]> jfr.im git - yt-dlp.git/blame - yt_dlp/YoutubeDL.py
[jsinterp] Some optimizations and refactoring
[yt-dlp.git] / yt_dlp / YoutubeDL.py
CommitLineData
cc52de43 1#!/usr/bin/env python3
26e63931 2import collections
31bd3925 3import contextlib
9d2ecdbc 4import datetime
c1c9a79c 5import errno
31bd3925 6import fileinput
b5ae35ee 7import functools
8222d8de 8import io
b82f815f 9import itertools
8694c600 10import json
62fec3b2 11import locale
083c9df9 12import operator
8222d8de 13import os
dca08720 14import platform
f8271158 15import random
8222d8de
JMF
16import re
17import shutil
dca08720 18import subprocess
8222d8de 19import sys
21cd8fae 20import tempfile
8222d8de 21import time
67134eab 22import tokenize
8222d8de 23import traceback
524e2e4f 24import unicodedata
f9934b96 25import urllib.request
961ea474
S
26from string import ascii_letters
27
f8271158 28from .cache import Cache
8c25f81b 29from .compat import (
8a82af35 30 HAS_LEGACY as compat_has_legacy,
003c69a8 31 compat_get_terminal_size,
e9c0cdd3 32 compat_os_name,
7d1eb38a 33 compat_shlex_quote,
ce02ed60
PH
34 compat_str,
35 compat_urllib_error,
36 compat_urllib_request,
8c25f81b 37)
982ee69a 38from .cookies import load_cookies
f8271158 39from .downloader import FFmpegFD, get_suitable_downloader, shorten_protocol_name
40from .downloader.rtmp import rtmpdump_version
f8271158 41from .extractor import gen_extractor_classes, get_info_extractor
42from .extractor.openload import PhantomJSwrapper
43from .minicurses import format_text
44from .postprocessor import _PLUGIN_CLASSES as plugin_postprocessors
45from .postprocessor import (
46 EmbedThumbnailPP,
47 FFmpegFixupDuplicateMoovPP,
48 FFmpegFixupDurationPP,
49 FFmpegFixupM3u8PP,
50 FFmpegFixupM4aPP,
51 FFmpegFixupStretchedPP,
52 FFmpegFixupTimestampPP,
53 FFmpegMergerPP,
54 FFmpegPostProcessor,
55 MoveFilesAfterDownloadPP,
56 get_postprocessor,
57)
58from .update import detect_variant
8c25f81b 59from .utils import (
f8271158 60 DEFAULT_OUTTMPL,
7b2c3f47 61 IDENTITY,
f8271158 62 LINK_TEMPLATES,
63 NO_DEFAULT,
1d485a1a 64 NUMBER_RE,
f8271158 65 OUTTMPL_TYPES,
66 POSTPROCESS_WHEN,
67 STR_FORMAT_RE_TMPL,
68 STR_FORMAT_TYPES,
69 ContentTooShortError,
70 DateRange,
71 DownloadCancelled,
72 DownloadError,
73 EntryNotInPlaylist,
74 ExistingVideoReached,
75 ExtractorError,
76 GeoRestrictedError,
77 HEADRequest,
f8271158 78 ISO3166Utils,
79 LazyList,
80 MaxDownloadsReached,
19a03940 81 Namespace,
f8271158 82 PagedList,
83 PerRequestProxyHandler,
7e88d7d7 84 PlaylistEntries,
f8271158 85 Popen,
86 PostProcessingError,
87 ReExtractInfo,
88 RejectedVideoReached,
89 SameFileError,
90 UnavailableVideoError,
91 YoutubeDLCookieProcessor,
92 YoutubeDLHandler,
93 YoutubeDLRedirectHandler,
eedb7ba5
S
94 age_restricted,
95 args_to_str,
ce02ed60 96 date_from_str,
ce02ed60 97 determine_ext,
b5559424 98 determine_protocol,
c0384f22 99 encode_compat_str,
ce02ed60 100 encodeFilename,
a06916d9 101 error_to_compat_str,
590bc6f6 102 expand_path,
90137ca4 103 filter_dict,
e29663c6 104 float_or_none,
02dbf93f 105 format_bytes,
e0fd9573 106 format_decimal_suffix,
f8271158 107 format_field,
525ef922 108 formatSeconds,
0bb322b9 109 get_domain,
c9969434 110 int_or_none,
732044af 111 iri_to_uri,
34921b43 112 join_nonempty,
ce02ed60 113 locked_file,
0202b52a 114 make_dir,
dca08720 115 make_HTTPS_handler,
8b7539d2 116 merge_headers,
3158150c 117 network_exceptions,
ec11a9f4 118 number_of_digits,
cd6fc19e 119 orderedSet,
083c9df9 120 parse_filesize,
dca08720 121 platform_name,
ce02ed60 122 preferredencoding,
eedb7ba5 123 prepend_extension,
51fb4995 124 register_socks_protocols,
3efb96a6 125 remove_terminal_sequences,
cfb56d1a 126 render_table,
eedb7ba5 127 replace_extension,
ce02ed60 128 sanitize_filename,
1bb5c511 129 sanitize_path,
dcf77cf1 130 sanitize_url,
67dda517 131 sanitized_Request,
e5660ee6 132 std_headers,
1211bb6d 133 str_or_none,
e29663c6 134 strftime_or_none,
ce02ed60 135 subtitles_filename,
819e0531 136 supports_terminal_sequences,
f2ebc5c7 137 timetuple_from_msec,
732044af 138 to_high_limit_path,
324ad820 139 traverse_obj,
6033d980 140 try_get,
29eb5174 141 url_basename,
7d1eb38a 142 variadic,
58b1f00d 143 version_tuple,
53973b4d 144 windows_enable_vt_mode,
ce02ed60
PH
145 write_json_file,
146 write_string,
4f026faf 147)
f8271158 148from .version import RELEASE_GIT_HEAD, __version__
8222d8de 149
e9c0cdd3
YCH
150if compat_os_name == 'nt':
151 import ctypes
152
2459b6e1 153
86e5f3ed 154class YoutubeDL:
8222d8de
JMF
155 """YoutubeDL class.
156
157 YoutubeDL objects are the ones responsible of downloading the
158 actual video file and writing it to disk if the user has requested
159 it, among some other tasks. In most cases there should be one per
160 program. As, given a video URL, the downloader doesn't know how to
161 extract all the needed information, task that InfoExtractors do, it
162 has to pass the URL to one of them.
163
164 For this, YoutubeDL objects have a method that allows
165 InfoExtractors to be registered in a given order. When it is passed
166 a URL, the YoutubeDL object handles it to the first InfoExtractor it
167 finds that reports being able to handle it. The InfoExtractor extracts
168 all the information about the video or videos the URL refers to, and
169 YoutubeDL process the extracted information, possibly using a File
170 Downloader to download the video.
171
172 YoutubeDL objects accept a lot of parameters. In order not to saturate
173 the object constructor with arguments, it receives a dictionary of
174 options instead. These options are available through the params
175 attribute for the InfoExtractors to use. The YoutubeDL also
176 registers itself as the downloader in charge for the InfoExtractors
177 that are added to it, so this is a "mutual registration".
178
179 Available options:
180
181 username: Username for authentication purposes.
182 password: Password for authentication purposes.
180940e0 183 videopassword: Password for accessing a video.
1da50aa3
S
184 ap_mso: Adobe Pass multiple-system operator identifier.
185 ap_username: Multiple-system operator account username.
186 ap_password: Multiple-system operator account password.
8222d8de
JMF
187 usenetrc: Use netrc for authentication instead.
188 verbose: Print additional info to stdout.
189 quiet: Do not print messages to stdout.
ad8915b7 190 no_warnings: Do not print out anything for warnings.
bb66c247 191 forceprint: A dict with keys WHEN mapped to a list of templates to
192 print to stdout. The allowed keys are video or any of the
193 items in utils.POSTPROCESS_WHEN.
ca30f449 194 For compatibility, a single list is also accepted
bb66c247 195 print_to_file: A dict with keys WHEN (same as forceprint) mapped to
196 a list of tuples with (template, filename)
8694c600 197 forcejson: Force printing info_dict as JSON.
63e0be34
PH
198 dump_single_json: Force printing the info_dict of the whole playlist
199 (or video) as a single JSON line.
c25228e5 200 force_write_download_archive: Force writing download archive regardless
201 of 'skip_download' or 'simulate'.
b7b04c78 202 simulate: Do not download the video files. If unset (or None),
203 simulate only if listsubtitles, listformats or list_thumbnails is used
eb8a4433 204 format: Video format code. see "FORMAT SELECTION" for more details.
093a1710 205 You can also pass a function. The function takes 'ctx' as
206 argument and returns the formats to download.
207 See "build_format_selector" for an implementation
63ad4d43 208 allow_unplayable_formats: Allow unplayable formats to be extracted and downloaded.
b7da73eb 209 ignore_no_formats_error: Ignore "No video formats" error. Usefull for
210 extracting metadata even if the video is not actually
211 available for download (experimental)
0930b11f 212 format_sort: A list of fields by which to sort the video formats.
213 See "Sorting Formats" for more details.
c25228e5 214 format_sort_force: Force the given format_sort. see "Sorting Formats"
215 for more details.
08d30158 216 prefer_free_formats: Whether to prefer video formats with free containers
217 over non-free ones of same quality.
c25228e5 218 allow_multiple_video_streams: Allow multiple video streams to be merged
219 into a single file
220 allow_multiple_audio_streams: Allow multiple audio streams to be merged
221 into a single file
0ba692ac 222 check_formats Whether to test if the formats are downloadable.
9f1a1c36 223 Can be True (check all), False (check none),
224 'selected' (check selected formats),
0ba692ac 225 or None (check only if requested by extractor)
4524baf0 226 paths: Dictionary of output paths. The allowed keys are 'home'
227 'temp' and the keys of OUTTMPL_TYPES (in utils.py)
de6000d9 228 outtmpl: Dictionary of templates for output names. Allowed keys
4524baf0 229 are 'default' and the keys of OUTTMPL_TYPES (in utils.py).
34488702 230 For compatibility with youtube-dl, a single string can also be used
a820dc72
RA
231 outtmpl_na_placeholder: Placeholder for unavailable meta fields.
232 restrictfilenames: Do not allow "&" and spaces in file names
233 trim_file_name: Limit length of filename (extension excluded)
4524baf0 234 windowsfilenames: Force the filenames to be windows compatible
b1940459 235 ignoreerrors: Do not stop on download/postprocessing errors.
236 Can be 'only_download' to ignore only download errors.
237 Default is 'only_download' for CLI, but False for API
26e2805c 238 skip_playlist_after_errors: Number of allowed failures until the rest of
239 the playlist is skipped
d22dec74 240 force_generic_extractor: Force downloader to use the generic extractor
0c3d0f51 241 overwrites: Overwrite all video and metadata files if True,
242 overwrite only non-video files if None
243 and don't overwrite any file if False
34488702 244 For compatibility with youtube-dl,
245 "nooverwrites" may also be used instead
c14e88f0 246 playlist_items: Specific indices of playlist to download.
75822ca7 247 playlistrandom: Download playlist items in random order.
7e9a6125 248 lazy_playlist: Process playlist entries as they are received.
8222d8de
JMF
249 matchtitle: Download only matching titles.
250 rejecttitle: Reject downloads for matching titles.
8bf9319e 251 logger: Log messages to a logging.Logger instance.
8222d8de 252 logtostderr: Log messages to stderr instead of stdout.
819e0531 253 consoletitle: Display progress in console window's titlebar.
8222d8de
JMF
254 writedescription: Write the video description to a .description file
255 writeinfojson: Write the video description to a .info.json file
75d43ca0 256 clean_infojson: Remove private fields from the infojson
34488702 257 getcomments: Extract video comments. This will not be written to disk
06167fbb 258 unless writeinfojson is also given
1fb07d10 259 writeannotations: Write the video annotations to a .annotations.xml file
8222d8de 260 writethumbnail: Write the thumbnail image to a file
c25228e5 261 allow_playlist_files: Whether to write playlists' description, infojson etc
262 also to disk when using the 'write*' options
ec82d85a 263 write_all_thumbnails: Write all thumbnail formats to files
732044af 264 writelink: Write an internet shortcut file, depending on the
265 current platform (.url/.webloc/.desktop)
266 writeurllink: Write a Windows internet shortcut file (.url)
267 writewebloclink: Write a macOS internet shortcut file (.webloc)
268 writedesktoplink: Write a Linux internet shortcut file (.desktop)
8222d8de 269 writesubtitles: Write the video subtitles to a file
741dd8ea 270 writeautomaticsub: Write the automatically generated subtitles to a file
8222d8de 271 listsubtitles: Lists all available subtitles for the video
a504ced0 272 subtitlesformat: The format code for subtitles
c32b0aab 273 subtitleslangs: List of languages of the subtitles to download (can be regex).
274 The list may contain "all" to refer to all the available
275 subtitles. The language can be prefixed with a "-" to
276 exclude it from the requested languages. Eg: ['all', '-live_chat']
8222d8de
JMF
277 keepvideo: Keep the video file after post-processing
278 daterange: A DateRange object, download only if the upload_date is in the range.
279 skip_download: Skip the actual download of the video file
c35f9e72 280 cachedir: Location of the cache files in the filesystem.
a0e07d31 281 False to disable filesystem cache.
47192f92 282 noplaylist: Download single video instead of a playlist if in doubt.
8dbe9899
PH
283 age_limit: An integer representing the user's age in years.
284 Unsuitable videos for the given age are skipped.
5fe18bdb
PH
285 min_views: An integer representing the minimum view count the video
286 must have in order to not be skipped.
287 Videos without view count information are always
288 downloaded. None for no limit.
289 max_views: An integer representing the maximum view count.
290 Videos that are more popular than that are not
291 downloaded.
292 Videos without view count information are always
293 downloaded. None for no limit.
294 download_archive: File name of a file where all downloads are recorded.
c1c9a79c
PH
295 Videos already present in the file are not downloaded
296 again.
8a51f564 297 break_on_existing: Stop the download process after attempting to download a
298 file that is in the archive.
