dumps(self.sanitize_info(res)))
3319 return wrapper
3320
8222d8de
JMF
3321 def download(self, url_list):
3322 """Download a given list of URLs."""
aa9369a2 3323 url_list = variadic(url_list) # Passing a single URL is a common mistake
bf1824b3 3324 outtmpl = self.params['outtmpl']['default']
3089bc74
S
3325 if (len(url_list) > 1
3326 and outtmpl != '-'
3327 and '%' not in outtmpl
3328 and self.params.get('max_downloads') != 1):
acd69589 3329 raise SameFileError(outtmpl)
8222d8de
JMF
3330
3331 for url in url_list:
aa9369a2 3332 self.__download_wrapper(self.extract_info)(
3333 url, force_generic_extractor=self.params.get('force_generic_extractor', False))
8222d8de
JMF
3334
3335 return self._download_retcode
3336
1dcc4c0c 3337 def download_with_info_file(self, info_filename):
31bd3925
JMF
3338 with contextlib.closing(fileinput.FileInput(
3339 [info_filename], mode='r',
3340 openhook=fileinput.hook_encoded('utf-8'))) as f:
3341 # FileInput doesn't have a read method, we can't call json.load
8012d892 3342 info = self.sanitize_info(json.loads('\n'.join(f)), self.params.get('clean_infojson', True))
d4943898 3343 try:
aa9369a2 3344 self.__download_wrapper(self.process_ie_result)(info, download=True)
f2ebc5c7 3345 except (DownloadError, EntryNotInPlaylist, ReExtractInfo) as e:
bf5f605e 3346 if not isinstance(e, EntryNotInPlaylist):
3347 self.to_stderr('\r')
d4943898
JMF
3348 webpage_url = info.get('webpage_url')
3349 if webpage_url is not None:
aa9369a2 3350 self.report_warning(f'The info failed to download: {e}; trying with URL {webpage_url}')
d4943898
JMF
3351 return self.download([webpage_url])
3352 else:
3353 raise
3354 return self._download_retcode
1dcc4c0c 3355
cb202fd2 3356 @staticmethod
8012d892 3357 def sanitize_info(info_dict, remove_private_keys=False):
3358 ''' Sanitize the infodict for converting to json '''
3ad56b42 3359 if info_dict is None:
3360 return info_dict
6e84b215 3361 info_dict.setdefault('epoch', int(time.time()))
6a5a30f9 3362 info_dict.setdefault('_type', 'video')
b5e7a2e6 3363 info_dict.setdefault('_version', {
3364 'version': __version__,
3365 'current_git_head': current_git_head(),
3366 'release_git_head': RELEASE_GIT_HEAD,
3367 'repository': REPOSITORY,
3368 })
09b49e1f 3369
8012d892 3370 if remove_private_keys:
0a5a191a 3371 reject = lambda k, v: v is None or k.startswith('__') or k in {
f46e2f9d 3372 'requested_downloads', 'requested_formats', 'requested_subtitles', 'requested_entries',
0a5a191a 3373 'entries', 'filepath', '_filename', 'infojson_filename', 'original_url', 'playlist_autonumber',
6e84b215 3374 }
ae8f99e6 3375 else:
09b49e1f 3376 reject = lambda k, v: False
adbc4ec4
THD
3377
3378 def filter_fn(obj):
3379 if isinstance(obj, dict):
3380 return {k: filter_fn(v) for k, v in obj.items() if not reject(k, v)}
3381 elif isinstance(obj, (list, tuple, set, LazyList)):
3382 return list(map(filter_fn, obj))
3383 elif obj is None or isinstance(obj, (str, int, float, bool)):
3384 return obj
3385 else:
3386 return repr(obj)
3387
5226731e 3388 return filter_fn(info_dict)
cb202fd2 3389
8012d892 3390 @staticmethod
3391 def filter_requested_info(info_dict, actually_filter=True):
3392 ''' Alias of sanitize_info for backward compatibility '''
3393 return YoutubeDL.sanitize_info(info_dict, actually_filter)
3394
43d7f5a5 3395 def _delete_downloaded_files(self, *files_to_delete, info={}, msg=None):
3396 for filename in set(filter(None, files_to_delete)):
3397 if msg:
3398 self.to_screen(msg % filename)
3399 try:
3400 os.remove(filename)
3401 except OSError:
3402 self.report_warning(f'Unable to delete file {filename}')
3403 if filename in info.get('__files_to_move', []): # NB: Delete even if None
3404 del info['__files_to_move'][filename]
3405
ed5835b4 3406 @staticmethod
3407 def post_extract(info_dict):
3408 def actual_post_extract(info_dict):
3409 if info_dict.get('_type') in ('playlist', 'multi_video'):
