]> jfr.im git - yt-dlp.git/blame - yt_dlp/YoutubeDL.py
[extractor/rcs] Fix embed extraction
[yt-dlp.git] / yt_dlp / YoutubeDL.py
CommitLineData
26e63931 1import collections
31bd3925 2import contextlib
9d2ecdbc 3import datetime
c1c9a79c 4import errno
31bd3925 5import fileinput
b5ae35ee 6import functools
8222d8de 7import io
b82f815f 8import itertools
8694c600 9import json
62fec3b2 10import locale
083c9df9 11import operator
8222d8de 12import os
f8271158 13import random
8222d8de
JMF
14import re
15import shutil
dca08720 16import subprocess
8222d8de 17import sys
21cd8fae 18import tempfile
8222d8de 19import time
67134eab 20import tokenize
8222d8de 21import traceback
524e2e4f 22import unicodedata
f9934b96 23import urllib.request
961ea474
S
24from string import ascii_letters
25
f8271158 26from .cache import Cache
14f25df2 27from .compat import compat_os_name, compat_shlex_quote
982ee69a 28from .cookies import load_cookies
f8271158 29from .downloader import FFmpegFD, get_suitable_downloader, shorten_protocol_name
30from .downloader.rtmp import rtmpdump_version
f8271158 31from .extractor import gen_extractor_classes, get_info_extractor
fe7866d0 32from .extractor.common import UnsupportedURLIE
f8271158 33from .extractor.openload import PhantomJSwrapper
34from .minicurses import format_text
35from .postprocessor import _PLUGIN_CLASSES as plugin_postprocessors
36from .postprocessor import (
37 EmbedThumbnailPP,
38 FFmpegFixupDuplicateMoovPP,
39 FFmpegFixupDurationPP,
40 FFmpegFixupM3u8PP,
41 FFmpegFixupM4aPP,
42 FFmpegFixupStretchedPP,
43 FFmpegFixupTimestampPP,
44 FFmpegMergerPP,
45 FFmpegPostProcessor,
ca9def71 46 FFmpegVideoConvertorPP,
f8271158 47 MoveFilesAfterDownloadPP,
48 get_postprocessor,
49)
ca9def71 50from .postprocessor.ffmpeg import resolve_mapping as resolve_recode_mapping
b5e7a2e6 51from .update import REPOSITORY, current_git_head, detect_variant
8c25f81b 52from .utils import (
f8271158 53 DEFAULT_OUTTMPL,
7b2c3f47 54 IDENTITY,
f8271158 55 LINK_TEMPLATES,
8dc59305 56 MEDIA_EXTENSIONS,
f8271158 57 NO_DEFAULT,
1d485a1a 58 NUMBER_RE,
f8271158 59 OUTTMPL_TYPES,
60 POSTPROCESS_WHEN,
61 STR_FORMAT_RE_TMPL,
62 STR_FORMAT_TYPES,
63 ContentTooShortError,
64 DateRange,
65 DownloadCancelled,
66 DownloadError,
67 EntryNotInPlaylist,
68 ExistingVideoReached,
69 ExtractorError,
70 GeoRestrictedError,
71 HEADRequest,
f8271158 72 ISO3166Utils,
73 LazyList,
74 MaxDownloadsReached,
19a03940 75 Namespace,
f8271158 76 PagedList,
77 PerRequestProxyHandler,
7e88d7d7 78 PlaylistEntries,
f8271158 79 Popen,
80 PostProcessingError,
81 ReExtractInfo,
82 RejectedVideoReached,
83 SameFileError,
84 UnavailableVideoError,
693f0600 85 UserNotLive,
f8271158 86 YoutubeDLCookieProcessor,
87 YoutubeDLHandler,
88 YoutubeDLRedirectHandler,
eedb7ba5
S
89 age_restricted,
90 args_to_str,
cb794ee0 91 bug_reports_message,
ce02ed60 92 date_from_str,
da4db748 93 deprecation_warning,
ce02ed60 94 determine_ext,
b5559424 95 determine_protocol,
c0384f22 96 encode_compat_str,
ce02ed60 97 encodeFilename,
a06916d9 98 error_to_compat_str,
47cdc68e 99 escapeHTML,
590bc6f6 100 expand_path,
90137ca4 101 filter_dict,
e29663c6 102 float_or_none,
02dbf93f 103 format_bytes,
e0fd9573 104 format_decimal_suffix,
f8271158 105 format_field,
525ef922 106 formatSeconds,
fc61aff4 107 get_compatible_ext,
0bb322b9 108 get_domain,
c9969434 109 int_or_none,
732044af 110 iri_to_uri,
941e881e 111 is_path_like,
34921b43 112 join_nonempty,
ce02ed60 113 locked_file,
0647d925 114 make_archive_id,
0202b52a 115 make_dir,
dca08720 116 make_HTTPS_handler,
8b7539d2 117 merge_headers,
3158150c 118 network_exceptions,
ec11a9f4 119 number_of_digits,
cd6fc19e 120 orderedSet,
5314b521 121 orderedSet_from_options,
083c9df9 122 parse_filesize,
ce02ed60 123 preferredencoding,
eedb7ba5 124 prepend_extension,
51fb4995 125 register_socks_protocols,
3efb96a6 126 remove_terminal_sequences,
cfb56d1a 127 render_table,
eedb7ba5 128 replace_extension,
ce02ed60 129 sanitize_filename,
1bb5c511 130 sanitize_path,
dcf77cf1 131 sanitize_url,
67dda517 132 sanitized_Request,
e5660ee6 133 std_headers,
1211bb6d 134 str_or_none,
e29663c6 135 strftime_or_none,
ce02ed60 136 subtitles_filename,
819e0531 137 supports_terminal_sequences,
b1f94422 138 system_identifier,
f2ebc5c7 139 timetuple_from_msec,
732044af 140 to_high_limit_path,
324ad820 141 traverse_obj,
fc61aff4 142 try_call,
6033d980 143 try_get,
29eb5174 144 url_basename,
7d1eb38a 145 variadic,
58b1f00d 146 version_tuple,
53973b4d 147 windows_enable_vt_mode,
ce02ed60
PH
148 write_json_file,
149 write_string,
4f026faf 150)
70b23409 151from .version import RELEASE_GIT_HEAD, VARIANT, __version__
8222d8de 152
e9c0cdd3
YCH
153if compat_os_name == 'nt':
154 import ctypes
155
2459b6e1 156
86e5f3ed 157class YoutubeDL:
8222d8de
JMF
158 """YoutubeDL class.
159
160 YoutubeDL objects are the ones responsible of downloading the
161 actual video file and writing it to disk if the user has requested
162 it, among some other tasks. In most cases there should be one per
163 program. As, given a video URL, the downloader doesn't know how to
164 extract all the needed information, task that InfoExtractors do, it
165 has to pass the URL to one of them.
166
167 For this, YoutubeDL objects have a method that allows
168 InfoExtractors to be registered in a given order. When it is passed
169 a URL, the YoutubeDL object handles it to the first InfoExtractor it
170 finds that reports being able to handle it. The InfoExtractor extracts
171 all the information about the video or videos the URL refers to, and
172 YoutubeDL process the extracted information, possibly using a File
173 Downloader to download the video.
174
175 YoutubeDL objects accept a lot of parameters. In order not to saturate
176 the object constructor with arguments, it receives a dictionary of
177 options instead. These options are available through the params
178 attribute for the InfoExtractors to use. The YoutubeDL also
179 registers itself as the downloader in charge for the InfoExtractors
180 that are added to it, so this is a "mutual registration".
181
182 Available options:
183
184 username: Username for authentication purposes.
185 password: Password for authentication purposes.
180940e0 186 videopassword: Password for accessing a video.
1da50aa3
S
187 ap_mso: Adobe Pass multiple-system operator identifier.
188 ap_username: Multiple-system operator account username.
189 ap_password: Multiple-system operator account password.
8222d8de
JMF
190 usenetrc: Use netrc for authentication instead.
191 verbose: Print additional info to stdout.
192 quiet: Do not print messages to stdout.
ad8915b7 193 no_warnings: Do not print out anything for warnings.
bb66c247 194 forceprint: A dict with keys WHEN mapped to a list of templates to
195 print to stdout. The allowed keys are video or any of the
196 items in utils.POSTPROCESS_WHEN.
ca30f449 197 For compatibility, a single list is also accepted
bb66c247 198 print_to_file: A dict with keys WHEN (same as forceprint) mapped to
199 a list of tuples with (template, filename)
8694c600 200 forcejson: Force printing info_dict as JSON.
63e0be34
PH
201 dump_single_json: Force printing the info_dict of the whole playlist
202 (or video) as a single JSON line.
c25228e5 203 force_write_download_archive: Force writing download archive regardless
204 of 'skip_download' or 'simulate'.
b7b04c78 205 simulate: Do not download the video files. If unset (or None),
206 simulate only if listsubtitles, listformats or list_thumbnails is used
eb8a4433 207 format: Video format code. see "FORMAT SELECTION" for more details.
093a1710 208 You can also pass a function. The function takes 'ctx' as
209 argument and returns the formats to download.
210 See "build_format_selector" for an implementation
63ad4d43 211 allow_unplayable_formats: Allow unplayable formats to be extracted and downloaded.
b7da73eb 212 ignore_no_formats_error: Ignore "No video formats" error. Usefull for
213 extracting metadata even if the video is not actually
214 available for download (experimental)
0930b11f 215 format_sort: A list of fields by which to sort the video formats.
216 See "Sorting Formats" for more details.
c25228e5 217 format_sort_force: Force the given format_sort. see "Sorting Formats"
218 for more details.
08d30158 219 prefer_free_formats: Whether to prefer video formats with free containers
220 over non-free ones of same quality.
c25228e5 221 allow_multiple_video_streams: Allow multiple video streams to be merged
222 into a single file
223 allow_multiple_audio_streams: Allow multiple audio streams to be merged
224 into a single file
0ba692ac 225 check_formats Whether to test if the formats are downloadable.
9f1a1c36 226 Can be True (check all), False (check none),
227 'selected' (check selected formats),
0ba692ac 228 or None (check only if requested by extractor)
4524baf0 229 paths: Dictionary of output paths. The allowed keys are 'home'
230 'temp' and the keys of OUTTMPL_TYPES (in utils.py)
de6000d9 231 outtmpl: Dictionary of templates for output names. Allowed keys
4524baf0 232 are 'default' and the keys of OUTTMPL_TYPES (in utils.py).
34488702 233 For compatibility with youtube-dl, a single string can also be used
a820dc72
RA
234 outtmpl_na_placeholder: Placeholder for unavailable meta fields.
235 restrictfilenames: Do not allow "&" and spaces in file names
236 trim_file_name: Limit length of filename (extension excluded)
4524baf0 237 windowsfilenames: Force the filenames to be windows compatible
b1940459 238 ignoreerrors: Do not stop on download/postprocessing errors.
239 Can be 'only_download' to ignore only download errors.
240 Default is 'only_download' for CLI, but False for API
26e2805c 241 skip_playlist_after_errors: Number of allowed failures until the rest of
242 the playlist is skipped
fe7866d0 243 allowed_extractors: List of regexes to match against extractor names that are allowed
0c3d0f51 244 overwrites: Overwrite all video and metadata files if True,
245 overwrite only non-video files if None
246 and don't overwrite any file if False
34488702 247 For compatibility with youtube-dl,
248 "nooverwrites" may also be used instead
c14e88f0 249 playlist_items: Specific indices of playlist to download.
75822ca7 250 playlistrandom: Download playlist items in random order.
7e9a6125 251 lazy_playlist: Process playlist entries as they are received.
8222d8de
JMF
252 matchtitle: Download only matching titles.
253 rejecttitle: Reject downloads for matching titles.
8bf9319e 254 logger: Log messages to a logging.Logger instance.
17ffed18 255 logtostderr: Print everything to stderr instead of stdout.
256 consoletitle: Display progress in console window's titlebar.
8222d8de
JMF
257 writedescription: Write the video description to a .description file
258 writeinfojson: Write the video description to a .info.json file
75d43ca0 259 clean_infojson: Remove private fields from the infojson
34488702 260 getcomments: Extract video comments. This will not be written to disk
06167fbb 261 unless writeinfojson is also given
1fb07d10 262 writeannotations: Write the video annotations to a .annotations.xml file
8222d8de 263 writethumbnail: Write the thumbnail image to a file
c25228e5 264 allow_playlist_files: Whether to write playlists' description, infojson etc
265 also to disk when using the 'write*' options
ec82d85a 266 write_all_thumbnails: Write all thumbnail formats to files
732044af 267 writelink: Write an internet shortcut file, depending on the
268 current platform (.url/.webloc/.desktop)
269 writeurllink: Write a Windows internet shortcut file (.url)
270 writewebloclink: Write a macOS internet shortcut file (.webloc)
271 writedesktoplink: Write a Linux internet shortcut file (.desktop)
8222d8de 272 writesubtitles: Write the video subtitles to a file
741dd8ea 273 writeautomaticsub: Write the automatically generated subtitles to a file
8222d8de 274 listsubtitles: Lists all available subtitles for the video
a504ced0 275 subtitlesformat: The format code for subtitles
c32b0aab 276 subtitleslangs: List of languages of the subtitles to download (can be regex).
277 The list may contain "all" to refer to all the available
278 subtitles. The language can be prefixed with a "-" to
62b58c09 279 exclude it from the requested languages, e.g. ['all', '-live_chat']
8222d8de
JMF
280 keepvideo: Keep the video file after post-processing
281 daterange: A DateRange object, download only if the upload_date is in the range.
282 skip_download: Skip the actual download of the video file
c35f9e72 283 cachedir: Location of the cache files in the filesystem.
a0e07d31 284 False to disable filesystem cache.
47192f92 285 noplaylist: Download single video instead of a playlist if in doubt.
8dbe9899
PH
286 age_limit: An integer representing the user's age in years.
287 Unsuitable videos for the given age are skipped.
5fe18bdb
PH
288 min_views: An integer representing the minimum view count the video
289 must have in order to not be skipped.
290 Videos without view count information are always
291 downloaded. None for no limit.
292 max_views: An integer representing the maximum view count.
293 Videos that are more popular than that are not
294 downloaded.
295 Videos without view count information are always
296 downloaded. None for no limit.
ae103564 297 download_archive: A set, or the name of a file where all downloads are recorded.
298 Videos already present in the file are not downloaded again.
8a51f564 299 break_on_existing: Stop the download process after attempting to download a
300 file that is in the archive.
301 break_on_reject: Stop the download process when encountering a video that
302 has been filtered out.
b222c271 303 break_per_url: Whether break_on_reject and break_on_existing
304 should act on each input URL as opposed to for the entire queue
d76fa1f3 305 cookiefile: File name or text stream from where cookies should be read and dumped to
f59f5ef8 306 cookiesfrombrowser: A tuple containing the name of the browser, the profile
9bd13fe5 307 name/path from where cookies are loaded, the name of the keyring,
308 and the container name, e.g. ('chrome', ) or
309 ('vivaldi', 'default', 'BASICTEXT') or ('firefox', 'default', None, 'Meta')
f81c62a6 310 legacyserverconnect: Explicitly allow HTTPS connection to servers that do not
311 support RFC 5746 secure renegotiation
f59f5ef8 312 nocheckcertificate: Do not verify SSL certificates
bb58c9ed 313 client_certificate: Path to client certificate file in PEM format. May include the private key
314 client_certificate_key: Path to private key file for client certificate
315 client_certificate_password: Password for client certificate private key, if encrypted.
316 If not provided and the key is encrypted, yt-dlp will ask interactively
7e8c0af0 317 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
c6e07cf1 318 (Only supported by some extractors)
8b7539d2 319 http_headers: A dictionary of custom headers to be used for all requests
a1ee09e8 320 proxy: URL of the proxy server to use
38cce791 321 geo_verification_proxy: URL of the proxy to use for IP address verification
504f20dd 322 on geo-restricted sites.
e344693b 323 socket_timeout: Time to wait for unresponsive hosts, in seconds
0783b09b
PH
324 bidi_workaround: Work around buggy terminals without bidirectional text
325 support, using fridibi
a0ddb8a2 326 debug_printtraffic:Print out sent and received HTTP traffic
04b4d394
PH
327 default_search: Prepend this string if an input url is not valid.
328 'auto' for elaborate guessing
62fec3b2 329 encoding: Use this encoding instead of the system-specified.
134c913c 330 extract_flat: Whether to resolve and process url_results further
331 * False: Always process (default)
332 * True: Never process
333 * 'in_playlist': Do not process inside playlist/multi_video
334 * 'discard': Always process, but don't return the result
335 from inside playlist/multi_video
336 * 'discard_in_playlist': Same as "discard", but only for
337 playlists (not multi_video)
f2ebc5c7 338 wait_for_video: If given, wait for scheduled streams to become available.
339 The value should be a tuple containing the range
340 (min_secs, max_secs) to wait between retries
4f026faf 341 postprocessors: A list of dictionaries, each with an entry
71b640cc 342 * key: The name of the postprocessor. See
7a5c1cfe 343 yt_dlp/postprocessor/__init__.py for a list.
bb66c247 344 * when: When to run the postprocessor. Allowed values are
345 the entries of utils.POSTPROCESS_WHEN
56d868db 346 Assumed to be 'post_process' if not given
71b640cc
PH
347 progress_hooks: A list of functions that get called on download
348 progress, with a dictionary with the entries
5cda4eda 349 * status: One of "downloading", "error", or "finished".
ee69b99a 350 Check this first and ignore unknown values.
3ba7740d 351 * info_dict: The extracted info_dict
71b640cc 352
5cda4eda 353 If status is one of "downloading", or "finished", the
ee69b99a
PH
354 following properties may also be present:
355 * filename: The final filename (always present)
5cda4eda 356 * tmpfilename: The filename we're currently writing to
71b640cc
PH
357 * downloaded_bytes: Bytes on disk
358 * total_bytes: Size of the whole file, None if unknown
5cda4eda
PH
359 * total_bytes_estimate: Guess of the eventual file size,
360 None if unavailable.
361 * elapsed: The number of seconds since download started.
71b640cc
PH
362 * eta: The estimated time in seconds, None if unknown
363 * speed: The download speed in bytes/second, None if
364 unknown
5cda4eda
PH
365 * fragment_index: The counter of the currently
366 downloaded video fragment.
367 * fragment_count: The number of fragments (= individual
368 files that will be merged)
71b640cc
PH
369
370 Progress hooks are guaranteed to be called at least once
371 (with status "finished") if the download is successful.
819e0531 372 postprocessor_hooks: A list of functions that get called on postprocessing
373 progress, with a dictionary with the entries
374 * status: One of "started", "processing", or "finished".
375 Check this first and ignore unknown values.
376 * postprocessor: Name of the postprocessor
377 * info_dict: The extracted info_dict
378
379 Progress hooks are guaranteed to be called at least twice
380 (with status "started" and "finished") if the processing is successful.
fc61aff4 381 merge_output_format: "/" separated list of extensions to use when merging formats.
6b591b29 382 final_ext: Expected final extension; used to detect when the file was
59a7a13e 383 already downloaded and converted
6271f1ca
PH
384 fixup: Automatically correct known faults of the file.
385 One of:
386 - "never": do nothing
387 - "warn": only emit a warning
388 - "detect_or_warn": check whether we can do anything
62cd676c 389 about it, warn otherwise (default)
504f20dd 390 source_address: Client-side IP address to bind to.
1cf376f5 391 sleep_interval_requests: Number of seconds to sleep between requests
392 during extraction
7aa589a5
S
393 sleep_interval: Number of seconds to sleep before each download when
394 used alone or a lower bound of a range for randomized
395 sleep before each download (minimum possible number
396 of seconds to sleep) when used along with
397 max_sleep_interval.
398 max_sleep_interval:Upper bound of a range for randomized sleep before each
399 download (maximum possible number of seconds to sleep).
400 Must only be used along with sleep_interval.
401 Actual sleep time will be a random float from range
402 [sleep_interval; max_sleep_interval].
1cf376f5 403 sleep_interval_subtitles: Number of seconds to sleep before each subtitle download
cfb56d1a
PH
404 listformats: Print an overview of available video formats and exit.
405 list_thumbnails: Print a table of all thumbnails and exit.
0a41f331 406 match_filter: A function that gets called for every video with the signature
407 (info_dict, *, incomplete: bool) -> Optional[str]
408 For backward compatibility with youtube-dl, the signature
409 (info_dict) -> Optional[str] is also allowed.
410 - If it returns a message, the video is ignored.
411 - If it returns None, the video is downloaded.
412 - If it returns utils.NO_DEFAULT, the user is interactively
413 asked whether to download the video.
347de493 414 match_filter_func in utils.py is one example for this.
7e5db8c9 415 no_color: Do not emit color codes in output.
0a840f58 416 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
504f20dd 417 HTTP header
0a840f58 418 geo_bypass_country:
773f291d
S
419 Two-letter ISO 3166-2 country code that will be used for
420 explicit geographic restriction bypassing via faking
504f20dd 421 X-Forwarded-For HTTP header
5f95927a
S
422 geo_bypass_ip_block:
423 IP range in CIDR notation that will be used similarly to
504f20dd 424 geo_bypass_country
52a8a1e1 425 external_downloader: A dictionary of protocol keys and the executable of the
426 external downloader to use for it. The allowed protocols
427 are default|http|ftp|m3u8|dash|rtsp|rtmp|mms.
428 Set the value to 'native' to use the native downloader
53ed7066 429 compat_opts: Compatibility options. See "Differences in default behavior".
3acf6d38 430 The following options do not work when used through the API:
b5ae35ee 431 filename, abort-on-error, multistreams, no-live-chat, format-sort
dac5df5a 432 no-clean-infojson, no-playlist-metafiles, no-keep-subs, no-attach-info-json.
e4f02757 433 Refer __init__.py for their implementation
819e0531 434 progress_template: Dictionary of templates for progress outputs.
435 Allowed keys are 'download', 'postprocess',
436 'download-title' (console title) and 'postprocess-title'.
437 The template is mapped on a dictionary with keys 'progress' and 'info'
23326151 438 retry_sleep_functions: Dictionary of functions that takes the number of attempts
439 as argument and returns the time to sleep in seconds.
440 Allowed keys are 'http', 'fragment', 'file_access'
0f446365
SW
441 download_ranges: A callback function that gets called for every video with
442 the signature (info_dict, ydl) -> Iterable[Section].
443 Only the returned sections will be downloaded.
444 Each Section is a dict with the following keys:
5ec1b6b7 445 * start_time: Start time of the section in seconds
446 * end_time: End time of the section in seconds
447 * title: Section title (Optional)
448 * index: Section number (Optional)
0f446365 449 force_keyframes_at_cuts: Re-encode the video when downloading ranges to get precise cuts
a7dc6a89 450 noprogress: Do not print the progress bar
a831c2ea 451 live_from_start: Whether to download livestreams videos from the start
fe7e0c98 452
8222d8de 453 The following parameters are not used by YoutubeDL itself, they are used by
7a5c1cfe 454 the downloader (see yt_dlp/downloader/common.py):
51d9739f 455 nopart, updatetime, buffersize, ratelimit, throttledratelimit, min_filesize,
205a0654 456 max_filesize, test, noresizebuffer, retries, file_access_retries, fragment_retries,
a7dc6a89 457 continuedl, xattr_set_filesize, hls_use_mpegts, http_chunk_size,
59a7a13e 458 external_downloader_args, concurrent_fragment_downloads.
76b1bd67
JMF
459
460 The following options are used by the post processors:
c0b7d117
S
461 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
462 to the binary or its containing directory.
43820c03 463 postprocessor_args: A dictionary of postprocessor/executable keys (in lower case)
34488702 464 and a list of additional command-line arguments for the
465 postprocessor/executable. The dict can also have "PP+EXE" keys
466 which are used when the given exe is used by the given PP.
467 Use 'default' as the name for arguments to passed to all PP
468 For compatibility with youtube-dl, a single list of args
469 can also be used
e409895f 470
471 The following options are used by the extractors:
62bff2c1 472 extractor_retries: Number of times to retry for known errors
473 dynamic_mpd: Whether to process dynamic DASH manifests (default: True)
e409895f 474 hls_split_discontinuity: Split HLS playlists to different formats at
62bff2c1 475 discontinuities such as ad breaks (default: False)
5d3a0e79 476 extractor_args: A dictionary of arguments to be passed to the extractors.
477 See "EXTRACTOR ARGUMENTS" for details.
62b58c09 478 E.g. {'youtube': {'skip': ['dash', 'hls']}}
88f23a18 479 mark_watched: Mark videos watched (even with --simulate). Only for YouTube
1890fc63 480
481 The following options are deprecated and may be removed in the future:
482
fe7866d0 483 force_generic_extractor: Force downloader to use the generic extractor
484 - Use allowed_extractors = ['generic', 'default']
7e9a6125 485 playliststart: - Use playlist_items
486 Playlist item to start at.
487 playlistend: - Use playlist_items
488 Playlist item to end at.
489 playlistreverse: - Use playlist_items
490 Download playlist items in reverse order.
1890fc63 491 forceurl: - Use forceprint
492 Force printing final URL.
493 forcetitle: - Use forceprint
494 Force printing title.
495 forceid: - Use forceprint
496 Force printing ID.
497 forcethumbnail: - Use forceprint
498 Force printing thumbnail URL.
499 forcedescription: - Use forceprint
500 Force printing description.
501 forcefilename: - Use forceprint
502 Force printing final filename.
503 forceduration: - Use forceprint
504 Force printing duration.
505 allsubtitles: - Use subtitleslangs = ['all']
506 Downloads all the subtitles of the video
507 (requires writesubtitles or writeautomaticsub)
508 include_ads: - Doesn't work
509 Download ads as well
510 call_home: - Not implemented
511 Boolean, true iff we are allowed to contact the
512 yt-dlp servers for debugging.