299 break_on_reject: Stop the download process when encountering a video that
300 has been filtered out.
b222c271 301 break_per_url: Whether break_on_reject and break_on_existing
302 should act on each input URL as opposed to for the entire queue
d76fa1f3 303 cookiefile: File name or text stream from where cookies should be read and dumped to
f59f5ef8
MB
304 cookiesfrombrowser: A tuple containing the name of the browser, the profile
305 name/pathfrom where cookies are loaded, and the name of the
306 keyring. Eg: ('chrome', ) or ('vivaldi', 'default', 'BASICTEXT')
f81c62a6 307 legacyserverconnect: Explicitly allow HTTPS connection to servers that do not
308 support RFC 5746 secure renegotiation
f59f5ef8 309 nocheckcertificate: Do not verify SSL certificates
bb58c9ed 310 client_certificate: Path to client certificate file in PEM format. May include the private key
311 client_certificate_key: Path to private key file for client certificate
312 client_certificate_password: Password for client certificate private key, if encrypted.
313 If not provided and the key is encrypted, yt-dlp will ask interactively
7e8c0af0
PH
314 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
315 At the moment, this is only supported by YouTube.
8b7539d2 316 http_headers: A dictionary of custom headers to be used for all requests
a1ee09e8 317 proxy: URL of the proxy server to use
38cce791 318 geo_verification_proxy: URL of the proxy to use for IP address verification
504f20dd 319 on geo-restricted sites.
e344693b 320 socket_timeout: Time to wait for unresponsive hosts, in seconds
0783b09b
PH
321 bidi_workaround: Work around buggy terminals without bidirectional text
322 support, using fridibi
a0ddb8a2 323 debug_printtraffic:Print out sent and received HTTP traffic
04b4d394
PH
324 default_search: Prepend this string if an input url is not valid.
325 'auto' for elaborate guessing
62fec3b2 326 encoding: Use this encoding instead of the system-specified.
e8ee972c 327 extract_flat: Do not resolve URLs, return the immediate result.
057a5206
PH
328 Pass in 'in_playlist' to only show this behavior for
329 playlist items.
f2ebc5c7 330 wait_for_video: If given, wait for scheduled streams to become available.
331 The value should be a tuple containing the range
332 (min_secs, max_secs) to wait between retries
4f026faf 333 postprocessors: A list of dictionaries, each with an entry
71b640cc 334 * key: The name of the postprocessor. See
7a5c1cfe 335 yt_dlp/postprocessor/__init__.py for a list.
bb66c247 336 * when: When to run the postprocessor. Allowed values are
337 the entries of utils.POSTPROCESS_WHEN
56d868db 338 Assumed to be 'post_process' if not given
71b640cc
PH
339 progress_hooks: A list of functions that get called on download
340 progress, with a dictionary with the entries
5cda4eda 341 * status: One of "downloading", "error", or "finished".
ee69b99a 342 Check this first and ignore unknown values.
3ba7740d 343 * info_dict: The extracted info_dict
71b640cc 344
5cda4eda 345 If status is one of "downloading", or "finished", the
ee69b99a
PH
346 following properties may also be present:
347 * filename: The final filename (always present)
5cda4eda 348 * tmpfilename: The filename we're currently writing to
71b640cc
PH
349 * downloaded_bytes: Bytes on disk
350 * total_bytes: Size of the whole file, None if unknown
5cda4eda
PH
351 * total_bytes_estimate: Guess of the eventual file size,
352 None if unavailable.
353 * elapsed: The number of seconds since download started.
71b640cc
PH
354 * eta: The estimated time in seconds, None if unknown
355 * speed: The download speed in bytes/second, None if
356 unknown
5cda4eda
PH
357 * fragment_index: The counter of the currently
358 downloaded video fragment.
359 * fragment_count: The number of fragments (= individual
360 files that will be merged)
71b640cc
PH
361
362 Progress hooks are guaranteed to be called at least once
363 (with status "finished") if the download is successful.
819e0531 364 postprocessor_hooks: A list of functions that get called on postprocessing
365 progress, with a dictionary with the entries
366 * status: One of "started", "processing", or "finished".
367 Check this first and ignore unknown values.
368 * postprocessor: Name of the postprocessor
369 * info_dict: The extracted info_dict
370
371 Progress hooks are guaranteed to be called at least twice
372 (with status "started" and "finished") if the processing is successful.
45598f15 373 merge_output_format: Extension to use when merging formats.
6b591b29 374 final_ext: Expected final extension; used to detect when the file was
59a7a13e 375 already downloaded and converted
6271f1ca
PH
376 fixup: Automatically correct known faults of the file.
377 One of:
378 - "never": do nothing
379 - "warn": only emit a warning
380 - "detect_or_warn": check whether we can do anything
62cd676c 381 about it, warn otherwise (default)
504f20dd 382 source_address: Client-side IP address to bind to.
1cf376f5 383 sleep_interval_requests: Number of seconds to sleep between requests
384 during extraction
7aa589a5
S
385 sleep_interval: Number of seconds to sleep before each download when
386 used alone or a lower bound of a range for randomized
387 sleep before each download (minimum possible number
388 of seconds to sleep) when used along with
389 max_sleep_interval.
390 max_sleep_interval:Upper bound of a range for randomized sleep before each
391 download (maximum possible number of seconds to sleep).
392 Must only be used along with sleep_interval.
393 Actual sleep time will be a random float from range
394 [sleep_interval; max_sleep_interval].
1cf376f5 395 sleep_interval_subtitles: Number of seconds to sleep before each subtitle download
cfb56d1a
PH
396 listformats: Print an overview of available video formats and exit.
397 list_thumbnails: Print a table of all thumbnails and exit.
0a41f331 398 match_filter: A function that gets called for every video with the signature
399 (info_dict, *, incomplete: bool) -> Optional[str]
400 For backward compatibility with youtube-dl, the signature
401 (info_dict) -> Optional[str] is also allowed.
402 - If it returns a message, the video is ignored.
403 - If it returns None, the video is downloaded.
404 - If it returns utils.NO_DEFAULT, the user is interactively
405 asked whether to download the video.
347de493 406 match_filter_func in utils.py is one example for this.
7e5db8c9 407 no_color: Do not emit color codes in output.
0a840f58 408 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
504f20dd 409 HTTP header
0a840f58 410 geo_bypass_country:
773f291d
S
411 Two-letter ISO 3166-2 country code that will be used for
412 explicit geographic restriction bypassing via faking
504f20dd 413 X-Forwarded-For HTTP header
5f95927a
S
414 geo_bypass_ip_block:
415 IP range in CIDR notation that will be used similarly to
504f20dd 416 geo_bypass_country
52a8a1e1 417 external_downloader: A dictionary of protocol keys and the executable of the
418 external downloader to use for it. The allowed protocols
419 are default|http|ftp|m3u8|dash|rtsp|rtmp|mms.
420 Set the value to 'native' to use the native downloader
53ed7066 421 compat_opts: Compatibility options. See "Differences in default behavior".
3acf6d38 422 The following options do not work when used through the API:
b5ae35ee 423 filename, abort-on-error, multistreams, no-live-chat, format-sort
dac5df5a 424 no-clean-infojson, no-playlist-metafiles, no-keep-subs, no-attach-info-json.
e4f02757 425 Refer __init__.py for their implementation
819e0531 426 progress_template: Dictionary of templates for progress outputs.
427 Allowed keys are 'download', 'postprocess',
428 'download-title' (console title) and 'postprocess-title'.
429 The template is mapped on a dictionary with keys 'progress' and 'info'
23326151 430 retry_sleep_functions: Dictionary of functions that takes the number of attempts
431 as argument and returns the time to sleep in seconds.
432 Allowed keys are 'http', 'fragment', 'file_access'
5ec1b6b7 433 download_ranges: A function that gets called for every video with the signature
434 (info_dict, *, ydl) -> Iterable[Section].
435 Only the returned sections will be downloaded. Each Section contains:
436 * start_time: Start time of the section in seconds
437 * end_time: End time of the section in seconds
438 * title: Section title (Optional)
439 * index: Section number (Optional)
fe7e0c98 440
8222d8de 441 The following parameters are not used by YoutubeDL itself, they are used by
7a5c1cfe 442 the downloader (see yt_dlp/downloader/common.py):
51d9739f 443 nopart, updatetime, buffersize, ratelimit, throttledratelimit, min_filesize,
205a0654
EH
444 max_filesize, test, noresizebuffer, retries, file_access_retries, fragment_retries,
445 continuedl, noprogress, xattr_set_filesize, hls_use_mpegts, http_chunk_size,
59a7a13e 446 external_downloader_args, concurrent_fragment_downloads.
76b1bd67
JMF
447
448 The following options are used by the post processors:
c0b7d117
S
449 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
450 to the binary or its containing directory.
43820c03 451 postprocessor_args: A dictionary of postprocessor/executable keys (in lower case)
34488702 452 and a list of additional command-line arguments for the
453 postprocessor/executable. The dict can also have "PP+EXE" keys
454 which are used when the given exe is used by the given PP.
455 Use 'default' as the name for arguments to passed to all PP
456 For compatibility with youtube-dl, a single list of args
457 can also be used
e409895f 458
459 The following options are used by the extractors:
62bff2c1 460 extractor_retries: Number of times to retry for known errors
461 dynamic_mpd: Whether to process dynamic DASH manifests (default: True)
e409895f 462 hls_split_discontinuity: Split HLS playlists to different formats at
62bff2c1 463 discontinuities such as ad breaks (default: False)
5d3a0e79 464 extractor_args: A dictionary of arguments to be passed to the extractors.
465 See "EXTRACTOR ARGUMENTS" for details.
466 Eg: {'youtube': {'skip': ['dash', 'hls']}}
88f23a18 467 mark_watched: Mark videos watched (even with --simulate). Only for YouTube
1890fc63 468
469 The following options are deprecated and may be removed in the future:
470
7e9a6125 471 playliststart: - Use playlist_items
472 Playlist item to start at.
473 playlistend: - Use playlist_items
474 Playlist item to end at.
475 playlistreverse: - Use playlist_items
476 Download playlist items in reverse order.
1890fc63 477 forceurl: - Use forceprint
478 Force printing final URL.
479 forcetitle: - Use forceprint
480 Force printing title.
481 forceid: - Use forceprint
482 Force printing ID.
483 forcethumbnail: - Use forceprint
484 Force printing thumbnail URL.
485 forcedescription: - Use forceprint
486 Force printing description.
487 forcefilename: - Use forceprint
488 Force printing final filename.
489 forceduration: - Use forceprint
490 Force printing duration.
491 allsubtitles: - Use subtitleslangs = ['all']
492 Downloads all the subtitles of the video
493 (requires writesubtitles or writeautomaticsub)
494 include_ads: - Doesn't work
495 Download ads as well
496 call_home: - Not implemented
497 Boolean, true iff we are allowed to contact the
498 yt-dlp servers for debugging.
499 post_hooks: - Register a custom postprocessor
500 A list of functions that get called as the final step
501 for each video file, after all postprocessors have been
502 called. The filename will be passed as the only argument.
503 hls_prefer_native: - Use external_downloader = {'m3u8': 'native'} or {'m3u8': 'ffmpeg'}.
504 Use the native HLS downloader instead of ffmpeg/avconv
505 if True, otherwise use ffmpeg/avconv if False, otherwise
506 use downloader suggested by extractor if None.
507 prefer_ffmpeg: - avconv support is deprecated
508 If False, use avconv instead of ffmpeg if both are available,
509 otherwise prefer ffmpeg.
510 youtube_include_dash_manifest: - Use extractor_args
5d3a0e79 511 If True (default), DASH manifests and related
62bff2c1 512 data will be downloaded and processed by extractor.
513 You can reduce network I/O by disabling it if you don't
514 care about DASH. (only for youtube)
1890fc63 515 youtube_include_hls_manifest: - Use extractor_args
5d3a0e79 516 If True (default), HLS manifests and related
62bff2c1 517 data will be downloaded and processed by extractor.
518 You can reduce network I/O by disabling it if you don't
519 care about HLS. (only for youtube)
8222d8de
JMF
520 """
521
86e5f3ed 522 _NUMERIC_FIELDS = {
c9969434 523 'width', 'height', 'tbr', 'abr', 'asr', 'vbr', 'fps', 'filesize', 'filesize_approx',
e6f21b3d 524 'timestamp', 'release_timestamp',
c9969434
S
525 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
526 'average_rating', 'comment_count', 'age_limit',
527 'start_time', 'end_time',
528 'chapter_number', 'season_number', 'episode_number',
529 'track_number', 'disc_number', 'release_year',
86e5f3ed 530 }
c9969434 531
6db9c4d5 532 _format_fields = {
533 # NB: Keep in sync with the docstring of extractor/common.py
a44ca5a4 534 'url', 'manifest_url', 'manifest_stream_number', 'ext', 'format', 'format_id', 'format_note',
6db9c4d5 535 'width', 'height', 'resolution', 'dynamic_range', 'tbr', 'abr', 'acodec', 'asr',
536 'vbr', 'fps', 'vcodec', 'container', 'filesize', 'filesize_approx',
537 'player_url', 'protocol', 'fragment_base_url', 'fragments', 'is_from_start',
538 'preference', 'language', 'language_preference', 'quality', 'source_preference',
539 'http_headers', 'stretched_ratio', 'no_resume', 'has_drm', 'downloader_options',
540 'page_url', 'app', 'play_path', 'tc_url', 'flash_version', 'rtmp_live', 'rtmp_conn', 'rtmp_protocol', 'rtmp_real_time'
541 }
48ee10ee 542 _format_selection_exts = {
543 'audio': {'m4a', 'mp3', 'ogg', 'aac'},
544 'video': {'mp4', 'flv', 'webm', '3gp'},
545 'storyboards': {'mhtml'},
546 }
547
3511266b 548 def __init__(self, params=None, auto_init=True):
883d4b1e 549 """Create a FileDownloader object with the given options.
550 @param auto_init Whether to load the default extractors and print header (if verbose).