3410 for video_dict in info_dict.get('entries', {}):
3411 actual_post_extract(video_dict or {})
3412 return
3413
09b49e1f 3414 post_extractor = info_dict.pop('__post_extractor', None) or (lambda: {})
3415 info_dict.update(post_extractor())
ed5835b4 3416
3417 actual_post_extract(info_dict or {})
3418
dcf64d43 3419 def run_pp(self, pp, infodict):
5bfa4862 3420 files_to_delete = []
dcf64d43 3421 if '__files_to_move' not in infodict:
3422 infodict['__files_to_move'] = {}
b1940459 3423 try:
3424 files_to_delete, infodict = pp.run(infodict)
3425 except PostProcessingError as e:
3426 # Must be True and not 'only_download'
3427 if self.params.get('ignoreerrors') is True:
3428 self.report_error(e)
3429 return infodict
3430 raise
3431
5bfa4862 3432 if not files_to_delete:
dcf64d43 3433 return infodict
5bfa4862 3434 if self.params.get('keepvideo', False):
3435 for f in files_to_delete:
dcf64d43 3436 infodict['__files_to_move'].setdefault(f, '')
5bfa4862 3437 else:
43d7f5a5 3438 self._delete_downloaded_files(
3439 *files_to_delete, info=infodict, msg='Deleting original file %s (pass -k to keep)')
dcf64d43 3440 return infodict
5bfa4862 3441
ed5835b4 3442 def run_all_pps(self, key, info, *, additional_pps=None):
bb66c247 3443 self._forceprint(key, info)
ed5835b4 3444 for pp in (additional_pps or []) + self._pps[key]:
dc5f409c 3445 info = self.run_pp(pp, info)
ed5835b4 3446 return info
277d6ff5 3447
56d868db 3448 def pre_process(self, ie_info, key='pre_process', files_to_move=None):
5bfa4862 3449 info = dict(ie_info)
56d868db 3450 info['__files_to_move'] = files_to_move or {}
415f8d51 3451 try:
3452 info = self.run_all_pps(key, info)
3453 except PostProcessingError as err:
3454 msg = f'Preprocessing: {err}'
3455 info.setdefault('__pending_error', msg)
3456 self.report_error(msg, is_error=False)
56d868db 3457 return info, info.pop('__files_to_move', None)
5bfa4862 3458
f46e2f9d 3459 def post_process(self, filename, info, files_to_move=None):
8222d8de 3460 """Run all the postprocessors on the given file."""
8222d8de 3461 info['filepath'] = filename
dcf64d43 3462 info['__files_to_move'] = files_to_move or {}
ed5835b4 3463 info = self.run_all_pps('post_process', info, additional_pps=info.get('__postprocessors'))
dcf64d43 3464 info = self.run_pp(MoveFilesAfterDownloadPP(self), info)
3465 del info['__files_to_move']
ed5835b4 3466 return self.run_all_pps('after_move', info)
c1c9a79c 3467
5db07df6 3468 def _make_archive_id(self, info_dict):
e9fef7ee
S
3469 video_id = info_dict.get('id')
3470 if not video_id:
3471 return
5db07df6
PH
3472 # Future-proof against any change in case
3473 # and backwards compatibility with prior versions
e9fef7ee 3474 extractor = info_dict.get('extractor_key') or info_dict.get('ie_key') # key in a playlist
7012b23c 3475 if extractor is None:
1211bb6d
S
3476 url = str_or_none(info_dict.get('url'))
3477 if not url:
3478 return
e9fef7ee 3479 # Try to find matching extractor for the URL and take its ie_key
8b7491c8 3480 for ie_key, ie in self._ies.items():
1211bb6d 3481 if ie.suitable(url):
8b7491c8 3482 extractor = ie_key
e9fef7ee
S
3483 break
3484 else:
3485 return
0647d925 3486 return make_archive_id(extractor, video_id)
5db07df6
PH
3487
3488 def in_download_archive(self, info_dict):
ae103564 3489 if not self.archive:
5db07df6
PH
3490 return False
3491
1e8fe57e 3492 vid_ids = [self._make_archive_id(info_dict)]
c200096c 3493 vid_ids.extend(info_dict.get('_old_archive_ids') or [])
1e8fe57e 3494 return any(id_ in self.archive for id_ in vid_ids)
c1c9a79c
PH
3495
3496 def record_download_archive(self, info_dict):
3497 fn = self.params.get('download_archive')
3498 if fn is None:
3499 return
5db07df6
PH
3500 vid_id = self._make_archive_id(info_dict)
3501 assert vid_id
ae103564 3502
a13e6848 3503 self.write_debug(f'Adding to archive: {vid_id}')
9c935fbc 3504 if is_path_like(fn):