513 post_hooks: - Register a custom postprocessor
514 A list of functions that get called as the final step
515 for each video file, after all postprocessors have been
516 called. The filename will be passed as the only argument.
517 hls_prefer_native: - Use external_downloader = {'m3u8': 'native'} or {'m3u8': 'ffmpeg'}.
518 Use the native HLS downloader instead of ffmpeg/avconv
519 if True, otherwise use ffmpeg/avconv if False, otherwise
520 use downloader suggested by extractor if None.
521 prefer_ffmpeg: - avconv support is deprecated
522 If False, use avconv instead of ffmpeg if both are available,
523 otherwise prefer ffmpeg.
524 youtube_include_dash_manifest: - Use extractor_args
5d3a0e79 525 If True (default), DASH manifests and related
62bff2c1 526 data will be downloaded and processed by extractor.
527 You can reduce network I/O by disabling it if you don't
528 care about DASH. (only for youtube)
1890fc63 529 youtube_include_hls_manifest: - Use extractor_args
5d3a0e79 530 If True (default), HLS manifests and related
62bff2c1 531 data will be downloaded and processed by extractor.
532 You can reduce network I/O by disabling it if you don't
533 care about HLS. (only for youtube)
8222d8de
JMF
534 """
535
86e5f3ed 536 _NUMERIC_FIELDS = {
b8ed0f15 537 'width', 'height', 'asr', 'audio_channels', 'fps',
538 'tbr', 'abr', 'vbr', 'filesize', 'filesize_approx',
e6f21b3d 539 'timestamp', 'release_timestamp',
c9969434
S
540 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
541 'average_rating', 'comment_count', 'age_limit',
542 'start_time', 'end_time',
543 'chapter_number', 'season_number', 'episode_number',
544 'track_number', 'disc_number', 'release_year',
86e5f3ed 545 }
c9969434 546
6db9c4d5 547 _format_fields = {
548 # NB: Keep in sync with the docstring of extractor/common.py
a44ca5a4 549 'url', 'manifest_url', 'manifest_stream_number', 'ext', 'format', 'format_id', 'format_note',
b8ed0f15 550 'width', 'height', 'resolution', 'dynamic_range', 'tbr', 'abr', 'acodec', 'asr', 'audio_channels',
6db9c4d5 551 'vbr', 'fps', 'vcodec', 'container', 'filesize', 'filesize_approx',
552 'player_url', 'protocol', 'fragment_base_url', 'fragments', 'is_from_start',
553 'preference', 'language', 'language_preference', 'quality', 'source_preference',
554 'http_headers', 'stretched_ratio', 'no_resume', 'has_drm', 'downloader_options',
555 'page_url', 'app', 'play_path', 'tc_url', 'flash_version', 'rtmp_live', 'rtmp_conn', 'rtmp_protocol', 'rtmp_real_time'
556 }
48ee10ee 557 _format_selection_exts = {
8dc59305 558 'audio': set(MEDIA_EXTENSIONS.common_audio),
559 'video': set(MEDIA_EXTENSIONS.common_video + ('3gp', )),
560 'storyboards': set(MEDIA_EXTENSIONS.storyboards),
48ee10ee 561 }
562
3511266b 563 def __init__(self, params=None, auto_init=True):
883d4b1e 564 """Create a FileDownloader object with the given options.
565 @param auto_init Whether to load the default extractors and print header (if verbose).
49a57e70 566 Set to 'no_verbose_header' to not print the header
883d4b1e 567 """
e9f9a10f
JMF
568 if params is None:
569 params = {}
592b7485 570 self.params = params
8b7491c8 571 self._ies = {}
56c73665 572 self._ies_instances = {}
1e43a6f7 573 self._pps = {k: [] for k in POSTPROCESS_WHEN}
b35496d8 574 self._printed_messages = set()
1cf376f5 575 self._first_webpage_request = True
ab8e5e51 576 self._post_hooks = []
933605d7 577 self._progress_hooks = []
819e0531 578 self._postprocessor_hooks = []
8222d8de
JMF
579 self._download_retcode = 0
580 self._num_downloads = 0
9c906919 581 self._num_videos = 0
592b7485 582 self._playlist_level = 0
583 self._playlist_urls = set()
a0e07d31 584 self.cache = Cache(self)
34308b30 585
819e0531 586 windows_enable_vt_mode()
591bb9d3 587 stdout = sys.stderr if self.params.get('logtostderr') else sys.stdout
588 self._out_files = Namespace(
589 out=stdout,
590 error=sys.stderr,
591 screen=sys.stderr if self.params.get('quiet') else stdout,
592 console=None if compat_os_name == 'nt' else next(
cf4f42cb 593 filter(supports_terminal_sequences, (sys.stderr, sys.stdout)), None)
591bb9d3 594 )
595 self._allow_colors = Namespace(**{
596 type_: not self.params.get('no_color') and supports_terminal_sequences(stream)
64fa820c 597 for type_, stream in self._out_files.items_ if type_ != 'console'
591bb9d3 598 })
819e0531 599
6929b41a 600 # The code is left like this to be reused for future deprecations
601 MIN_SUPPORTED, MIN_RECOMMENDED = (3, 7), (3, 7)
eff42759 602 current_version = sys.version_info[:2]
603 if current_version < MIN_RECOMMENDED:
9d339c41 604 msg = ('Support for Python version %d.%d has been deprecated. '
24093d52 605 'See https://github.com/yt-dlp/yt-dlp/issues/3764 for more details.'
c6e07cf1 606 '\n You will no longer receive updates on this version')
eff42759 607 if current_version < MIN_SUPPORTED:
608 msg = 'Python version %d.%d is no longer supported'
609 self.deprecation_warning(
610 f'{msg}! Please update to Python %d.%d or above' % (*current_version, *MIN_RECOMMENDED))
a61f4b28 611
88acdbc2 612 if self.params.get('allow_unplayable_formats'):
613 self.report_warning(
ec11a9f4 614 f'You have asked for {self._format_err("UNPLAYABLE", self.Styles.EMPHASIS)} formats to be listed/downloaded. '
819e0531 615 'This is a developer option intended for debugging. \n'
616 ' If you experience any issues while using this option, '
ec11a9f4 617 f'{self._format_err("DO NOT", self.Styles.ERROR)} open a bug report')
88acdbc2 618
be5df5ee
S
619 def check_deprecated(param, option, suggestion):
620 if self.params.get(param) is not None:
86e5f3ed 621 self.report_warning(f'{option} is deprecated. Use {suggestion} instead')
be5df5ee
S
622 return True
623 return False
624
625 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
38cce791
YCH
626 if self.params.get('geo_verification_proxy') is None:
627 self.params['geo_verification_proxy'] = self.params['cn_verification_proxy']
628
0d1bb027 629 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
630 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
53ed7066 631 check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"')
0d1bb027 632
49a57e70 633 for msg in self.params.get('_warnings', []):
0d1bb027 634 self.report_warning(msg)
ee8dd27a 635 for msg in self.params.get('_deprecation_warnings', []):
da4db748 636 self.deprecated_feature(msg)
0d1bb027 637
8a82af35 638 self.params['compat_opts'] = set(self.params.get('compat_opts', ()))
8a82af35 639 if 'list-formats' in self.params['compat_opts']:
ec11a9f4 640 self.params['listformats_table'] = False
641
b5ae35ee 642 if 'overwrites' not in self.params and self.params.get('nooverwrites') is not None:
b868936c 643 # nooverwrites was unnecessarily changed to overwrites
644 # in 0c3d0f51778b153f65c21906031c2e091fcfb641
645 # This ensures compatibility with both keys
646 self.params['overwrites'] = not self.params['nooverwrites']
b5ae35ee 647 elif self.params.get('overwrites') is None:
648 self.params.pop('overwrites', None)
b868936c 649 else:
650 self.params['nooverwrites'] = not self.params['overwrites']
b9d973be 651
455a15e2 652 self.params.setdefault('forceprint', {})
653 self.params.setdefault('print_to_file', {})
bb66c247 654
655 # Compatibility with older syntax
ca30f449 656 if not isinstance(params['forceprint'], dict):
455a15e2 657 self.params['forceprint'] = {'video': params['forceprint']}
ca30f449 658
455a15e2 659 if self.params.get('bidi_workaround', False):
1c088fa8
PH
660 try:
661 import pty
662 master, slave = pty.openpty()
ac668111 663 width = shutil.get_terminal_size().columns
591bb9d3 664 width_args = [] if width is None else ['-w', str(width)]
665 sp_kwargs = {'stdin': subprocess.PIPE, 'stdout': slave, 'stderr': self._out_files.error}
5d681e96 666 try:
d3c93ec2 667 self._output_process = Popen(['bidiv'] + width_args, **sp_kwargs)
5d681e96 668 except OSError:
d3c93ec2 669 self._output_process = Popen(['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
5d681e96 670 self._output_channel = os.fdopen(master, 'rb')
1c088fa8 671 except OSError as ose:
66e7ace1 672 if ose.errno == errno.ENOENT:
49a57e70 673 self.report_warning(
674 'Could not find fribidi executable, ignoring --bidi-workaround. '
675 'Make sure that fribidi is an executable file in one of the directories in your $PATH.')
1c088fa8
PH
676 else:
677 raise
0783b09b 678
97ec5bc5 679 if auto_init:
680 if auto_init != 'no_verbose_header':
681 self.print_debug_header()
682 self.add_default_info_extractors()
683
3089bc74
S
684 if (sys.platform != 'win32'
685 and sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
455a15e2 686 and not self.params.get('restrictfilenames', False)):
e9137224 687 # Unicode filesystem API will throw errors (#1474, #13027)
34308b30 688 self.report_warning(
6febd1c1 689 'Assuming --restrict-filenames since file system encoding '
1b725173 690 'cannot encode all characters. '
6febd1c1 691 'Set the LC_ALL environment variable to fix this.')
4a98cdbf 692 self.params['restrictfilenames'] = True
34308b30 693
bf1824b3 694 self._parse_outtmpl()
486dd09e 695
187986a8 696 # Creating format selector here allows us to catch syntax errors before the extraction
697 self.format_selector = (
fa9f30b8 698 self.params.get('format') if self.params.get('format') in (None, '-')
093a1710 699 else self.params['format'] if callable(self.params['format'])
187986a8 700 else self.build_format_selector(self.params['format']))
701
8b7539d2 702 # Set http_headers defaults according to std_headers
703 self.params['http_headers'] = merge_headers(std_headers, self.params.get('http_headers', {}))
704
013b50b7 705 hooks = {
706 'post_hooks': self.add_post_hook,
707 'progress_hooks': self.add_progress_hook,
708 'postprocessor_hooks': self.add_postprocessor_hook,
709 }
710 for opt, fn in hooks.items():
711 for ph in self.params.get(opt, []):
712 fn(ph)
71b640cc 713
5bfc8bee 714 for pp_def_raw in self.params.get('postprocessors', []):
715 pp_def = dict(pp_def_raw)
716 when = pp_def.pop('when', 'post_process')
717 self.add_post_processor(
f9934b96 718 get_postprocessor(pp_def.pop('key'))(self, **pp_def),
5bfc8bee 719 when=when)
720
97ec5bc5 721 self._setup_opener()
51fb4995
YCH
722 register_socks_protocols()
723
ed39cac5 724 def preload_download_archive(fn):
725 """Preload the archive, if any is specified"""
ae103564 726 archive = set()
ed39cac5 727 if fn is None:
ae103564 728 return archive
941e881e 729 elif not is_path_like(fn):
ae103564 730 return fn
731
49a57e70 732 self.write_debug(f'Loading archive file {fn!r}')
ed39cac5 733 try:
734 with locked_file(fn, 'r', encoding='utf-8') as archive_file:
735 for line in archive_file:
ae103564 736 archive.add(line.strip())
86e5f3ed 737 except OSError as ioe:
ed39cac5 738 if ioe.errno != errno.ENOENT:
739 raise
ae103564 740 return archive
ed39cac5 741
ae103564 742 self.archive = preload_download_archive(self.params.get('download_archive'))
ed39cac5 743
7d4111ed
PH
744 def warn_if_short_id(self, argv):
745 # short YouTube ID starting with dash?
746 idxs = [
747 i for i, a in enumerate(argv)
748 if re.match(r'^-[0-9A-Za-z_-]{10}$', a)]
749 if idxs:
750 correct_argv = (
7a5c1cfe 751 ['yt-dlp']
3089bc74
S
752 + [a for i, a in enumerate(argv) if i not in idxs]
753 + ['--'] + [argv[i] for i in idxs]
7d4111ed
PH
754 )
755 self.report_warning(
756 'Long argument string detected. '
49a57e70 757 'Use -- to separate parameters and URLs, like this:\n%s' %
7d4111ed
PH
758 args_to_str(correct_argv))
759
8222d8de
JMF
760 def add_info_extractor(self, ie):
761 """Add an InfoExtractor object to the end of the list."""
8b7491c8 762 ie_key = ie.ie_key()
763 self._ies[ie_key] = ie
e52d7f85 764 if not isinstance(ie, type):
8b7491c8 765 self._ies_instances[ie_key] = ie
e52d7f85 766 ie.set_downloader(self)
8222d8de 767
56c73665
JMF
768 def get_info_extractor(self, ie_key):
769 """
770 Get an instance of an IE with name ie_key, it will try to get one from
771 the _ies list, if there's no instance it will create a new one and add
772 it to the extractor list.
773 """
774 ie = self._ies_instances.get(ie_key)
775 if ie is None:
776 ie = get_info_extractor(ie_key)()
777 self.add_info_extractor(ie)
778 return ie
779
023fa8c4
JMF
780 def add_default_info_extractors(self):
781 """
782 Add the InfoExtractors returned by gen_extractors to the end of the list
783 """
fe7866d0 784 all_ies = {ie.IE_NAME.lower(): ie for ie in gen_extractor_classes()}
785 all_ies['end'] = UnsupportedURLIE()
786 try:
787 ie_names = orderedSet_from_options(
788 self.params.get('allowed_extractors', ['default']), {
789 'all': list(all_ies),
790 'default': [name for name, ie in all_ies.items() if ie._ENABLED],
791 }, use_regex=True)
792 except re.error as e:
793 raise ValueError(f'Wrong regex for allowed_extractors: {e.pattern}')
794 for name in ie_names:
795 self.add_info_extractor(all_ies[name])
796 self.write_debug(f'Loaded {len(ie_names)} extractors')
023fa8c4 797
56d868db 798 def add_post_processor(self, pp, when='post_process'):
8222d8de 799 """Add a PostProcessor object to the end of the chain."""
8aa0e7cd 800 assert when in POSTPROCESS_WHEN, f'Invalid when={when}'
5bfa4862 801 self._pps[when].append(pp)
8222d8de
JMF
802 pp.set_downloader(self)
803
ab8e5e51
AM
804 def add_post_hook(self, ph):
805 """Add the post hook"""
806 self._post_hooks.append(ph)
807
933605d7 808 def add_progress_hook(self, ph):
819e0531 809 """Add the download progress hook"""
933605d7 810 self._progress_hooks.append(ph)
8ab470f1 811
819e0531 812 def add_postprocessor_hook(self, ph):
813 """Add the postprocessing progress hook"""
814 self._postprocessor_hooks.append(ph)
5bfc8bee 815 for pps in self._pps.values():
816 for pp in pps:
817 pp.add_progress_hook(ph)
819e0531 818
1c088fa8 819 def _bidi_workaround(self, message):
5d681e96 820 if not hasattr(self, '_output_channel'):
1c088fa8
PH
821 return message
822
5d681e96 823 assert hasattr(self, '_output_process')
14f25df2 824 assert isinstance(message, str)
6febd1c1 825 line_count = message.count('\n') + 1
0f06bcd7 826 self._output_process.stdin.write((message + '\n').encode())
5d681e96 827 self._output_process.stdin.flush()
0f06bcd7 828 res = ''.join(self._output_channel.readline().decode()
9e1a5b84 829 for _ in range(line_count))
6febd1c1 830 return res[:-len('\n')]
1c088fa8 831
b35496d8 832 def _write_string(self, message, out=None, only_once=False):
833 if only_once:
834 if message in self._printed_messages:
835 return
836 self._printed_messages.add(message)
837 write_string(message, out=out, encoding=self.params.get('encoding'))
734f90bb 838
cf4f42cb 839 def to_stdout(self, message, skip_eol=False, quiet=None):
0760b0a7 840 """Print message to stdout"""
cf4f42cb 841 if quiet is not None:
da4db748 842 self.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument quiet. '
843 'Use "YoutubeDL.to_screen" instead')
8a82af35 844 if skip_eol is not False:
da4db748 845 self.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument skip_eol. '
846 'Use "YoutubeDL.to_screen" instead')
0bf9dc1e 847 self._write_string(f'{self._bidi_workaround(message)}\n', self._out_files.out)
cf4f42cb 848
849 def to_screen(self, message, skip_eol=False, quiet=None):
850 """Print message to screen if not in quiet mode"""
8bf9319e 851 if self.params.get('logger'):
43afe285 852 self.params['logger'].debug(message)
cf4f42cb 853 return
854 if (self.params.get('quiet') if quiet is None else quiet) and not self.params.get('verbose'):
855 return
856 self._write_string(
857 '%s%s' % (self._bidi_workaround(message), ('' if skip_eol else '\n')),
591bb9d3 858 self._out_files.screen)
8222d8de 859
b35496d8 860 def to_stderr(self, message, only_once=False):
0760b0a7 861 """Print message to stderr"""
14f25df2 862 assert isinstance(message, str)
8bf9319e 863 if self.params.get('logger'):
43afe285
IB
864 self.params['logger'].error(message)
865 else:
5792c950 866 self._write_string(f'{self._bidi_workaround(message)}\n', self._out_files.error, only_once=only_once)
cf4f42cb 867
868 def _send_console_code(self, code):
591bb9d3 869 if compat_os_name == 'nt' or not self._out_files.console:
cf4f42cb 870 return
591bb9d3 871 self._write_string(code, self._out_files.console)
8222d8de 872
1e5b9a95
PH
873 def to_console_title(self, message):
874 if not self.params.get('consoletitle', False):
875 return
3efb96a6 876 message = remove_terminal_sequences(message)
4bede0d8
C
877 if compat_os_name == 'nt':
878 if ctypes.windll.kernel32.GetConsoleWindow():
879 # c_wchar_p() might not be necessary if `message` is
880 # already of type unicode()
881 ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
cf4f42cb 882 else:
883 self._send_console_code(f'\033]0;{message}\007')
1e5b9a95 884
bdde425c 885 def save_console_title(self):
cf4f42cb 886 if not self.params.get('consoletitle') or self.params.get('simulate'):
bdde425c 887 return
592b7485 888 self._send_console_code('\033[22;0t') # Save the title on stack
bdde425c
PH
889
890 def restore_console_title(self):
cf4f42cb 891 if not self.params.get('consoletitle') or self.params.get('simulate'):
bdde425c 892 return
592b7485 893 self._send_console_code('\033[23;0t') # Restore the title from stack
bdde425c
PH
894
895 def __enter__(self):
896 self.save_console_title()
897 return self
898
899 def __exit__(self, *args):
900 self.restore_console_title()
f89197d7 901
dca08720 902 if self.params.get('cookiefile') is not None:
1bab3437 903 self.cookiejar.save(ignore_discard=True, ignore_expires=True)
bdde425c 904
fa9f30b8 905 def trouble(self, message=None, tb=None, is_error=True):
8222d8de
JMF
906 """Determine action to take when a download problem appears.
907
908 Depending on if the downloader has been configured to ignore
909 download errors or not, this method may throw an exception or
910 not when errors are found, after printing the message.
911
fa9f30b8 912 @param tb If given, is additional traceback information
913 @param is_error Whether to raise error according to ignorerrors
8222d8de
JMF
914 """
915 if message is not None:
916 self.to_stderr(message)
917 if self.params.get('verbose'):
918 if tb is None:
919 if sys.exc_info()[0]: # if .trouble has been called from an except block
6febd1c1 920 tb = ''
8222d8de 921 if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
6febd1c1 922 tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
c0384f22 923 tb += encode_compat_str(traceback.format_exc())
8222d8de
JMF
924 else:
925 tb_data = traceback.format_list(traceback.extract_stack())
6febd1c1 926 tb = ''.join(tb_data)
c19bc311 927 if tb:
928 self.to_stderr(tb)
fa9f30b8 929 if not is_error:
930 return
b1940459 931 if not self.params.get('ignoreerrors'):
8222d8de
JMF
932 if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
933 exc_info = sys.exc_info()[1].exc_info
934 else:
935 exc_info = sys.exc_info()
936 raise DownloadError(message, exc_info)
937 self._download_retcode = 1
938
19a03940 939 Styles = Namespace(
940 HEADERS='yellow',
941 EMPHASIS='light blue',
492272fe 942 FILENAME='green',
19a03940 943 ID='green',
944 DELIM='blue',
945 ERROR='red',
946 WARNING='yellow',
947 SUPPRESS='light black',
948 )
ec11a9f4 949
7578d77d 950 def _format_text(self, handle, allow_colors, text, f, fallback=None, *, test_encoding=False):
e5a998f3 951 text = str(text)
ec11a9f4 952 if test_encoding:
953 original_text = text
5c104538 954 # handle.encoding can be None. See https://github.com/yt-dlp/yt-dlp/issues/2711
955 encoding = self.params.get('encoding') or getattr(handle, 'encoding', None) or 'ascii'
ec11a9f4 956 text = text.encode(encoding, 'ignore').decode(encoding)
957 if fallback is not None and text != original_text:
958 text = fallback
7578d77d 959 return format_text(text, f) if allow_colors else text if fallback is None else fallback
ec11a9f4 960
591bb9d3 961 def _format_out(self, *args, **kwargs):
962 return self._format_text(self._out_files.out, self._allow_colors.out, *args, **kwargs)
963
ec11a9f4 964 def _format_screen(self, *args, **kwargs):
591bb9d3 965 return self._format_text(self._out_files.screen, self._allow_colors.screen, *args, **kwargs)
ec11a9f4 966
967 def _format_err(self, *args, **kwargs):
591bb9d3 968 return self._format_text(self._out_files.error, self._allow_colors.error, *args, **kwargs)
819e0531 969
c84aeac6 970 def report_warning(self, message, only_once=False):
8222d8de
JMF
971 '''
972 Print the message to stderr, it will be prefixed with 'WARNING:'
973 If stderr is a tty file the 'WARNING:' will be colored
974 '''
6d07ce01
JMF
975 if self.params.get('logger') is not None:
976 self.params['logger'].warning(message)
8222d8de 977 else:
ad8915b7
PH
978 if self.params.get('no_warnings'):
979 return
ec11a9f4 980 self.to_stderr(f'{self._format_err("WARNING:", self.Styles.WARNING)} {message}', only_once)
8222d8de 981
da4db748 982 def deprecation_warning(self, message, *, stacklevel=0):
983 deprecation_warning(
984 message, stacklevel=stacklevel + 1, printer=self.report_error, is_error=False)
985
986 def deprecated_feature(self, message):
ee8dd27a 987 if self.params.get('logger') is not None:
da4db748 988 self.params['logger'].warning(f'Deprecated Feature: {message}')
989 self.to_stderr(f'{self._format_err("Deprecated Feature:", self.Styles.ERROR)} {message}', True)
ee8dd27a 990
fa9f30b8 991 def report_error(self, message, *args, **kwargs):
8222d8de
JMF
992 '''
993 Do the same as trouble, but prefixes the message with 'ERROR:', colored
994 in red if stderr is a tty file.
995 '''
fa9f30b8 996 self.trouble(f'{self._format_err("ERROR:", self.Styles.ERROR)} {message}', *args, **kwargs)
8222d8de 997
b35496d8 998 def write_debug(self, message, only_once=False):
0760b0a7 999 '''Log debug message or Print message to stderr'''
1000 if not self.params.get('verbose', False):
1001 return
8a82af35 1002 message = f'[debug] {message}'
0760b0a7 1003 if self.params.get('logger'):
1004 self.params['logger'].debug(message)
1005 else:
b35496d8 1006 self.to_stderr(message, only_once)
0760b0a7 1007
8222d8de
JMF
1008 def report_file_already_downloaded(self, file_name):
1009 """Report file has already been fully downloaded."""
1010 try:
6febd1c1 1011 self.to_screen('[download] %s has already been downloaded' % file_name)
ce02ed60 1012 except UnicodeEncodeError:
6febd1c1 1013 self.to_screen('[download] The file has already been downloaded')
8222d8de 1014
0c3d0f51 1015 def report_file_delete(self, file_name):
1016 """Report that existing file will be deleted."""