49a57e70 551 Set to 'no_verbose_header' to not print the header
883d4b1e 552 """
e9f9a10f
JMF
553 if params is None:
554 params = {}
592b7485 555 self.params = params
8b7491c8 556 self._ies = {}
56c73665 557 self._ies_instances = {}
1e43a6f7 558 self._pps = {k: [] for k in POSTPROCESS_WHEN}
b35496d8 559 self._printed_messages = set()
1cf376f5 560 self._first_webpage_request = True
ab8e5e51 561 self._post_hooks = []
933605d7 562 self._progress_hooks = []
819e0531 563 self._postprocessor_hooks = []
8222d8de
JMF
564 self._download_retcode = 0
565 self._num_downloads = 0
9c906919 566 self._num_videos = 0
592b7485 567 self._playlist_level = 0
568 self._playlist_urls = set()
a0e07d31 569 self.cache = Cache(self)
34308b30 570
819e0531 571 windows_enable_vt_mode()
591bb9d3 572 stdout = sys.stderr if self.params.get('logtostderr') else sys.stdout
573 self._out_files = Namespace(
574 out=stdout,
575 error=sys.stderr,
576 screen=sys.stderr if self.params.get('quiet') else stdout,
577 console=None if compat_os_name == 'nt' else next(
cf4f42cb 578 filter(supports_terminal_sequences, (sys.stderr, sys.stdout)), None)
591bb9d3 579 )
580 self._allow_colors = Namespace(**{
581 type_: not self.params.get('no_color') and supports_terminal_sequences(stream)
64fa820c 582 for type_, stream in self._out_files.items_ if type_ != 'console'
591bb9d3 583 })
819e0531 584
a61f4b28 585 if sys.version_info < (3, 6):
586 self.report_warning(
0181adef 587 'Python version %d.%d is not supported! Please update to Python 3.6 or above' % sys.version_info[:2])
a61f4b28 588
88acdbc2 589 if self.params.get('allow_unplayable_formats'):
590 self.report_warning(
ec11a9f4 591 f'You have asked for {self._format_err("UNPLAYABLE", self.Styles.EMPHASIS)} formats to be listed/downloaded. '
819e0531 592 'This is a developer option intended for debugging. \n'
593 ' If you experience any issues while using this option, '
ec11a9f4 594 f'{self._format_err("DO NOT", self.Styles.ERROR)} open a bug report')
88acdbc2 595
be5df5ee
S
596 def check_deprecated(param, option, suggestion):
597 if self.params.get(param) is not None:
86e5f3ed 598 self.report_warning(f'{option} is deprecated. Use {suggestion} instead')
be5df5ee
S
599 return True
600 return False
601
602 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
38cce791
YCH
603 if self.params.get('geo_verification_proxy') is None:
604 self.params['geo_verification_proxy'] = self.params['cn_verification_proxy']
605
0d1bb027 606 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
607 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
53ed7066 608 check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"')
0d1bb027 609
49a57e70 610 for msg in self.params.get('_warnings', []):
0d1bb027 611 self.report_warning(msg)
ee8dd27a 612 for msg in self.params.get('_deprecation_warnings', []):
613 self.deprecation_warning(msg)
0d1bb027 614
8a82af35 615 self.params['compat_opts'] = set(self.params.get('compat_opts', ()))
616 if not compat_has_legacy:
617 self.params['compat_opts'].add('no-compat-legacy')
618 if 'list-formats' in self.params['compat_opts']:
ec11a9f4 619 self.params['listformats_table'] = False
620
b5ae35ee 621 if 'overwrites' not in self.params and self.params.get('nooverwrites') is not None:
b868936c 622 # nooverwrites was unnecessarily changed to overwrites
623 # in 0c3d0f51778b153f65c21906031c2e091fcfb641
624 # This ensures compatibility with both keys
625 self.params['overwrites'] = not self.params['nooverwrites']
b5ae35ee 626 elif self.params.get('overwrites') is None:
627 self.params.pop('overwrites', None)
b868936c 628 else:
629 self.params['nooverwrites'] = not self.params['overwrites']
b9d973be 630
455a15e2 631 self.params.setdefault('forceprint', {})
632 self.params.setdefault('print_to_file', {})
bb66c247 633
634 # Compatibility with older syntax
ca30f449 635 if not isinstance(params['forceprint'], dict):
455a15e2 636 self.params['forceprint'] = {'video': params['forceprint']}
ca30f449 637
455a15e2 638 if self.params.get('bidi_workaround', False):
1c088fa8
PH
639 try:
640 import pty
641 master, slave = pty.openpty()
003c69a8 642 width = compat_get_terminal_size().columns
591bb9d3 643 width_args = [] if width is None else ['-w', str(width)]
644 sp_kwargs = {'stdin': subprocess.PIPE, 'stdout': slave, 'stderr': self._out_files.error}
5d681e96 645 try:
d3c93ec2 646 self._output_process = Popen(['bidiv'] + width_args, **sp_kwargs)
5d681e96 647 except OSError:
d3c93ec2 648 self._output_process = Popen(['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
5d681e96 649 self._output_channel = os.fdopen(master, 'rb')
1c088fa8 650 except OSError as ose:
66e7ace1 651 if ose.errno == errno.ENOENT:
49a57e70 652 self.report_warning(
653 'Could not find fribidi executable, ignoring --bidi-workaround. '
654 'Make sure that fribidi is an executable file in one of the directories in your $PATH.')
1c088fa8
PH
655 else:
656 raise
0783b09b 657
97ec5bc5 658 if auto_init:
659 if auto_init != 'no_verbose_header':
660 self.print_debug_header()
661 self.add_default_info_extractors()
662
3089bc74
S
663 if (sys.platform != 'win32'
664 and sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
455a15e2 665 and not self.params.get('restrictfilenames', False)):
e9137224 666 # Unicode filesystem API will throw errors (#1474, #13027)
34308b30 667 self.report_warning(
6febd1c1 668 'Assuming --restrict-filenames since file system encoding '
1b725173 669 'cannot encode all characters. '
6febd1c1 670 'Set the LC_ALL environment variable to fix this.')
4a98cdbf 671 self.params['restrictfilenames'] = True
34308b30 672
bf1824b3 673 self._parse_outtmpl()
486dd09e 674
187986a8 675 # Creating format selector here allows us to catch syntax errors before the extraction
676 self.format_selector = (
fa9f30b8 677 self.params.get('format') if self.params.get('format') in (None, '-')
093a1710 678 else self.params['format'] if callable(self.params['format'])
187986a8 679 else self.build_format_selector(self.params['format']))
680
8b7539d2 681 # Set http_headers defaults according to std_headers
682 self.params['http_headers'] = merge_headers(std_headers, self.params.get('http_headers', {}))
683
013b50b7 684 hooks = {
685 'post_hooks': self.add_post_hook,
686 'progress_hooks': self.add_progress_hook,
687 'postprocessor_hooks': self.add_postprocessor_hook,
688 }
689 for opt, fn in hooks.items():
690 for ph in self.params.get(opt, []):
691 fn(ph)
71b640cc 692
5bfc8bee 693 for pp_def_raw in self.params.get('postprocessors', []):
694 pp_def = dict(pp_def_raw)
695 when = pp_def.pop('when', 'post_process')
696 self.add_post_processor(
f9934b96 697 get_postprocessor(pp_def.pop('key'))(self, **pp_def),
5bfc8bee 698 when=when)
699
97ec5bc5 700 self._setup_opener()
51fb4995
YCH
701 register_socks_protocols()
702
ed39cac5 703 def preload_download_archive(fn):
704 """Preload the archive, if any is specified"""
705 if fn is None:
706 return False
49a57e70 707 self.write_debug(f'Loading archive file {fn!r}')
ed39cac5 708 try:
709 with locked_file(fn, 'r', encoding='utf-8') as archive_file:
710 for line in archive_file:
711 self.archive.add(line.strip())
86e5f3ed 712 except OSError as ioe:
ed39cac5 713 if ioe.errno != errno.ENOENT:
714 raise
715 return False
716 return True
717
718 self.archive = set()
719 preload_download_archive(self.params.get('download_archive'))
720
7d4111ed
PH
721 def warn_if_short_id(self, argv):
722 # short YouTube ID starting with dash?
723 idxs = [
724 i for i, a in enumerate(argv)
725 if re.match(r'^-[0-9A-Za-z_-]{10}$', a)]
726 if idxs:
727 correct_argv = (
7a5c1cfe 728 ['yt-dlp']
3089bc74
S
729 + [a for i, a in enumerate(argv) if i not in idxs]
730 + ['--'] + [argv[i] for i in idxs]
7d4111ed
PH
731 )
732 self.report_warning(
733 'Long argument string detected. '
49a57e70 734 'Use -- to separate parameters and URLs, like this:\n%s' %
7d4111ed
PH
735 args_to_str(correct_argv))
736
8222d8de
JMF
737 def add_info_extractor(self, ie):
738 """Add an InfoExtractor object to the end of the list."""
8b7491c8 739 ie_key = ie.ie_key()
740 self._ies[ie_key] = ie
e52d7f85 741 if not isinstance(ie, type):
8b7491c8 742 self._ies_instances[ie_key] = ie
e52d7f85 743 ie.set_downloader(self)
8222d8de 744
8b7491c8 745 def _get_info_extractor_class(self, ie_key):
746 ie = self._ies.get(ie_key)
747 if ie is None:
748 ie = get_info_extractor(ie_key)
749 self.add_info_extractor(ie)
750 return ie
751
56c73665
JMF
752 def get_info_extractor(self, ie_key):
753 """
754 Get an instance of an IE with name ie_key, it will try to get one from
755 the _ies list, if there's no instance it will create a new one and add
756 it to the extractor list.
757 """
758 ie = self._ies_instances.get(ie_key)
759 if ie is None:
760 ie = get_info_extractor(ie_key)()
761 self.add_info_extractor(ie)
762 return ie
763
023fa8c4
JMF
764 def add_default_info_extractors(self):
765 """
766 Add the InfoExtractors returned by gen_extractors to the end of the list
767 """
e52d7f85 768 for ie in gen_extractor_classes():
023fa8c4
JMF
769 self.add_info_extractor(ie)
770
56d868db 771 def add_post_processor(self, pp, when='post_process'):
8222d8de 772 """Add a PostProcessor object to the end of the chain."""
8aa0e7cd 773 assert when in POSTPROCESS_WHEN, f'Invalid when={when}'
5bfa4862 774 self._pps[when].append(pp)
8222d8de
JMF
775 pp.set_downloader(self)
776
ab8e5e51
AM
777 def add_post_hook(self, ph):
778 """Add the post hook"""
779 self._post_hooks.append(ph)
780
933605d7 781 def add_progress_hook(self, ph):
819e0531 782 """Add the download progress hook"""
933605d7 783 self._progress_hooks.append(ph)
8ab470f1 784
819e0531 785 def add_postprocessor_hook(self, ph):
786 """Add the postprocessing progress hook"""
787 self._postprocessor_hooks.append(ph)
5bfc8bee 788 for pps in self._pps.values():
789 for pp in pps:
790 pp.add_progress_hook(ph)
819e0531 791
1c088fa8 792 def _bidi_workaround(self, message):
5d681e96 793 if not hasattr(self, '_output_channel'):
1c088fa8
PH
794 return message
795
5d681e96 796 assert hasattr(self, '_output_process')
11b85ce6 797 assert isinstance(message, compat_str)
6febd1c1 798 line_count = message.count('\n') + 1
0f06bcd7 799 self._output_process.stdin.write((message + '\n').encode())
5d681e96 800 self._output_process.stdin.flush()
0f06bcd7 801 res = ''.join(self._output_channel.readline().decode()
9e1a5b84 802 for _ in range(line_count))
6febd1c1 803 return res[:-len('\n')]
1c088fa8 804
b35496d8 805 def _write_string(self, message, out=None, only_once=False):
806 if only_once:
807 if message in self._printed_messages:
808 return
809 self._printed_messages.add(message)
810 write_string(message, out=out, encoding=self.params.get('encoding'))
734f90bb 811
cf4f42cb 812 def to_stdout(self, message, skip_eol=False, quiet=None):
0760b0a7 813 """Print message to stdout"""
cf4f42cb 814 if quiet is not None:
ae6a1b95 815 self.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument quiet. Use "YoutubeDL.to_screen" instead')
8a82af35 816 if skip_eol is not False:
817 self.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument skip_eol. Use "YoutubeDL.to_screen" instead')
0bf9dc1e 818 self._write_string(f'{self._bidi_workaround(message)}\n', self._out_files.out)
cf4f42cb 819
820 def to_screen(self, message, skip_eol=False, quiet=None):
821 """Print message to screen if not in quiet mode"""
8bf9319e 822 if self.params.get('logger'):
43afe285 823 self.params['logger'].debug(message)
cf4f42cb 824 return
825 if (self.params.get('quiet') if quiet is None else quiet) and not self.params.get('verbose'):
826 return
827 self._write_string(
828 '%s%s' % (self._bidi_workaround(message), ('' if skip_eol else '\n')),
591bb9d3 829 self._out_files.screen)
8222d8de 830
b35496d8 831 def to_stderr(self, message, only_once=False):
0760b0a7 832 """Print message to stderr"""
11b85ce6 833 assert isinstance(message, compat_str)
8bf9319e 834 if self.params.get('logger'):
43afe285
IB
835 self.params['logger'].error(message)
836 else:
5792c950 837 self._write_string(f'{self._bidi_workaround(message)}\n', self._out_files.error, only_once=only_once)
cf4f42cb 838
839 def _send_console_code(self, code):
591bb9d3 840 if compat_os_name == 'nt' or not self._out_files.console:
cf4f42cb 841 return
591bb9d3 842 self._write_string(code, self._out_files.console)
8222d8de 843
1e5b9a95
PH
844 def to_console_title(self, message):
845 if not self.params.get('consoletitle', False):
846 return
3efb96a6 847 message = remove_terminal_sequences(message)
4bede0d8
C
848 if compat_os_name == 'nt':
849 if ctypes.windll.kernel32.GetConsoleWindow():
850 # c_wchar_p() might not be necessary if `message` is
851 # already of type unicode()
852 ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
cf4f42cb 853 else:
854 self._send_console_code(f'\033]0;{message}\007')
1e5b9a95 855
bdde425c 856 def save_console_title(self):
cf4f42cb 857 if not self.params.get('consoletitle') or self.params.get('simulate'):
bdde425c 858 return
592b7485 859 self._send_console_code('\033[22;0t') # Save the title on stack
bdde425c
PH
860
861 def restore_console_title(self):
cf4f42cb 862 if not self.params.get('consoletitle') or self.params.get('simulate'):
bdde425c 863 return
592b7485 864 self._send_console_code('\033[23;0t') # Restore the title from stack
bdde425c
PH
865
866 def __enter__(self):
867 self.save_console_title()
868 return self
869
870 def __exit__(self, *args):
871 self.restore_console_title()
f89197d7 872
dca08720 873 if self.params.get('cookiefile') is not None:
1bab3437 874 self.cookiejar.save(ignore_discard=True, ignore_expires=True)
bdde425c 875
fa9f30b8 876 def trouble(self, message=None, tb=None, is_error=True):
8222d8de
JMF
877 """Determine action to take when a download problem appears.