ae103564 3505 with locked_file(fn, 'a', encoding='utf-8') as archive_file:
3506 archive_file.write(vid_id + '\n')
a45e8619 3507 self.archive.add(vid_id)
dd82ffea 3508
8c51aa65 3509 @staticmethod
8abeeb94 3510 def format_resolution(format, default='unknown'):
9359f3d4 3511 if format.get('vcodec') == 'none' and format.get('acodec') != 'none':
fb04e403 3512 return 'audio only'
f49d89ee
PH
3513 if format.get('resolution') is not None:
3514 return format['resolution']
35615307 3515 if format.get('width') and format.get('height'):
ff51ed58 3516 return '%dx%d' % (format['width'], format['height'])
35615307 3517 elif format.get('height'):
ff51ed58 3518 return '%sp' % format['height']
35615307 3519 elif format.get('width'):
ff51ed58 3520 return '%dx?' % format['width']
3521 return default
8c51aa65 3522
8130779d 3523 def _list_format_headers(self, *headers):
3524 if self.params.get('listformats_table', True) is not False:
591bb9d3 3525 return [self._format_out(header, self.Styles.HEADERS) for header in headers]
8130779d 3526 return headers
3527
c57f7757
PH
3528 def _format_note(self, fdict):
3529 res = ''
3530 if fdict.get('ext') in ['f4f', 'f4m']:
f304da8a 3531 res += '(unsupported)'
32f90364
PH
3532 if fdict.get('language'):
3533 if res:
3534 res += ' '
f304da8a 3535 res += '[%s]' % fdict['language']
c57f7757 3536 if fdict.get('format_note') is not None:
f304da8a 3537 if res:
3538 res += ' '
3539 res += fdict['format_note']
c57f7757 3540 if fdict.get('tbr') is not None:
f304da8a 3541 if res:
3542 res += ', '
3543 res += '%4dk' % fdict['tbr']
c57f7757
PH
3544 if fdict.get('container') is not None:
3545 if res:
3546 res += ', '
3547 res += '%s container' % fdict['container']
3089bc74
S
3548 if (fdict.get('vcodec') is not None
3549 and fdict.get('vcodec') != 'none'):
c57f7757
PH
3550 if res:
3551 res += ', '
3552 res += fdict['vcodec']
91c7271a 3553 if fdict.get('vbr') is not None:
c57f7757
PH
3554 res += '@'
3555 elif fdict.get('vbr') is not None and fdict.get('abr') is not None:
3556 res += 'video@'
3557 if fdict.get('vbr') is not None:
3558 res += '%4dk' % fdict['vbr']
fbb21cf5 3559 if fdict.get('fps') is not None:
5d583bdf
S
3560 if res:
3561 res += ', '
3562 res += '%sfps' % fdict['fps']
c57f7757
PH
3563 if fdict.get('acodec') is not None:
3564 if res:
3565 res += ', '
3566 if fdict['acodec'] == 'none':
3567 res += 'video only'
3568 else:
3569 res += '%-5s' % fdict['acodec']
3570 elif fdict.get('abr') is not None:
3571 if res:
3572 res += ', '
3573 res += 'audio'
3574 if fdict.get('abr') is not None:
3575 res += '@%3dk' % fdict['abr']
3576 if fdict.get('asr') is not None:
3577 res += ' (%5dHz)' % fdict['asr']
3578 if fdict.get('filesize') is not None:
3579 if res:
3580 res += ', '
3581 res += format_bytes(fdict['filesize'])
9732d77e
PH
3582 elif fdict.get('filesize_approx') is not None:
3583 if res:
3584 res += ', '
3585 res += '~' + format_bytes(fdict['filesize_approx'])
c57f7757 3586 return res
91c7271a 3587
aebb4f4b 3588 def _get_formats(self, info_dict):
3589 if info_dict.get('formats') is None:
3590 if info_dict.get('url') and info_dict.get('_type', 'video') == 'video':
3591 return [info_dict]
3592 return []
3593 return info_dict['formats']
b69fd25c 3594
aebb4f4b 3595 def render_formats_table(self, info_dict):
3596 formats = self._get_formats(info_dict)
3597 if not formats:
3598 return
8130779d 3599 if not self.params.get('listformats_table', True) is not False:
76d321f6 3600 table = [
3601 [
3602 format_field(f, 'format_id'),
3603 format_field(f, 'ext'),
3604 self.format_resolution(f),
8130779d 3605 self._format_note(f)
d5d1df8a 3606 ] for f in formats if (f.get('preference') or 0) >= -1000]
8130779d 3607 return render_table(['format code', 'extension', 'resolution', 'note'], table, extra_gap=1)
3608
d816f61f 3609 def simplified_codec(f, field):
3610 assert field in ('acodec', 'vcodec')
3611 codec = f.get(field, 'unknown')