1017 try:
c25228e5 1018 self.to_screen('Deleting existing file %s' % file_name)
0c3d0f51 1019 except UnicodeEncodeError:
c25228e5 1020 self.to_screen('Deleting existing file')
0c3d0f51 1021
319b6059 1022 def raise_no_formats(self, info, forced=False, *, msg=None):
0a5a191a 1023 has_drm = info.get('_has_drm')
319b6059 1024 ignored, expected = self.params.get('ignore_no_formats_error'), bool(msg)
1025 msg = msg or has_drm and 'This video is DRM protected' or 'No video formats found!'
1026 if forced or not ignored:
1151c407 1027 raise ExtractorError(msg, video_id=info['id'], ie=info['extractor'],
319b6059 1028 expected=has_drm or ignored or expected)
88acdbc2 1029 else:
1030 self.report_warning(msg)
1031
de6000d9 1032 def parse_outtmpl(self):
bf1824b3 1033 self.deprecation_warning('"YoutubeDL.parse_outtmpl" is deprecated and may be removed in a future version')
1034 self._parse_outtmpl()
1035 return self.params['outtmpl']
1036
1037 def _parse_outtmpl(self):
7b2c3f47 1038 sanitize = IDENTITY
bf1824b3 1039 if self.params.get('restrictfilenames'): # Remove spaces in the default template
71ce444a 1040 sanitize = lambda x: x.replace(' - ', ' ').replace(' ', '-')
bf1824b3 1041
1042 outtmpl = self.params.setdefault('outtmpl', {})
1043 if not isinstance(outtmpl, dict):
1044 self.params['outtmpl'] = outtmpl = {'default': outtmpl}
1045 outtmpl.update({k: sanitize(v) for k, v in DEFAULT_OUTTMPL.items() if outtmpl.get(k) is None})
de6000d9 1046
21cd8fae 1047 def get_output_path(self, dir_type='', filename=None):
1048 paths = self.params.get('paths', {})
d2c8aadf 1049 assert isinstance(paths, dict), '"paths" parameter must be a dictionary'
21cd8fae 1050 path = os.path.join(
1051 expand_path(paths.get('home', '').strip()),
1052 expand_path(paths.get(dir_type, '').strip()) if dir_type else '',
1053 filename or '')
21cd8fae 1054 return sanitize_path(path, force=self.params.get('windowsfilenames'))
1055
76a264ac 1056 @staticmethod
901130bb 1057 def _outtmpl_expandpath(outtmpl):
1058 # expand_path translates '%%' into '%' and '$$' into '$'
1059 # correspondingly that is not what we want since we need to keep
1060 # '%%' intact for template dict substitution step. Working around
1061 # with boundary-alike separator hack.
1062 sep = ''.join([random.choice(ascii_letters) for _ in range(32)])
86e5f3ed 1063 outtmpl = outtmpl.replace('%%', f'%{sep}%').replace('$$', f'${sep}$')
901130bb 1064
1065 # outtmpl should be expand_path'ed before template dict substitution
1066 # because meta fields may contain env variables we don't want to
62b58c09 1067 # be expanded. E.g. for outtmpl "%(title)s.%(ext)s" and
901130bb 1068 # title "Hello $PATH", we don't want `$PATH` to be expanded.
1069 return expand_path(outtmpl).replace(sep, '')
1070
1071 @staticmethod
1072 def escape_outtmpl(outtmpl):
1073 ''' Escape any remaining strings like %s, %abc% etc. '''
1074 return re.sub(
1075 STR_FORMAT_RE_TMPL.format('', '(?![%(\0])'),
1076 lambda mobj: ('' if mobj.group('has_key') else '%') + mobj.group(0),
1077 outtmpl)
1078
1079 @classmethod
1080 def validate_outtmpl(cls, outtmpl):
76a264ac 1081 ''' @return None or Exception object '''
7d1eb38a 1082 outtmpl = re.sub(
47cdc68e 1083 STR_FORMAT_RE_TMPL.format('[^)]*', '[ljhqBUDS]'),
7d1eb38a 1084 lambda mobj: f'{mobj.group(0)[:-1]}s',
1085 cls._outtmpl_expandpath(outtmpl))
76a264ac 1086 try:
7d1eb38a 1087 cls.escape_outtmpl(outtmpl) % collections.defaultdict(int)
76a264ac 1088 return None
1089 except ValueError as err:
1090 return err
1091
03b4de72 1092 @staticmethod
1093 def _copy_infodict(info_dict):
1094 info_dict = dict(info_dict)
09b49e1f 1095 info_dict.pop('__postprocessors', None)
415f8d51 1096 info_dict.pop('__pending_error', None)
03b4de72 1097 return info_dict
1098
e0fd9573 1099 def prepare_outtmpl(self, outtmpl, info_dict, sanitize=False):
1100 """ Make the outtmpl and info_dict suitable for substitution: ydl.escape_outtmpl(outtmpl) % info_dict
1101 @param sanitize Whether to sanitize the output as a filename.
1102 For backward compatibility, a function can also be passed
1103 """
1104
6e84b215 1105 info_dict.setdefault('epoch', int(time.time())) # keep epoch consistent once set
143db31d 1106
03b4de72 1107 info_dict = self._copy_infodict(info_dict)
752cda38 1108 info_dict['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
53c18592 1109 formatSeconds(info_dict['duration'], '-' if sanitize else ':')
143db31d 1110 if info_dict.get('duration', None) is not None
1111 else None)
1d485a1a 1112 info_dict['autonumber'] = int(self.params.get('autonumber_start', 1) - 1 + self._num_downloads)
9c906919 1113 info_dict['video_autonumber'] = self._num_videos
752cda38 1114 if info_dict.get('resolution') is None:
1115 info_dict['resolution'] = self.format_resolution(info_dict, default=None)
143db31d 1116
e6f21b3d 1117 # For fields playlist_index, playlist_autonumber and autonumber convert all occurrences
143db31d 1118 # of %(field)s to %(field)0Nd for backward compatibility
1119 field_size_compat_map = {
0a5a191a 1120 'playlist_index': number_of_digits(info_dict.get('__last_playlist_index') or 0),
ec11a9f4 1121 'playlist_autonumber': number_of_digits(info_dict.get('n_entries') or 0),
752cda38 1122 'autonumber': self.params.get('autonumber_size') or 5,
143db31d 1123 }
752cda38 1124
385a27fa 1125 TMPL_DICT = {}
47cdc68e 1126 EXTERNAL_FORMAT_RE = re.compile(STR_FORMAT_RE_TMPL.format('[^)]*', f'[{STR_FORMAT_TYPES}ljhqBUDS]'))
385a27fa 1127 MATH_FUNCTIONS = {
1128 '+': float.__add__,
1129 '-': float.__sub__,
1130 }
e625be0d 1131 # Field is of the form key1.key2...
07a1250e 1132 # where keys (except first) can be string, int, slice or "{field, ...}"
1133 FIELD_INNER_RE = r'(?:\w+|%(num)s|%(num)s?(?::%(num)s?){1,2})' % {'num': r'(?:-?\d+)'}
1134 FIELD_RE = r'\w*(?:\.(?:%(inner)s|{%(field)s(?:,%(field)s)*}))*' % {
1135 'inner': FIELD_INNER_RE,
1136 'field': rf'\w*(?:\.{FIELD_INNER_RE})*'
1137 }
1d485a1a 1138 MATH_FIELD_RE = rf'(?:{FIELD_RE}|-?{NUMBER_RE})'
385a27fa 1139 MATH_OPERATORS_RE = r'(?:%s)' % '|'.join(map(re.escape, MATH_FUNCTIONS.keys()))
1d485a1a 1140 INTERNAL_FORMAT_RE = re.compile(rf'''(?x)
e625be0d 1141 (?P<negate>-)?
1d485a1a 1142 (?P<fields>{FIELD_RE})
1143 (?P<maths>(?:{MATH_OPERATORS_RE}{MATH_FIELD_RE})*)
e625be0d 1144 (?:>(?P<strf_format>.+?))?
34baa9fd 1145 (?P<remaining>
1146 (?P<alternate>(?<!\\),[^|&)]+)?
1147 (?:&(?P<replacement>.*?))?
1148 (?:\|(?P<default>.*?))?
1d485a1a 1149 )$''')
752cda38 1150
07a1250e 1151 def _traverse_infodict(fields):
1152 fields = [f for x in re.split(r'\.({.+?})\.?', fields)
1153 for f in ([x] if x.startswith('{') else x.split('.'))]
1154 for i in (0, -1):
1155 if fields and not fields[i]:
1156 fields.pop(i)
1157
1158 for i, f in enumerate(fields):
1159 if not f.startswith('{'):
1160 continue
1161 assert f.endswith('}'), f'No closing brace for {f} in {fields}'
1162 fields[i] = {k: k.split('.') for k in f[1:-1].split(',')}
1163
1164 return traverse_obj(info_dict, fields, is_user_input=True, traverse_string=True)
76a264ac 1165
752cda38 1166 def get_value(mdict):
1167 # Object traversal
2b8a2973 1168 value = _traverse_infodict(mdict['fields'])
752cda38 1169 # Negative
1170 if mdict['negate']:
1171 value = float_or_none(value)
1172 if value is not None:
1173 value *= -1
1174 # Do maths
385a27fa 1175 offset_key = mdict['maths']
1176 if offset_key:
752cda38 1177 value = float_or_none(value)
1178 operator = None
385a27fa 1179 while offset_key:
1180 item = re.match(
1181 MATH_FIELD_RE if operator else MATH_OPERATORS_RE,
1182 offset_key).group(0)
1183 offset_key = offset_key[len(item):]
1184 if operator is None:
752cda38 1185 operator = MATH_FUNCTIONS[item]
385a27fa 1186 continue
1187 item, multiplier = (item[1:], -1) if item[0] == '-' else (item, 1)
1188 offset = float_or_none(item)
1189 if offset is None:
2b8a2973 1190 offset = float_or_none(_traverse_infodict(item))
385a27fa 1191 try:
1192 value = operator(value, multiplier * offset)
1193 except (TypeError, ZeroDivisionError):
1194 return None
1195 operator = None
752cda38 1196 # Datetime formatting
1197 if mdict['strf_format']:
7c37ff97 1198 value = strftime_or_none(value, mdict['strf_format'].replace('\\,', ','))
752cda38 1199
a6bcaf71 1200 # XXX: Workaround for https://github.com/yt-dlp/yt-dlp/issues/4485
1201 if sanitize and value == '':
1202 value = None
752cda38 1203 return value
1204
b868936c 1205 na = self.params.get('outtmpl_na_placeholder', 'NA')
1206
e0fd9573 1207 def filename_sanitizer(key, value, restricted=self.params.get('restrictfilenames')):
5c3895ff 1208 return sanitize_filename(str(value), restricted=restricted, is_id=(
1209 bool(re.search(r'(^|[_.])id(\.|$)', key))
8a82af35 1210 if 'filename-sanitization' in self.params['compat_opts']
5c3895ff 1211 else NO_DEFAULT))
e0fd9573 1212
1213 sanitizer = sanitize if callable(sanitize) else filename_sanitizer
1214 sanitize = bool(sanitize)
1215
6e84b215 1216 def _dumpjson_default(obj):
1217 if isinstance(obj, (set, LazyList)):
1218 return list(obj)
adbc4ec4 1219 return repr(obj)
6e84b215 1220
752cda38 1221 def create_key(outer_mobj):
1222 if not outer_mobj.group('has_key'):
b836dc94 1223 return outer_mobj.group(0)
752cda38 1224 key = outer_mobj.group('key')
752cda38 1225 mobj = re.match(INTERNAL_FORMAT_RE, key)
e0fd9573 1226 initial_field = mobj.group('fields') if mobj else ''
e978789f 1227 value, replacement, default = None, None, na
7c37ff97 1228 while mobj:
e625be0d 1229 mobj = mobj.groupdict()
7c37ff97 1230 default = mobj['default'] if mobj['default'] is not None else default
752cda38 1231 value = get_value(mobj)
e978789f 1232 replacement = mobj['replacement']
7c37ff97 1233 if value is None and mobj['alternate']:
34baa9fd 1234 mobj = re.match(INTERNAL_FORMAT_RE, mobj['remaining'][1:])
7c37ff97 1235 else:
1236 break
752cda38 1237
b868936c 1238 fmt = outer_mobj.group('format')
752cda38 1239 if fmt == 's' and value is not None and key in field_size_compat_map.keys():
86e5f3ed 1240 fmt = f'0{field_size_compat_map[key]:d}d'
752cda38 1241
e978789f 1242 value = default if value is None else value if replacement is None else replacement
752cda38 1243
4476d2c7 1244 flags = outer_mobj.group('conversion') or ''
7d1eb38a 1245 str_fmt = f'{fmt[:-1]}s'
524e2e4f 1246 if fmt[-1] == 'l': # list
4476d2c7 1247 delim = '\n' if '#' in flags else ', '
9e907ebd 1248 value, fmt = delim.join(map(str, variadic(value, allowed_types=(str, bytes)))), str_fmt
524e2e4f 1249 elif fmt[-1] == 'j': # json
deae7c17 1250 value, fmt = json.dumps(
1251 value, default=_dumpjson_default,
1252 indent=4 if '#' in flags else None, ensure_ascii=False), str_fmt
47cdc68e 1253 elif fmt[-1] == 'h': # html
deae7c17 1254 value, fmt = escapeHTML(str(value)), str_fmt
524e2e4f 1255 elif fmt[-1] == 'q': # quoted
4476d2c7 1256 value = map(str, variadic(value) if '#' in flags else [value])
1257 value, fmt = ' '.join(map(compat_shlex_quote, value)), str_fmt
524e2e4f 1258 elif fmt[-1] == 'B': # bytes
0f06bcd7 1259 value = f'%{str_fmt}'.encode() % str(value).encode()
f5aa5cfb 1260 value, fmt = value.decode('utf-8', 'ignore'), 's'
524e2e4f 1261 elif fmt[-1] == 'U': # unicode normalized
524e2e4f 1262 value, fmt = unicodedata.normalize(
1263 # "+" = compatibility equivalence, "#" = NFD
4476d2c7 1264 'NF%s%s' % ('K' if '+' in flags else '', 'D' if '#' in flags else 'C'),
524e2e4f 1265 value), str_fmt
e0fd9573 1266 elif fmt[-1] == 'D': # decimal suffix
abbeeebc 1267 num_fmt, fmt = fmt[:-1].replace('#', ''), 's'
1268 value = format_decimal_suffix(value, f'%{num_fmt}f%s' if num_fmt else '%d%s',
1269 factor=1024 if '#' in flags else 1000)
37893bb0 1270 elif fmt[-1] == 'S': # filename sanitization
e0fd9573 1271 value, fmt = filename_sanitizer(initial_field, value, restricted='#' in flags), str_fmt
7d1eb38a 1272 elif fmt[-1] == 'c':
524e2e4f 1273 if value:
1274 value = str(value)[0]
76a264ac 1275 else:
524e2e4f 1276 fmt = str_fmt
76a264ac 1277 elif fmt[-1] not in 'rs': # numeric
a439a3a4 1278 value = float_or_none(value)
752cda38 1279 if value is None:
1280 value, fmt = default, 's'
901130bb 1281
752cda38 1282 if sanitize:
1283 if fmt[-1] == 'r':
1284 # If value is an object, sanitize might convert it to a string
1285 # So we convert it to repr first
7d1eb38a 1286 value, fmt = repr(value), str_fmt
639f1cea 1287 if fmt[-1] in 'csr':
e0fd9573 1288 value = sanitizer(initial_field, value)
901130bb 1289
b868936c 1290 key = '%s\0%s' % (key.replace('%', '%\0'), outer_mobj.group('format'))
385a27fa 1291 TMPL_DICT[key] = value
b868936c 1292 return '{prefix}%({key}){fmt}'.format(key=key, fmt=fmt, prefix=outer_mobj.group('prefix'))
752cda38 1293
385a27fa 1294 return EXTERNAL_FORMAT_RE.sub(create_key, outtmpl), TMPL_DICT
143db31d 1295
819e0531 1296 def evaluate_outtmpl(self, outtmpl, info_dict, *args, **kwargs):
1297 outtmpl, info_dict = self.prepare_outtmpl(outtmpl, info_dict, *args, **kwargs)
1298 return self.escape_outtmpl(outtmpl) % info_dict
1299
5127e92a 1300 def _prepare_filename(self, info_dict, *, outtmpl=None, tmpl_type=None):
1301 assert None in (outtmpl, tmpl_type), 'outtmpl and tmpl_type are mutually exclusive'
1302 if outtmpl is None:
bf1824b3 1303 outtmpl = self.params['outtmpl'].get(tmpl_type or 'default', self.params['outtmpl']['default'])
8222d8de 1304 try:
5127e92a 1305 outtmpl = self._outtmpl_expandpath(outtmpl)
e0fd9573 1306 filename = self.evaluate_outtmpl(outtmpl, info_dict, True)
6a0546e3 1307 if not filename:
1308 return None
15da37c7 1309
5127e92a 1310 if tmpl_type in ('', 'temp'):
6a0546e3 1311 final_ext, ext = self.params.get('final_ext'), info_dict.get('ext')
1312 if final_ext and ext and final_ext != ext and filename.endswith(f'.{final_ext}'):
1313 filename = replace_extension(filename, ext, final_ext)
5127e92a 1314 elif tmpl_type:
6a0546e3 1315 force_ext = OUTTMPL_TYPES[tmpl_type]
1316 if force_ext:
1317 filename = replace_extension(filename, force_ext, info_dict.get('ext'))
de6000d9 1318
bdc3fd2f
U
1319 # https://github.com/blackjack4494/youtube-dlc/issues/85
1320 trim_file_name = self.params.get('trim_file_name', False)
1321 if trim_file_name:
5c22c63d 1322 no_ext, *ext = filename.rsplit('.', 2)
1323 filename = join_nonempty(no_ext[:trim_file_name], *ext, delim='.')
bdc3fd2f 1324
0202b52a 1325 return filename
8222d8de 1326 except ValueError as err:
6febd1c1 1327 self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')')
8222d8de
JMF
1328 return None
1329
5127e92a 1330 def prepare_filename(self, info_dict, dir_type='', *, outtmpl=None, warn=False):
1331 """Generate the output filename"""
1332 if outtmpl:
1333 assert not dir_type, 'outtmpl and dir_type are mutually exclusive'
1334 dir_type = None
1335 filename = self._prepare_filename(info_dict, tmpl_type=dir_type, outtmpl=outtmpl)
80c03fa9 1336 if not filename and dir_type not in ('', 'temp'):
1337 return ''
de6000d9 1338
c84aeac6 1339 if warn:
21cd8fae 1340 if not self.params.get('paths'):
de6000d9 1341 pass
1342 elif filename == '-':
c84aeac6 1343 self.report_warning('--paths is ignored when an outputting to stdout', only_once=True)
de6000d9 1344 elif os.path.isabs(filename):
c84aeac6 1345 self.report_warning('--paths is ignored since an absolute path is given in output template', only_once=True)
de6000d9 1346 if filename == '-' or not filename:
1347 return filename
1348
21cd8fae 1349 return self.get_output_path(dir_type, filename)
0202b52a 1350
120fe513 1351 def _match_entry(self, info_dict, incomplete=False, silent=False):
ecdec191 1352 """ Returns None if the file should be downloaded """
8222d8de 1353
3bec830a 1354 video_title = info_dict.get('title', info_dict.get('id', 'entry'))
c77495e3 1355
8b0d7497 1356 def check_filter():
8b0d7497 1357 if 'title' in info_dict:
1358 # This can happen when we're just evaluating the playlist
1359 title = info_dict['title']
1360 matchtitle = self.params.get('matchtitle', False)
1361 if matchtitle:
1362 if not re.search(matchtitle, title, re.IGNORECASE):
1363 return '"' + title + '" title did not match pattern "' + matchtitle + '"'
1364 rejecttitle = self.params.get('rejecttitle', False)
1365 if rejecttitle:
1366 if re.search(rejecttitle, title, re.IGNORECASE):
1367 return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
1368 date = info_dict.get('upload_date')
1369 if date is not None:
1370 dateRange = self.params.get('daterange', DateRange())
1371 if date not in dateRange:
86e5f3ed 1372 return f'{date_from_str(date).isoformat()} upload date is not in range {dateRange}'
8b0d7497 1373 view_count = info_dict.get('view_count')
1374 if view_count is not None:
1375 min_views = self.params.get('min_views')
1376 if min_views is not None and view_count < min_views:
1377 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
1378 max_views = self.params.get('max_views')
1379 if max_views is not None and view_count > max_views:
1380 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
1381 if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')):
1382 return 'Skipping "%s" because it is age restricted' % video_title
8b0d7497 1383
8f18aca8 1384 match_filter = self.params.get('match_filter')
1385 if match_filter is not None:
1386 try:
1387 ret = match_filter(info_dict, incomplete=incomplete)
1388 except TypeError:
1389 # For backward compatibility
1390 ret = None if incomplete else match_filter(info_dict)
492272fe 1391 if ret is NO_DEFAULT:
1392 while True:
1393 filename = self._format_screen(self.prepare_filename(info_dict), self.Styles.FILENAME)
1394 reply = input(self._format_screen(
1395 f'Download "{filename}"? (Y/n): ', self.Styles.EMPHASIS)).lower().strip()
1396 if reply in {'y', ''}:
1397 return None
1398 elif reply == 'n':
1399 return f'Skipping {video_title}'
492272fe 1400 elif ret is not None:
8f18aca8 1401 return ret
8b0d7497 1402 return None
1403
c77495e3 1404 if self.in_download_archive(info_dict):
1405 reason = '%s has already been recorded in the archive' % video_title
1406 break_opt, break_err = 'break_on_existing', ExistingVideoReached
1407 else:
1408 reason = check_filter()
1409 break_opt, break_err = 'break_on_reject', RejectedVideoReached
8b0d7497 1410 if reason is not None:
120fe513 1411 if not silent:
1412 self.to_screen('[download] ' + reason)
c77495e3 1413 if self.params.get(break_opt, False):
1414 raise break_err()
8b0d7497 1415 return reason
fe7e0c98 1416
b6c45014
JMF
1417 @staticmethod
1418 def add_extra_info(info_dict, extra_info):
1419 '''Set the keys from extra_info in info dict if they are missing'''
1420 for key, value in extra_info.items():
1421 info_dict.setdefault(key, value)
1422
409e1828 1423 def extract_info(self, url, download=True, ie_key=None, extra_info=None,
61aa5ba3 1424 process=True, force_generic_extractor=False):
41d1cca3 1425 """
17ffed18 1426 Extract and return the information dictionary of the URL
41d1cca3 1427
1428 Arguments:
17ffed18 1429 @param url URL to extract
41d1cca3 1430
1431 Keyword arguments:
17ffed18 1432 @param download Whether to download videos
1433 @param process Whether to resolve all unresolved references (URLs, playlist items).
1434 Must be True for download to work
1435 @param ie_key Use only the extractor with this key
1436
1437 @param extra_info Dictionary containing the extra values to add to the info (For internal use only)
1438 @force_generic_extractor Force using the generic extractor (Deprecated; use ie_key='Generic')
41d1cca3 1439 """
fe7e0c98 1440
409e1828 1441 if extra_info is None:
1442 extra_info = {}
1443
61aa5ba3 1444 if not ie_key and force_generic_extractor:
d22dec74
S
1445 ie_key = 'Generic'
1446
8222d8de 1447 if ie_key:
fe7866d0 1448 ies = {ie_key: self._ies[ie_key]} if ie_key in self._ies else {}
8222d8de
JMF
1449 else:
1450 ies = self._ies
1451
fe7866d0 1452 for key, ie in ies.items():
8222d8de
JMF
1453 if not ie.suitable(url):
1454 continue
1455
1456 if not ie.working():
6febd1c1
PH
1457 self.report_warning('The program functionality for this site has been marked as broken, '
1458 'and will probably not work.')