878
879 Depending on if the downloader has been configured to ignore
880 download errors or not, this method may throw an exception or
881 not when errors are found, after printing the message.
882
fa9f30b8 883 @param tb If given, is additional traceback information
884 @param is_error Whether to raise error according to ignorerrors
8222d8de
JMF
885 """
886 if message is not None:
887 self.to_stderr(message)
888 if self.params.get('verbose'):
889 if tb is None:
890 if sys.exc_info()[0]: # if .trouble has been called from an except block
6febd1c1 891 tb = ''
8222d8de 892 if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
6febd1c1 893 tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
c0384f22 894 tb += encode_compat_str(traceback.format_exc())
8222d8de
JMF
895 else:
896 tb_data = traceback.format_list(traceback.extract_stack())
6febd1c1 897 tb = ''.join(tb_data)
c19bc311 898 if tb:
899 self.to_stderr(tb)
fa9f30b8 900 if not is_error:
901 return
b1940459 902 if not self.params.get('ignoreerrors'):
8222d8de
JMF
903 if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
904 exc_info = sys.exc_info()[1].exc_info
905 else:
906 exc_info = sys.exc_info()
907 raise DownloadError(message, exc_info)
908 self._download_retcode = 1
909
19a03940 910 Styles = Namespace(
911 HEADERS='yellow',
912 EMPHASIS='light blue',
492272fe 913 FILENAME='green',
19a03940 914 ID='green',
915 DELIM='blue',
916 ERROR='red',
917 WARNING='yellow',
918 SUPPRESS='light black',
919 )
ec11a9f4 920
7578d77d 921 def _format_text(self, handle, allow_colors, text, f, fallback=None, *, test_encoding=False):
e5a998f3 922 text = str(text)
ec11a9f4 923 if test_encoding:
924 original_text = text
5c104538 925 # handle.encoding can be None. See https://github.com/yt-dlp/yt-dlp/issues/2711
926 encoding = self.params.get('encoding') or getattr(handle, 'encoding', None) or 'ascii'
ec11a9f4 927 text = text.encode(encoding, 'ignore').decode(encoding)
928 if fallback is not None and text != original_text:
929 text = fallback
7578d77d 930 return format_text(text, f) if allow_colors else text if fallback is None else fallback
ec11a9f4 931
591bb9d3 932 def _format_out(self, *args, **kwargs):
933 return self._format_text(self._out_files.out, self._allow_colors.out, *args, **kwargs)
934
ec11a9f4 935 def _format_screen(self, *args, **kwargs):
591bb9d3 936 return self._format_text(self._out_files.screen, self._allow_colors.screen, *args, **kwargs)
ec11a9f4 937
938 def _format_err(self, *args, **kwargs):
591bb9d3 939 return self._format_text(self._out_files.error, self._allow_colors.error, *args, **kwargs)
819e0531 940
c84aeac6 941 def report_warning(self, message, only_once=False):
8222d8de
JMF
942 '''
943 Print the message to stderr, it will be prefixed with 'WARNING:'
944 If stderr is a tty file the 'WARNING:' will be colored
945 '''
6d07ce01
JMF
946 if self.params.get('logger') is not None:
947 self.params['logger'].warning(message)
8222d8de 948 else:
ad8915b7
PH
949 if self.params.get('no_warnings'):
950 return
ec11a9f4 951 self.to_stderr(f'{self._format_err("WARNING:", self.Styles.WARNING)} {message}', only_once)
8222d8de 952
ee8dd27a 953 def deprecation_warning(self, message):
954 if self.params.get('logger') is not None:
a44ca5a4 955 self.params['logger'].warning(f'DeprecationWarning: {message}')
ee8dd27a 956 else:
957 self.to_stderr(f'{self._format_err("DeprecationWarning:", self.Styles.ERROR)} {message}', True)
958
fa9f30b8 959 def report_error(self, message, *args, **kwargs):
8222d8de
JMF
960 '''
961 Do the same as trouble, but prefixes the message with 'ERROR:', colored
962 in red if stderr is a tty file.
963 '''
fa9f30b8 964 self.trouble(f'{self._format_err("ERROR:", self.Styles.ERROR)} {message}', *args, **kwargs)
8222d8de 965
b35496d8 966 def write_debug(self, message, only_once=False):
0760b0a7 967 '''Log debug message or Print message to stderr'''
968 if not self.params.get('verbose', False):
969 return
8a82af35 970 message = f'[debug] {message}'
0760b0a7 971 if self.params.get('logger'):
972 self.params['logger'].debug(message)
973 else:
b35496d8 974 self.to_stderr(message, only_once)
0760b0a7 975
8222d8de
JMF
976 def report_file_already_downloaded(self, file_name):
977 """Report file has already been fully downloaded."""
978 try:
6febd1c1 979 self.to_screen('[download] %s has already been downloaded' % file_name)
ce02ed60 980 except UnicodeEncodeError:
6febd1c1 981 self.to_screen('[download] The file has already been downloaded')
8222d8de 982
0c3d0f51 983 def report_file_delete(self, file_name):
984 """Report that existing file will be deleted."""
985 try:
c25228e5 986 self.to_screen('Deleting existing file %s' % file_name)
0c3d0f51 987 except UnicodeEncodeError:
c25228e5 988 self.to_screen('Deleting existing file')
0c3d0f51 989
319b6059 990 def raise_no_formats(self, info, forced=False, *, msg=None):
0a5a191a 991 has_drm = info.get('_has_drm')
319b6059 992 ignored, expected = self.params.get('ignore_no_formats_error'), bool(msg)
993 msg = msg or has_drm and 'This video is DRM protected' or 'No video formats found!'
994 if forced or not ignored:
1151c407 995 raise ExtractorError(msg, video_id=info['id'], ie=info['extractor'],
319b6059 996 expected=has_drm or ignored or expected)
88acdbc2 997 else:
998 self.report_warning(msg)
999
de6000d9 1000 def parse_outtmpl(self):
bf1824b3 1001 self.deprecation_warning('"YoutubeDL.parse_outtmpl" is deprecated and may be removed in a future version')
1002 self._parse_outtmpl()
1003 return self.params['outtmpl']
1004
1005 def _parse_outtmpl(self):
7b2c3f47 1006 sanitize = IDENTITY
bf1824b3 1007 if self.params.get('restrictfilenames'): # Remove spaces in the default template
71ce444a 1008 sanitize = lambda x: x.replace(' - ', ' ').replace(' ', '-')
bf1824b3 1009
1010 outtmpl = self.params.setdefault('outtmpl', {})
1011 if not isinstance(outtmpl, dict):
1012 self.params['outtmpl'] = outtmpl = {'default': outtmpl}
1013 outtmpl.update({k: sanitize(v) for k, v in DEFAULT_OUTTMPL.items() if outtmpl.get(k) is None})
de6000d9 1014
21cd8fae 1015 def get_output_path(self, dir_type='', filename=None):
1016 paths = self.params.get('paths', {})
1017 assert isinstance(paths, dict)
1018 path = os.path.join(
1019 expand_path(paths.get('home', '').strip()),
1020 expand_path(paths.get(dir_type, '').strip()) if dir_type else '',
1021 filename or '')
21cd8fae 1022 return sanitize_path(path, force=self.params.get('windowsfilenames'))
1023
76a264ac 1024 @staticmethod
901130bb 1025 def _outtmpl_expandpath(outtmpl):
1026 # expand_path translates '%%' into '%' and '$$' into '$'
1027 # correspondingly that is not what we want since we need to keep
1028 # '%%' intact for template dict substitution step. Working around
1029 # with boundary-alike separator hack.
1030 sep = ''.join([random.choice(ascii_letters) for _ in range(32)])
86e5f3ed 1031 outtmpl = outtmpl.replace('%%', f'%{sep}%').replace('$$', f'${sep}$')
901130bb 1032
1033 # outtmpl should be expand_path'ed before template dict substitution
1034 # because meta fields may contain env variables we don't want to
1035 # be expanded. For example, for outtmpl "%(title)s.%(ext)s" and
1036 # title "Hello $PATH", we don't want `$PATH` to be expanded.
1037 return expand_path(outtmpl).replace(sep, '')
1038
1039 @staticmethod
1040 def escape_outtmpl(outtmpl):
1041 ''' Escape any remaining strings like %s, %abc% etc. '''
1042 return re.sub(
1043 STR_FORMAT_RE_TMPL.format('', '(?![%(\0])'),
1044 lambda mobj: ('' if mobj.group('has_key') else '%') + mobj.group(0),
1045 outtmpl)
1046
1047 @classmethod
1048 def validate_outtmpl(cls, outtmpl):
76a264ac 1049 ''' @return None or Exception object '''
7d1eb38a 1050 outtmpl = re.sub(
37893bb0 1051 STR_FORMAT_RE_TMPL.format('[^)]*', '[ljqBUDS]'),
7d1eb38a 1052 lambda mobj: f'{mobj.group(0)[:-1]}s',
1053 cls._outtmpl_expandpath(outtmpl))
76a264ac 1054 try:
7d1eb38a 1055 cls.escape_outtmpl(outtmpl) % collections.defaultdict(int)
76a264ac 1056 return None
1057 except ValueError as err:
1058 return err
1059
03b4de72 1060 @staticmethod
1061 def _copy_infodict(info_dict):
1062 info_dict = dict(info_dict)
09b49e1f 1063 info_dict.pop('__postprocessors', None)
415f8d51 1064 info_dict.pop('__pending_error', None)
03b4de72 1065 return info_dict
1066
e0fd9573 1067 def prepare_outtmpl(self, outtmpl, info_dict, sanitize=False):
1068 """ Make the outtmpl and info_dict suitable for substitution: ydl.escape_outtmpl(outtmpl) % info_dict
1069 @param sanitize Whether to sanitize the output as a filename.
1070 For backward compatibility, a function can also be passed
1071 """
1072
6e84b215 1073 info_dict.setdefault('epoch', int(time.time())) # keep epoch consistent once set
143db31d 1074
03b4de72 1075 info_dict = self._copy_infodict(info_dict)
752cda38 1076 info_dict['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
53c18592 1077 formatSeconds(info_dict['duration'], '-' if sanitize else ':')
143db31d 1078 if info_dict.get('duration', None) is not None
1079 else None)
1d485a1a 1080 info_dict['autonumber'] = int(self.params.get('autonumber_start', 1) - 1 + self._num_downloads)
9c906919 1081 info_dict['video_autonumber'] = self._num_videos
752cda38 1082 if info_dict.get('resolution') is None:
1083 info_dict['resolution'] = self.format_resolution(info_dict, default=None)
143db31d 1084
e6f21b3d 1085 # For fields playlist_index, playlist_autonumber and autonumber convert all occurrences
143db31d 1086 # of %(field)s to %(field)0Nd for backward compatibility
1087 field_size_compat_map = {
0a5a191a 1088 'playlist_index': number_of_digits(info_dict.get('__last_playlist_index') or 0),
ec11a9f4 1089 'playlist_autonumber': number_of_digits(info_dict.get('n_entries') or 0),
752cda38 1090 'autonumber': self.params.get('autonumber_size') or 5,
143db31d 1091 }
752cda38 1092
385a27fa 1093 TMPL_DICT = {}
37893bb0 1094 EXTERNAL_FORMAT_RE = re.compile(STR_FORMAT_RE_TMPL.format('[^)]*', f'[{STR_FORMAT_TYPES}ljqBUDS]'))
385a27fa 1095 MATH_FUNCTIONS = {
1096 '+': float.__add__,
1097 '-': float.__sub__,
1098 }
e625be0d 1099 # Field is of the form key1.key2...
1100 # where keys (except first) can be string, int or slice
2b8a2973 1101 FIELD_RE = r'\w*(?:\.(?:\w+|{num}|{num}?(?::{num}?){{1,2}}))*'.format(num=r'(?:-?\d+)')
1d485a1a 1102 MATH_FIELD_RE = rf'(?:{FIELD_RE}|-?{NUMBER_RE})'
385a27fa 1103 MATH_OPERATORS_RE = r'(?:%s)' % '|'.join(map(re.escape, MATH_FUNCTIONS.keys()))
1d485a1a 1104 INTERNAL_FORMAT_RE = re.compile(rf'''(?x)
e625be0d 1105 (?P<negate>-)?
1d485a1a 1106 (?P<fields>{FIELD_RE})
1107 (?P<maths>(?:{MATH_OPERATORS_RE}{MATH_FIELD_RE})*)
e625be0d 1108 (?:>(?P<strf_format>.+?))?
34baa9fd 1109 (?P<remaining>
1110 (?P<alternate>(?<!\\),[^|&)]+)?
1111 (?:&(?P<replacement>.*?))?
1112 (?:\|(?P<default>.*?))?
1d485a1a 1113 )$''')
752cda38 1114
2b8a2973 1115 def _traverse_infodict(k):
1116 k = k.split('.')