f5ea4748 3612 if not codec:
3613 return 'unknown'
3614 elif codec != 'none':
d816f61f 3615 return '.'.join(codec.split('.')[:4])
3616
3617 if field == 'vcodec' and f.get('acodec') == 'none':
3618 return 'images'
3619 elif field == 'acodec' and f.get('vcodec') == 'none':
3620 return ''
3621 return self._format_out('audio only' if field == 'vcodec' else 'video only',
3622 self.Styles.SUPPRESS)
3623
591bb9d3 3624 delim = self._format_out('\u2502', self.Styles.DELIM, '|', test_encoding=True)
8130779d 3625 table = [
3626 [
591bb9d3 3627 self._format_out(format_field(f, 'format_id'), self.Styles.ID),
8130779d 3628 format_field(f, 'ext'),
3629 format_field(f, func=self.format_resolution, ignore=('audio only', 'images')),
563e0bf8 3630 format_field(f, 'fps', '\t%d', func=round),
8130779d 3631 format_field(f, 'dynamic_range', '%s', ignore=(None, 'SDR')).replace('HDR', ''),
b8ed0f15 3632 format_field(f, 'audio_channels', '\t%s'),
8130779d 3633 delim,
3634 format_field(f, 'filesize', ' \t%s', func=format_bytes) + format_field(f, 'filesize_approx', '~\t%s', func=format_bytes),
563e0bf8 3635 format_field(f, 'tbr', '\t%dk', func=round),
8130779d 3636 shorten_protocol_name(f.get('protocol', '')),
3637 delim,
d816f61f 3638 simplified_codec(f, 'vcodec'),
563e0bf8 3639 format_field(f, 'vbr', '\t%dk', func=round),
d816f61f 3640 simplified_codec(f, 'acodec'),
563e0bf8 3641 format_field(f, 'abr', '\t%dk', func=round),
ae61d108 3642 format_field(f, 'asr', '\t%s', func=format_decimal_suffix),
8130779d 3643 join_nonempty(
591bb9d3 3644 self._format_out('UNSUPPORTED', 'light red') if f.get('ext') in ('f4f', 'f4m') else None,
8130779d 3645 format_field(f, 'language', '[%s]'),
3646 join_nonempty(format_field(f, 'format_note'),
3647 format_field(f, 'container', ignore=(None, f.get('ext'))),
3648 delim=', '),
3649 delim=' '),
3650 ] for f in formats if f.get('preference') is None or f['preference'] >= -1000]
3651 header_line = self._list_format_headers(
b8ed0f15 3652 'ID', 'EXT', 'RESOLUTION', '\tFPS', 'HDR', 'CH', delim, '\tFILESIZE', '\tTBR', 'PROTO',
8130779d 3653 delim, 'VCODEC', '\tVBR', 'ACODEC', '\tABR', '\tASR', 'MORE INFO')
3654
3655 return render_table(
3656 header_line, table, hide_empty=True,
591bb9d3 3657 delim=self._format_out('\u2500', self.Styles.DELIM, '-', test_encoding=True))
8130779d 3658
3659 def render_thumbnails_table(self, info_dict):
88f23a18 3660 thumbnails = list(info_dict.get('thumbnails') or [])
cfb56d1a 3661 if not thumbnails:
8130779d 3662 return None
3663 return render_table(
ec11a9f4 3664 self._list_format_headers('ID', 'Width', 'Height', 'URL'),
177662e0 3665 [[t.get('id'), t.get('width') or 'unknown', t.get('height') or 'unknown', t['url']] for t in thumbnails])
2412044c 3666
8130779d 3667 def render_subtitles_table(self, video_id, subtitles):
2412044c 3668 def _row(lang, formats):
49c258e1 3669 exts, names = zip(*((f['ext'], f.get('name') or 'unknown') for f in reversed(formats)))
2412044c 3670 if len(set(names)) == 1:
7aee40c1 3671 names = [] if names[0] == 'unknown' else names[:1]
2412044c 3672 return [lang, ', '.join(names), ', '.join(exts)]
3673
8130779d 3674 if not subtitles:
3675 return None
3676 return render_table(
ec11a9f4 3677 self._list_format_headers('Language', 'Name', 'Formats'),
2412044c 3678 [_row(lang, formats) for lang, formats in subtitles.items()],
8130779d 3679 hide_empty=True)
3680
3681 def __list_table(self, video_id, name, func, *args):
3682 table = func(*args)
3683 if not table:
3684 self.to_screen(f'{video_id} has no {name}')
3685 return
3686 self.to_screen(f'[info] Available {name} for {video_id}:')
3687 self.to_stdout(table)
3688
3689 def list_formats(self, info_dict):
3690 self.__list_table(info_dict['id'], 'formats', self.render_formats_table, info_dict)
3691
3692 def list_thumbnails(self, info_dict):
3693 self.__list_table(info_dict['id'], 'thumbnails', self.render_thumbnails_table, info_dict)