8222d8de 1459
1151c407 1460 temp_id = ie.get_temp_id(url)
fe7866d0 1461 if temp_id is not None and self.in_download_archive({'id': temp_id, 'ie_key': key}):
1462 self.to_screen(f'[{key}] {temp_id}: has already been recorded in the archive')
5e5be0c0 1463 if self.params.get('break_on_existing', False):
1464 raise ExistingVideoReached()
a0566bbf 1465 break
fe7866d0 1466 return self.__extract_info(url, self.get_info_extractor(key), download, extra_info, process)
a0566bbf 1467 else:
fe7866d0 1468 extractors_restricted = self.params.get('allowed_extractors') not in (None, ['default'])
1469 self.report_error(f'No suitable extractor{format_field(ie_key, None, " (%s)")} found for URL {url}',
1470 tb=False if extractors_restricted else None)
a0566bbf 1471
7e88d7d7 1472 def _handle_extraction_exceptions(func):
b5ae35ee 1473 @functools.wraps(func)
a0566bbf 1474 def wrapper(self, *args, **kwargs):
6da22e7d 1475 while True:
1476 try:
1477 return func(self, *args, **kwargs)
1478 except (DownloadCancelled, LazyList.IndexError, PagedList.IndexError):
8222d8de 1479 raise
6da22e7d 1480 except ReExtractInfo as e:
1481 if e.expected:
1482 self.to_screen(f'{e}; Re-extracting data')
1483 else:
1484 self.to_stderr('\r')
1485 self.report_warning(f'{e}; Re-extracting data')
1486 continue
1487 except GeoRestrictedError as e:
1488 msg = e.msg
1489 if e.countries:
1490 msg += '\nThis video is available in %s.' % ', '.join(
1491 map(ISO3166Utils.short2full, e.countries))
1492 msg += '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1493 self.report_error(msg)
1494 except ExtractorError as e: # An error we somewhat expected
1495 self.report_error(str(e), e.format_traceback())
1496 except Exception as e:
1497 if self.params.get('ignoreerrors'):
1498 self.report_error(str(e), tb=encode_compat_str(traceback.format_exc()))
1499 else:
1500 raise
1501 break
a0566bbf 1502 return wrapper
1503
693f0600 1504 def _wait_for_video(self, ie_result={}):
f2ebc5c7 1505 if (not self.params.get('wait_for_video')
1506 or ie_result.get('_type', 'video') != 'video'
1507 or ie_result.get('formats') or ie_result.get('url')):
1508 return
1509
1510 format_dur = lambda dur: '%02d:%02d:%02d' % timetuple_from_msec(dur * 1000)[:-1]
1511 last_msg = ''
1512
1513 def progress(msg):
1514 nonlocal last_msg
a7dc6a89 1515 full_msg = f'{msg}\n'
1516 if not self.params.get('noprogress'):
1517 full_msg = msg + ' ' * (len(last_msg) - len(msg)) + '\r'
1518 elif last_msg:
1519 return
1520 self.to_screen(full_msg, skip_eol=True)
f2ebc5c7 1521 last_msg = msg
1522
1523 min_wait, max_wait = self.params.get('wait_for_video')
1524 diff = try_get(ie_result, lambda x: x['release_timestamp'] - time.time())
1525 if diff is None and ie_result.get('live_status') == 'is_upcoming':
16c620bc 1526 diff = round(random.uniform(min_wait, max_wait) if (max_wait and min_wait) else (max_wait or min_wait), 0)
f2ebc5c7 1527 self.report_warning('Release time of video is not known')
693f0600 1528 elif ie_result and (diff or 0) <= 0:
f2ebc5c7 1529 self.report_warning('Video should already be available according to extracted info')
38d79fd1 1530 diff = min(max(diff or 0, min_wait or 0), max_wait or float('inf'))
f2ebc5c7 1531 self.to_screen(f'[wait] Waiting for {format_dur(diff)} - Press Ctrl+C to try now')
1532
1533 wait_till = time.time() + diff
1534 try:
1535 while True:
1536 diff = wait_till - time.time()
1537 if diff <= 0:
1538 progress('')
1539 raise ReExtractInfo('[wait] Wait period ended', expected=True)
1540 progress(f'[wait] Remaining time until next attempt: {self._format_screen(format_dur(diff), self.Styles.EMPHASIS)}')
1541 time.sleep(1)
1542 except KeyboardInterrupt:
1543 progress('')
1544 raise ReExtractInfo('[wait] Interrupted by user', expected=True)
1545 except BaseException as e:
1546 if not isinstance(e, ReExtractInfo):
1547 self.to_screen('')
1548 raise
1549
7e88d7d7 1550 @_handle_extraction_exceptions
58f197b7 1551 def __extract_info(self, url, ie, download, extra_info, process):
693f0600 1552 try:
1553 ie_result = ie.extract(url)
1554 except UserNotLive as e:
1555 if process:
1556 if self.params.get('wait_for_video'):
1557 self.report_warning(e)
1558 self._wait_for_video()
1559 raise
a0566bbf 1560 if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
cb794ee0 1561 self.report_warning(f'Extractor {ie.IE_NAME} returned nothing{bug_reports_message()}')
a0566bbf 1562 return
1563 if isinstance(ie_result, list):
1564 # Backwards compatibility: old IE result format
1565 ie_result = {
1566 '_type': 'compat_list',
1567 'entries': ie_result,
1568 }
e37d0efb 1569 if extra_info.get('original_url'):
1570 ie_result.setdefault('original_url', extra_info['original_url'])
a0566bbf 1571 self.add_default_extra_info(ie_result, ie, url)
1572 if process:
f2ebc5c7 1573 self._wait_for_video(ie_result)
a0566bbf 1574 return self.process_ie_result(ie_result, download, extra_info)
8222d8de 1575 else:
a0566bbf 1576 return ie_result
fe7e0c98 1577
ea38e55f 1578 def add_default_extra_info(self, ie_result, ie, url):
6033d980 1579 if url is not None:
1580 self.add_extra_info(ie_result, {
1581 'webpage_url': url,
1582 'original_url': url,
57ebfca3 1583 })
1584 webpage_url = ie_result.get('webpage_url')
1585 if webpage_url:
1586 self.add_extra_info(ie_result, {
1587 'webpage_url_basename': url_basename(webpage_url),
1588 'webpage_url_domain': get_domain(webpage_url),
6033d980 1589 })
1590 if ie is not None:
1591 self.add_extra_info(ie_result, {
1592 'extractor': ie.IE_NAME,
1593 'extractor_key': ie.ie_key(),
1594 })
ea38e55f 1595
58adec46 1596 def process_ie_result(self, ie_result, download=True, extra_info=None):
8222d8de
JMF
1597 """
1598 Take the result of the ie(may be modified) and resolve all unresolved
1599 references (URLs, playlist items).
1600
1601 It will also download the videos if 'download'.
1602 Returns the resolved ie_result.
1603 """
58adec46 1604 if extra_info is None:
1605 extra_info = {}
e8ee972c
PH
1606 result_type = ie_result.get('_type', 'video')
1607
057a5206 1608 if result_type in ('url', 'url_transparent'):
8f97a15d 1609 ie_result['url'] = sanitize_url(
1610 ie_result['url'], scheme='http' if self.params.get('prefer_insecure') else 'https')
e37d0efb 1611 if ie_result.get('original_url'):
1612 extra_info.setdefault('original_url', ie_result['original_url'])
1613
057a5206 1614 extract_flat = self.params.get('extract_flat', False)
3089bc74
S
1615 if ((extract_flat == 'in_playlist' and 'playlist' in extra_info)
1616 or extract_flat is True):
ecb54191 1617 info_copy = ie_result.copy()
6033d980 1618 ie = try_get(ie_result.get('ie_key'), self.get_info_extractor)
360167b9 1619 if ie and not ie_result.get('id'):
4614bc22 1620 info_copy['id'] = ie.get_temp_id(ie_result['url'])
6033d980 1621 self.add_default_extra_info(info_copy, ie, ie_result['url'])
4614bc22 1622 self.add_extra_info(info_copy, extra_info)
b5475f11 1623 info_copy, _ = self.pre_process(info_copy)
ecb54191 1624 self.__forced_printings(info_copy, self.prepare_filename(info_copy), incomplete=True)
415f8d51 1625 self._raise_pending_errors(info_copy)
4614bc22 1626 if self.params.get('force_write_download_archive', False):
1627 self.record_download_archive(info_copy)
e8ee972c
PH
1628 return ie_result
1629
8222d8de 1630 if result_type == 'video':
b6c45014 1631 self.add_extra_info(ie_result, extra_info)
9c2b75b5 1632 ie_result = self.process_video_result(ie_result, download=download)
415f8d51 1633 self._raise_pending_errors(ie_result)
28b0eb0f 1634 additional_urls = (ie_result or {}).get('additional_urls')
9c2b75b5 1635 if additional_urls:
e9f4ccd1 1636 # TODO: Improve MetadataParserPP to allow setting a list
14f25df2 1637 if isinstance(additional_urls, str):
9c2b75b5 1638 additional_urls = [additional_urls]
1639 self.to_screen(
1640 '[info] %s: %d additional URL(s) requested' % (ie_result['id'], len(additional_urls)))
1641 self.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls))
1642 ie_result['additional_entries'] = [
1643 self.extract_info(
b69fd25c 1644 url, download, extra_info=extra_info,
9c2b75b5 1645 force_generic_extractor=self.params.get('force_generic_extractor'))
1646 for url in additional_urls
1647 ]
1648 return ie_result
8222d8de
JMF
1649 elif result_type == 'url':
1650 # We have to add extra_info to the results because it may be
1651 # contained in a playlist
07cce701 1652 return self.extract_info(
1653 ie_result['url'], download,
1654 ie_key=ie_result.get('ie_key'),
1655 extra_info=extra_info)
7fc3fa05
PH
1656 elif result_type == 'url_transparent':
1657 # Use the information from the embedding page
1658 info = self.extract_info(
1659 ie_result['url'], ie_key=ie_result.get('ie_key'),
1660 extra_info=extra_info, download=False, process=False)
1661
1640eb09
S
1662 # extract_info may return None when ignoreerrors is enabled and
1663 # extraction failed with an error, don't crash and return early
1664 # in this case
1665 if not info:
1666 return info
1667
3975b4d2 1668 exempted_fields = {'_type', 'url', 'ie_key'}
1669 if not ie_result.get('section_end') and ie_result.get('section_start') is None:
1670 # For video clips, the id etc of the clip extractor should be used
1671 exempted_fields |= {'id', 'extractor', 'extractor_key'}
1672
412c617d 1673 new_result = info.copy()
3975b4d2 1674 new_result.update(filter_dict(ie_result, lambda k, v: v is not None and k not in exempted_fields))
7fc3fa05 1675
0563f7ac
S
1676 # Extracted info may not be a video result (i.e.
1677 # info.get('_type', 'video') != video) but rather an url or
1678 # url_transparent. In such cases outer metadata (from ie_result)
1679 # should be propagated to inner one (info). For this to happen
1680 # _type of info should be overridden with url_transparent. This
067aa17e 1681 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
0563f7ac
S
1682 if new_result.get('_type') == 'url':
1683 new_result['_type'] = 'url_transparent'
7fc3fa05
PH
1684
1685 return self.process_ie_result(
1686 new_result, download=download, extra_info=extra_info)
40fcba5e 1687 elif result_type in ('playlist', 'multi_video'):
30a074c2 1688 # Protect from infinite recursion due to recursively nested playlists
1689 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
0bd5a039 1690 webpage_url = ie_result.get('webpage_url') # Playlists maynot have webpage_url
1691 if webpage_url and webpage_url in self._playlist_urls:
7e85e872 1692 self.to_screen(
30a074c2 1693 '[download] Skipping already downloaded playlist: %s'
1694 % ie_result.get('title') or ie_result.get('id'))
1695 return
7e85e872 1696
30a074c2 1697 self._playlist_level += 1
1698 self._playlist_urls.add(webpage_url)
03f83004 1699 self._fill_common_fields(ie_result, False)
bc516a3f 1700 self._sanitize_thumbnails(ie_result)
30a074c2 1701 try:
1702 return self.__process_playlist(ie_result, download)
1703 finally:
1704 self._playlist_level -= 1
1705 if not self._playlist_level:
1706 self._playlist_urls.clear()
8222d8de 1707 elif result_type == 'compat_list':
c9bf4114
PH
1708 self.report_warning(
1709 'Extractor %s returned a compat_list result. '
1710 'It needs to be updated.' % ie_result.get('extractor'))
5f6a1245 1711
8222d8de 1712 def _fixup(r):
b868936c 1713 self.add_extra_info(r, {
1714 'extractor': ie_result['extractor'],
1715 'webpage_url': ie_result['webpage_url'],
1716 'webpage_url_basename': url_basename(ie_result['webpage_url']),
0bb322b9 1717 'webpage_url_domain': get_domain(ie_result['webpage_url']),
b868936c 1718 'extractor_key': ie_result['extractor_key'],
1719 })
8222d8de
JMF
1720 return r
1721 ie_result['entries'] = [
b6c45014 1722 self.process_ie_result(_fixup(r), download, extra_info)
8222d8de
JMF
1723 for r in ie_result['entries']
1724 ]
1725 return ie_result
1726 else:
1727 raise Exception('Invalid result type: %s' % result_type)
1728
e92caff5 1729 def _ensure_dir_exists(self, path):
1730 return make_dir(path, self.report_error)
1731
3b603dbd 1732 @staticmethod
3bec830a 1733 def _playlist_infodict(ie_result, strict=False, **kwargs):
1734 info = {
1735 'playlist_count': ie_result.get('playlist_count'),
3b603dbd 1736 'playlist': ie_result.get('title') or ie_result.get('id'),
1737 'playlist_id': ie_result.get('id'),
1738 'playlist_title': ie_result.get('title'),
1739 'playlist_uploader': ie_result.get('uploader'),
1740 'playlist_uploader_id': ie_result.get('uploader_id'),
3b603dbd 1741 **kwargs,
1742 }
3bec830a 1743 if strict:
1744 return info
0bd5a039 1745 if ie_result.get('webpage_url'):
1746 info.update({
1747 'webpage_url': ie_result['webpage_url'],
1748 'webpage_url_basename': url_basename(ie_result['webpage_url']),
1749 'webpage_url_domain': get_domain(ie_result['webpage_url']),
1750 })
3bec830a 1751 return {
1752 **info,
1753 'playlist_index': 0,
1754 '__last_playlist_index': max(ie_result['requested_entries'] or (0, 0)),
1755 'extractor': ie_result['extractor'],
3bec830a 1756 'extractor_key': ie_result['extractor_key'],
1757 }
3b603dbd 1758
30a074c2 1759 def __process_playlist(self, ie_result, download):
7e88d7d7 1760 """Process each entry in the playlist"""
f5ea4748 1761 assert ie_result['_type'] in ('playlist', 'multi_video')
1762
3bec830a 1763 common_info = self._playlist_infodict(ie_result, strict=True)
3955b207 1764 title = common_info.get('playlist') or '<Untitled>'
3bec830a 1765 if self._match_entry(common_info, incomplete=True) is not None:
1766 return
c6e07cf1 1767 self.to_screen(f'[download] Downloading {ie_result["_type"]}: {title}')
f0d785d3 1768
7e88d7d7 1769 all_entries = PlaylistEntries(self, ie_result)
7e9a6125 1770 entries = orderedSet(all_entries.get_requested_items(), lazy=True)
1771
1772 lazy = self.params.get('lazy_playlist')
1773 if lazy:
1774 resolved_entries, n_entries = [], 'N/A'
1775 ie_result['requested_entries'], ie_result['entries'] = None, None
1776 else:
1777 entries = resolved_entries = list(entries)
1778 n_entries = len(resolved_entries)
1779 ie_result['requested_entries'], ie_result['entries'] = tuple(zip(*resolved_entries)) or ([], [])
1780 if not ie_result.get('playlist_count'):
1781 # Better to do this after potentially exhausting entries
1782 ie_result['playlist_count'] = all_entries.get_full_count()
498f5606 1783
0647d925 1784 extra = self._playlist_infodict(ie_result, n_entries=int_or_none(n_entries))
1785 ie_copy = collections.ChainMap(ie_result, extra)
3bec830a 1786
e08a85d8 1787 _infojson_written = False
0bfc53d0 1788 write_playlist_files = self.params.get('allow_playlist_files', True)
1789 if write_playlist_files and self.params.get('list_thumbnails'):
1790 self.list_thumbnails(ie_result)
1791 if write_playlist_files and not self.params.get('simulate'):
e08a85d8 1792 _infojson_written = self._write_info_json(
1793 'playlist', ie_result, self.prepare_filename(ie_copy, 'pl_infojson'))
1794 if _infojson_written is None:
80c03fa9 1795 return
1796 if self._write_description('playlist', ie_result,
1797 self.prepare_filename(ie_copy, 'pl_description')) is None:
1798 return
681de68e 1799 # TODO: This should be passed to ThumbnailsConvertor if necessary
3bec830a 1800 self._write_thumbnails('playlist', ie_result, self.prepare_filename(ie_copy, 'pl_thumbnail'))
30a074c2 1801
7e9a6125 1802 if lazy:
1803 if self.params.get('playlistreverse') or self.params.get('playlistrandom'):
1804 self.report_warning('playlistreverse and playlistrandom are not supported with lazy_playlist', only_once=True)
1805 elif self.params.get('playlistreverse'):
1806 entries.reverse()
1807 elif self.params.get('playlistrandom'):
30a074c2 1808 random.shuffle(entries)
1809
7e88d7d7 1810 self.to_screen(f'[{ie_result["extractor"]}] Playlist {title}: Downloading {n_entries} videos'
1811 f'{format_field(ie_result, "playlist_count", " of %s")}')
30a074c2 1812
134c913c 1813 keep_resolved_entries = self.params.get('extract_flat') != 'discard'
1814 if self.params.get('extract_flat') == 'discard_in_playlist':
1815 keep_resolved_entries = ie_result['_type'] != 'playlist'
1816 if keep_resolved_entries:
1817 self.write_debug('The information of all playlist entries will be held in memory')
1818
26e2805c 1819 failures = 0
1820 max_failures = self.params.get('skip_playlist_after_errors') or float('inf')
7e9a6125 1821 for i, (playlist_index, entry) in enumerate(entries):
1822 if lazy:
1823 resolved_entries.append((playlist_index, entry))
3bec830a 1824 if not entry:
7e88d7d7 1825 continue
1826
7e88d7d7 1827 entry['__x_forwarded_for_ip'] = ie_result.get('__x_forwarded_for_ip')
7e9a6125 1828 if not lazy and 'playlist-index' in self.params.get('compat_opts', []):
1829 playlist_index = ie_result['requested_entries'][i]
1830
0647d925 1831 entry_copy = collections.ChainMap(entry, {
3bec830a 1832 **common_info,
3955b207 1833 'n_entries': int_or_none(n_entries),
71729754 1834 'playlist_index': playlist_index,
7e9a6125 1835 'playlist_autonumber': i + 1,
0647d925 1836 })
3bec830a 1837
0647d925 1838 if self._match_entry(entry_copy, incomplete=True) is not None:
f0ad6f8c 1839 # For compatabilty with youtube-dl. See https://github.com/yt-dlp/yt-dlp/issues/4369
1840 resolved_entries[i] = (playlist_index, NO_DEFAULT)
3bec830a 1841 continue
1842
1843 self.to_screen('[download] Downloading video %s of %s' % (
1844 self._format_screen(i + 1, self.Styles.ID), self._format_screen(n_entries, self.Styles.EMPHASIS)))
1845
a6ca61d4 1846 extra.update({
1847 'playlist_index': playlist_index,
1848 'playlist_autonumber': i + 1,
1849 })
3bec830a 1850 entry_result = self.__process_iterable_entry(entry, download, extra)
26e2805c 1851 if not entry_result:
1852 failures += 1
1853 if failures >= max_failures:
1854 self.report_error(
7e88d7d7 1855 f'Skipping the remaining entries in playlist "{title}" since {failures} items failed extraction')
26e2805c 1856 break
134c913c 1857 if keep_resolved_entries:
1858 resolved_entries[i] = (playlist_index, entry_result)
7e88d7d7 1859
1860 # Update with processed data
f0ad6f8c 1861 ie_result['requested_entries'] = [i for i, e in resolved_entries if e is not NO_DEFAULT]
1862 ie_result['entries'] = [e for _, e in resolved_entries if e is not NO_DEFAULT]
e08a85d8 1863
1864 # Write the updated info to json
cb96c5be 1865 if _infojson_written is True and self._write_info_json(
e08a85d8 1866 'updated playlist', ie_result,
1867 self.prepare_filename(ie_copy, 'pl_infojson'), overwrite=True) is None:
1868 return
ca30f449 1869
ed5835b4 1870 ie_result = self.run_all_pps('playlist', ie_result)
7e88d7d7 1871 self.to_screen(f'[download] Finished downloading playlist: {title}')
30a074c2 1872 return ie_result
1873
7e88d7d7 1874 @_handle_extraction_exceptions
a0566bbf 1875 def __process_iterable_entry(self, entry, download, extra_info):
1876 return self.process_ie_result(
1877 entry, download=download, extra_info=extra_info)
1878
67134eab
JMF
1879 def _build_format_filter(self, filter_spec):
1880 " Returns a function to filter the formats according to the filter_spec "
083c9df9
PH
1881
1882 OPERATORS = {
1883 '<': operator.lt,
1884 '<=': operator.le,
1885 '>': operator.gt,
1886 '>=': operator.ge,
1887 '=': operator.eq,
1888 '!=': operator.ne,
1889 }
67134eab 1890 operator_rex = re.compile(r'''(?x)\s*
187986a8 1891 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)\s*
1892 (?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1893 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)\s*
083c9df9 1894 ''' % '|'.join(map(re.escape, OPERATORS.keys())))
187986a8 1895 m = operator_rex.fullmatch(filter_spec)
9ddb6925
S
1896 if m:
1897 try:
1898 comparison_value = int(m.group('value'))
1899 except ValueError:
1900 comparison_value = parse_filesize(m.group('value'))
1901 if comparison_value is None:
1902 comparison_value = parse_filesize(m.group('value') + 'B')
1903 if comparison_value is None:
1904 raise ValueError(
1905 'Invalid value %r in format specification %r' % (
67134eab 1906 m.group('value'), filter_spec))
9ddb6925
S
1907 op = OPERATORS[m.group('op')]
1908
083c9df9 1909 if not m:
9ddb6925
S
1910 STR_OPERATORS = {
1911 '=': operator.eq,
10d33b34
YCH
1912 '^=': lambda attr, value: attr.startswith(value),
1913 '$=': lambda attr, value: attr.endswith(value),
1914 '*=': lambda attr, value: value in attr,
1ce9a3cb 1915 '~=': lambda attr, value: value.search(attr) is not None
9ddb6925 1916 }
187986a8 1917 str_operator_rex = re.compile(r'''(?x)\s*
1918 (?P<key>[a-zA-Z0-9._-]+)\s*
1ce9a3cb
LF
1919 (?P<negation>!\s*)?(?P<op>%s)\s*(?P<none_inclusive>\?\s*)?
1920 (?P<quote>["'])?
1921 (?P<value>(?(quote)(?:(?!(?P=quote))[^\\]|\\.)+|[\w.-]+))
1922 (?(quote)(?P=quote))\s*
9ddb6925 1923 ''' % '|'.join(map(re.escape, STR_OPERATORS.keys())))
187986a8 1924 m = str_operator_rex.fullmatch(filter_spec)
9ddb6925 1925 if m:
1ce9a3cb
LF
1926 if m.group('op') == '~=':
1927 comparison_value = re.compile(m.group('value'))
1928 else:
1929 comparison_value = re.sub(r'''\\([\\"'])''', r'\1', m.group('value'))
2cc779f4
S
1930 str_op = STR_OPERATORS[m.group('op')]
1931 if m.group('negation'):
e118a879 1932 op = lambda attr, value: not str_op(attr, value)
2cc779f4
S
1933 else:
1934 op = str_op
083c9df9 1935
9ddb6925 1936 if not m:
187986a8 1937 raise SyntaxError('Invalid filter specification %r' % filter_spec)
083c9df9
PH
1938
1939 def _filter(f):
1940 actual_value = f.get(m.group('key'))
1941 if actual_value is None:
1942 return m.group('none_inclusive')
1943 return op(actual_value, comparison_value)
67134eab
JMF
1944 return _filter
1945
9f1a1c36 1946 def _check_formats(self, formats):
1947 for f in formats:
1948 self.to_screen('[info] Testing format %s' % f['format_id'])
75689fe5 1949 path = self.get_output_path('temp')
1950 if not self._ensure_dir_exists(f'{path}/'):
1951 continue
1952 temp_file = tempfile.NamedTemporaryFile(suffix='.tmp', delete=False, dir=path or None)
9f1a1c36 1953 temp_file.close()
1954 try:
1955 success, _ = self.dl(temp_file.name, f, test=True)
8a82af35 1956 except (DownloadError, OSError, ValueError) + network_exceptions:
9f1a1c36 1957 success = False
1958 finally:
1959 if os.path.exists(temp_file.name):
1960 try:
1961 os.remove(temp_file.name)
1962 except OSError:
1963 self.report_warning('Unable to delete temporary file "%s"' % temp_file.name)
1964 if success:
1965 yield f
1966 else:
1967 self.to_screen('[info] Unable to download format %s. Skipping...' % f['format_id'])
1968
0017d9ad 1969 def _default_format_spec(self, info_dict, download=True):
0017d9ad 1970
af0f7428
S
1971 def can_merge():
1972 merger = FFmpegMergerPP(self)
1973 return merger.available and merger.can_merge()
1974
91ebc640 1975 prefer_best = (
b7b04c78 1976 not self.params.get('simulate')
91ebc640 1977 and download
1978 and (
1979 not can_merge()
21633673 1980 or info_dict.get('is_live') and not self.params.get('live_from_start')
bf1824b3 1981 or self.params['outtmpl']['default'] == '-'))
53ed7066 1982 compat = (
1983 prefer_best
1984 or self.params.get('allow_multiple_audio_streams', False)
8a82af35 1985 or 'format-spec' in self.params['compat_opts'])
91ebc640 1986
1987 return (
53ed7066 1988 'best/bestvideo+bestaudio' if prefer_best
1989 else 'bestvideo*+bestaudio/best' if not compat
91ebc640 1990 else 'bestvideo+bestaudio/best')
0017d9ad 1991
67134eab
JMF
1992 def build_format_selector(self, format_spec):
1993 def syntax_error(note, start):
1994 message = (
1995 'Invalid format specification: '
86e5f3ed 1996 '{}\n\t{}\n\t{}^'.format(note, format_spec, ' ' * start[1]))
67134eab
JMF
1997 return SyntaxError(message)
1998
1999 PICKFIRST = 'PICKFIRST'
2000 MERGE = 'MERGE'
2001 SINGLE = 'SINGLE'
0130afb7 2002 GROUP = 'GROUP'
67134eab
JMF
2003 FormatSelector = collections.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
2004
91ebc640 2005 allow_multiple_streams = {'audio': self.params.get('allow_multiple_audio_streams', False),
2006 'video': self.params.get('allow_multiple_video_streams', False)}
909d24dd 2007
9f1a1c36 2008 check_formats = self.params.get('check_formats') == 'selected'
e8e73840 2009
67134eab
JMF
2010 def _parse_filter(tokens):
2011 filter_parts = []
2012 for type, string, start, _, _ in tokens:
2013 if type == tokenize.OP and string == ']':
2014 return ''.join(filter_parts)
2015 else:
2016 filter_parts.append(string)
2017
232541df 2018 def _remove_unused_ops(tokens):
62b58c09
L
2019 # Remove operators that we don't use and join them with the surrounding strings.