1117 if k[0] == '':
1118 k.pop(0)
1119 return traverse_obj(info_dict, k, is_user_input=True, traverse_string=True)
76a264ac 1120
752cda38 1121 def get_value(mdict):
1122 # Object traversal
2b8a2973 1123 value = _traverse_infodict(mdict['fields'])
752cda38 1124 # Negative
1125 if mdict['negate']:
1126 value = float_or_none(value)
1127 if value is not None:
1128 value *= -1
1129 # Do maths
385a27fa 1130 offset_key = mdict['maths']
1131 if offset_key:
752cda38 1132 value = float_or_none(value)
1133 operator = None
385a27fa 1134 while offset_key:
1135 item = re.match(
1136 MATH_FIELD_RE if operator else MATH_OPERATORS_RE,
1137 offset_key).group(0)
1138 offset_key = offset_key[len(item):]
1139 if operator is None:
752cda38 1140 operator = MATH_FUNCTIONS[item]
385a27fa 1141 continue
1142 item, multiplier = (item[1:], -1) if item[0] == '-' else (item, 1)
1143 offset = float_or_none(item)
1144 if offset is None:
2b8a2973 1145 offset = float_or_none(_traverse_infodict(item))
385a27fa 1146 try:
1147 value = operator(value, multiplier * offset)
1148 except (TypeError, ZeroDivisionError):
1149 return None
1150 operator = None
752cda38 1151 # Datetime formatting
1152 if mdict['strf_format']:
7c37ff97 1153 value = strftime_or_none(value, mdict['strf_format'].replace('\\,', ','))
752cda38 1154
1155 return value
1156
b868936c 1157 na = self.params.get('outtmpl_na_placeholder', 'NA')
1158
e0fd9573 1159 def filename_sanitizer(key, value, restricted=self.params.get('restrictfilenames')):
5c3895ff 1160 return sanitize_filename(str(value), restricted=restricted, is_id=(
1161 bool(re.search(r'(^|[_.])id(\.|$)', key))
8a82af35 1162 if 'filename-sanitization' in self.params['compat_opts']
5c3895ff 1163 else NO_DEFAULT))
e0fd9573 1164
1165 sanitizer = sanitize if callable(sanitize) else filename_sanitizer
1166 sanitize = bool(sanitize)
1167
6e84b215 1168 def _dumpjson_default(obj):
1169 if isinstance(obj, (set, LazyList)):
1170 return list(obj)
adbc4ec4 1171 return repr(obj)
6e84b215 1172
752cda38 1173 def create_key(outer_mobj):
1174 if not outer_mobj.group('has_key'):
b836dc94 1175 return outer_mobj.group(0)
752cda38 1176 key = outer_mobj.group('key')
752cda38 1177 mobj = re.match(INTERNAL_FORMAT_RE, key)
e0fd9573 1178 initial_field = mobj.group('fields') if mobj else ''
e978789f 1179 value, replacement, default = None, None, na
7c37ff97 1180 while mobj:
e625be0d 1181 mobj = mobj.groupdict()
7c37ff97 1182 default = mobj['default'] if mobj['default'] is not None else default
752cda38 1183 value = get_value(mobj)
e978789f 1184 replacement = mobj['replacement']
7c37ff97 1185 if value is None and mobj['alternate']:
34baa9fd 1186 mobj = re.match(INTERNAL_FORMAT_RE, mobj['remaining'][1:])
7c37ff97 1187 else:
1188 break
752cda38 1189
b868936c 1190 fmt = outer_mobj.group('format')
752cda38 1191 if fmt == 's' and value is not None and key in field_size_compat_map.keys():
86e5f3ed 1192 fmt = f'0{field_size_compat_map[key]:d}d'
752cda38 1193
e978789f 1194 value = default if value is None else value if replacement is None else replacement
752cda38 1195
4476d2c7 1196 flags = outer_mobj.group('conversion') or ''
7d1eb38a 1197 str_fmt = f'{fmt[:-1]}s'
524e2e4f 1198 if fmt[-1] == 'l': # list
4476d2c7 1199 delim = '\n' if '#' in flags else ', '
9e907ebd 1200 value, fmt = delim.join(map(str, variadic(value, allowed_types=(str, bytes)))), str_fmt
524e2e4f 1201 elif fmt[-1] == 'j': # json
4476d2c7 1202 value, fmt = json.dumps(value, default=_dumpjson_default, indent=4 if '#' in flags else None), str_fmt
524e2e4f 1203 elif fmt[-1] == 'q': # quoted
4476d2c7 1204 value = map(str, variadic(value) if '#' in flags else [value])
1205 value, fmt = ' '.join(map(compat_shlex_quote, value)), str_fmt
524e2e4f 1206 elif fmt[-1] == 'B': # bytes
0f06bcd7 1207 value = f'%{str_fmt}'.encode() % str(value).encode()
f5aa5cfb 1208 value, fmt = value.decode('utf-8', 'ignore'), 's'
524e2e4f 1209 elif fmt[-1] == 'U': # unicode normalized
524e2e4f 1210 value, fmt = unicodedata.normalize(
1211 # "+" = compatibility equivalence, "#" = NFD
4476d2c7 1212 'NF%s%s' % ('K' if '+' in flags else '', 'D' if '#' in flags else 'C'),
524e2e4f 1213 value), str_fmt
e0fd9573 1214 elif fmt[-1] == 'D': # decimal suffix
abbeeebc 1215 num_fmt, fmt = fmt[:-1].replace('#', ''), 's'
1216 value = format_decimal_suffix(value, f'%{num_fmt}f%s' if num_fmt else '%d%s',
1217 factor=1024 if '#' in flags else 1000)
37893bb0 1218 elif fmt[-1] == 'S': # filename sanitization
e0fd9573 1219 value, fmt = filename_sanitizer(initial_field, value, restricted='#' in flags), str_fmt
7d1eb38a 1220 elif fmt[-1] == 'c':
524e2e4f 1221 if value:
1222 value = str(value)[0]
76a264ac 1223 else:
524e2e4f 1224 fmt = str_fmt
76a264ac 1225 elif fmt[-1] not in 'rs': # numeric
a439a3a4 1226 value = float_or_none(value)
752cda38 1227 if value is None:
1228 value, fmt = default, 's'
901130bb 1229
752cda38 1230 if sanitize:
1231 if fmt[-1] == 'r':
1232 # If value is an object, sanitize might convert it to a string
1233 # So we convert it to repr first
7d1eb38a 1234 value, fmt = repr(value), str_fmt
639f1cea 1235 if fmt[-1] in 'csr':
e0fd9573 1236 value = sanitizer(initial_field, value)
901130bb 1237
b868936c 1238 key = '%s\0%s' % (key.replace('%', '%\0'), outer_mobj.group('format'))
385a27fa 1239 TMPL_DICT[key] = value
b868936c 1240 return '{prefix}%({key}){fmt}'.format(key=key, fmt=fmt, prefix=outer_mobj.group('prefix'))
752cda38 1241
385a27fa 1242 return EXTERNAL_FORMAT_RE.sub(create_key, outtmpl), TMPL_DICT
143db31d 1243
819e0531 1244 def evaluate_outtmpl(self, outtmpl, info_dict, *args, **kwargs):
1245 outtmpl, info_dict = self.prepare_outtmpl(outtmpl, info_dict, *args, **kwargs)
1246 return self.escape_outtmpl(outtmpl) % info_dict
1247
5127e92a 1248 def _prepare_filename(self, info_dict, *, outtmpl=None, tmpl_type=None):
1249 assert None in (outtmpl, tmpl_type), 'outtmpl and tmpl_type are mutually exclusive'
1250 if outtmpl is None:
bf1824b3 1251 outtmpl = self.params['outtmpl'].get(tmpl_type or 'default', self.params['outtmpl']['default'])
8222d8de 1252 try:
5127e92a 1253 outtmpl = self._outtmpl_expandpath(outtmpl)
e0fd9573 1254 filename = self.evaluate_outtmpl(outtmpl, info_dict, True)
6a0546e3 1255 if not filename:
1256 return None
15da37c7 1257
5127e92a 1258 if tmpl_type in ('', 'temp'):
6a0546e3 1259 final_ext, ext = self.params.get('final_ext'), info_dict.get('ext')
1260 if final_ext and ext and final_ext != ext and filename.endswith(f'.{final_ext}'):
1261 filename = replace_extension(filename, ext, final_ext)
5127e92a 1262 elif tmpl_type:
6a0546e3 1263 force_ext = OUTTMPL_TYPES[tmpl_type]
1264 if force_ext:
1265 filename = replace_extension(filename, force_ext, info_dict.get('ext'))
de6000d9 1266
bdc3fd2f
U
1267 # https://github.com/blackjack4494/youtube-dlc/issues/85
1268 trim_file_name = self.params.get('trim_file_name', False)
1269 if trim_file_name:
5c22c63d 1270 no_ext, *ext = filename.rsplit('.', 2)
1271 filename = join_nonempty(no_ext[:trim_file_name], *ext, delim='.')
bdc3fd2f 1272
0202b52a 1273 return filename
8222d8de 1274 except ValueError as err:
6febd1c1 1275 self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')')
8222d8de
JMF
1276 return None
1277
5127e92a 1278 def prepare_filename(self, info_dict, dir_type='', *, outtmpl=None, warn=False):
1279 """Generate the output filename"""
1280 if outtmpl:
1281 assert not dir_type, 'outtmpl and dir_type are mutually exclusive'
1282 dir_type = None
1283 filename = self._prepare_filename(info_dict, tmpl_type=dir_type, outtmpl=outtmpl)
80c03fa9 1284 if not filename and dir_type not in ('', 'temp'):
1285 return ''
de6000d9 1286
c84aeac6 1287 if warn:
21cd8fae 1288 if not self.params.get('paths'):
de6000d9 1289 pass
1290 elif filename == '-':
c84aeac6 1291 self.report_warning('--paths is ignored when an outputting to stdout', only_once=True)
de6000d9 1292 elif os.path.isabs(filename):
c84aeac6 1293 self.report_warning('--paths is ignored since an absolute path is given in output template', only_once=True)
de6000d9 1294 if filename == '-' or not filename:
1295 return filename
1296
21cd8fae 1297 return self.get_output_path(dir_type, filename)
0202b52a 1298
120fe513 1299 def _match_entry(self, info_dict, incomplete=False, silent=False):
ecdec191 1300 """ Returns None if the file should be downloaded """
8222d8de 1301
c77495e3 1302 video_title = info_dict.get('title', info_dict.get('id', 'video'))
1303
8b0d7497 1304 def check_filter():
8b0d7497 1305 if 'title' in info_dict:
1306 # This can happen when we're just evaluating the playlist
1307 title = info_dict['title']
1308 matchtitle = self.params.get('matchtitle', False)
1309 if matchtitle:
1310 if not re.search(matchtitle, title, re.IGNORECASE):
1311 return '"' + title + '" title did not match pattern "' + matchtitle + '"'
1312 rejecttitle = self.params.get('rejecttitle', False)
1313 if rejecttitle:
1314 if re.search(rejecttitle, title, re.IGNORECASE):
1315 return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
1316 date = info_dict.get('upload_date')
1317 if date is not None:
1318 dateRange = self.params.get('daterange', DateRange())
1319 if date not in dateRange:
86e5f3ed 1320 return f'{date_from_str(date).isoformat()} upload date is not in range {dateRange}'
8b0d7497 1321 view_count = info_dict.get('view_count')
1322 if view_count is not None:
1323 min_views = self.params.get('min_views')
1324 if min_views is not None and view_count < min_views:
1325 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
1326 max_views = self.params.get('max_views')
1327 if max_views is not None and view_count > max_views:
1328 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
1329 if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')):
1330 return 'Skipping "%s" because it is age restricted' % video_title
8b0d7497 1331
8f18aca8 1332 match_filter = self.params.get('match_filter')
1333 if match_filter is not None:
1334 try:
1335 ret = match_filter(info_dict, incomplete=incomplete)
1336 except TypeError:
1337 # For backward compatibility
1338 ret = None if incomplete else match_filter(info_dict)
492272fe 1339 if ret is NO_DEFAULT:
1340 while True:
1341 filename = self._format_screen(self.prepare_filename(info_dict), self.Styles.FILENAME)
1342 reply = input(self._format_screen(
1343 f'Download "{filename}"? (Y/n): ', self.Styles.EMPHASIS)).lower().strip()
1344 if reply in {'y', ''}:
1345 return None
1346 elif reply == 'n':
1347 return f'Skipping {video_title}'
492272fe 1348 elif ret is not None:
8f18aca8 1349 return ret
8b0d7497 1350 return None
1351
c77495e3 1352 if self.in_download_archive(info_dict):
1353 reason = '%s has already been recorded in the archive' % video_title
1354 break_opt, break_err = 'break_on_existing', ExistingVideoReached
1355 else:
1356 reason = check_filter()
1357 break_opt, break_err = 'break_on_reject', RejectedVideoReached
8b0d7497 1358 if reason is not None:
120fe513 1359 if not silent:
1360 self.to_screen('[download] ' + reason)
c77495e3 1361 if self.params.get(break_opt, False):
1362 raise break_err()
8b0d7497 1363 return reason
fe7e0c98 1364
b6c45014
JMF
1365 @staticmethod
1366 def add_extra_info(info_dict, extra_info):
1367 '''Set the keys from extra_info in info dict if they are missing'''
1368 for key, value in extra_info.items():
1369 info_dict.setdefault(key, value)
1370
409e1828 1371 def extract_info(self, url, download=True, ie_key=None, extra_info=None,
61aa5ba3 1372 process=True, force_generic_extractor=False):
41d1cca3 1373 """
1374 Return a list with a dictionary for each video extracted.
1375
1376 Arguments:
1377 url -- URL to extract
1378
1379 Keyword arguments:
1380 download -- whether to download videos during extraction
1381 ie_key -- extractor key hint
1382 extra_info -- dictionary containing the extra values to add to each result
1383 process -- whether to resolve all unresolved references (URLs, playlist items),
1384 must be True for download to work.
1385 force_generic_extractor -- force using the generic extractor
1386 """
fe7e0c98 1387
409e1828 1388 if extra_info is None:
1389 extra_info = {}
1390
61aa5ba3 1391 if not ie_key and force_generic_extractor:
d22dec74
S
1392 ie_key = 'Generic'
1393
8222d8de 1394 if ie_key:
8b7491c8 1395 ies = {ie_key: self._get_info_extractor_class(ie_key)}
8222d8de
JMF
1396 else:
1397 ies = self._ies
1398
8b7491c8 1399 for ie_key, ie in ies.items():
8222d8de
JMF
1400 if not ie.suitable(url):
1401 continue
1402
1403 if not ie.working():
6febd1c1
PH
1404 self.report_warning('The program functionality for this site has been marked as broken, '
1405 'and will probably not work.')