3694
3695 def list_subtitles(self, video_id, subtitles, name='subtitles'):
3696 self.__list_table(video_id, name, self.render_subtitles_table, video_id, subtitles)
a504ced0 3697
dca08720
PH
3698 def urlopen(self, req):
3699 """ Start an HTTP download """
f9934b96 3700 if isinstance(req, str):
67dda517 3701 req = sanitized_Request(req)
19a41fc6 3702 return self._opener.open(req, timeout=self._socket_timeout)
dca08720
PH
3703
3704 def print_debug_header(self):
3705 if not self.params.get('verbose'):
3706 return
49a57e70 3707
a057779d 3708 from . import _IN_CLI # Must be delayed import
3709
560738f3 3710 # These imports can be slow. So import them only as needed
3711 from .extractor.extractors import _LAZY_LOADER
3712 from .extractor.extractors import _PLUGIN_CLASSES as plugin_extractors
3713
49a57e70 3714 def get_encoding(stream):
2a938746 3715 ret = str(getattr(stream, 'encoding', 'missing (%s)' % type(stream).__name__))
49a57e70 3716 if not supports_terminal_sequences(stream):
53973b4d 3717 from .utils import WINDOWS_VT_MODE # Must be imported locally
e3c7d495 3718 ret += ' (No VT)' if WINDOWS_VT_MODE is False else ' (No ANSI)'
49a57e70 3719 return ret
3720
591bb9d3 3721 encoding_str = 'Encodings: locale %s, fs %s, pref %s, %s' % (
49a57e70 3722 locale.getpreferredencoding(),
3723 sys.getfilesystemencoding(),
591bb9d3 3724 self.get_encoding(),
3725 ', '.join(
64fa820c 3726 f'{key} {get_encoding(stream)}' for key, stream in self._out_files.items_
591bb9d3 3727 if stream is not None and key != 'console')
3728 )
883d4b1e 3729
3730 logger = self.params.get('logger')
3731 if logger:
3732 write_debug = lambda msg: logger.debug(f'[debug] {msg}')
3733 write_debug(encoding_str)
3734 else:
96565c7e 3735 write_string(f'[debug] {encoding_str}\n', encoding=None)
49a57e70 3736 write_debug = lambda msg: self._write_string(f'[debug] {msg}\n')
734f90bb 3737
4c88ff87 3738 source = detect_variant()
70b23409 3739 if VARIANT not in (None, 'pip'):
3740 source += '*'
36eaf303 3741 write_debug(join_nonempty(
b5e7a2e6 3742 f'{"yt-dlp" if REPOSITORY == "yt-dlp/yt-dlp" else REPOSITORY} version',
3743 __version__,
36eaf303 3744 f'[{RELEASE_GIT_HEAD}]' if RELEASE_GIT_HEAD else '',
3745 '' if source == 'unknown' else f'({source})',
a057779d 3746 '' if _IN_CLI else 'API',
36eaf303 3747 delim=' '))
497074f0 3748
3749 if not _IN_CLI:
3750 write_debug(f'params: {self.params}')
3751
6e21fdd2 3752 if not _LAZY_LOADER:
3753 if os.environ.get('YTDLP_NO_LAZY_EXTRACTORS'):
49a57e70 3754 write_debug('Lazy loading extractors is forcibly disabled')
6e21fdd2 3755 else:
49a57e70 3756 write_debug('Lazy loading extractors is disabled')
3ae5e797 3757 if plugin_extractors or plugin_postprocessors:
49a57e70 3758 write_debug('Plugins: %s' % [
3ae5e797 3759 '%s%s' % (klass.__name__, '' if klass.__name__ == name else f' as {name}')
3760 for name, klass in itertools.chain(plugin_extractors.items(), plugin_postprocessors.items())])
8a82af35 3761 if self.params['compat_opts']:
3762 write_debug('Compatibility options: %s' % ', '.join(self.params['compat_opts']))
36eaf303 3763
b5e7a2e6 3764 if current_git_head():
3765 write_debug(f'Git HEAD: {current_git_head()}')
b1f94422 3766 write_debug(system_identifier())
d28b5171 3767
8913ef74 3768 exe_versions, ffmpeg_features = FFmpegPostProcessor.get_versions_and_features(self)
3769 ffmpeg_features = {key for key, val in ffmpeg_features.items() if val}
3770 if ffmpeg_features:
19a03940 3771 exe_versions['ffmpeg'] += ' (%s)' % ','.join(sorted(ffmpeg_features))
8913ef74 3772
4c83c967 3773 exe_versions['rtmpdump'] = rtmpdump_version()
feee8d32 3774 exe_versions['phantomjs'] = PhantomJSwrapper._version()
d28b5171 3775 exe_str = ', '.join(
2831b468 3776 f'{exe} {v}' for exe, v in sorted(exe_versions.items()) if v
3777 ) or 'none'
49a57e70 3778 write_debug('exe versions: %s' % exe_str)
dca08720 3779