2020 # E.g. 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
232541df
JMF
2021 ALLOWED_OPS = ('/', '+', ',', '(', ')')
2022 last_string, last_start, last_end, last_line = None, None, None, None
2023 for type, string, start, end, line in tokens:
2024 if type == tokenize.OP and string == '[':
2025 if last_string:
2026 yield tokenize.NAME, last_string, last_start, last_end, last_line
2027 last_string = None
2028 yield type, string, start, end, line
2029 # everything inside brackets will be handled by _parse_filter
2030 for type, string, start, end, line in tokens:
2031 yield type, string, start, end, line
2032 if type == tokenize.OP and string == ']':
2033 break
2034 elif type == tokenize.OP and string in ALLOWED_OPS:
2035 if last_string:
2036 yield tokenize.NAME, last_string, last_start, last_end, last_line
2037 last_string = None
2038 yield type, string, start, end, line
2039 elif type in [tokenize.NAME, tokenize.NUMBER, tokenize.OP]:
2040 if not last_string:
2041 last_string = string
2042 last_start = start
2043 last_end = end
2044 else:
2045 last_string += string
2046 if last_string:
2047 yield tokenize.NAME, last_string, last_start, last_end, last_line
2048
cf2ac6df 2049 def _parse_format_selection(tokens, inside_merge=False, inside_choice=False, inside_group=False):
67134eab
JMF
2050 selectors = []
2051 current_selector = None
2052 for type, string, start, _, _ in tokens:
2053 # ENCODING is only defined in python 3.x
2054 if type == getattr(tokenize, 'ENCODING', None):
2055 continue
2056 elif type in [tokenize.NAME, tokenize.NUMBER]:
2057 current_selector = FormatSelector(SINGLE, string, [])
2058 elif type == tokenize.OP:
cf2ac6df
JMF
2059 if string == ')':
2060 if not inside_group:
2061 # ')' will be handled by the parentheses group
2062 tokens.restore_last_token()
67134eab 2063 break
cf2ac6df 2064 elif inside_merge and string in ['/', ',']:
0130afb7
JMF
2065 tokens.restore_last_token()
2066 break
cf2ac6df
JMF
2067 elif inside_choice and string == ',':
2068 tokens.restore_last_token()
2069 break
2070 elif string == ',':
0a31a350
JMF
2071 if not current_selector:
2072 raise syntax_error('"," must follow a format selector', start)
67134eab
JMF
2073 selectors.append(current_selector)
2074 current_selector = None
2075 elif string == '/':
d96d604e
JMF
2076 if not current_selector:
2077 raise syntax_error('"/" must follow a format selector', start)
67134eab 2078 first_choice = current_selector
cf2ac6df 2079 second_choice = _parse_format_selection(tokens, inside_choice=True)
f5f4a27a 2080 current_selector = FormatSelector(PICKFIRST, (first_choice, second_choice), [])
67134eab
JMF
2081 elif string == '[':
2082 if not current_selector:
2083 current_selector = FormatSelector(SINGLE, 'best', [])
2084 format_filter = _parse_filter(tokens)
2085 current_selector.filters.append(format_filter)
0130afb7
JMF
2086 elif string == '(':
2087 if current_selector:
2088 raise syntax_error('Unexpected "("', start)
cf2ac6df
JMF
2089 group = _parse_format_selection(tokens, inside_group=True)
2090 current_selector = FormatSelector(GROUP, group, [])
67134eab 2091 elif string == '+':
d03cfdce 2092 if not current_selector:
2093 raise syntax_error('Unexpected "+"', start)
2094 selector_1 = current_selector
2095 selector_2 = _parse_format_selection(tokens, inside_merge=True)
2096 if not selector_2:
2097 raise syntax_error('Expected a selector', start)
2098 current_selector = FormatSelector(MERGE, (selector_1, selector_2), [])
67134eab 2099 else:
86e5f3ed 2100 raise syntax_error(f'Operator not recognized: "{string}"', start)
67134eab
JMF
2101 elif type == tokenize.ENDMARKER:
2102 break
2103 if current_selector:
2104 selectors.append(current_selector)
2105 return selectors
2106
f8d4ad9a 2107 def _merge(formats_pair):
2108 format_1, format_2 = formats_pair
2109
2110 formats_info = []
2111 formats_info.extend(format_1.get('requested_formats', (format_1,)))
2112 formats_info.extend(format_2.get('requested_formats', (format_2,)))
2113
2114 if not allow_multiple_streams['video'] or not allow_multiple_streams['audio']:
551f9388 2115 get_no_more = {'video': False, 'audio': False}
f8d4ad9a 2116 for (i, fmt_info) in enumerate(formats_info):
551f9388 2117 if fmt_info.get('acodec') == fmt_info.get('vcodec') == 'none':
2118 formats_info.pop(i)
2119 continue
2120 for aud_vid in ['audio', 'video']:
f8d4ad9a 2121 if not allow_multiple_streams[aud_vid] and fmt_info.get(aud_vid[0] + 'codec') != 'none':
2122 if get_no_more[aud_vid]:
2123 formats_info.pop(i)
f5510afe 2124 break
f8d4ad9a 2125 get_no_more[aud_vid] = True
2126
2127 if len(formats_info) == 1:
2128 return formats_info[0]
2129
2130 video_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('vcodec') != 'none']
2131 audio_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('acodec') != 'none']
2132
2133 the_only_video = video_fmts[0] if len(video_fmts) == 1 else None
2134 the_only_audio = audio_fmts[0] if len(audio_fmts) == 1 else None
2135
fc61aff4
LL
2136 output_ext = get_compatible_ext(
2137 vcodecs=[f.get('vcodec') for f in video_fmts],
2138 acodecs=[f.get('acodec') for f in audio_fmts],
2139 vexts=[f['ext'] for f in video_fmts],
2140 aexts=[f['ext'] for f in audio_fmts],
2141 preferences=(try_call(lambda: self.params['merge_output_format'].split('/'))
2142 or self.params.get('prefer_free_formats') and ('webm', 'mkv')))
f8d4ad9a 2143
975a0d0d 2144 filtered = lambda *keys: filter(None, (traverse_obj(fmt, *keys) for fmt in formats_info))
2145
f8d4ad9a 2146 new_dict = {
2147 'requested_formats': formats_info,
975a0d0d 2148 'format': '+'.join(filtered('format')),
2149 'format_id': '+'.join(filtered('format_id')),
f8d4ad9a 2150 'ext': output_ext,
975a0d0d 2151 'protocol': '+'.join(map(determine_protocol, formats_info)),
093a1710 2152 'language': '+'.join(orderedSet(filtered('language'))) or None,
2153 'format_note': '+'.join(orderedSet(filtered('format_note'))) or None,
2154 'filesize_approx': sum(filtered('filesize', 'filesize_approx')) or None,
975a0d0d 2155 'tbr': sum(filtered('tbr', 'vbr', 'abr')),
f8d4ad9a 2156 }
2157
2158 if the_only_video:
2159 new_dict.update({
2160 'width': the_only_video.get('width'),
2161 'height': the_only_video.get('height'),
2162 'resolution': the_only_video.get('resolution') or self.format_resolution(the_only_video),
2163 'fps': the_only_video.get('fps'),
49a57e70 2164 'dynamic_range': the_only_video.get('dynamic_range'),
f8d4ad9a 2165 'vcodec': the_only_video.get('vcodec'),
2166 'vbr': the_only_video.get('vbr'),
2167 'stretched_ratio': the_only_video.get('stretched_ratio'),
2168 })
2169
2170 if the_only_audio:
2171 new_dict.update({
2172 'acodec': the_only_audio.get('acodec'),
2173 'abr': the_only_audio.get('abr'),
975a0d0d 2174 'asr': the_only_audio.get('asr'),
b8ed0f15 2175 'audio_channels': the_only_audio.get('audio_channels')
f8d4ad9a 2176 })
2177
2178 return new_dict
2179
e8e73840 2180 def _check_formats(formats):
981052c9 2181 if not check_formats:
2182 yield from formats
b5ac45b1 2183 return
9f1a1c36 2184 yield from self._check_formats(formats)
e8e73840 2185
67134eab 2186 def _build_selector_function(selector):
909d24dd 2187 if isinstance(selector, list): # ,
67134eab
JMF
2188 fs = [_build_selector_function(s) for s in selector]
2189
317f7ab6 2190 def selector_function(ctx):
67134eab 2191 for f in fs:
981052c9 2192 yield from f(ctx)
67134eab 2193 return selector_function
909d24dd 2194
2195 elif selector.type == GROUP: # ()
0130afb7 2196 selector_function = _build_selector_function(selector.selector)
909d24dd 2197
2198 elif selector.type == PICKFIRST: # /
67134eab
JMF
2199 fs = [_build_selector_function(s) for s in selector.selector]
2200
317f7ab6 2201 def selector_function(ctx):
67134eab 2202 for f in fs:
317f7ab6 2203 picked_formats = list(f(ctx))
67134eab
JMF
2204 if picked_formats:
2205 return picked_formats
2206 return []
67134eab 2207
981052c9 2208 elif selector.type == MERGE: # +
2209 selector_1, selector_2 = map(_build_selector_function, selector.selector)
2210
2211 def selector_function(ctx):
adbc4ec4 2212 for pair in itertools.product(selector_1(ctx), selector_2(ctx)):
981052c9 2213 yield _merge(pair)
2214
909d24dd 2215 elif selector.type == SINGLE: # atom
598d185d 2216 format_spec = selector.selector or 'best'
909d24dd 2217
f8d4ad9a 2218 # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
909d24dd 2219 if format_spec == 'all':
2220 def selector_function(ctx):
9222c381 2221 yield from _check_formats(ctx['formats'][::-1])
f8d4ad9a 2222 elif format_spec == 'mergeall':
2223 def selector_function(ctx):
316f2650 2224 formats = list(_check_formats(
2225 f for f in ctx['formats'] if f.get('vcodec') != 'none' or f.get('acodec') != 'none'))
e01d6aa4 2226 if not formats:
2227 return
921b76ca 2228 merged_format = formats[-1]
2229 for f in formats[-2::-1]:
f8d4ad9a 2230 merged_format = _merge((merged_format, f))
2231 yield merged_format
909d24dd 2232
2233 else:
85e801a9 2234 format_fallback, seperate_fallback, format_reverse, format_idx = False, None, True, 1
eff63539 2235 mobj = re.match(
2236 r'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
2237 format_spec)
2238 if mobj is not None:
2239 format_idx = int_or_none(mobj.group('n'), default=1)
e8e73840 2240 format_reverse = mobj.group('bw')[0] == 'b'
eff63539 2241 format_type = (mobj.group('type') or [None])[0]
2242 not_format_type = {'v': 'a', 'a': 'v'}.get(format_type)
2243 format_modified = mobj.group('mod') is not None
909d24dd 2244
2245 format_fallback = not format_type and not format_modified # for b, w
8326b00a 2246 _filter_f = (
eff63539 2247 (lambda f: f.get('%scodec' % format_type) != 'none')
2248 if format_type and format_modified # bv*, ba*, wv*, wa*
2249 else (lambda f: f.get('%scodec' % not_format_type) == 'none')
2250 if format_type # bv, ba, wv, wa
2251 else (lambda f: f.get('vcodec') != 'none' and f.get('acodec') != 'none')
2252 if not format_modified # b, w
8326b00a 2253 else lambda f: True) # b*, w*
2254 filter_f = lambda f: _filter_f(f) and (
2255 f.get('vcodec') != 'none' or f.get('acodec') != 'none')
67134eab 2256 else:
48ee10ee 2257 if format_spec in self._format_selection_exts['audio']:
b11c04a8 2258 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none'
48ee10ee 2259 elif format_spec in self._format_selection_exts['video']:
b11c04a8 2260 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none' and f.get('vcodec') != 'none'
85e801a9 2261 seperate_fallback = lambda f: f.get('ext') == format_spec and f.get('vcodec') != 'none'
48ee10ee 2262 elif format_spec in self._format_selection_exts['storyboards']:
b11c04a8 2263 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') == 'none' and f.get('vcodec') == 'none'
2264 else:
b5ae35ee 2265 filter_f = lambda f: f.get('format_id') == format_spec # id
909d24dd 2266
2267 def selector_function(ctx):
2268 formats = list(ctx['formats'])
909d24dd 2269 matches = list(filter(filter_f, formats)) if filter_f is not None else formats
85e801a9 2270 if not matches:
2271 if format_fallback and ctx['incomplete_formats']:
2272 # for extractors with incomplete formats (audio only (soundcloud)
2273 # or video only (imgur)) best/worst will fallback to
2274 # best/worst {video,audio}-only format
2275 matches = formats
2276 elif seperate_fallback and not ctx['has_merged_format']:
2277 # for compatibility with youtube-dl when there is no pre-merged format
2278 matches = list(filter(seperate_fallback, formats))
981052c9 2279 matches = LazyList(_check_formats(matches[::-1 if format_reverse else 1]))
2280 try:
e8e73840 2281 yield matches[format_idx - 1]
4abea8ca 2282 except LazyList.IndexError:
981052c9 2283 return
083c9df9 2284
67134eab 2285 filters = [self._build_format_filter(f) for f in selector.filters]
083c9df9 2286
317f7ab6 2287 def final_selector(ctx):
adbc4ec4 2288 ctx_copy = dict(ctx)
67134eab 2289 for _filter in filters:
317f7ab6
S
2290 ctx_copy['formats'] = list(filter(_filter, ctx_copy['formats']))
2291 return selector_function(ctx_copy)
67134eab 2292 return final_selector
083c9df9 2293
0f06bcd7 2294 stream = io.BytesIO(format_spec.encode())
0130afb7 2295 try:
f9934b96 2296 tokens = list(_remove_unused_ops(tokenize.tokenize(stream.readline)))
0130afb7
JMF
2297 except tokenize.TokenError:
2298 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec)))
2299
86e5f3ed 2300 class TokenIterator:
0130afb7
JMF
2301 def __init__(self, tokens):
2302 self.tokens = tokens
2303 self.counter = 0
2304
2305 def __iter__(self):
2306 return self
2307
2308 def __next__(self):
2309 if self.counter >= len(self.tokens):
2310 raise StopIteration()
2311 value = self.tokens[self.counter]
2312 self.counter += 1
2313 return value
2314
2315 next = __next__
2316
2317 def restore_last_token(self):
2318 self.counter -= 1
2319
2320 parsed_selector = _parse_format_selection(iter(TokenIterator(tokens)))
67134eab 2321 return _build_selector_function(parsed_selector)
a9c58ad9 2322
e5660ee6 2323 def _calc_headers(self, info_dict):
8b7539d2 2324 res = merge_headers(self.params['http_headers'], info_dict.get('http_headers') or {})
e5660ee6 2325
c487cf00 2326 cookies = self._calc_cookies(info_dict['url'])
e5660ee6
JMF
2327 if cookies:
2328 res['Cookie'] = cookies
2329
0016b84e
S
2330 if 'X-Forwarded-For' not in res:
2331 x_forwarded_for_ip = info_dict.get('__x_forwarded_for_ip')
2332 if x_forwarded_for_ip:
2333 res['X-Forwarded-For'] = x_forwarded_for_ip
2334
e5660ee6
JMF
2335 return res
2336
c487cf00 2337 def _calc_cookies(self, url):
2338 pr = sanitized_Request(url)
e5660ee6 2339 self.cookiejar.add_cookie_header(pr)
662435f7 2340 return pr.get_header('Cookie')
e5660ee6 2341
9f1a1c36 2342 def _sort_thumbnails(self, thumbnails):
2343 thumbnails.sort(key=lambda t: (
2344 t.get('preference') if t.get('preference') is not None else -1,
2345 t.get('width') if t.get('width') is not None else -1,
2346 t.get('height') if t.get('height') is not None else -1,
2347 t.get('id') if t.get('id') is not None else '',
2348 t.get('url')))
2349
b0249bca 2350 def _sanitize_thumbnails(self, info_dict):
bc516a3f 2351 thumbnails = info_dict.get('thumbnails')
2352 if thumbnails is None:
2353 thumbnail = info_dict.get('thumbnail')
2354 if thumbnail:
2355 info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}]
9f1a1c36 2356 if not thumbnails:
2357 return
2358
2359 def check_thumbnails(thumbnails):
2360 for t in thumbnails:
2361 self.to_screen(f'[info] Testing thumbnail {t["id"]}')
2362 try:
2363 self.urlopen(HEADRequest(t['url']))
2364 except network_exceptions as err:
2365 self.to_screen(f'[info] Unable to connect to thumbnail {t["id"]} URL {t["url"]!r} - {err}. Skipping...')
2366 continue
2367 yield t
2368
2369 self._sort_thumbnails(thumbnails)
2370 for i, t in enumerate(thumbnails):
2371 if t.get('id') is None:
2372 t['id'] = '%d' % i
2373 if t.get('width') and t.get('height'):
2374 t['resolution'] = '%dx%d' % (t['width'], t['height'])
2375 t['url'] = sanitize_url(t['url'])
2376
2377 if self.params.get('check_formats') is True:
282f5709 2378 info_dict['thumbnails'] = LazyList(check_thumbnails(thumbnails[::-1]), reverse=True)
9f1a1c36 2379 else:
2380 info_dict['thumbnails'] = thumbnails
bc516a3f 2381
03f83004
LNO
2382 def _fill_common_fields(self, info_dict, is_video=True):
2383 # TODO: move sanitization here
2384 if is_video:
2385 # playlists are allowed to lack "title"
d4736fdb 2386 title = info_dict.get('title', NO_DEFAULT)
2387 if title is NO_DEFAULT:
03f83004
LNO
2388 raise ExtractorError('Missing "title" field in extractor result',
2389 video_id=info_dict['id'], ie=info_dict['extractor'])
d4736fdb 2390 info_dict['fulltitle'] = title
2391 if not title:
2392 if title == '':
2393 self.write_debug('Extractor gave empty title. Creating a generic title')
2394 else:
2395 self.report_warning('Extractor failed to obtain "title". Creating a generic title instead')
1d485a1a 2396 info_dict['title'] = f'{info_dict["extractor"].replace(":", "-")} video #{info_dict["id"]}'
03f83004
LNO
2397
2398 if info_dict.get('duration') is not None:
2399 info_dict['duration_string'] = formatSeconds(info_dict['duration'])
2400
2401 for ts_key, date_key in (
2402 ('timestamp', 'upload_date'),
2403 ('release_timestamp', 'release_date'),
2404 ('modified_timestamp', 'modified_date'),
2405 ):
2406 if info_dict.get(date_key) is None and info_dict.get(ts_key) is not None:
2407 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
2408 # see http://bugs.python.org/issue1646728)
19a03940 2409 with contextlib.suppress(ValueError, OverflowError, OSError):
03f83004
LNO
2410 upload_date = datetime.datetime.utcfromtimestamp(info_dict[ts_key])
2411 info_dict[date_key] = upload_date.strftime('%Y%m%d')
03f83004
LNO
2412
2413 live_keys = ('is_live', 'was_live')
2414 live_status = info_dict.get('live_status')
2415 if live_status is None:
2416 for key in live_keys:
2417 if info_dict.get(key) is False:
2418 continue
2419 if info_dict.get(key):
2420 live_status = key
2421 break
2422 if all(info_dict.get(key) is False for key in live_keys):
2423 live_status = 'not_live'
2424 if live_status:
2425 info_dict['live_status'] = live_status
2426 for key in live_keys:
2427 if info_dict.get(key) is None:
2428 info_dict[key] = (live_status == key)
2429
2430 # Auto generate title fields corresponding to the *_number fields when missing
2431 # in order to always have clean titles. This is very common for TV series.