8222d8de 1406
1151c407 1407 temp_id = ie.get_temp_id(url)
a0566bbf 1408 if temp_id is not None and self.in_download_archive({'id': temp_id, 'ie_key': ie_key}):
5e5be0c0 1409 self.to_screen(f'[{ie_key}] {temp_id}: has already been recorded in the archive')
1410 if self.params.get('break_on_existing', False):
1411 raise ExistingVideoReached()
a0566bbf 1412 break
8b7491c8 1413 return self.__extract_info(url, self.get_info_extractor(ie_key), download, extra_info, process)
a0566bbf 1414 else:
1415 self.report_error('no suitable InfoExtractor for URL %s' % url)
1416
7e88d7d7 1417 def _handle_extraction_exceptions(func):
b5ae35ee 1418 @functools.wraps(func)
a0566bbf 1419 def wrapper(self, *args, **kwargs):
6da22e7d 1420 while True:
1421 try:
1422 return func(self, *args, **kwargs)
1423 except (DownloadCancelled, LazyList.IndexError, PagedList.IndexError):
8222d8de 1424 raise
6da22e7d 1425 except ReExtractInfo as e:
1426 if e.expected:
1427 self.to_screen(f'{e}; Re-extracting data')
1428 else:
1429 self.to_stderr('\r')
1430 self.report_warning(f'{e}; Re-extracting data')
1431 continue
1432 except GeoRestrictedError as e:
1433 msg = e.msg
1434 if e.countries:
1435 msg += '\nThis video is available in %s.' % ', '.join(
1436 map(ISO3166Utils.short2full, e.countries))
1437 msg += '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1438 self.report_error(msg)
1439 except ExtractorError as e: # An error we somewhat expected
1440 self.report_error(str(e), e.format_traceback())
1441 except Exception as e:
1442 if self.params.get('ignoreerrors'):
1443 self.report_error(str(e), tb=encode_compat_str(traceback.format_exc()))
1444 else:
1445 raise
1446 break
a0566bbf 1447 return wrapper
1448
f2ebc5c7 1449 def _wait_for_video(self, ie_result):
1450 if (not self.params.get('wait_for_video')
1451 or ie_result.get('_type', 'video') != 'video'
1452 or ie_result.get('formats') or ie_result.get('url')):
1453 return
1454
1455 format_dur = lambda dur: '%02d:%02d:%02d' % timetuple_from_msec(dur * 1000)[:-1]
1456 last_msg = ''
1457
1458 def progress(msg):
1459 nonlocal last_msg
1460 self.to_screen(msg + ' ' * (len(last_msg) - len(msg)) + '\r', skip_eol=True)
1461 last_msg = msg
1462
1463 min_wait, max_wait = self.params.get('wait_for_video')
1464 diff = try_get(ie_result, lambda x: x['release_timestamp'] - time.time())
1465 if diff is None and ie_result.get('live_status') == 'is_upcoming':
16c620bc 1466 diff = round(random.uniform(min_wait, max_wait) if (max_wait and min_wait) else (max_wait or min_wait), 0)
f2ebc5c7 1467 self.report_warning('Release time of video is not known')
1468 elif (diff or 0) <= 0:
1469 self.report_warning('Video should already be available according to extracted info')
38d79fd1 1470 diff = min(max(diff or 0, min_wait or 0), max_wait or float('inf'))
f2ebc5c7 1471 self.to_screen(f'[wait] Waiting for {format_dur(diff)} - Press Ctrl+C to try now')
1472
1473 wait_till = time.time() + diff
1474 try:
1475 while True:
1476 diff = wait_till - time.time()
1477 if diff <= 0:
1478 progress('')
1479 raise ReExtractInfo('[wait] Wait period ended', expected=True)
1480 progress(f'[wait] Remaining time until next attempt: {self._format_screen(format_dur(diff), self.Styles.EMPHASIS)}')
1481 time.sleep(1)
1482 except KeyboardInterrupt:
1483 progress('')
1484 raise ReExtractInfo('[wait] Interrupted by user', expected=True)
1485 except BaseException as e:
1486 if not isinstance(e, ReExtractInfo):
1487 self.to_screen('')
1488 raise
1489
7e88d7d7 1490 @_handle_extraction_exceptions
58f197b7 1491 def __extract_info(self, url, ie, download, extra_info, process):
a0566bbf 1492 ie_result = ie.extract(url)
1493 if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
1494 return
1495 if isinstance(ie_result, list):
1496 # Backwards compatibility: old IE result format
1497 ie_result = {
1498 '_type': 'compat_list',
1499 'entries': ie_result,
1500 }
e37d0efb 1501 if extra_info.get('original_url'):
1502 ie_result.setdefault('original_url', extra_info['original_url'])
a0566bbf 1503 self.add_default_extra_info(ie_result, ie, url)
1504 if process:
f2ebc5c7 1505 self._wait_for_video(ie_result)
a0566bbf 1506 return self.process_ie_result(ie_result, download, extra_info)
8222d8de 1507 else:
a0566bbf 1508 return ie_result
fe7e0c98 1509
ea38e55f 1510 def add_default_extra_info(self, ie_result, ie, url):
6033d980 1511 if url is not None:
1512 self.add_extra_info(ie_result, {
1513 'webpage_url': url,
1514 'original_url': url,
57ebfca3 1515 })
1516 webpage_url = ie_result.get('webpage_url')
1517 if webpage_url:
1518 self.add_extra_info(ie_result, {
1519 'webpage_url_basename': url_basename(webpage_url),
1520 'webpage_url_domain': get_domain(webpage_url),
6033d980 1521 })
1522 if ie is not None:
1523 self.add_extra_info(ie_result, {
1524 'extractor': ie.IE_NAME,
1525 'extractor_key': ie.ie_key(),
1526 })
ea38e55f 1527
58adec46 1528 def process_ie_result(self, ie_result, download=True, extra_info=None):
8222d8de
JMF
1529 """
1530 Take the result of the ie(may be modified) and resolve all unresolved
1531 references (URLs, playlist items).
1532
1533 It will also download the videos if 'download'.
1534 Returns the resolved ie_result.
1535 """
58adec46 1536 if extra_info is None:
1537 extra_info = {}
e8ee972c
PH
1538 result_type = ie_result.get('_type', 'video')
1539
057a5206 1540 if result_type in ('url', 'url_transparent'):
134c6ea8 1541 ie_result['url'] = sanitize_url(ie_result['url'])
e37d0efb 1542 if ie_result.get('original_url'):
1543 extra_info.setdefault('original_url', ie_result['original_url'])
1544
057a5206 1545 extract_flat = self.params.get('extract_flat', False)
3089bc74
S
1546 if ((extract_flat == 'in_playlist' and 'playlist' in extra_info)
1547 or extract_flat is True):
ecb54191 1548 info_copy = ie_result.copy()
6033d980 1549 ie = try_get(ie_result.get('ie_key'), self.get_info_extractor)
360167b9 1550 if ie and not ie_result.get('id'):
4614bc22 1551 info_copy['id'] = ie.get_temp_id(ie_result['url'])
6033d980 1552 self.add_default_extra_info(info_copy, ie, ie_result['url'])
4614bc22 1553 self.add_extra_info(info_copy, extra_info)
b5475f11 1554 info_copy, _ = self.pre_process(info_copy)
ecb54191 1555 self.__forced_printings(info_copy, self.prepare_filename(info_copy), incomplete=True)
415f8d51 1556 self._raise_pending_errors(info_copy)
4614bc22 1557 if self.params.get('force_write_download_archive', False):
1558 self.record_download_archive(info_copy)
e8ee972c
PH
1559 return ie_result
1560
8222d8de 1561 if result_type == 'video':
b6c45014 1562 self.add_extra_info(ie_result, extra_info)
9c2b75b5 1563 ie_result = self.process_video_result(ie_result, download=download)
415f8d51 1564 self._raise_pending_errors(ie_result)
28b0eb0f 1565 additional_urls = (ie_result or {}).get('additional_urls')
9c2b75b5 1566 if additional_urls:
e9f4ccd1 1567 # TODO: Improve MetadataParserPP to allow setting a list
9c2b75b5 1568 if isinstance(additional_urls, compat_str):
1569 additional_urls = [additional_urls]
1570 self.to_screen(
1571 '[info] %s: %d additional URL(s) requested' % (ie_result['id'], len(additional_urls)))
1572 self.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls))
1573 ie_result['additional_entries'] = [
1574 self.extract_info(
b69fd25c 1575 url, download, extra_info=extra_info,
9c2b75b5 1576 force_generic_extractor=self.params.get('force_generic_extractor'))
1577 for url in additional_urls
1578 ]
1579 return ie_result
8222d8de
JMF
1580 elif result_type == 'url':
1581 # We have to add extra_info to the results because it may be
1582 # contained in a playlist
07cce701 1583 return self.extract_info(
1584 ie_result['url'], download,
1585 ie_key=ie_result.get('ie_key'),
1586 extra_info=extra_info)
7fc3fa05
PH
1587 elif result_type == 'url_transparent':
1588 # Use the information from the embedding page
1589 info = self.extract_info(
1590 ie_result['url'], ie_key=ie_result.get('ie_key'),
1591 extra_info=extra_info, download=False, process=False)
1592
1640eb09
S
1593 # extract_info may return None when ignoreerrors is enabled and
1594 # extraction failed with an error, don't crash and return early
1595 # in this case
1596 if not info:
1597 return info
1598
412c617d 1599 new_result = info.copy()
90137ca4 1600 new_result.update(filter_dict(ie_result, lambda k, v: (
1601 v is not None and k not in {'_type', 'url', 'id', 'extractor', 'extractor_key', 'ie_key'})))
7fc3fa05 1602
0563f7ac
S
1603 # Extracted info may not be a video result (i.e.
1604 # info.get('_type', 'video') != video) but rather an url or
1605 # url_transparent. In such cases outer metadata (from ie_result)
1606 # should be propagated to inner one (info). For this to happen
1607 # _type of info should be overridden with url_transparent. This
067aa17e 1608 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
0563f7ac
S
1609 if new_result.get('_type') == 'url':
1610 new_result['_type'] = 'url_transparent'
7fc3fa05
PH
1611
1612 return self.process_ie_result(
1613 new_result, download=download, extra_info=extra_info)
40fcba5e 1614 elif result_type in ('playlist', 'multi_video'):
30a074c2 1615 # Protect from infinite recursion due to recursively nested playlists
1616 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
1617 webpage_url = ie_result['webpage_url']
1618 if webpage_url in self._playlist_urls:
7e85e872 1619 self.to_screen(
30a074c2 1620 '[download] Skipping already downloaded playlist: %s'
1621 % ie_result.get('title') or ie_result.get('id'))
1622 return
7e85e872 1623
30a074c2 1624 self._playlist_level += 1
1625 self._playlist_urls.add(webpage_url)
03f83004 1626 self._fill_common_fields(ie_result, False)
bc516a3f 1627 self._sanitize_thumbnails(ie_result)
30a074c2 1628 try:
1629 return self.__process_playlist(ie_result, download)
1630 finally:
1631 self._playlist_level -= 1
1632 if not self._playlist_level:
1633 self._playlist_urls.clear()
8222d8de 1634 elif result_type == 'compat_list':
c9bf4114
PH
1635 self.report_warning(
1636 'Extractor %s returned a compat_list result. '
1637 'It needs to be updated.' % ie_result.get('extractor'))
5f6a1245 1638
8222d8de 1639 def _fixup(r):
b868936c 1640 self.add_extra_info(r, {
1641 'extractor': ie_result['extractor'],
1642 'webpage_url': ie_result['webpage_url'],
1643 'webpage_url_basename': url_basename(ie_result['webpage_url']),
0bb322b9 1644 'webpage_url_domain': get_domain(ie_result['webpage_url']),
b868936c 1645 'extractor_key': ie_result['extractor_key'],
1646 })
8222d8de
JMF
1647 return r
1648 ie_result['entries'] = [
b6c45014 1649 self.process_ie_result(_fixup(r), download, extra_info)
8222d8de
JMF
1650 for r in ie_result['entries']
1651 ]
1652 return ie_result
1653 else:
1654 raise Exception('Invalid result type: %s' % result_type)
1655
e92caff5 1656 def _ensure_dir_exists(self, path):
1657 return make_dir(path, self.report_error)
1658
3b603dbd 1659 @staticmethod
1660 def _playlist_infodict(ie_result, **kwargs):
1661 return {
1662 **ie_result,
1663 'playlist': ie_result.get('title') or ie_result.get('id'),
1664 'playlist_id': ie_result.get('id'),
1665 'playlist_title': ie_result.get('title'),
1666 'playlist_uploader': ie_result.get('uploader'),
1667 'playlist_uploader_id': ie_result.get('uploader_id'),
1668 'playlist_index': 0,
1669 **kwargs,
1670 }
1671
30a074c2 1672 def __process_playlist(self, ie_result, download):
7e88d7d7 1673 """Process each entry in the playlist"""
1674 title = ie_result.get('title') or ie_result.get('id') or '<Untitled>'
1675 self.to_screen(f'[download] Downloading playlist: {title}')
f0d785d3 1676
7e88d7d7 1677 all_entries = PlaylistEntries(self, ie_result)
7e9a6125 1678 entries = orderedSet(all_entries.get_requested_items(), lazy=True)
1679
1680 lazy = self.params.get('lazy_playlist')
1681 if lazy:
1682 resolved_entries, n_entries = [], 'N/A'
1683 ie_result['requested_entries'], ie_result['entries'] = None, None
1684 else:
1685 entries = resolved_entries = list(entries)
1686 n_entries = len(resolved_entries)
1687 ie_result['requested_entries'], ie_result['entries'] = tuple(zip(*resolved_entries)) or ([], [])
1688 if not ie_result.get('playlist_count'):
1689 # Better to do this after potentially exhausting entries
1690 ie_result['playlist_count'] = all_entries.get_full_count()
498f5606 1691
e08a85d8 1692 _infojson_written = False
0bfc53d0 1693 write_playlist_files = self.params.get('allow_playlist_files', True)
1694 if write_playlist_files and self.params.get('list_thumbnails'):
1695 self.list_thumbnails(ie_result)
1696 if write_playlist_files and not self.params.get('simulate'):
7e9a6125 1697 ie_copy = self._playlist_infodict(ie_result, n_entries=int_or_none(n_entries))
e08a85d8 1698 _infojson_written = self._write_info_json(
1699 'playlist', ie_result, self.prepare_filename(ie_copy, 'pl_infojson'))
1700 if _infojson_written is None:
80c03fa9 1701 return
1702 if self._write_description('playlist', ie_result,
1703 self.prepare_filename(ie_copy, 'pl_description')) is None:
1704 return
681de68e 1705 # TODO: This should be passed to ThumbnailsConvertor if necessary
80c03fa9 1706 self._write_thumbnails('playlist', ie_copy, self.prepare_filename(ie_copy, 'pl_thumbnail'))
30a074c2 1707
7e9a6125 1708 if lazy:
1709 if self.params.get('playlistreverse') or self.params.get('playlistrandom'):
1710 self.report_warning('playlistreverse and playlistrandom are not supported with lazy_playlist', only_once=True)
1711 elif self.params.get('playlistreverse'):
1712 entries.reverse()
1713 elif self.params.get('playlistrandom'):
30a074c2 1714 random.shuffle(entries)
1715
7e88d7d7 1716 self.to_screen(f'[{ie_result["extractor"]}] Playlist {title}: Downloading {n_entries} videos'
1717 f'{format_field(ie_result, "playlist_count", " of %s")}')
30a074c2 1718
26e2805c 1719 failures = 0
1720 max_failures = self.params.get('skip_playlist_after_errors') or float('inf')
7e9a6125 1721 for i, (playlist_index, entry) in enumerate(entries):
1722 if lazy:
1723 resolved_entries.append((playlist_index, entry))
1724
7e88d7d7 1725 # TODO: Add auto-generated fields
1726 if self._match_entry(entry, incomplete=True) is not None:
1727 continue
1728
19a03940 1729 self.to_screen('[download] Downloading video %s of %s' % (
7e9a6125 1730 self._format_screen(i + 1, self.Styles.ID), self._format_screen(n_entries, self.Styles.EMPHASIS)))
7e88d7d7 1731
1732 entry['__x_forwarded_for_ip'] = ie_result.get('__x_forwarded_for_ip')
7e9a6125 1733 if not lazy and 'playlist-index' in self.params.get('compat_opts', []):
1734 playlist_index = ie_result['requested_entries'][i]
1735
7e88d7d7 1736 entry_result = self.__process_iterable_entry(entry, download, {
7e9a6125 1737 'n_entries': int_or_none(n_entries),
1738 '__last_playlist_index': max(ie_result['requested_entries'] or (0, 0)),
f0d785d3 1739 'playlist_count': ie_result.get('playlist_count'),
71729754 1740 'playlist_index': playlist_index,
7e9a6125 1741 'playlist_autonumber': i + 1,
7e88d7d7 1742 'playlist': title,
30a074c2 1743 'playlist_id': ie_result.get('id'),
1744 'playlist_title': ie_result.get('title'),
1745 'playlist_uploader': ie_result.get('uploader'),
1746 'playlist_uploader_id': ie_result.get('uploader_id'),
30a074c2 1747 'extractor': ie_result['extractor'],
1748 'webpage_url': ie_result['webpage_url'],
1749 'webpage_url_basename': url_basename(ie_result['webpage_url']),
0bb322b9 1750 'webpage_url_domain': get_domain(ie_result['webpage_url']),
30a074c2 1751 'extractor_key': ie_result['extractor_key'],
7e88d7d7 1752 })
26e2805c 1753 if not entry_result:
1754 failures += 1
1755 if failures >= max_failures:
1756 self.report_error(
7e88d7d7 1757 f'Skipping the remaining entries in playlist "{title}" since {failures} items failed extraction')
26e2805c 1758 break
7e9a6125 1759 resolved_entries[i] = (playlist_index, entry_result)
7e88d7d7 1760
1761 # Update with processed data
7e9a6125 1762 ie_result['requested_entries'], ie_result['entries'] = tuple(zip(*resolved_entries)) or ([], [])
e08a85d8 1763
1764 # Write the updated info to json
cb96c5be 1765 if _infojson_written is True and self._write_info_json(
e08a85d8 1766 'updated playlist', ie_result,
1767 self.prepare_filename(ie_copy, 'pl_infojson'), overwrite=True) is None:
1768 return
ca30f449 1769
ed5835b4 1770 ie_result = self.run_all_pps('playlist', ie_result)
7e88d7d7 1771 self.to_screen(f'[download] Finished downloading playlist: {title}')
30a074c2 1772 return ie_result
1773
7e88d7d7 1774 @_handle_extraction_exceptions
a0566bbf 1775 def __process_iterable_entry(self, entry, download, extra_info):
1776 return self.process_ie_result(
1777 entry, download=download, extra_info=extra_info)
1778
67134eab
JMF
1779 def _build_format_filter(self, filter_spec):
1780 " Returns a function to filter the formats according to the filter_spec "
083c9df9
PH
1781
1782 OPERATORS = {
1783 '<': operator.lt,
1784 '<=': operator.le,
1785 '>': operator.gt,
1786 '>=': operator.ge,
1787 '=': operator.eq,
1788 '!=': operator.ne,
1789 }
67134eab 1790 operator_rex = re.compile(r'''(?x)\s*
187986a8 1791 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)\s*
1792 (?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1793 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)\s*
083c9df9 1794 ''' % '|'.join(map(re.escape, OPERATORS.keys())))
187986a8 1795 m = operator_rex.fullmatch(filter_spec)
9ddb6925
S
1796 if m:
1797 try:
1798 comparison_value = int(m.group('value'))
1799 except ValueError:
1800 comparison_value = parse_filesize(m.group('value'))
1801 if comparison_value is None:
1802 comparison_value = parse_filesize(m.group('value') + 'B')
1803 if comparison_value is None:
1804 raise ValueError(
1805 'Invalid value %r in format specification %r' % (
67134eab 1806 m.group('value'), filter_spec))
9ddb6925
S
1807 op = OPERATORS[m.group('op')]
1808
083c9df9 1809 if not m:
9ddb6925
S
1810 STR_OPERATORS = {
1811 '=': operator.eq,
10d33b34
YCH
1812 '^=': lambda attr, value: attr.startswith(value),
1813 '$=': lambda attr, value: attr.endswith(value),
1814 '*=': lambda attr, value: value in attr,
1ce9a3cb 1815 '~=': lambda attr, value: value.search(attr) is not None
9ddb6925 1816 }
187986a8 1817 str_operator_rex = re.compile(r'''(?x)\s*
1818 (?P<key>[a-zA-Z0-9._-]+)\s*
1ce9a3cb
LF
1819 (?P<negation>!\s*)?(?P<op>%s)\s*(?P<none_inclusive>\?\s*)?
1820 (?P<quote>["'])?
1821 (?P<value>(?(quote)(?:(?!(?P=quote))[^\\]|\\.)+|[\w.-]+))
1822 (?(quote)(?P=quote))\s*
9ddb6925 1823 ''' % '|'.join(map(re.escape, STR_OPERATORS.keys())))
187986a8 1824 m = str_operator_rex.fullmatch(filter_spec)
9ddb6925 1825 if m:
1ce9a3cb
LF
1826 if m.group('op') == '~=':
1827 comparison_value = re.compile(m.group('value'))
1828 else:
1829 comparison_value = re.sub(r'''\\([\\"'])''', r'\1', m.group('value'))
2cc779f4
S
1830 str_op = STR_OPERATORS[m.group('op')]
1831 if m.group('negation'):
e118a879 1832 op = lambda attr, value: not str_op(attr, value)
2cc779f4
S
1833 else:
1834 op = str_op
083c9df9 1835
9ddb6925 1836 if not m:
187986a8 1837 raise SyntaxError('Invalid filter specification %r' % filter_spec)
083c9df9
PH
1838
1839 def _filter(f):
1840 actual_value = f.get(m.group('key'))
1841 if actual_value is None:
1842 return m.group('none_inclusive')
1843 return op(actual_value, comparison_value)
67134eab
JMF
1844 return _filter
1845
9f1a1c36 1846 def _check_formats(self, formats):
1847 for f in formats:
1848 self.to_screen('[info] Testing format %s' % f['format_id'])
75689fe5 1849 path = self.get_output_path('temp')
1850 if not self._ensure_dir_exists(f'{path}/'):
1851 continue
1852 temp_file = tempfile.NamedTemporaryFile(suffix='.tmp', delete=False, dir=path or None)
9f1a1c36 1853 temp_file.close()
1854 try:
1855 success, _ = self.dl(temp_file.name, f, test=True)
8a82af35 1856 except (DownloadError, OSError, ValueError) + network_exceptions:
9f1a1c36 1857 success = False
1858 finally:
1859 if os.path.exists(temp_file.name):
1860 try:
1861 os.remove(temp_file.name)
1862 except OSError:
1863 self.report_warning('Unable to delete temporary file "%s"' % temp_file.name)
1864 if success:
1865 yield f
1866 else:
1867 self.to_screen('[info] Unable to download format %s. Skipping...' % f['format_id'])
1868
0017d9ad 1869 def _default_format_spec(self, info_dict, download=True):
0017d9ad 1870
af0f7428
S
1871 def can_merge():
1872 merger = FFmpegMergerPP(self)
1873 return merger.available and merger.can_merge()
1874
91ebc640 1875 prefer_best = (
b7b04c78 1876 not self.params.get('simulate')
91ebc640 1877 and download
1878 and (
1879 not can_merge()
21633673 1880 or info_dict.get('is_live') and not self.params.get('live_from_start')
bf1824b3 1881 or self.params['outtmpl']['default'] == '-'))
53ed7066 1882 compat = (
1883 prefer_best
1884 or self.params.get('allow_multiple_audio_streams', False)
8a82af35 1885 or 'format-spec' in self.params['compat_opts'])
91ebc640 1886
1887 return (
53ed7066 1888 'best/bestvideo+bestaudio' if prefer_best
1889 else 'bestvideo*+bestaudio/best' if not compat
91ebc640 1890 else 'bestvideo+bestaudio/best')
0017d9ad 1891
67134eab
JMF
1892 def build_format_selector(self, format_spec):
1893 def syntax_error(note, start):
1894 message = (
1895 'Invalid format specification: '
86e5f3ed 1896 '{}\n\t{}\n\t{}^'.format(note, format_spec, ' ' * start[1]))
67134eab
JMF
1897 return SyntaxError(message)
1898
1899 PICKFIRST = 'PICKFIRST'
1900 MERGE = 'MERGE'
1901 SINGLE = 'SINGLE'
0130afb7 1902 GROUP = 'GROUP'
67134eab
JMF
1903 FormatSelector = collections.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
1904
91ebc640 1905 allow_multiple_streams = {'audio': self.params.get('allow_multiple_audio_streams', False),
1906 'video': self.params.get('allow_multiple_video_streams', False)}
909d24dd 1907
9f1a1c36 1908 check_formats = self.params.get('check_formats') == 'selected'
e8e73840 1909
67134eab
JMF
1910 def _parse_filter(tokens):
1911 filter_parts = []
1912 for type, string, start, _, _ in tokens:
1913 if type == tokenize.OP and string == ']':
1914 return ''.join(filter_parts)
1915 else:
1916 filter_parts.append(string)
1917
232541df 1918 def _remove_unused_ops(tokens):
17cc1534 1919 # Remove operators that we don't use and join them with the surrounding strings
232541df
JMF
1920 # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
1921 ALLOWED_OPS = ('/', '+', ',', '(', ')')
1922 last_string, last_start, last_end, last_line = None, None, None, None
1923 for type, string, start, end, line in tokens:
1924 if type == tokenize.OP and string == '[':
1925 if last_string:
1926 yield tokenize.NAME, last_string, last_start, last_end, last_line
1927 last_string = None
1928 yield type, string, start, end, line
1929 # everything inside brackets will be handled by _parse_filter
1930 for type, string, start, end, line in tokens:
1931 yield type, string, start, end, line
1932 if type == tokenize.OP and string == ']':
1933 break
1934 elif type == tokenize.OP and string in ALLOWED_OPS:
1935 if last_string:
1936 yield tokenize.NAME, last_string, last_start, last_end, last_line
1937 last_string = None
1938 yield type, string, start, end, line
1939 elif type in [tokenize.NAME, tokenize.NUMBER, tokenize.OP]:
1940 if not last_string:
1941 last_string = string
1942 last_start = start
1943 last_end = end
1944 else:
1945 last_string += string
1946 if last_string:
1947 yield tokenize.NAME, last_string, last_start, last_end, last_line
1948
cf2ac6df 1949 def _parse_format_selection(tokens, inside_merge=False, inside_choice=False, inside_group=False):
67134eab
JMF
1950 selectors = []
1951 current_selector = None
1952 for type, string, start, _, _ in tokens:
1953 # ENCODING is only defined in python 3.x
1954 if type == getattr(tokenize, 'ENCODING', None):
1955 continue
1956 elif type in [tokenize.NAME, tokenize.NUMBER]:
1957 current_selector = FormatSelector(SINGLE, string, [])
1958 elif type == tokenize.OP:
cf2ac6df
JMF
1959 if string == ')':
1960 if not inside_group:
1961 # ')' will be handled by the parentheses group
1962 tokens.restore_last_token()
67134eab 1963 break
cf2ac6df 1964 elif inside_merge and string in ['/', ',']:
0130afb7
JMF
1965 tokens.restore_last_token()
1966 break
cf2ac6df
JMF
1967 elif inside_choice and string == ',':
1968 tokens.restore_last_token()
1969 break
1970 elif string == ',':
0a31a350
JMF
1971 if not current_selector:
1972 raise syntax_error('"," must follow a format selector', start)
67134eab
JMF
1973 selectors.append(current_selector)
1974 current_selector = None
1975 elif string == '/':
d96d604e
JMF
1976 if not current_selector:
1977 raise syntax_error('"/" must follow a format selector', start)
67134eab 1978 first_choice = current_selector
cf2ac6df 1979 second_choice = _parse_format_selection(tokens, inside_choice=True)
f5f4a27a 1980 current_selector = FormatSelector(PICKFIRST, (first_choice, second_choice), [])
67134eab
JMF
1981 elif string == '[':
1982 if not current_selector:
1983 current_selector = FormatSelector(SINGLE, 'best', [])
1984 format_filter = _parse_filter(tokens)
1985 current_selector.filters.append(format_filter)
0130afb7
JMF
1986 elif string == '(':
1987 if current_selector:
1988 raise syntax_error('Unexpected "("', start)
cf2ac6df
JMF
1989 group = _parse_format_selection(tokens, inside_group=True)
1990 current_selector = FormatSelector(GROUP, group, [])
67134eab 1991 elif string == '+':
d03cfdce 1992 if not current_selector:
1993 raise syntax_error('Unexpected "+"', start)
1994 selector_1 = current_selector
1995 selector_2 = _parse_format_selection(tokens, inside_merge=True)
1996 if not selector_2:
1997 raise syntax_error('Expected a selector', start)
1998 current_selector = FormatSelector(MERGE, (selector_1, selector_2), [])
67134eab 1999 else:
86e5f3ed 2000 raise syntax_error(f'Operator not recognized: "{string}"', start)
67134eab
JMF
2001 elif type == tokenize.ENDMARKER:
2002 break
2003 if current_selector:
2004 selectors.append(current_selector)
2005 return selectors
2006
f8d4ad9a 2007 def _merge(formats_pair):