1d485a1a 3780 from .compat.compat_utils import get_package_info
9b8ee23b 3781 from .dependencies import available_dependencies
3782
3783 write_debug('Optional libraries: %s' % (', '.join(sorted({
1d485a1a 3784 join_nonempty(*get_package_info(m)) for m in available_dependencies.values()
9b8ee23b 3785 })) or 'none'))
2831b468 3786
97ec5bc5 3787 self._setup_opener()
dca08720
PH
3788 proxy_map = {}
3789 for handler in self._opener.handlers:
3790 if hasattr(handler, 'proxies'):
3791 proxy_map.update(handler.proxies)
49a57e70 3792 write_debug(f'Proxy map: {proxy_map}')
dca08720 3793
49a57e70 3794 # Not implemented
3795 if False and self.params.get('call_home'):
0f06bcd7 3796 ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode()
49a57e70 3797 write_debug('Public IP address: %s' % ipaddr)
58b1f00d 3798 latest_version = self.urlopen(
0f06bcd7 3799 'https://yt-dl.org/latest/version').read().decode()
58b1f00d
PH
3800 if version_tuple(latest_version) > version_tuple(__version__):
3801 self.report_warning(
3802 'You are using an outdated version (newest version: %s)! '
3803 'See https://yt-dl.org/update if you need help updating.' %
3804 latest_version)
3805
e344693b 3806 def _setup_opener(self):
97ec5bc5 3807 if hasattr(self, '_opener'):
3808 return
6ad14cab 3809 timeout_val = self.params.get('socket_timeout')
17bddf3e 3810 self._socket_timeout = 20 if timeout_val is None else float(timeout_val)
6ad14cab 3811
982ee69a 3812 opts_cookiesfrombrowser = self.params.get('cookiesfrombrowser')
dca08720
PH
3813 opts_cookiefile = self.params.get('cookiefile')
3814 opts_proxy = self.params.get('proxy')
3815
982ee69a 3816 self.cookiejar = load_cookies(opts_cookiefile, opts_cookiesfrombrowser, self)
dca08720 3817
6a3f4c3f 3818 cookie_processor = YoutubeDLCookieProcessor(self.cookiejar)
dca08720
PH
3819 if opts_proxy is not None:
3820 if opts_proxy == '':
3821 proxies = {}
3822 else:
3823 proxies = {'http': opts_proxy, 'https': opts_proxy}
3824 else:
ac668111 3825 proxies = urllib.request.getproxies()
067aa17e 3826 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
dca08720
PH
3827 if 'http' in proxies and 'https' not in proxies:
3828 proxies['https'] = proxies['http']
91410c9b 3829 proxy_handler = PerRequestProxyHandler(proxies)
a0ddb8a2
PH
3830
3831 debuglevel = 1 if self.params.get('debug_printtraffic') else 0
be4a824d
PH
3832 https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel)
3833 ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel)
fca6dba8 3834 redirect_handler = YoutubeDLRedirectHandler()
f9934b96 3835 data_handler = urllib.request.DataHandler()
6240b0a2
JMF
3836
3837 # When passing our own FileHandler instance, build_opener won't add the
3838 # default FileHandler and allows us to disable the file protocol, which
3839 # can be used for malicious purposes (see
067aa17e 3840 # https://github.com/ytdl-org/youtube-dl/issues/8227)
ac668111 3841 file_handler = urllib.request.FileHandler()
6240b0a2
JMF
3842
3843 def file_open(*args, **kwargs):
ac668111 3844 raise urllib.error.URLError('file:// scheme is explicitly disabled in yt-dlp for security reasons')
6240b0a2
JMF
3845 file_handler.file_open = file_open
3846
ac668111 3847 opener = urllib.request.build_opener(
fca6dba8 3848 proxy_handler, https_handler, cookie_processor, ydlh, redirect_handler, data_handler, file_handler)
2461f79d 3849
dca08720
PH
3850 # Delete the default user-agent header, which would otherwise apply in
3851 # cases where our custom HTTP handler doesn't come into play
067aa17e 3852 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
dca08720
PH
3853 opener.addheaders = []
3854 self._opener = opener
62fec3b2
PH
3855
3856 def encode(self, s):
3857 if isinstance(s, bytes):
3858 return s # Already encoded
3859
3860 try:
3861 return s.encode(self.get_encoding())
3862 except UnicodeEncodeError as err:
3863 err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.'