2432 for field in ('chapter', 'season', 'episode'):
2433 if info_dict.get('%s_number' % field) is not None and not info_dict.get(field):
2434 info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field])
2435
415f8d51 2436 def _raise_pending_errors(self, info):
2437 err = info.pop('__pending_error', None)
2438 if err:
2439 self.report_error(err, tb=False)
2440
dd82ffea
JMF
2441 def process_video_result(self, info_dict, download=True):
2442 assert info_dict.get('_type', 'video') == 'video'
9c906919 2443 self._num_videos += 1
dd82ffea 2444
bec1fad2 2445 if 'id' not in info_dict:
fc08bdd6 2446 raise ExtractorError('Missing "id" field in extractor result', ie=info_dict['extractor'])
2447 elif not info_dict.get('id'):
2448 raise ExtractorError('Extractor failed to obtain "id"', ie=info_dict['extractor'])
455a15e2 2449
c9969434
S
2450 def report_force_conversion(field, field_not, conversion):
2451 self.report_warning(
2452 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
2453 % (field, field_not, conversion))
2454
2455 def sanitize_string_field(info, string_field):
2456 field = info.get(string_field)
14f25df2 2457 if field is None or isinstance(field, str):
c9969434
S
2458 return
2459 report_force_conversion(string_field, 'a string', 'string')
14f25df2 2460 info[string_field] = str(field)
c9969434
S
2461
2462 def sanitize_numeric_fields(info):
2463 for numeric_field in self._NUMERIC_FIELDS:
2464 field = info.get(numeric_field)
f9934b96 2465 if field is None or isinstance(field, (int, float)):
c9969434
S
2466 continue
2467 report_force_conversion(numeric_field, 'numeric', 'int')
2468 info[numeric_field] = int_or_none(field)
2469
2470 sanitize_string_field(info_dict, 'id')
2471 sanitize_numeric_fields(info_dict)
3975b4d2 2472 if info_dict.get('section_end') and info_dict.get('section_start') is not None:
2473 info_dict['duration'] = round(info_dict['section_end'] - info_dict['section_start'], 3)
4c3f8c3f 2474 if (info_dict.get('duration') or 0) <= 0 and info_dict.pop('duration', None):
50e93e03 2475 self.report_warning('"duration" field is negative, there is an error in extractor')
be6217b2 2476
9eef7c4e 2477 chapters = info_dict.get('chapters') or []
a3976e07 2478 if chapters and chapters[0].get('start_time'):
2479 chapters.insert(0, {'start_time': 0})
2480
9eef7c4e 2481 dummy_chapter = {'end_time': 0, 'start_time': info_dict.get('duration')}
a3976e07 2482 for idx, (prev, current, next_) in enumerate(zip(
2483 (dummy_chapter, *chapters), chapters, (*chapters[1:], dummy_chapter)), 1):
9eef7c4e 2484 if current.get('start_time') is None:
2485 current['start_time'] = prev.get('end_time')
2486 if not current.get('end_time'):
2487 current['end_time'] = next_.get('start_time')
a3976e07 2488 if not current.get('title'):
2489 current['title'] = f'<Untitled Chapter {idx}>'
9eef7c4e 2490
dd82ffea
JMF
2491 if 'playlist' not in info_dict:
2492 # It isn't part of a playlist
2493 info_dict['playlist'] = None
2494 info_dict['playlist_index'] = None
2495
bc516a3f 2496 self._sanitize_thumbnails(info_dict)
d5519808 2497
536a55da 2498 thumbnail = info_dict.get('thumbnail')
bc516a3f 2499 thumbnails = info_dict.get('thumbnails')
536a55da
S
2500 if thumbnail:
2501 info_dict['thumbnail'] = sanitize_url(thumbnail)
2502 elif thumbnails:
d5519808
PH
2503 info_dict['thumbnail'] = thumbnails[-1]['url']
2504
ae30b840 2505 if info_dict.get('display_id') is None and 'id' in info_dict:
0afef30b
PH
2506 info_dict['display_id'] = info_dict['id']
2507
03f83004 2508 self._fill_common_fields(info_dict)
33d2fc2f 2509
05108a49
S
2510 for cc_kind in ('subtitles', 'automatic_captions'):
2511 cc = info_dict.get(cc_kind)
2512 if cc:
2513 for _, subtitle in cc.items():
2514 for subtitle_format in subtitle:
2515 if subtitle_format.get('url'):
2516 subtitle_format['url'] = sanitize_url(subtitle_format['url'])
2517 if subtitle_format.get('ext') is None:
2518 subtitle_format['ext'] = determine_ext(subtitle_format['url']).lower()
2519
2520 automatic_captions = info_dict.get('automatic_captions')
4bba3716 2521 subtitles = info_dict.get('subtitles')
4bba3716 2522
360e1ca5 2523 info_dict['requested_subtitles'] = self.process_subtitles(
05108a49 2524 info_dict['id'], subtitles, automatic_captions)
a504ced0 2525
dd82ffea
JMF
2526 if info_dict.get('formats') is None:
2527 # There's only one format available
2528 formats = [info_dict]
2529 else:
2530 formats = info_dict['formats']
2531
0a5a191a 2532 # or None ensures --clean-infojson removes it
2533 info_dict['_has_drm'] = any(f.get('has_drm') for f in formats) or None
88acdbc2 2534 if not self.params.get('allow_unplayable_formats'):
2535 formats = [f for f in formats if not f.get('has_drm')]
17ffed18 2536
2537 if formats and all(f.get('acodec') == f.get('vcodec') == 'none' for f in formats):
2538 self.report_warning(
2539 f'{"This video is DRM protected and " if info_dict["_has_drm"] else ""}'
2540 'only images are available for download. Use --list-formats to see them'.capitalize())
88acdbc2 2541
319b6059 2542 get_from_start = not info_dict.get('is_live') or bool(self.params.get('live_from_start'))
2543 if not get_from_start:
2544 info_dict['title'] += ' ' + datetime.datetime.now().strftime('%Y-%m-%d %H:%M')
2545 if info_dict.get('is_live') and formats:
adbc4ec4 2546 formats = [f for f in formats if bool(f.get('is_from_start')) == get_from_start]
319b6059 2547 if get_from_start and not formats:
a44ca5a4 2548 self.raise_no_formats(info_dict, msg=(
2549 '--live-from-start is passed, but there are no formats that can be downloaded from the start. '
2550 'If you want to download from the current time, use --no-live-from-start'))
adbc4ec4 2551
73af5cc8
S
2552 def is_wellformed(f):
2553 url = f.get('url')
a5ac0c47 2554 if not url:
73af5cc8
S
2555 self.report_warning(
2556 '"url" field is missing or empty - skipping format, '
2557 'there is an error in extractor')
a5ac0c47
S
2558 return False
2559 if isinstance(url, bytes):
2560 sanitize_string_field(f, 'url')
2561 return True
73af5cc8
S
2562
2563 # Filter out malformed formats for better extraction robustness
1ac7f461 2564 formats = list(filter(is_wellformed, formats or []))
2565
2566 if not formats:
2567 self.raise_no_formats(info_dict)
73af5cc8 2568
181c7053
S
2569 formats_dict = {}
2570
dd82ffea 2571 # We check that all the formats have the format and format_id fields
db95dc13 2572 for i, format in enumerate(formats):
c9969434
S
2573 sanitize_string_field(format, 'format_id')
2574 sanitize_numeric_fields(format)
dcf77cf1 2575 format['url'] = sanitize_url(format['url'])
e74e3b63 2576 if not format.get('format_id'):
14f25df2 2577 format['format_id'] = str(i)
e2effb08
S
2578 else:
2579 # Sanitize format_id from characters used in format selector expression
ec85ded8 2580 format['format_id'] = re.sub(r'[\s,/+\[\]()]', '_', format['format_id'])
181c7053
S
2581 format_id = format['format_id']
2582 if format_id not in formats_dict:
2583 formats_dict[format_id] = []
2584 formats_dict[format_id].append(format)
2585
2586 # Make sure all formats have unique format_id
03b4de72 2587 common_exts = set(itertools.chain(*self._format_selection_exts.values()))
181c7053 2588 for format_id, ambiguous_formats in formats_dict.items():
48ee10ee 2589 ambigious_id = len(ambiguous_formats) > 1
2590 for i, format in enumerate(ambiguous_formats):
2591 if ambigious_id:
181c7053 2592 format['format_id'] = '%s-%d' % (format_id, i)
48ee10ee 2593 if format.get('ext') is None:
2594 format['ext'] = determine_ext(format['url']).lower()
2595 # Ensure there is no conflict between id and ext in format selection
2596 # See https://github.com/yt-dlp/yt-dlp/issues/1282
2597 if format['format_id'] != format['ext'] and format['format_id'] in common_exts:
2598 format['format_id'] = 'f%s' % format['format_id']
181c7053
S
2599
2600 for i, format in enumerate(formats):
8c51aa65 2601 if format.get('format') is None:
6febd1c1 2602 format['format'] = '{id} - {res}{note}'.format(
8c51aa65
JMF
2603 id=format['format_id'],
2604 res=self.format_resolution(format),
b868936c 2605 note=format_field(format, 'format_note', ' (%s)'),
8c51aa65 2606 )
6f0be937 2607 if format.get('protocol') is None:
b5559424 2608 format['protocol'] = determine_protocol(format)
239df021 2609 if format.get('resolution') is None:
2610 format['resolution'] = self.format_resolution(format, default=None)
176f1866 2611 if format.get('dynamic_range') is None and format.get('vcodec') != 'none':
2612 format['dynamic_range'] = 'SDR'
f2fe69c7 2613 if (info_dict.get('duration') and format.get('tbr')
2614 and not format.get('filesize') and not format.get('filesize_approx')):
56ba69e4 2615 format['filesize_approx'] = int(info_dict['duration'] * format['tbr'] * (1024 / 8))
f2fe69c7 2616
e5660ee6
JMF
2617 # Add HTTP headers, so that external programs can use them from the
2618 # json output
2619 full_format_info = info_dict.copy()
2620 full_format_info.update(format)
2621 format['http_headers'] = self._calc_headers(full_format_info)
0016b84e
S
2622 # Remove private housekeeping stuff
2623 if '__x_forwarded_for_ip' in info_dict:
2624 del info_dict['__x_forwarded_for_ip']
dd82ffea 2625
9f1a1c36 2626 if self.params.get('check_formats') is True:
282f5709 2627 formats = LazyList(self._check_formats(formats[::-1]), reverse=True)
9f1a1c36 2628
88acdbc2 2629 if not formats or formats[0] is not info_dict:
b3d9ef88
JMF
2630 # only set the 'formats' fields if the original info_dict list them
2631 # otherwise we end up with a circular reference, the first (and unique)
f89197d7 2632 # element in the 'formats' field in info_dict is info_dict itself,
dfb1b146 2633 # which can't be exported to json
b3d9ef88 2634 info_dict['formats'] = formats
4ec82a72 2635
2636 info_dict, _ = self.pre_process(info_dict)
2637
6db9c4d5 2638 if self._match_entry(info_dict, incomplete=self._format_fields) is not None:
09b49e1f 2639 return info_dict
2640
2641 self.post_extract(info_dict)
2642 info_dict, _ = self.pre_process(info_dict, 'after_filter')
2643
093a1710 2644 # The pre-processors may have modified the formats
2645 formats = info_dict.get('formats', [info_dict])
2646
fa9f30b8 2647 list_only = self.params.get('simulate') is None and (
2648 self.params.get('list_thumbnails') or self.params.get('listformats') or self.params.get('listsubtitles'))
2649 interactive_format_selection = not list_only and self.format_selector == '-'
b7b04c78 2650 if self.params.get('list_thumbnails'):
2651 self.list_thumbnails(info_dict)
b7b04c78 2652 if self.params.get('listsubtitles'):
2653 if 'automatic_captions' in info_dict:
2654 self.list_subtitles(
2655 info_dict['id'], automatic_captions, 'automatic captions')
2656 self.list_subtitles(info_dict['id'], subtitles, 'subtitles')
fa9f30b8 2657 if self.params.get('listformats') or interactive_format_selection:
b69fd25c 2658 self.list_formats(info_dict)
169dbde9 2659 if list_only:
b7b04c78 2660 # Without this printing, -F --print-json will not work
169dbde9 2661 self.__forced_printings(info_dict, self.prepare_filename(info_dict), incomplete=True)
c487cf00 2662 return info_dict
bfaae0a7 2663
187986a8 2664 format_selector = self.format_selector
2665 if format_selector is None:
0017d9ad 2666 req_format = self._default_format_spec(info_dict, download=download)
0760b0a7 2667 self.write_debug('Default format spec: %s' % req_format)
187986a8 2668 format_selector = self.build_format_selector(req_format)
317f7ab6 2669
fa9f30b8 2670 while True:
2671 if interactive_format_selection:
2672 req_format = input(
2673 self._format_screen('\nEnter format selector: ', self.Styles.EMPHASIS))
2674 try:
2675 format_selector = self.build_format_selector(req_format)
2676 except SyntaxError as err:
2677 self.report_error(err, tb=False, is_error=False)
2678 continue
2679
85e801a9 2680 formats_to_download = list(format_selector({
fa9f30b8 2681 'formats': formats,
85e801a9 2682 'has_merged_format': any('none' not in (f.get('acodec'), f.get('vcodec')) for f in formats),
2683 'incomplete_formats': (
2684 # All formats are video-only or
2685 all(f.get('vcodec') != 'none' and f.get('acodec') == 'none' for f in formats)
2686 # all formats are audio-only
2687 or all(f.get('vcodec') == 'none' and f.get('acodec') != 'none' for f in formats)),
2688 }))
fa9f30b8 2689 if interactive_format_selection and not formats_to_download:
2690 self.report_error('Requested format is not available', tb=False, is_error=False)
2691 continue
2692 break
317f7ab6 2693
dd82ffea 2694 if not formats_to_download:
b7da73eb 2695 if not self.params.get('ignore_no_formats_error'):
c0b6e5c7 2696 raise ExtractorError(
2697 'Requested format is not available. Use --list-formats for a list of available formats',
2698 expected=True, video_id=info_dict['id'], ie=info_dict['extractor'])
b62fa6d7 2699 self.report_warning('Requested format is not available')
2700 # Process what we can, even without any available formats.
2701 formats_to_download = [{}]
a13e6848 2702
0500ee3d 2703 requested_ranges = tuple(self.params.get('download_ranges', lambda *_: [{}])(info_dict, self))
5ec1b6b7 2704 best_format, downloaded_formats = formats_to_download[-1], []
b62fa6d7 2705 if download:
0500ee3d 2706 if best_format and requested_ranges:
5ec1b6b7 2707 def to_screen(*msg):
2708 self.to_screen(f'[info] {info_dict["id"]}: {" ".join(", ".join(variadic(m)) for m in msg)}')
2709
2710 to_screen(f'Downloading {len(formats_to_download)} format(s):',
2711 (f['format_id'] for f in formats_to_download))
0500ee3d 2712 if requested_ranges != ({}, ):
5ec1b6b7 2713 to_screen(f'Downloading {len(requested_ranges)} time ranges:',
fc2ba496 2714 (f'{c["start_time"]:.1f}-{c["end_time"]:.1f}' for c in requested_ranges))
a13e6848 2715 max_downloads_reached = False
5ec1b6b7 2716
0500ee3d 2717 for fmt, chapter in itertools.product(formats_to_download, requested_ranges):
5ec1b6b7 2718 new_info = self._copy_infodict(info_dict)
b7da73eb 2719 new_info.update(fmt)
3975b4d2 2720 offset, duration = info_dict.get('section_start') or 0, info_dict.get('duration') or float('inf')
fc2ba496 2721 end_time = offset + min(chapter.get('end_time', duration), duration)
3975b4d2 2722 if chapter or offset:
5ec1b6b7 2723 new_info.update({
3975b4d2 2724 'section_start': offset + chapter.get('start_time', 0),
fc2ba496 2725 'section_end': end_time if end_time < offset + duration else None,
5ec1b6b7 2726 'section_title': chapter.get('title'),
2727 'section_number': chapter.get('index'),
2728 })
2729 downloaded_formats.append(new_info)
a13e6848 2730 try:
2731 self.process_info(new_info)
2732 except MaxDownloadsReached:
2733 max_downloads_reached = True
415f8d51 2734 self._raise_pending_errors(new_info)
f46e2f9d 2735 # Remove copied info
2736 for key, val in tuple(new_info.items()):
2737 if info_dict.get(key) == val:
2738 new_info.pop(key)
a13e6848 2739 if max_downloads_reached:
2740 break
ebed8b37 2741
5ec1b6b7 2742 write_archive = {f.get('__write_download_archive', False) for f in downloaded_formats}
a13e6848 2743 assert write_archive.issubset({True, False, 'ignore'})
2744 if True in write_archive and False not in write_archive:
2745 self.record_download_archive(info_dict)
be72c624 2746
5ec1b6b7 2747 info_dict['requested_downloads'] = downloaded_formats
ed5835b4 2748 info_dict = self.run_all_pps('after_video', info_dict)
a13e6848 2749 if max_downloads_reached:
2750 raise MaxDownloadsReached()
ebed8b37 2751
49a57e70 2752 # We update the info dict with the selected best quality format (backwards compatibility)
be72c624 2753 info_dict.update(best_format)
dd82ffea
JMF
2754 return info_dict
2755
98c70d6f 2756 def process_subtitles(self, video_id, normal_subtitles, automatic_captions):
a504ced0 2757 """Select the requested subtitles and their format"""
d8a58ddc 2758 available_subs, normal_sub_langs = {}, []
98c70d6f
JMF
2759 if normal_subtitles and self.params.get('writesubtitles'):
2760 available_subs.update(normal_subtitles)
d8a58ddc 2761 normal_sub_langs = tuple(normal_subtitles.keys())
98c70d6f
JMF
2762 if automatic_captions and self.params.get('writeautomaticsub'):
2763 for lang, cap_info in automatic_captions.items():
360e1ca5
JMF
2764 if lang not in available_subs:
2765 available_subs[lang] = cap_info
2766
d2c8aadf 2767 if not available_subs or (
2768 not self.params.get('writesubtitles')
2769 and not self.params.get('writeautomaticsub')):
4d171848 2770 return None
a504ced0 2771
d8a58ddc 2772 all_sub_langs = tuple(available_subs.keys())
a504ced0 2773 if self.params.get('allsubtitles', False):
c32b0aab 2774 requested_langs = all_sub_langs
2775 elif self.params.get('subtitleslangs', False):
5314b521 2776 try:
2777 requested_langs = orderedSet_from_options(
2778 self.params.get('subtitleslangs'), {'all': all_sub_langs}, use_regex=True)
2779 except re.error as e:
2780 raise ValueError(f'Wrong regex for subtitlelangs: {e.pattern}')
d8a58ddc 2781 elif normal_sub_langs:
2782 requested_langs = ['en'] if 'en' in normal_sub_langs else normal_sub_langs[:1]
a504ced0 2783 else:
d8a58ddc 2784 requested_langs = ['en'] if 'en' in all_sub_langs else all_sub_langs[:1]
ad3dc496 2785 if requested_langs:
d2c8aadf 2786 self.to_screen(f'[info] {video_id}: Downloading subtitles: {", ".join(requested_langs)}')
a504ced0
JMF
2787
2788 formats_query = self.params.get('subtitlesformat', 'best')
2789 formats_preference = formats_query.split('/') if formats_query else []
2790 subs = {}
2791 for lang in requested_langs:
2792 formats = available_subs.get(lang)
2793 if formats is None:
86e5f3ed 2794 self.report_warning(f'{lang} subtitles not available for {video_id}')
a504ced0 2795 continue
a504ced0
JMF
2796 for ext in formats_preference:
2797 if ext == 'best':
2798 f = formats[-1]
2799 break
2800 matches = list(filter(lambda f: f['ext'] == ext, formats))
2801 if matches:
2802 f = matches[-1]
2803 break
2804 else:
2805 f = formats[-1]
2806 self.report_warning(
2807 'No subtitle format found matching "%s" for language %s, '
2808 'using %s' % (formats_query, lang, f['ext']))
2809 subs[lang] = f
2810 return subs
2811
bb66c247 2812 def _forceprint(self, key, info_dict):
2813 if info_dict is None:
2814 return
2815 info_copy = info_dict.copy()
2816 info_copy['formats_table'] = self.render_formats_table(info_dict)
2817 info_copy['thumbnails_table'] = self.render_thumbnails_table(info_dict)
2818 info_copy['subtitles_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('subtitles'))
2819 info_copy['automatic_captions_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('automatic_captions'))
2820
2821 def format_tmpl(tmpl):
48c8424b 2822 mobj = re.fullmatch(r'([\w.:,]|-\d|(?P<dict>{([\w.:,]|-\d)+}))+=?', tmpl)
07a1250e 2823 if not mobj:
2824 return tmpl
48c8424b 2825
2826 fmt = '%({})s'
2827 if tmpl.startswith('{'):
2828 tmpl = f'.{tmpl}'
2829 if tmpl.endswith('='):
2830 tmpl, fmt = tmpl[:-1], '{0} = %({0})#j'
2831 return '\n'.join(map(fmt.format, [tmpl] if mobj.group('dict') else tmpl.split(',')))
8130779d 2832
bb66c247 2833 for tmpl in self.params['forceprint'].get(key, []):
2834 self.to_stdout(self.evaluate_outtmpl(format_tmpl(tmpl), info_copy))
2835
2836 for tmpl, file_tmpl in self.params['print_to_file'].get(key, []):
5127e92a 2837 filename = self.prepare_filename(info_dict, outtmpl=file_tmpl)
bb66c247 2838 tmpl = format_tmpl(tmpl)
2839 self.to_screen(f'[info] Writing {tmpl!r} to: {filename}')
8d93e69d 2840 if self._ensure_dir_exists(filename):
86e5f3ed 2841 with open(filename, 'a', encoding='utf-8') as f:
8d93e69d 2842 f.write(self.evaluate_outtmpl(tmpl, info_copy) + '\n')
ca30f449 2843
d06daf23 2844 def __forced_printings(self, info_dict, filename, incomplete):
53c18592 2845 def print_mandatory(field, actual_field=None):
2846 if actual_field is None:
2847 actual_field = field
d06daf23 2848 if (self.params.get('force%s' % field, False)
53c18592 2849 and (not incomplete or info_dict.get(actual_field) is not None)):
2850 self.to_stdout(info_dict[actual_field])
d06daf23
S
2851
2852 def print_optional(field):
2853 if (self.params.get('force%s' % field, False)
2854 and info_dict.get(field) is not None):
2855 self.to_stdout(info_dict[field])
2856
53c18592 2857 info_dict = info_dict.copy()
2858 if filename is not None:
2859 info_dict['filename'] = filename
2860 if info_dict.get('requested_formats') is not None:
2861 # For RTMP URLs, also include the playpath
2862 info_dict['urls'] = '\n'.join(f['url'] + f.get('play_path', '') for f in info_dict['requested_formats'])
10331a26 2863 elif info_dict.get('url'):
53c18592 2864 info_dict['urls'] = info_dict['url'] + info_dict.get('play_path', '')
2865
bb66c247 2866 if (self.params.get('forcejson')
2867 or self.params['forceprint'].get('video')
2868 or self.params['print_to_file'].get('video')):
2b8a2973 2869 self.post_extract(info_dict)
bb66c247 2870 self._forceprint('video', info_dict)
53c18592 2871
d06daf23
S
2872 print_mandatory('title')
2873 print_mandatory('id')
53c18592 2874 print_mandatory('url', 'urls')
d06daf23
S
2875 print_optional('thumbnail')
2876 print_optional('description')
53c18592 2877 print_optional('filename')
b868936c 2878 if self.params.get('forceduration') and info_dict.get('duration') is not None:
d06daf23
S
2879 self.to_stdout(formatSeconds(info_dict['duration']))
2880 print_mandatory('format')
53c18592 2881
2b8a2973 2882 if self.params.get('forcejson'):
6e84b215 2883 self.to_stdout(json.dumps(self.sanitize_info(info_dict)))
d06daf23 2884
e8e73840 2885 def dl(self, name, info, subtitle=False, test=False):
88acdbc2 2886 if not info.get('url'):
1151c407 2887 self.raise_no_formats(info, True)
e8e73840 2888
2889 if test:
2890 verbose = self.params.get('verbose')
2891 params = {
2892 'test': True,
a169858f 2893 'quiet': self.params.get('quiet') or not verbose,
e8e73840 2894 'verbose': verbose,
2895 'noprogress': not verbose,
2896 'nopart': True,
2897 'skip_unavailable_fragments': False,
2898 'keep_fragments': False,
2899 'overwrites': True,
2900 '_no_ytdl_file': True,
2901 }
2902 else:
2903 params = self.params
96fccc10 2904 fd = get_suitable_downloader(info, params, to_stdout=(name == '-'))(self, params)
e8e73840 2905 if not test:
2906 for ph in self._progress_hooks:
2907 fd.add_progress_hook(ph)
42676437
M
2908 urls = '", "'.join(
2909 (f['url'].split(',')[0] + ',<data>' if f['url'].startswith('data:') else f['url'])
2910 for f in info.get('requested_formats', []) or [info])
3a408f9d 2911 self.write_debug(f'Invoking {fd.FD_NAME} downloader on "{urls}"')
03b4de72 2912
adbc4ec4
THD
2913 # Note: Ideally info should be a deep-copied so that hooks cannot modify it.
2914 # But it may contain objects that are not deep-copyable
2915 new_info = self._copy_infodict(info)
e8e73840 2916 if new_info.get('http_headers') is None:
2917 new_info['http_headers'] = self._calc_headers(new_info)
2918 return fd.download(name, new_info, subtitle)
2919
e04938ab 2920 def existing_file(self, filepaths, *, default_overwrite=True):
2921 existing_files = list(filter(os.path.exists, orderedSet(filepaths)))
2922 if existing_files and not self.params.get('overwrites', default_overwrite):
2923 return existing_files[0]
2924
2925 for file in existing_files:
2926 self.report_file_delete(file)
2927 os.remove(file)
2928 return None
2929
8222d8de 2930 def process_info(self, info_dict):
09b49e1f 2931 """Process a single resolved IE result. (Modifies it in-place)"""
8222d8de
JMF
2932
2933 assert info_dict.get('_type', 'video') == 'video'
f46e2f9d 2934 original_infodict = info_dict
fd288278 2935
4513a41a 2936 if 'format' not in info_dict and 'ext' in info_dict:
8222d8de
JMF
2937 info_dict['format'] = info_dict['ext']
2938
09b49e1f 2939 # This is mostly just for backward compatibility of process_info
2940 # As a side-effect, this allows for format-specific filters
c77495e3 2941 if self._match_entry(info_dict) is not None:
9e907ebd 2942 info_dict['__write_download_archive'] = 'ignore'
8222d8de
JMF
2943 return
2944
09b49e1f 2945 # Does nothing under normal operation - for backward compatibility of process_info
277d6ff5 2946 self.post_extract(info_dict)
0c14d66a 2947 self._num_downloads += 1
8222d8de 2948
dcf64d43 2949 # info_dict['_filename'] needs to be set for backward compatibility
de6000d9 2950 info_dict['_filename'] = full_filename = self.prepare_filename(info_dict, warn=True)
2951 temp_filename = self.prepare_filename(info_dict, 'temp')
0202b52a 2952 files_to_move = {}
8222d8de
JMF
2953
2954 # Forced printings
4513a41a 2955 self.__forced_printings(info_dict, full_filename, incomplete=('format' not in info_dict))
8222d8de 2956
ca6d59d2 2957 def check_max_downloads():
2958 if self._num_downloads >= float(self.params.get('max_downloads') or 'inf'):
2959 raise MaxDownloadsReached()
2960
b7b04c78 2961 if self.params.get('simulate'):
9e907ebd 2962 info_dict['__write_download_archive'] = self.params.get('force_write_download_archive')
ca6d59d2 2963 check_max_downloads()
8222d8de
JMF
2964 return
2965
de6000d9 2966 if full_filename is None:
8222d8de 2967 return
e92caff5 2968 if not self._ensure_dir_exists(encodeFilename(full_filename)):
0202b52a 2969 return
e92caff5 2970 if not self._ensure_dir_exists(encodeFilename(temp_filename)):
8222d8de
JMF
2971 return
2972
80c03fa9 2973 if self._write_description('video', info_dict,
2974 self.prepare_filename(info_dict, 'description')) is None:
2975 return
2976
2977 sub_files = self._write_subtitles(info_dict, temp_filename)
2978 if sub_files is None:
2979 return
2980 files_to_move.update(dict(sub_files))
2981
2982 thumb_files = self._write_thumbnails(
2983 'video', info_dict, temp_filename, self.prepare_filename(info_dict, 'thumbnail'))
2984 if thumb_files is None:
2985 return
2986 files_to_move.update(dict(thumb_files))
8222d8de 2987
80c03fa9 2988 infofn = self.prepare_filename(info_dict, 'infojson')
2989 _infojson_written = self._write_info_json('video', info_dict, infofn)
2990 if _infojson_written:
dac5df5a 2991 info_dict['infojson_filename'] = infofn
e75bb0d6 2992 # For backward compatibility, even though it was a private field
80c03fa9 2993 info_dict['__infojson_filename'] = infofn
2994 elif _infojson_written is None:
2995 return
2996
2997 # Note: Annotations are deprecated
2998 annofn = None
1fb07d10 2999 if self.params.get('writeannotations', False):
de6000d9 3000 annofn = self.prepare_filename(info_dict, 'annotation')
80c03fa9 3001 if annofn:
e92caff5 3002 if not self._ensure_dir_exists(encodeFilename(annofn)):
0202b52a 3003 return
0c3d0f51 3004 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(annofn)):
6febd1c1 3005 self.to_screen('[info] Video annotations are already present')
ffddb112
RA
3006 elif not info_dict.get('annotations'):
3007 self.report_warning('There are no annotations to write.')
7b6fefc9
PH
3008 else:
3009 try:
6febd1c1 3010 self.to_screen('[info] Writing video annotations to: ' + annofn)
86e5f3ed 3011 with open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
7b6fefc9
PH
3012 annofile.write(info_dict['annotations'])
3013 except (KeyError, TypeError):
6febd1c1 3014 self.report_warning('There are no annotations to write.')
86e5f3ed 3015 except OSError:
6febd1c1 3016 self.report_error('Cannot write annotations file: ' + annofn)
7b6fefc9 3017 return
1fb07d10 3018
732044af 3019 # Write internet shortcut files
08438d2c 3020 def _write_link_file(link_type):
60f3e995 3021 url = try_get(info_dict['webpage_url'], iri_to_uri)
3022 if not url:
3023 self.report_warning(
3024 f'Cannot write internet shortcut file because the actual URL of "{info_dict["webpage_url"]}" is unknown')
3025 return True
08438d2c 3026 linkfn = replace_extension(self.prepare_filename(info_dict, 'link'), link_type, info_dict.get('ext'))
0e6b018a
Z
3027 if not self._ensure_dir_exists(encodeFilename(linkfn)):
3028 return False
10e3742e 3029 if self.params.get('overwrites', True) and os.path.exists(encodeFilename(linkfn)):
08438d2c 3030 self.to_screen(f'[info] Internet shortcut (.{link_type}) is already present')
3031 return True
3032 try:
3033 self.to_screen(f'[info] Writing internet shortcut (.{link_type}) to: {linkfn}')
86e5f3ed 3034 with open(encodeFilename(to_high_limit_path(linkfn)), 'w', encoding='utf-8',
3035 newline='\r\n' if link_type == 'url' else '\n') as linkfile:
60f3e995 3036 template_vars = {'url': url}
08438d2c 3037 if link_type == 'desktop':
3038 template_vars['filename'] = linkfn[:-(len(link_type) + 1)]
3039 linkfile.write(LINK_TEMPLATES[link_type] % template_vars)
86e5f3ed 3040 except OSError:
08438d2c 3041 self.report_error(f'Cannot write internet shortcut {linkfn}')
3042 return False
732044af 3043 return True
3044
08438d2c 3045 write_links = {
3046 'url': self.params.get('writeurllink'),
3047 'webloc': self.params.get('writewebloclink'),
3048 'desktop': self.params.get('writedesktoplink'),
3049 }
3050 if self.params.get('writelink'):
3051 link_type = ('webloc' if sys.platform == 'darwin'
3052 else 'desktop' if sys.platform.startswith('linux')
3053 else 'url')
3054 write_links[link_type] = True
3055
3056 if any(should_write and not _write_link_file(link_type)
3057 for link_type, should_write in write_links.items()):
3058 return
732044af 3059
f46e2f9d 3060 def replace_info_dict(new_info):
3061 nonlocal info_dict
3062 if new_info == info_dict:
3063 return
3064 info_dict.clear()
3065 info_dict.update(new_info)
3066
415f8d51 3067 new_info, files_to_move = self.pre_process(info_dict, 'before_dl', files_to_move)
3068 replace_info_dict(new_info)
56d868db 3069
a13e6848 3070 if self.params.get('skip_download'):
56d868db 3071 info_dict['filepath'] = temp_filename
3072 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
3073 info_dict['__files_to_move'] = files_to_move
f46e2f9d 3074 replace_info_dict(self.run_pp(MoveFilesAfterDownloadPP(self, False), info_dict))
9e907ebd 3075 info_dict['__write_download_archive'] = self.params.get('force_write_download_archive')
56d868db 3076 else:
3077 # Download
b868936c 3078 info_dict.setdefault('__postprocessors', [])
4340deca 3079 try:
0202b52a 3080
e04938ab 3081 def existing_video_file(*filepaths):
6b591b29 3082 ext = info_dict.get('ext')
e04938ab 3083 converted = lambda file: replace_extension(file, self.params.get('final_ext') or ext, ext)
3084 file = self.existing_file(itertools.chain(*zip(map(converted, filepaths), filepaths)),
3085 default_overwrite=False)
3086 if file:
3087 info_dict['ext'] = os.path.splitext(file)[1][1:]
3088 return file
0202b52a 3089
7b2c3f47 3090 fd, success = None, True
fccf90e7 3091 if info_dict.get('protocol') or info_dict.get('url'):
56ba69e4 3092 fd = get_suitable_downloader(info_dict, self.params, to_stdout=temp_filename == '-')
3093 if fd is not FFmpegFD and (
3094 info_dict.get('section_start') or info_dict.get('section_end')):
7b2c3f47 3095 msg = ('This format cannot be partially downloaded' if FFmpegFD.available()
56ba69e4 3096 else 'You have requested downloading the video partially, but ffmpeg is not installed')
3097 self.report_error(f'{msg}. Aborting')
5ec1b6b7 3098 return
5ec1b6b7 3099
4340deca 3100 if info_dict.get('requested_formats') is not None:
81cd954a 3101 requested_formats = info_dict['requested_formats']
0202b52a 3102 old_ext = info_dict['ext']
4e3b637d 3103 if self.params.get('merge_output_format') is None:
4e3b637d 3104 if (info_dict['ext'] == 'webm'
3105 and info_dict.get('thumbnails')
3106 # check with type instead of pp_key, __name__, or isinstance
3107 # since we dont want any custom PPs to trigger this
c487cf00 3108 and any(type(pp) == EmbedThumbnailPP for pp in self._pps['post_process'])): # noqa: E721
4e3b637d 3109 info_dict['ext'] = 'mkv'
3110 self.report_warning(
3111 'webm doesn\'t support embedding a thumbnail, mkv will be used')
124bc071 3112 new_ext = info_dict['ext']
0202b52a 3113
124bc071 3114 def correct_ext(filename, ext=new_ext):
96fccc10 3115 if filename == '-':
3116 return filename
0202b52a 3117 filename_real_ext = os.path.splitext(filename)[1][1:]
3118 filename_wo_ext = (
3119 os.path.splitext(filename)[0]
124bc071 3120 if filename_real_ext in (old_ext, new_ext)
0202b52a 3121 else filename)
86e5f3ed 3122 return f'{filename_wo_ext}.{ext}'
0202b52a 3123
38c6902b 3124 # Ensure filename always has a correct extension for successful merge
0202b52a 3125 full_filename = correct_ext(full_filename)
3126 temp_filename = correct_ext(temp_filename)
e04938ab 3127 dl_filename = existing_video_file(full_filename, temp_filename)
1ea24129 3128 info_dict['__real_download'] = False
18e674b4 3129
7b2c3f47 3130 merger = FFmpegMergerPP(self)
adbc4ec4 3131 downloaded = []
dbf5416a 3132 if dl_filename is not None:
6c7274ec 3133 self.report_file_already_downloaded(dl_filename)
adbc4ec4
THD
3134 elif fd:
3135 for f in requested_formats if fd != FFmpegFD else []:
3136 f['filepath'] = fname = prepend_extension(
3137 correct_ext(temp_filename, info_dict['ext']),
3138 'f%s' % f['format_id'], info_dict['ext'])
3139 downloaded.append(fname)
dbf5416a 3140 info_dict['url'] = '\n'.join(f['url'] for f in requested_formats)
3141 success, real_download = self.dl(temp_filename, info_dict)
3142 info_dict['__real_download'] = real_download
18e674b4 3143 else:
18e674b4 3144 if self.params.get('allow_unplayable_formats'):
3145 self.report_warning(
3146 'You have requested merging of multiple formats '
3147 'while also allowing unplayable formats to be downloaded. '
3148 'The formats won\'t be merged to prevent data corruption.')
3149 elif not merger.available:
e8969bda 3150 msg = 'You have requested merging of multiple formats but ffmpeg is not installed'
3151 if not self.params.get('ignoreerrors'):
3152 self.report_error(f'{msg}. Aborting due to --abort-on-error')
3153 return
3154 self.report_warning(f'{msg}. The formats won\'t be merged')
18e674b4 3155
96fccc10 3156 if temp_filename == '-':
adbc4ec4 3157 reason = ('using a downloader other than ffmpeg' if FFmpegFD.can_merge_formats(info_dict, self.params)
96fccc10 3158 else 'but the formats are incompatible for simultaneous download' if merger.available
3159 else 'but ffmpeg is not installed')
3160 self.report_warning(
3161 f'You have requested downloading multiple formats to stdout {reason}. '
3162 'The formats will be streamed one after the other')
3163 fname = temp_filename
dbf5416a 3164 for f in requested_formats:
3165 new_info = dict(info_dict)
3166 del new_info['requested_formats']
3167 new_info.update(f)
96fccc10 3168 if temp_filename != '-':
124bc071 3169 fname = prepend_extension(
3170 correct_ext(temp_filename, new_info['ext']),
3171 'f%s' % f['format_id'], new_info['ext'])
96fccc10 3172 if not self._ensure_dir_exists(fname):
3173 return
a21e0ab1 3174 f['filepath'] = fname
96fccc10 3175 downloaded.append(fname)
dbf5416a 3176 partial_success, real_download = self.dl(fname, new_info)
3177 info_dict['__real_download'] = info_dict['__real_download'] or real_download
3178 success = success and partial_success
adbc4ec4
THD
3179
3180 if downloaded and merger.available and not self.params.get('allow_unplayable_formats'):
3181 info_dict['__postprocessors'].append(merger)
3182 info_dict['__files_to_merge'] = downloaded
3183 # Even if there were no downloads, it is being merged only now
3184 info_dict['__real_download'] = True
3185 else:
3186 for file in downloaded:
3187 files_to_move[file] = None
4340deca
P
3188 else:
3189 # Just a single file
e04938ab 3190 dl_filename = existing_video_file(full_filename, temp_filename)
6c7274ec 3191 if dl_filename is None or dl_filename == temp_filename:
3192 # dl_filename == temp_filename could mean that the file was partially downloaded with --no-part.
3193 # So we should try to resume the download
e8e73840 3194 success, real_download = self.dl(temp_filename, info_dict)
0202b52a 3195 info_dict['__real_download'] = real_download
6c7274ec 3196 else:
3197 self.report_file_already_downloaded(dl_filename)
0202b52a 3198
0202b52a 3199 dl_filename = dl_filename or temp_filename
c571435f 3200 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
0202b52a 3201
3158150c 3202 except network_exceptions as err:
7960b056 3203 self.report_error('unable to download video data: %s' % error_to_compat_str(err))
4340deca 3204 return
86e5f3ed 3205 except OSError as err:
4340deca
P
3206 raise UnavailableVideoError(err)
3207 except (ContentTooShortError, ) as err:
86e5f3ed 3208 self.report_error(f'content too short (expected {err.expected} bytes and served {err.downloaded})')
4340deca 3209 return
8222d8de 3210
415f8d51 3211 self._raise_pending_errors(info_dict)
de6000d9 3212 if success and full_filename != '-':
f17f8651 3213
fd7cfb64 3214 def fixup():
3215 do_fixup = True
3216 fixup_policy = self.params.get('fixup')
3217 vid = info_dict['id']
3218
3219 if fixup_policy in ('ignore', 'never'):
3220 return
3221 elif fixup_policy == 'warn':
3fe75fdc 3222 do_fixup = 'warn'
f89b3e2d 3223 elif fixup_policy != 'force':
3224 assert fixup_policy in ('detect_or_warn', None)
3225 if not info_dict.get('__real_download'):
3226 do_fixup = False
fd7cfb64 3227
3228 def ffmpeg_fixup(cndn, msg, cls):
3fe75fdc 3229 if not (do_fixup and cndn):
fd7cfb64 3230 return
3fe75fdc 3231 elif do_fixup == 'warn':
fd7cfb64 3232 self.report_warning(f'{vid}: {msg}')
3233 return
3234 pp = cls(self)
3235 if pp.available:
3236 info_dict['__postprocessors'].append(pp)
3237 else:
3238 self.report_warning(f'{vid}: {msg}. Install ffmpeg to fix this automatically')
3239
3240 stretched_ratio = info_dict.get('stretched_ratio')
ca9def71
LNO
3241 ffmpeg_fixup(stretched_ratio not in (1, None),
3242 f'Non-uniform pixel ratio {stretched_ratio}',
3243 FFmpegFixupStretchedPP)
fd7cfb64 3244
993191c0 3245 downloader = get_suitable_downloader(info_dict, self.params) if 'protocol' in info_dict else None
24146491 3246 downloader = downloader.FD_NAME if downloader else None
adbc4ec4 3247
ca9def71
LNO
3248 ext = info_dict.get('ext')
3249 postprocessed_by_ffmpeg = info_dict.get('requested_formats') or any((
3250 isinstance(pp, FFmpegVideoConvertorPP)
3251 and resolve_recode_mapping(ext, pp.mapping)[0] not in (ext, None)
3252 ) for pp in self._pps['post_process'])
3253
3254 if not postprocessed_by_ffmpeg:
3255 ffmpeg_fixup(ext == 'm4a' and info_dict.get('container') == 'm4a_dash',
f2df4071 3256 'writing DASH m4a. Only some players support this container',
3257 FFmpegFixupM4aPP)
24146491 3258 ffmpeg_fixup(downloader == 'hlsnative' and not self.params.get('hls_use_mpegts')
494f5230 3259 or info_dict.get('is_live') and self.params.get('hls_use_mpegts') is None,
adbc4ec4
THD
3260 'Possible MPEG-TS in MP4 container or malformed AAC timestamps',
3261 FFmpegFixupM3u8PP)
3262 ffmpeg_fixup(info_dict.get('is_live') and downloader == 'DashSegmentsFD',
3263 'Possible duplicate MOOV atoms', FFmpegFixupDuplicateMoovPP)
3264
24146491 3265 ffmpeg_fixup(downloader == 'web_socket_fragment', 'Malformed timestamps detected', FFmpegFixupTimestampPP)
3266 ffmpeg_fixup(downloader == 'web_socket_fragment', 'Malformed duration detected', FFmpegFixupDurationPP)
fd7cfb64 3267
3268 fixup()
8222d8de 3269 try:
f46e2f9d 3270 replace_info_dict(self.post_process(dl_filename, info_dict, files_to_move))
af819c21 3271 except PostProcessingError as err:
3272 self.report_error('Postprocessing: %s' % str(err))
8222d8de 3273 return
ab8e5e51
AM
3274 try:
3275 for ph in self._post_hooks:
23c1a667 3276 ph(info_dict['filepath'])
ab8e5e51
AM
3277 except Exception as err:
3278 self.report_error('post hooks: %s' % str(err))
3279 return
9e907ebd 3280 info_dict['__write_download_archive'] = True
2d30509f 3281
c487cf00 3282 assert info_dict is original_infodict # Make sure the info_dict was modified in-place
a13e6848 3283 if self.params.get('force_write_download_archive'):
9e907ebd 3284 info_dict['__write_download_archive'] = True
ca6d59d2 3285 check_max_downloads()
8222d8de 3286
aa9369a2 3287 def __download_wrapper(self, func):
3288 @functools.wraps(func)
3289 def wrapper(*args, **kwargs):
3290 try:
3291 res = func(*args, **kwargs)
3292 except UnavailableVideoError as e:
3293 self.report_error(e)
b222c271 3294 except DownloadCancelled as e:
3295 self.to_screen(f'[info] {e}')
3296 if not self.params.get('break_per_url'):
3297 raise
fd404bec 3298 self._num_downloads = 0
aa9369a2 3299 else:
3300 if self.params.get('dump_single_json', False):
3301 self.post_extract(res)
3302 self.to_stdout(json.dumps(self.sanitize_info(res)))
3303 return wrapper
3304
8222d8de
JMF
3305 def download(self, url_list):
3306 """Download a given list of URLs."""
aa9369a2 3307 url_list = variadic(url_list) # Passing a single URL is a common mistake
bf1824b3 3308 outtmpl = self.params['outtmpl']['default']
3089bc74
S
3309 if (len(url_list) > 1
3310 and outtmpl != '-'
3311 and '%' not in outtmpl
3312 and self.params.get('max_downloads') != 1):
acd69589 3313 raise SameFileError(outtmpl)
8222d8de
JMF
3314
3315 for url in url_list:
aa9369a2 3316 self.__download_wrapper(self.extract_info)(
3317 url, force_generic_extractor=self.params.get('force_generic_extractor', False))
8222d8de
JMF
3318
3319 return self._download_retcode
3320
1dcc4c0c 3321 def download_with_info_file(self, info_filename):
31bd3925
JMF
3322 with contextlib.closing(fileinput.FileInput(
3323 [info_filename], mode='r',
3324 openhook=fileinput.hook_encoded('utf-8'))) as f:
3325 # FileInput doesn't have a read method, we can't call json.load
8012d892 3326 info = self.sanitize_info(json.loads('\n'.join(f)), self.params.get('clean_infojson', True))
d4943898 3327 try:
aa9369a2 3328 self.__download_wrapper(self.process_ie_result)(info, download=True)
f2ebc5c7 3329 except (DownloadError, EntryNotInPlaylist, ReExtractInfo) as e:
bf5f605e 3330 if not isinstance(e, EntryNotInPlaylist):
3331 self.to_stderr('\r')
d4943898
JMF
3332 webpage_url = info.get('webpage_url')
3333 if webpage_url is not None:
aa9369a2 3334 self.report_warning(f'The info failed to download: {e}; trying with URL {webpage_url}')
d4943898
JMF
3335 return self.download([webpage_url])
3336 else:
3337 raise
3338 return self._download_retcode
1dcc4c0c 3339
cb202fd2 3340 @staticmethod
8012d892 3341 def sanitize_info(info_dict, remove_private_keys=False):
3342 ''' Sanitize the infodict for converting to json '''
3ad56b42 3343 if info_dict is None:
3344 return info_dict
6e84b215 3345 info_dict.setdefault('epoch', int(time.time()))
6a5a30f9 3346 info_dict.setdefault('_type', 'video')
b5e7a2e6 3347 info_dict.setdefault('_version', {
3348 'version': __version__,
3349 'current_git_head': current_git_head(),
3350 'release_git_head': RELEASE_GIT_HEAD,
3351 'repository': REPOSITORY,
3352 })
09b49e1f 3353
8012d892 3354 if remove_private_keys:
0a5a191a 3355 reject = lambda k, v: v is None or k.startswith('__') or k in {
f46e2f9d 3356 'requested_downloads', 'requested_formats', 'requested_subtitles', 'requested_entries',
0a5a191a 3357 'entries', 'filepath', '_filename', 'infojson_filename', 'original_url', 'playlist_autonumber',
6e84b215 3358 }
ae8f99e6 3359 else:
09b49e1f 3360 reject = lambda k, v: False
adbc4ec4
THD
3361
3362 def filter_fn(obj):
3363 if isinstance(obj, dict):
3364 return {k: filter_fn(v) for k, v in obj.items() if not reject(k, v)}
3365 elif isinstance(obj, (list, tuple, set, LazyList)):
3366 return list(map(filter_fn, obj))
3367 elif obj is None or isinstance(obj, (str, int, float, bool)):
3368 return obj
3369 else:
3370 return repr(obj)
3371
5226731e 3372 return filter_fn(info_dict)
cb202fd2 3373
8012d892 3374 @staticmethod
3375 def filter_requested_info(info_dict, actually_filter=True):
3376 ''' Alias of sanitize_info for backward compatibility '''
3377 return YoutubeDL.sanitize_info(info_dict, actually_filter)
3378
43d7f5a5 3379 def _delete_downloaded_files(self, *files_to_delete, info={}, msg=None):
3380 for filename in set(filter(None, files_to_delete)):
3381 if msg:
3382 self.to_screen(msg % filename)
3383 try:
3384 os.remove(filename)
3385 except OSError:
3386 self.report_warning(f'Unable to delete file {filename}')
3387 if filename in info.get('__files_to_move', []): # NB: Delete even if None
3388 del info['__files_to_move'][filename]
3389
ed5835b4 3390 @staticmethod
3391 def post_extract(info_dict):
3392 def actual_post_extract(info_dict):
3393 if info_dict.get('_type') in ('playlist', 'multi_video'):
3394 for video_dict in info_dict.get('entries', {}):
3395 actual_post_extract(video_dict or {})
3396 return
3397
09b49e1f 3398 post_extractor = info_dict.pop('__post_extractor', None) or (lambda: {})
3399 info_dict.update(post_extractor())
ed5835b4 3400
3401 actual_post_extract(info_dict or {})
3402
dcf64d43 3403 def run_pp(self, pp, infodict):
5bfa4862 3404 files_to_delete = []
dcf64d43 3405 if '__files_to_move' not in infodict:
3406 infodict['__files_to_move'] = {}
b1940459 3407 try:
3408 files_to_delete, infodict = pp.run(infodict)
3409 except PostProcessingError as e:
3410 # Must be True and not 'only_download'
3411 if self.params.get('ignoreerrors') is True:
3412 self.report_error(e)
3413 return infodict
3414 raise
3415
5bfa4862 3416 if not files_to_delete:
dcf64d43 3417 return infodict
5bfa4862 3418 if self.params.get('keepvideo', False):
3419 for f in files_to_delete:
dcf64d43 3420 infodict['__files_to_move'].setdefault(f, '')
5bfa4862 3421 else:
43d7f5a5 3422 self._delete_downloaded_files(
3423 *files_to_delete, info=infodict, msg='Deleting original file %s (pass -k to keep)')
dcf64d43 3424 return infodict
5bfa4862 3425
ed5835b4 3426 def run_all_pps(self, key, info, *, additional_pps=None):
bb66c247 3427 self._forceprint(key, info)
ed5835b4 3428 for pp in (additional_pps or []) + self._pps[key]:
dc5f409c 3429 info = self.run_pp(pp, info)
ed5835b4 3430 return info
277d6ff5 3431
56d868db 3432 def pre_process(self, ie_info, key='pre_process', files_to_move=None):
5bfa4862 3433 info = dict(ie_info)
56d868db 3434 info['__files_to_move'] = files_to_move or {}
415f8d51 3435 try:
3436 info = self.run_all_pps(key, info)
3437 except PostProcessingError as err:
3438 msg = f'Preprocessing: {err}'
3439 info.setdefault('__pending_error', msg)
3440 self.report_error(msg, is_error=False)
56d868db 3441 return info, info.pop('__files_to_move', None)
5bfa4862 3442
f46e2f9d 3443 def post_process(self, filename, info, files_to_move=None):
8222d8de 3444 """Run all the postprocessors on the given file."""