3864 raise
3865
3866 def get_encoding(self):
3867 encoding = self.params.get('encoding')
3868 if encoding is None:
3869 encoding = preferredencoding()
3870 return encoding
ec82d85a 3871
e08a85d8 3872 def _write_info_json(self, label, ie_result, infofn, overwrite=None):
cb96c5be 3873 ''' Write infojson and returns True = written, 'exists' = Already exists, False = skip, None = error '''
e08a85d8 3874 if overwrite is None:
3875 overwrite = self.params.get('overwrites', True)
80c03fa9 3876 if not self.params.get('writeinfojson'):
3877 return False
3878 elif not infofn:
3879 self.write_debug(f'Skipping writing {label} infojson')
3880 return False
3881 elif not self._ensure_dir_exists(infofn):
3882 return None
e08a85d8 3883 elif not overwrite and os.path.exists(infofn):
80c03fa9 3884 self.to_screen(f'[info] {label.title()} metadata is already present')
cb96c5be 3885 return 'exists'
3886
3887 self.to_screen(f'[info] Writing {label} metadata as JSON to: {infofn}')
3888 try:
3889 write_json_file(self.sanitize_info(ie_result, self.params.get('clean_infojson', True)), infofn)
3890 return True
86e5f3ed 3891 except OSError:
cb96c5be 3892 self.report_error(f'Cannot write {label} metadata to JSON file {infofn}')
3893 return None
80c03fa9 3894
3895 def _write_description(self, label, ie_result, descfn):
3896 ''' Write description and returns True = written, False = skip, None = error '''
3897 if not self.params.get('writedescription'):
3898 return False
3899 elif not descfn:
3900 self.write_debug(f'Skipping writing {label} description')
3901 return False
3902 elif not self._ensure_dir_exists(descfn):
3903 return None
3904 elif not self.params.get('overwrites', True) and os.path.exists(descfn):
3905 self.to_screen(f'[info] {label.title()} description is already present')
3906 elif ie_result.get('description') is None:
3907 self.report_warning(f'There\'s no {label} description to write')
3908 return False
3909 else:
3910 try:
3911 self.to_screen(f'[info] Writing {label} description to: {descfn}')
86e5f3ed 3912 with open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
80c03fa9 3913 descfile.write(ie_result['description'])
86e5f3ed 3914 except OSError:
80c03fa9 3915 self.report_error(f'Cannot write {label} description file {descfn}')
3916 return None
3917 return True
3918
3919 def _write_subtitles(self, info_dict, filename):
3920 ''' Write subtitles to file and return list of (sub_filename, final_sub_filename); or None if error'''
3921 ret = []
3922 subtitles = info_dict.get('requested_subtitles')
3923 if not subtitles or not (self.params.get('writesubtitles') or self.params.get('writeautomaticsub')):
3924 # subtitles download errors are already managed as troubles in relevant IE
3925 # that way it will silently go on when used with unsupporting IE
3926 return ret
3927
3928 sub_filename_base = self.prepare_filename(info_dict, 'subtitle')
3929 if not sub_filename_base:
3930 self.to_screen('[info] Skipping writing video subtitles')
3931 return ret
3932 for sub_lang, sub_info in subtitles.items():
3933 sub_format = sub_info['ext']
3934 sub_filename = subtitles_filename(filename, sub_lang, sub_format, info_dict.get('ext'))
3935 sub_filename_final = subtitles_filename(sub_filename_base, sub_lang, sub_format, info_dict.get('ext'))
e04938ab 3936 existing_sub = self.existing_file((sub_filename_final, sub_filename))
3937 if existing_sub:
80c03fa9 3938 self.to_screen(f'[info] Video subtitle {sub_lang}.{sub_format} is already present')
e04938ab 3939 sub_info['filepath'] = existing_sub
3940 ret.append((existing_sub, sub_filename_final))
80c03fa9 3941 continue
3942