8222d8de 3445 info['filepath'] = filename
dcf64d43 3446 info['__files_to_move'] = files_to_move or {}
ed5835b4 3447 info = self.run_all_pps('post_process', info, additional_pps=info.get('__postprocessors'))
dcf64d43 3448 info = self.run_pp(MoveFilesAfterDownloadPP(self), info)
3449 del info['__files_to_move']
ed5835b4 3450 return self.run_all_pps('after_move', info)
c1c9a79c 3451
5db07df6 3452 def _make_archive_id(self, info_dict):
e9fef7ee
S
3453 video_id = info_dict.get('id')
3454 if not video_id:
3455 return
5db07df6
PH
3456 # Future-proof against any change in case
3457 # and backwards compatibility with prior versions
e9fef7ee 3458 extractor = info_dict.get('extractor_key') or info_dict.get('ie_key') # key in a playlist
7012b23c 3459 if extractor is None:
1211bb6d
S
3460 url = str_or_none(info_dict.get('url'))
3461 if not url:
3462 return
e9fef7ee 3463 # Try to find matching extractor for the URL and take its ie_key
8b7491c8 3464 for ie_key, ie in self._ies.items():
1211bb6d 3465 if ie.suitable(url):
8b7491c8 3466 extractor = ie_key
e9fef7ee
S
3467 break
3468 else:
3469 return
0647d925 3470 return make_archive_id(extractor, video_id)
5db07df6
PH
3471
3472 def in_download_archive(self, info_dict):
ae103564 3473 if not self.archive:
5db07df6
PH
3474 return False
3475
1e8fe57e 3476 vid_ids = [self._make_archive_id(info_dict)]
c200096c 3477 vid_ids.extend(info_dict.get('_old_archive_ids') or [])
1e8fe57e 3478 return any(id_ in self.archive for id_ in vid_ids)
c1c9a79c
PH
3479
3480 def record_download_archive(self, info_dict):
3481 fn = self.params.get('download_archive')
3482 if fn is None:
3483 return
5db07df6
PH
3484 vid_id = self._make_archive_id(info_dict)
3485 assert vid_id
ae103564 3486
a13e6848 3487 self.write_debug(f'Adding to archive: {vid_id}')
9c935fbc 3488 if is_path_like(fn):
ae103564 3489 with locked_file(fn, 'a', encoding='utf-8') as archive_file:
3490 archive_file.write(vid_id + '\n')
a45e8619 3491 self.archive.add(vid_id)
dd82ffea 3492
8c51aa65 3493 @staticmethod
8abeeb94 3494 def format_resolution(format, default='unknown'):
9359f3d4 3495 if format.get('vcodec') == 'none' and format.get('acodec') != 'none':
fb04e403 3496 return 'audio only'
f49d89ee
PH
3497 if format.get('resolution') is not None:
3498 return format['resolution']
35615307 3499 if format.get('width') and format.get('height'):
ff51ed58 3500 return '%dx%d' % (format['width'], format['height'])
35615307 3501 elif format.get('height'):
ff51ed58 3502 return '%sp' % format['height']
35615307 3503 elif format.get('width'):
ff51ed58 3504 return '%dx?' % format['width']
3505 return default
8c51aa65 3506
8130779d 3507 def _list_format_headers(self, *headers):
3508 if self.params.get('listformats_table', True) is not False:
591bb9d3 3509 return [self._format_out(header, self.Styles.HEADERS) for header in headers]
8130779d 3510 return headers
3511
c57f7757
PH
3512 def _format_note(self, fdict):
3513 res = ''
3514 if fdict.get('ext') in ['f4f', 'f4m']:
f304da8a 3515 res += '(unsupported)'
32f90364
PH
3516 if fdict.get('language'):
3517 if res:
3518 res += ' '
f304da8a 3519 res += '[%s]' % fdict['language']
c57f7757 3520 if fdict.get('format_note') is not None:
f304da8a 3521 if res:
3522 res += ' '
3523 res += fdict['format_note']
c57f7757 3524 if fdict.get('tbr') is not None:
f304da8a 3525 if res:
3526 res += ', '
3527 res += '%4dk' % fdict['tbr']
c57f7757
PH
3528 if fdict.get('container') is not None:
3529 if res:
3530 res += ', '
3531 res += '%s container' % fdict['container']
3089bc74
S
3532 if (fdict.get('vcodec') is not None
3533 and fdict.get('vcodec') != 'none'):
c57f7757
PH
3534 if res:
3535 res += ', '
3536 res += fdict['vcodec']
91c7271a 3537 if fdict.get('vbr') is not None:
c57f7757
PH
3538 res += '@'
3539 elif fdict.get('vbr') is not None and fdict.get('abr') is not None:
3540 res += 'video@'
3541 if fdict.get('vbr') is not None:
3542 res += '%4dk' % fdict['vbr']
fbb21cf5 3543 if fdict.get('fps') is not None:
5d583bdf
S
3544 if res:
3545 res += ', '
3546 res += '%sfps' % fdict['fps']
c57f7757
PH
3547 if fdict.get('acodec') is not None:
3548 if res:
3549 res += ', '
3550 if fdict['acodec'] == 'none':
3551 res += 'video only'
3552 else:
3553 res += '%-5s' % fdict['acodec']
3554 elif fdict.get('abr') is not None:
3555 if res:
3556 res += ', '
3557 res += 'audio'
3558 if fdict.get('abr') is not None:
3559 res += '@%3dk' % fdict['abr']
3560 if fdict.get('asr') is not None:
3561 res += ' (%5dHz)' % fdict['asr']
3562 if fdict.get('filesize') is not None:
3563 if res:
3564 res += ', '
3565 res += format_bytes(fdict['filesize'])
9732d77e
PH
3566 elif fdict.get('filesize_approx') is not None:
3567 if res:
3568 res += ', '
3569 res += '~' + format_bytes(fdict['filesize_approx'])
c57f7757 3570 return res
91c7271a 3571
8130779d 3572 def render_formats_table(self, info_dict):
b69fd25c 3573 if not info_dict.get('formats') and not info_dict.get('url'):
8130779d 3574 return None
b69fd25c 3575
94badb25 3576 formats = info_dict.get('formats', [info_dict])
8130779d 3577 if not self.params.get('listformats_table', True) is not False:
76d321f6 3578 table = [
3579 [
3580 format_field(f, 'format_id'),
3581 format_field(f, 'ext'),
3582 self.format_resolution(f),
8130779d 3583 self._format_note(f)
3584 ] for f in formats if f.get('preference') is None or f['preference'] >= -1000]
3585 return render_table(['format code', 'extension', 'resolution', 'note'], table, extra_gap=1)
3586
d816f61f 3587 def simplified_codec(f, field):
3588 assert field in ('acodec', 'vcodec')
3589 codec = f.get(field, 'unknown')
f5ea4748 3590 if not codec:
3591 return 'unknown'
3592 elif codec != 'none':
d816f61f 3593 return '.'.join(codec.split('.')[:4])
3594
3595 if field == 'vcodec' and f.get('acodec') == 'none':
3596 return 'images'
3597 elif field == 'acodec' and f.get('vcodec') == 'none':
3598 return ''
3599 return self._format_out('audio only' if field == 'vcodec' else 'video only',
3600 self.Styles.SUPPRESS)
3601
591bb9d3 3602 delim = self._format_out('\u2502', self.Styles.DELIM, '|', test_encoding=True)
8130779d 3603 table = [
3604 [
591bb9d3 3605 self._format_out(format_field(f, 'format_id'), self.Styles.ID),
8130779d 3606 format_field(f, 'ext'),
3607 format_field(f, func=self.format_resolution, ignore=('audio only', 'images')),
563e0bf8 3608 format_field(f, 'fps', '\t%d', func=round),
8130779d 3609 format_field(f, 'dynamic_range', '%s', ignore=(None, 'SDR')).replace('HDR', ''),
b8ed0f15 3610 format_field(f, 'audio_channels', '\t%s'),
8130779d 3611 delim,
3612 format_field(f, 'filesize', ' \t%s', func=format_bytes) + format_field(f, 'filesize_approx', '~\t%s', func=format_bytes),
563e0bf8 3613 format_field(f, 'tbr', '\t%dk', func=round),
8130779d 3614 shorten_protocol_name(f.get('protocol', '')),
3615 delim,
d816f61f 3616 simplified_codec(f, 'vcodec'),
563e0bf8 3617 format_field(f, 'vbr', '\t%dk', func=round),
d816f61f 3618 simplified_codec(f, 'acodec'),
563e0bf8 3619 format_field(f, 'abr', '\t%dk', func=round),
ae61d108 3620 format_field(f, 'asr', '\t%s', func=format_decimal_suffix),
8130779d 3621 join_nonempty(
591bb9d3 3622 self._format_out('UNSUPPORTED', 'light red') if f.get('ext') in ('f4f', 'f4m') else None,
8130779d 3623 format_field(f, 'language', '[%s]'),
3624 join_nonempty(format_field(f, 'format_note'),
3625 format_field(f, 'container', ignore=(None, f.get('ext'))),
3626 delim=', '),
3627 delim=' '),
3628 ] for f in formats if f.get('preference') is None or f['preference'] >= -1000]
3629 header_line = self._list_format_headers(
b8ed0f15 3630 'ID', 'EXT', 'RESOLUTION', '\tFPS', 'HDR', 'CH', delim, '\tFILESIZE', '\tTBR', 'PROTO',
8130779d 3631 delim, 'VCODEC', '\tVBR', 'ACODEC', '\tABR', '\tASR', 'MORE INFO')
3632
3633 return render_table(
3634 header_line, table, hide_empty=True,
591bb9d3 3635 delim=self._format_out('\u2500', self.Styles.DELIM, '-', test_encoding=True))
8130779d 3636
3637 def render_thumbnails_table(self, info_dict):
88f23a18 3638 thumbnails = list(info_dict.get('thumbnails') or [])
cfb56d1a 3639 if not thumbnails:
8130779d 3640 return None
3641 return render_table(
ec11a9f4 3642 self._list_format_headers('ID', 'Width', 'Height', 'URL'),
6970b600 3643 [[t.get('id'), t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails])
2412044c 3644
8130779d 3645 def render_subtitles_table(self, video_id, subtitles):
2412044c 3646 def _row(lang, formats):
49c258e1 3647 exts, names = zip(*((f['ext'], f.get('name') or 'unknown') for f in reversed(formats)))
2412044c 3648 if len(set(names)) == 1:
7aee40c1 3649 names = [] if names[0] == 'unknown' else names[:1]
2412044c 3650 return [lang, ', '.join(names), ', '.join(exts)]
3651
8130779d 3652 if not subtitles:
3653 return None
3654 return render_table(
ec11a9f4 3655 self._list_format_headers('Language', 'Name', 'Formats'),
2412044c 3656 [_row(lang, formats) for lang, formats in subtitles.items()],
8130779d 3657 hide_empty=True)
3658
3659 def __list_table(self, video_id, name, func, *args):
3660 table = func(*args)
3661 if not table:
3662 self.to_screen(f'{video_id} has no {name}')
3663 return
3664 self.to_screen(f'[info] Available {name} for {video_id}:')
3665 self.to_stdout(table)
3666
3667 def list_formats(self, info_dict):
3668 self.__list_table(info_dict['id'], 'formats', self.render_formats_table, info_dict)
3669
3670 def list_thumbnails(self, info_dict):
3671 self.__list_table(info_dict['id'], 'thumbnails', self.render_thumbnails_table, info_dict)
3672
3673 def list_subtitles(self, video_id, subtitles, name='subtitles'):
3674 self.__list_table(video_id, name, self.render_subtitles_table, video_id, subtitles)
a504ced0 3675
dca08720
PH
3676 def urlopen(self, req):
3677 """ Start an HTTP download """
f9934b96 3678 if isinstance(req, str):
67dda517 3679 req = sanitized_Request(req)
19a41fc6 3680 return self._opener.open(req, timeout=self._socket_timeout)
dca08720
PH
3681
3682 def print_debug_header(self):
3683 if not self.params.get('verbose'):
3684 return
49a57e70 3685
560738f3 3686 # These imports can be slow. So import them only as needed
3687 from .extractor.extractors import _LAZY_LOADER
3688 from .extractor.extractors import _PLUGIN_CLASSES as plugin_extractors
3689
49a57e70 3690 def get_encoding(stream):
2a938746 3691 ret = str(getattr(stream, 'encoding', 'missing (%s)' % type(stream).__name__))
49a57e70 3692 if not supports_terminal_sequences(stream):
53973b4d 3693 from .utils import WINDOWS_VT_MODE # Must be imported locally
e3c7d495 3694 ret += ' (No VT)' if WINDOWS_VT_MODE is False else ' (No ANSI)'
49a57e70 3695 return ret
3696
591bb9d3 3697 encoding_str = 'Encodings: locale %s, fs %s, pref %s, %s' % (
49a57e70 3698 locale.getpreferredencoding(),
3699 sys.getfilesystemencoding(),
591bb9d3 3700 self.get_encoding(),
3701 ', '.join(
64fa820c 3702 f'{key} {get_encoding(stream)}' for key, stream in self._out_files.items_
591bb9d3 3703 if stream is not None and key != 'console')
3704 )
883d4b1e 3705
3706 logger = self.params.get('logger')
3707 if logger:
3708 write_debug = lambda msg: logger.debug(f'[debug] {msg}')
3709 write_debug(encoding_str)
3710 else:
96565c7e 3711 write_string(f'[debug] {encoding_str}\n', encoding=None)
49a57e70 3712 write_debug = lambda msg: self._write_string(f'[debug] {msg}\n')
734f90bb 3713
4c88ff87 3714 source = detect_variant()
70b23409 3715 if VARIANT not in (None, 'pip'):
3716 source += '*'
36eaf303 3717 write_debug(join_nonempty(
b5e7a2e6 3718 f'{"yt-dlp" if REPOSITORY == "yt-dlp/yt-dlp" else REPOSITORY} version',
3719 __version__,
36eaf303 3720 f'[{RELEASE_GIT_HEAD}]' if RELEASE_GIT_HEAD else '',
3721 '' if source == 'unknown' else f'({source})',
3722 delim=' '))
6e21fdd2 3723 if not _LAZY_LOADER:
3724 if os.environ.get('YTDLP_NO_LAZY_EXTRACTORS'):
49a57e70 3725 write_debug('Lazy loading extractors is forcibly disabled')
6e21fdd2 3726 else:
49a57e70 3727 write_debug('Lazy loading extractors is disabled')
3ae5e797 3728 if plugin_extractors or plugin_postprocessors:
49a57e70 3729 write_debug('Plugins: %s' % [
3ae5e797 3730 '%s%s' % (klass.__name__, '' if klass.__name__ == name else f' as {name}')
3731 for name, klass in itertools.chain(plugin_extractors.items(), plugin_postprocessors.items())])
8a82af35 3732 if self.params['compat_opts']:
3733 write_debug('Compatibility options: %s' % ', '.join(self.params['compat_opts']))
36eaf303 3734
b5e7a2e6 3735 if current_git_head():
3736 write_debug(f'Git HEAD: {current_git_head()}')
b1f94422 3737 write_debug(system_identifier())
d28b5171 3738
8913ef74 3739 exe_versions, ffmpeg_features = FFmpegPostProcessor.get_versions_and_features(self)
3740 ffmpeg_features = {key for key, val in ffmpeg_features.items() if val}
3741 if ffmpeg_features:
19a03940 3742 exe_versions['ffmpeg'] += ' (%s)' % ','.join(sorted(ffmpeg_features))
8913ef74 3743
4c83c967 3744 exe_versions['rtmpdump'] = rtmpdump_version()
feee8d32 3745 exe_versions['phantomjs'] = PhantomJSwrapper._version()
d28b5171 3746 exe_str = ', '.join(
2831b468 3747 f'{exe} {v}' for exe, v in sorted(exe_versions.items()) if v
3748 ) or 'none'
49a57e70 3749 write_debug('exe versions: %s' % exe_str)
dca08720 3750
1d485a1a 3751 from .compat.compat_utils import get_package_info
9b8ee23b 3752 from .dependencies import available_dependencies
3753
3754 write_debug('Optional libraries: %s' % (', '.join(sorted({
1d485a1a 3755 join_nonempty(*get_package_info(m)) for m in available_dependencies.values()
9b8ee23b 3756 })) or 'none'))
2831b468 3757
97ec5bc5 3758 self._setup_opener()
dca08720
PH
3759 proxy_map = {}
3760 for handler in self._opener.handlers:
3761 if hasattr(handler, 'proxies'):
3762 proxy_map.update(handler.proxies)
49a57e70 3763 write_debug(f'Proxy map: {proxy_map}')
dca08720 3764
49a57e70 3765 # Not implemented
3766 if False and self.params.get('call_home'):
0f06bcd7 3767 ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode()
49a57e70 3768 write_debug('Public IP address: %s' % ipaddr)
58b1f00d 3769 latest_version = self.urlopen(
0f06bcd7 3770 'https://yt-dl.org/latest/version').read().decode()
58b1f00d
PH
3771 if version_tuple(latest_version) > version_tuple(__version__):
3772 self.report_warning(
3773 'You are using an outdated version (newest version: %s)! '
3774 'See https://yt-dl.org/update if you need help updating.' %
3775 latest_version)
3776
e344693b 3777 def _setup_opener(self):
97ec5bc5 3778 if hasattr(self, '_opener'):
3779 return
6ad14cab 3780 timeout_val = self.params.get('socket_timeout')
17bddf3e 3781 self._socket_timeout = 20 if timeout_val is None else float(timeout_val)
6ad14cab 3782
982ee69a 3783 opts_cookiesfrombrowser = self.params.get('cookiesfrombrowser')
dca08720
PH
3784 opts_cookiefile = self.params.get('cookiefile')
3785 opts_proxy = self.params.get('proxy')
3786
982ee69a 3787 self.cookiejar = load_cookies(opts_cookiefile, opts_cookiesfrombrowser, self)
dca08720 3788
6a3f4c3f 3789 cookie_processor = YoutubeDLCookieProcessor(self.cookiejar)
dca08720
PH
3790 if opts_proxy is not None:
3791 if opts_proxy == '':
3792 proxies = {}
3793 else:
3794 proxies = {'http': opts_proxy, 'https': opts_proxy}
3795 else:
ac668111 3796 proxies = urllib.request.getproxies()
067aa17e 3797 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
dca08720
PH
3798 if 'http' in proxies and 'https' not in proxies:
3799 proxies['https'] = proxies['http']
91410c9b 3800 proxy_handler = PerRequestProxyHandler(proxies)
a0ddb8a2
PH
3801
3802 debuglevel = 1 if self.params.get('debug_printtraffic') else 0
be4a824d
PH
3803 https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel)
3804 ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel)
fca6dba8 3805 redirect_handler = YoutubeDLRedirectHandler()
f9934b96 3806 data_handler = urllib.request.DataHandler()
6240b0a2
JMF
3807
3808 # When passing our own FileHandler instance, build_opener won't add the
3809 # default FileHandler and allows us to disable the file protocol, which
3810 # can be used for malicious purposes (see
067aa17e 3811 # https://github.com/ytdl-org/youtube-dl/issues/8227)
ac668111 3812 file_handler = urllib.request.FileHandler()
6240b0a2
JMF
3813
3814 def file_open(*args, **kwargs):
ac668111 3815 raise urllib.error.URLError('file:// scheme is explicitly disabled in yt-dlp for security reasons')
6240b0a2
JMF
3816 file_handler.file_open = file_open
3817
ac668111 3818 opener = urllib.request.build_opener(
fca6dba8 3819 proxy_handler, https_handler, cookie_processor, ydlh, redirect_handler, data_handler, file_handler)
2461f79d 3820
dca08720
PH
3821 # Delete the default user-agent header, which would otherwise apply in
3822 # cases where our custom HTTP handler doesn't come into play
067aa17e 3823 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
dca08720
PH
3824 opener.addheaders = []
3825 self._opener = opener
62fec3b2
PH
3826
3827 def encode(self, s):
3828 if isinstance(s, bytes):
3829 return s # Already encoded
3830
3831 try:
3832 return s.encode(self.get_encoding())
3833 except UnicodeEncodeError as err:
3834 err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.'
3835 raise
3836
3837 def get_encoding(self):
3838 encoding = self.params.get('encoding')
3839 if encoding is None:
3840 encoding = preferredencoding()
3841 return encoding
ec82d85a 3842
e08a85d8 3843 def _write_info_json(self, label, ie_result, infofn, overwrite=None):
cb96c5be 3844 ''' Write infojson and returns True = written, 'exists' = Already exists, False = skip, None = error '''
e08a85d8 3845 if overwrite is None:
3846 overwrite = self.params.get('overwrites', True)
80c03fa9 3847 if not self.params.get('writeinfojson'):
3848 return False
3849 elif not infofn:
3850 self.write_debug(f'Skipping writing {label} infojson')
3851 return False
3852 elif not self._ensure_dir_exists(infofn):
3853 return None
e08a85d8 3854 elif not overwrite and os.path.exists(infofn):
80c03fa9 3855 self.to_screen(f'[info] {label.title()} metadata is already present')
cb96c5be 3856 return 'exists'
3857
3858 self.to_screen(f'[info] Writing {label} metadata as JSON to: {infofn}')
3859 try:
3860 write_json_file(self.sanitize_info(ie_result, self.params.get('clean_infojson', True)), infofn)
3861 return True
86e5f3ed 3862 except OSError:
cb96c5be 3863 self.report_error(f'Cannot write {label} metadata to JSON file {infofn}')
3864 return None
80c03fa9 3865
3866 def _write_description(self, label, ie_result, descfn):
3867 ''' Write description and returns True = written, False = skip, None = error '''
3868 if not self.params.get('writedescription'):
3869 return False
3870 elif not descfn:
3871 self.write_debug(f'Skipping writing {label} description')
3872 return False
3873 elif not self._ensure_dir_exists(descfn):
3874 return None
3875 elif not self.params.get('overwrites', True) and os.path.exists(descfn):
3876 self.to_screen(f'[info] {label.title()} description is already present')
3877 elif ie_result.get('description') is None:
3878 self.report_warning(f'There\'s no {label} description to write')
3879 return False
3880 else:
3881 try:
3882 self.to_screen(f'[info] Writing {label} description to: {descfn}')
86e5f3ed 3883 with open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
80c03fa9 3884 descfile.write(ie_result['description'])
86e5f3ed 3885 except OSError:
80c03fa9 3886 self.report_error(f'Cannot write {label} description file {descfn}')
3887 return None
3888 return True
3889
3890 def _write_subtitles(self, info_dict, filename):
3891 ''' Write subtitles to file and return list of (sub_filename, final_sub_filename); or None if error'''
3892 ret = []
3893 subtitles = info_dict.get('requested_subtitles')
3894 if not subtitles or not (self.params.get('writesubtitles') or self.params.get('writeautomaticsub')):
3895 # subtitles download errors are already managed as troubles in relevant IE
3896 # that way it will silently go on when used with unsupporting IE
3897 return ret
3898
3899 sub_filename_base = self.prepare_filename(info_dict, 'subtitle')
3900 if not sub_filename_base:
3901 self.to_screen('[info] Skipping writing video subtitles')
3902 return ret
3903 for sub_lang, sub_info in subtitles.items():
3904 sub_format = sub_info['ext']
3905 sub_filename = subtitles_filename(filename, sub_lang, sub_format, info_dict.get('ext'))
3906 sub_filename_final = subtitles_filename(sub_filename_base, sub_lang, sub_format, info_dict.get('ext'))
e04938ab 3907 existing_sub = self.existing_file((sub_filename_final, sub_filename))
3908 if existing_sub:
80c03fa9 3909 self.to_screen(f'[info] Video subtitle {sub_lang}.{sub_format} is already present')
e04938ab 3910 sub_info['filepath'] = existing_sub
3911 ret.append((existing_sub, sub_filename_final))
80c03fa9 3912 continue
3913
3914 self.to_screen(f'[info] Writing video subtitles to: {sub_filename}')
3915 if sub_info.get('data') is not None:
3916 try:
3917 # Use newline='' to prevent conversion of newline characters
3918 # See https://github.com/ytdl-org/youtube-dl/issues/10268
86e5f3ed 3919 with open(sub_filename, 'w', encoding='utf-8', newline='') as subfile:
80c03fa9 3920 subfile.write(sub_info['data'])
3921 sub_info['filepath'] = sub_filename
3922 ret.append((sub_filename, sub_filename_final))
3923 continue
86e5f3ed 3924 except OSError:
80c03fa9 3925 self.report_error(f'Cannot write video subtitles file {sub_filename}')
3926 return None
3927
3928 try:
3929 sub_copy = sub_info.copy()
3930 sub_copy.setdefault('http_headers', info_dict.get('http_headers'))
3931 self.dl(sub_filename, sub_copy, subtitle=True)
3932 sub_info['filepath'] = sub_filename
3933 ret.append((sub_filename, sub_filename_final))
6020e05d 3934 except (DownloadError, ExtractorError, IOError, OSError, ValueError) + network_exceptions as err:
c70c418d 3935 msg = f'Unable to download video subtitles for {sub_lang!r}: {err}'
6020e05d 3936 if self.params.get('ignoreerrors') is not True: # False or 'only_download'
c70c418d 3937 if not self.params.get('ignoreerrors'):
3938 self.report_error(msg)
3939 raise DownloadError(msg)
3940 self.report_warning(msg)
519804a9 3941 return ret
80c03fa9 3942
3943 def _write_thumbnails(self, label, info_dict, filename, thumb_filename_base=None):
3944 ''' Write thumbnails to file and return list of (thumb_filename, final_thumb_filename) '''
6c4fd172 3945 write_all = self.params.get('write_all_thumbnails', False)
80c03fa9 3946 thumbnails, ret = [], []
6c4fd172 3947 if write_all or self.params.get('writethumbnail', False):
0202b52a 3948 thumbnails = info_dict.get('thumbnails') or []
6c4fd172 3949 multiple = write_all and len(thumbnails) > 1
ec82d85a 3950
80c03fa9 3951 if thumb_filename_base is None:
3952 thumb_filename_base = filename
3953 if thumbnails and not thumb_filename_base:
3954 self.write_debug(f'Skipping writing {label} thumbnail')
3955 return ret
3956
dd0228ce 3957 for idx, t in list(enumerate(thumbnails))[::-1]:
80c03fa9 3958 thumb_ext = (f'{t["id"]}.' if multiple else '') + determine_ext(t['url'], 'jpg')
aa9369a2 3959 thumb_display_id = f'{label} thumbnail {t["id"]}'
80c03fa9 3960 thumb_filename = replace_extension(filename, thumb_ext, info_dict.get('ext'))
3961 thumb_filename_final = replace_extension(thumb_filename_base, thumb_ext, info_dict.get('ext'))
ec82d85a 3962
e04938ab 3963 existing_thumb = self.existing_file((thumb_filename_final, thumb_filename))
3964 if existing_thumb:
aa9369a2 3965 self.to_screen('[info] %s is already present' % (
3966 thumb_display_id if multiple else f'{label} thumbnail').capitalize())
e04938ab 3967 t['filepath'] = existing_thumb
3968 ret.append((existing_thumb, thumb_filename_final))
ec82d85a 3969 else:
80c03fa9 3970 self.to_screen(f'[info] Downloading {thumb_display_id} ...')
ec82d85a 3971 try:
297e9952 3972 uf = self.urlopen(sanitized_Request(t['url'], headers=t.get('http_headers', {})))
80c03fa9 3973 self.to_screen(f'[info] Writing {thumb_display_id} to: {thumb_filename}')
d3d89c32 3974 with open(encodeFilename(thumb_filename), 'wb') as thumbf:
ec82d85a 3975 shutil.copyfileobj(uf, thumbf)
80c03fa9 3976 ret.append((thumb_filename, thumb_filename_final))
885cc0b7 3977 t['filepath'] = thumb_filename
3158150c 3978 except network_exceptions as err:
dd0228ce 3979 thumbnails.pop(idx)
80c03fa9 3980 self.report_warning(f'Unable to download {thumb_display_id}: {err}')
6c4fd172 3981 if ret and not write_all:
3982 break
0202b52a 3983 return ret