3943 self.to_screen(f'[info] Writing video subtitles to: {sub_filename}')
3944 if sub_info.get('data') is not None:
3945 try:
3946 # Use newline='' to prevent conversion of newline characters
3947 # See https://github.com/ytdl-org/youtube-dl/issues/10268
86e5f3ed 3948 with open(sub_filename, 'w', encoding='utf-8', newline='') as subfile:
80c03fa9 3949 subfile.write(sub_info['data'])
3950 sub_info['filepath'] = sub_filename
3951 ret.append((sub_filename, sub_filename_final))
3952 continue
86e5f3ed 3953 except OSError:
80c03fa9 3954 self.report_error(f'Cannot write video subtitles file {sub_filename}')
3955 return None
3956
3957 try:
3958 sub_copy = sub_info.copy()
3959 sub_copy.setdefault('http_headers', info_dict.get('http_headers'))
3960 self.dl(sub_filename, sub_copy, subtitle=True)
3961 sub_info['filepath'] = sub_filename
3962 ret.append((sub_filename, sub_filename_final))
6020e05d 3963 except (DownloadError, ExtractorError, IOError, OSError, ValueError) + network_exceptions as err:
c70c418d 3964 msg = f'Unable to download video subtitles for {sub_lang!r}: {err}'
6020e05d 3965 if self.params.get('ignoreerrors') is not True: # False or 'only_download'
c70c418d 3966 if not self.params.get('ignoreerrors'):
3967 self.report_error(msg)
3968 raise DownloadError(msg)
3969 self.report_warning(msg)
519804a9 3970 return ret
80c03fa9 3971
3972 def _write_thumbnails(self, label, info_dict, filename, thumb_filename_base=None):
3973 ''' Write thumbnails to file and return list of (thumb_filename, final_thumb_filename) '''
6c4fd172 3974 write_all = self.params.get('write_all_thumbnails', False)
80c03fa9 3975 thumbnails, ret = [], []
6c4fd172 3976 if write_all or self.params.get('writethumbnail', False):
0202b52a 3977 thumbnails = info_dict.get('thumbnails') or []
6c4fd172 3978 multiple = write_all and len(thumbnails) > 1
ec82d85a 3979
80c03fa9 3980 if thumb_filename_base is None:
3981 thumb_filename_base = filename
3982 if thumbnails and not thumb_filename_base:
3983 self.write_debug(f'Skipping writing {label} thumbnail')
3984 return ret
3985
dd0228ce 3986 for idx, t in list(enumerate(thumbnails))[::-1]:
80c03fa9 3987 thumb_ext = (f'{t["id"]}.' if multiple else '') + determine_ext(t['url'], 'jpg')
aa9369a2 3988 thumb_display_id = f'{label} thumbnail {t["id"]}'
80c03fa9 3989 thumb_filename = replace_extension(filename, thumb_ext, info_dict.get('ext'))
3990 thumb_filename_final = replace_extension(thumb_filename_base, thumb_ext, info_dict.get('ext'))
ec82d85a 3991
e04938ab 3992 existing_thumb = self.existing_file((thumb_filename_final, thumb_filename))
3993 if existing_thumb:
aa9369a2 3994 self.to_screen('[info] %s is already present' % (
3995 thumb_display_id if multiple else f'{label} thumbnail').capitalize())
e04938ab 3996 t['filepath'] = existing_thumb
3997 ret.append((existing_thumb, thumb_filename_final))
ec82d85a 3998 else:
80c03fa9 3999 self.to_screen(f'[info] Downloading {thumb_display_id} ...')
ec82d85a 4000 try:
297e9952 4001 uf = self.urlopen(sanitized_Request(t['url'], headers=t.get('http_headers', {})))
80c03fa9 4002 self.to_screen(f'[info] Writing {thumb_display_id} to: {thumb_filename}')
d3d89c32 4003 with open(encodeFilename(thumb_filename), 'wb') as thumbf:
ec82d85a 4004 shutil.copyfileobj(uf, thumbf)
80c03fa9 4005 ret.append((thumb_filename, thumb_filename_final))
885cc0b7 4006 t['filepath'] = thumb_filename
3158150c 4007 except network_exceptions as err:
dd0228ce 4008 thumbnails.pop(idx)
80c03fa9 4009 self.report_warning(f'Unable to download {thumb_display_id}: {err}')
6c4fd172 4010 if ret and not write_all:
4011 break
0202b52a 4012 return ret