2008 format_1, format_2 = formats_pair
2009
2010 formats_info = []
2011 formats_info.extend(format_1.get('requested_formats', (format_1,)))
2012 formats_info.extend(format_2.get('requested_formats', (format_2,)))
2013
2014 if not allow_multiple_streams['video'] or not allow_multiple_streams['audio']:
551f9388 2015 get_no_more = {'video': False, 'audio': False}
f8d4ad9a 2016 for (i, fmt_info) in enumerate(formats_info):
551f9388 2017 if fmt_info.get('acodec') == fmt_info.get('vcodec') == 'none':
2018 formats_info.pop(i)
2019 continue
2020 for aud_vid in ['audio', 'video']:
f8d4ad9a 2021 if not allow_multiple_streams[aud_vid] and fmt_info.get(aud_vid[0] + 'codec') != 'none':
2022 if get_no_more[aud_vid]:
2023 formats_info.pop(i)
f5510afe 2024 break
f8d4ad9a 2025 get_no_more[aud_vid] = True
2026
2027 if len(formats_info) == 1:
2028 return formats_info[0]
2029
2030 video_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('vcodec') != 'none']
2031 audio_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('acodec') != 'none']
2032
2033 the_only_video = video_fmts[0] if len(video_fmts) == 1 else None
2034 the_only_audio = audio_fmts[0] if len(audio_fmts) == 1 else None
2035
2036 output_ext = self.params.get('merge_output_format')
2037 if not output_ext:
2038 if the_only_video:
2039 output_ext = the_only_video['ext']
2040 elif the_only_audio and not video_fmts:
2041 output_ext = the_only_audio['ext']
2042 else:
2043 output_ext = 'mkv'
2044
975a0d0d 2045 filtered = lambda *keys: filter(None, (traverse_obj(fmt, *keys) for fmt in formats_info))
2046
f8d4ad9a 2047 new_dict = {
2048 'requested_formats': formats_info,
975a0d0d 2049 'format': '+'.join(filtered('format')),
2050 'format_id': '+'.join(filtered('format_id')),
f8d4ad9a 2051 'ext': output_ext,
975a0d0d 2052 'protocol': '+'.join(map(determine_protocol, formats_info)),
093a1710 2053 'language': '+'.join(orderedSet(filtered('language'))) or None,
2054 'format_note': '+'.join(orderedSet(filtered('format_note'))) or None,
2055 'filesize_approx': sum(filtered('filesize', 'filesize_approx')) or None,
975a0d0d 2056 'tbr': sum(filtered('tbr', 'vbr', 'abr')),
f8d4ad9a 2057 }
2058
2059 if the_only_video:
2060 new_dict.update({
2061 'width': the_only_video.get('width'),
2062 'height': the_only_video.get('height'),
2063 'resolution': the_only_video.get('resolution') or self.format_resolution(the_only_video),
2064 'fps': the_only_video.get('fps'),
49a57e70 2065 'dynamic_range': the_only_video.get('dynamic_range'),
f8d4ad9a 2066 'vcodec': the_only_video.get('vcodec'),
2067 'vbr': the_only_video.get('vbr'),
2068 'stretched_ratio': the_only_video.get('stretched_ratio'),
2069 })
2070
2071 if the_only_audio:
2072 new_dict.update({
2073 'acodec': the_only_audio.get('acodec'),
2074 'abr': the_only_audio.get('abr'),
975a0d0d 2075 'asr': the_only_audio.get('asr'),
f8d4ad9a 2076 })
2077
2078 return new_dict
2079
e8e73840 2080 def _check_formats(formats):
981052c9 2081 if not check_formats:
2082 yield from formats
b5ac45b1 2083 return
9f1a1c36 2084 yield from self._check_formats(formats)
e8e73840 2085
67134eab 2086 def _build_selector_function(selector):
909d24dd 2087 if isinstance(selector, list): # ,
67134eab
JMF
2088 fs = [_build_selector_function(s) for s in selector]
2089
317f7ab6 2090 def selector_function(ctx):
67134eab 2091 for f in fs:
981052c9 2092 yield from f(ctx)
67134eab 2093 return selector_function
909d24dd 2094
2095 elif selector.type == GROUP: # ()
0130afb7 2096 selector_function = _build_selector_function(selector.selector)
909d24dd 2097
2098 elif selector.type == PICKFIRST: # /
67134eab
JMF
2099 fs = [_build_selector_function(s) for s in selector.selector]
2100
317f7ab6 2101 def selector_function(ctx):
67134eab 2102 for f in fs:
317f7ab6 2103 picked_formats = list(f(ctx))
67134eab
JMF
2104 if picked_formats:
2105 return picked_formats
2106 return []
67134eab 2107
981052c9 2108 elif selector.type == MERGE: # +
2109 selector_1, selector_2 = map(_build_selector_function, selector.selector)
2110
2111 def selector_function(ctx):
adbc4ec4 2112 for pair in itertools.product(selector_1(ctx), selector_2(ctx)):
981052c9 2113 yield _merge(pair)
2114
909d24dd 2115 elif selector.type == SINGLE: # atom
598d185d 2116 format_spec = selector.selector or 'best'
909d24dd 2117
f8d4ad9a 2118 # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
909d24dd 2119 if format_spec == 'all':
2120 def selector_function(ctx):
9222c381 2121 yield from _check_formats(ctx['formats'][::-1])
f8d4ad9a 2122 elif format_spec == 'mergeall':
2123 def selector_function(ctx):
316f2650 2124 formats = list(_check_formats(
2125 f for f in ctx['formats'] if f.get('vcodec') != 'none' or f.get('acodec') != 'none'))
e01d6aa4 2126 if not formats:
2127 return
921b76ca 2128 merged_format = formats[-1]
2129 for f in formats[-2::-1]:
f8d4ad9a 2130 merged_format = _merge((merged_format, f))
2131 yield merged_format
909d24dd 2132
2133 else:
85e801a9 2134 format_fallback, seperate_fallback, format_reverse, format_idx = False, None, True, 1
eff63539 2135 mobj = re.match(
2136 r'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
2137 format_spec)
2138 if mobj is not None:
2139 format_idx = int_or_none(mobj.group('n'), default=1)
e8e73840 2140 format_reverse = mobj.group('bw')[0] == 'b'
eff63539 2141 format_type = (mobj.group('type') or [None])[0]
2142 not_format_type = {'v': 'a', 'a': 'v'}.get(format_type)
2143 format_modified = mobj.group('mod') is not None
909d24dd 2144
2145 format_fallback = not format_type and not format_modified # for b, w
8326b00a 2146 _filter_f = (
eff63539 2147 (lambda f: f.get('%scodec' % format_type) != 'none')
2148 if format_type and format_modified # bv*, ba*, wv*, wa*
2149 else (lambda f: f.get('%scodec' % not_format_type) == 'none')
2150 if format_type # bv, ba, wv, wa
2151 else (lambda f: f.get('vcodec') != 'none' and f.get('acodec') != 'none')
2152 if not format_modified # b, w
8326b00a 2153 else lambda f: True) # b*, w*
2154 filter_f = lambda f: _filter_f(f) and (
2155 f.get('vcodec') != 'none' or f.get('acodec') != 'none')
67134eab 2156 else:
48ee10ee 2157 if format_spec in self._format_selection_exts['audio']:
b11c04a8 2158 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none'
48ee10ee 2159 elif format_spec in self._format_selection_exts['video']:
b11c04a8 2160 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none' and f.get('vcodec') != 'none'
85e801a9 2161 seperate_fallback = lambda f: f.get('ext') == format_spec and f.get('vcodec') != 'none'
48ee10ee 2162 elif format_spec in self._format_selection_exts['storyboards']:
b11c04a8 2163 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') == 'none' and f.get('vcodec') == 'none'
2164 else:
b5ae35ee 2165 filter_f = lambda f: f.get('format_id') == format_spec # id
909d24dd 2166
2167 def selector_function(ctx):
2168 formats = list(ctx['formats'])
909d24dd 2169 matches = list(filter(filter_f, formats)) if filter_f is not None else formats
85e801a9 2170 if not matches:
2171 if format_fallback and ctx['incomplete_formats']:
2172 # for extractors with incomplete formats (audio only (soundcloud)
2173 # or video only (imgur)) best/worst will fallback to
2174 # best/worst {video,audio}-only format
2175 matches = formats
2176 elif seperate_fallback and not ctx['has_merged_format']:
2177 # for compatibility with youtube-dl when there is no pre-merged format
2178 matches = list(filter(seperate_fallback, formats))
981052c9 2179 matches = LazyList(_check_formats(matches[::-1 if format_reverse else 1]))
2180 try:
e8e73840 2181 yield matches[format_idx - 1]
4abea8ca 2182 except LazyList.IndexError:
981052c9 2183 return
083c9df9 2184
67134eab 2185 filters = [self._build_format_filter(f) for f in selector.filters]
083c9df9 2186
317f7ab6 2187 def final_selector(ctx):
adbc4ec4 2188 ctx_copy = dict(ctx)
67134eab 2189 for _filter in filters:
317f7ab6
S
2190 ctx_copy['formats'] = list(filter(_filter, ctx_copy['formats']))
2191 return selector_function(ctx_copy)
67134eab 2192 return final_selector
083c9df9 2193
0f06bcd7 2194 stream = io.BytesIO(format_spec.encode())
0130afb7 2195 try:
f9934b96 2196 tokens = list(_remove_unused_ops(tokenize.tokenize(stream.readline)))
0130afb7
JMF
2197 except tokenize.TokenError:
2198 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec)))
2199
86e5f3ed 2200 class TokenIterator:
0130afb7
JMF
2201 def __init__(self, tokens):
2202 self.tokens = tokens
2203 self.counter = 0
2204
2205 def __iter__(self):
2206 return self
2207
2208 def __next__(self):
2209 if self.counter >= len(self.tokens):
2210 raise StopIteration()
2211 value = self.tokens[self.counter]
2212 self.counter += 1
2213 return value
2214
2215 next = __next__
2216
2217 def restore_last_token(self):
2218 self.counter -= 1
2219
2220 parsed_selector = _parse_format_selection(iter(TokenIterator(tokens)))
67134eab 2221 return _build_selector_function(parsed_selector)
a9c58ad9 2222
e5660ee6 2223 def _calc_headers(self, info_dict):
8b7539d2 2224 res = merge_headers(self.params['http_headers'], info_dict.get('http_headers') or {})
e5660ee6 2225
c487cf00 2226 cookies = self._calc_cookies(info_dict['url'])
e5660ee6
JMF
2227 if cookies:
2228 res['Cookie'] = cookies
2229
0016b84e
S
2230 if 'X-Forwarded-For' not in res:
2231 x_forwarded_for_ip = info_dict.get('__x_forwarded_for_ip')
2232 if x_forwarded_for_ip:
2233 res['X-Forwarded-For'] = x_forwarded_for_ip
2234
e5660ee6
JMF
2235 return res
2236
c487cf00 2237 def _calc_cookies(self, url):
2238 pr = sanitized_Request(url)
e5660ee6 2239 self.cookiejar.add_cookie_header(pr)
662435f7 2240 return pr.get_header('Cookie')
e5660ee6 2241
9f1a1c36 2242 def _sort_thumbnails(self, thumbnails):
2243 thumbnails.sort(key=lambda t: (
2244 t.get('preference') if t.get('preference') is not None else -1,
2245 t.get('width') if t.get('width') is not None else -1,
2246 t.get('height') if t.get('height') is not None else -1,
2247 t.get('id') if t.get('id') is not None else '',
2248 t.get('url')))
2249
b0249bca 2250 def _sanitize_thumbnails(self, info_dict):
bc516a3f 2251 thumbnails = info_dict.get('thumbnails')
2252 if thumbnails is None:
2253 thumbnail = info_dict.get('thumbnail')
2254 if thumbnail:
2255 info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}]
9f1a1c36 2256 if not thumbnails:
2257 return
2258
2259 def check_thumbnails(thumbnails):
2260 for t in thumbnails:
2261 self.to_screen(f'[info] Testing thumbnail {t["id"]}')
2262 try:
2263 self.urlopen(HEADRequest(t['url']))
2264 except network_exceptions as err:
2265 self.to_screen(f'[info] Unable to connect to thumbnail {t["id"]} URL {t["url"]!r} - {err}. Skipping...')
2266 continue
2267 yield t
2268
2269 self._sort_thumbnails(thumbnails)
2270 for i, t in enumerate(thumbnails):
2271 if t.get('id') is None:
2272 t['id'] = '%d' % i
2273 if t.get('width') and t.get('height'):
2274 t['resolution'] = '%dx%d' % (t['width'], t['height'])
2275 t['url'] = sanitize_url(t['url'])
2276
2277 if self.params.get('check_formats') is True:
282f5709 2278 info_dict['thumbnails'] = LazyList(check_thumbnails(thumbnails[::-1]), reverse=True)
9f1a1c36 2279 else:
2280 info_dict['thumbnails'] = thumbnails
bc516a3f 2281
03f83004
LNO
2282 def _fill_common_fields(self, info_dict, is_video=True):
2283 # TODO: move sanitization here
2284 if is_video:
2285 # playlists are allowed to lack "title"
d4736fdb 2286 title = info_dict.get('title', NO_DEFAULT)
2287 if title is NO_DEFAULT:
03f83004
LNO
2288 raise ExtractorError('Missing "title" field in extractor result',
2289 video_id=info_dict['id'], ie=info_dict['extractor'])
d4736fdb 2290 info_dict['fulltitle'] = title
2291 if not title:
2292 if title == '':
2293 self.write_debug('Extractor gave empty title. Creating a generic title')
2294 else:
2295 self.report_warning('Extractor failed to obtain "title". Creating a generic title instead')
1d485a1a 2296 info_dict['title'] = f'{info_dict["extractor"].replace(":", "-")} video #{info_dict["id"]}'
03f83004
LNO
2297
2298 if info_dict.get('duration') is not None:
2299 info_dict['duration_string'] = formatSeconds(info_dict['duration'])
2300
2301 for ts_key, date_key in (
2302 ('timestamp', 'upload_date'),
2303 ('release_timestamp', 'release_date'),
2304 ('modified_timestamp', 'modified_date'),
2305 ):
2306 if info_dict.get(date_key) is None and info_dict.get(ts_key) is not None:
2307 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
2308 # see http://bugs.python.org/issue1646728)
19a03940 2309 with contextlib.suppress(ValueError, OverflowError, OSError):
03f83004
LNO
2310 upload_date = datetime.datetime.utcfromtimestamp(info_dict[ts_key])
2311 info_dict[date_key] = upload_date.strftime('%Y%m%d')
03f83004
LNO
2312
2313 live_keys = ('is_live', 'was_live')
2314 live_status = info_dict.get('live_status')
2315 if live_status is None:
2316 for key in live_keys:
2317 if info_dict.get(key) is False:
2318 continue
2319 if info_dict.get(key):
2320 live_status = key
2321 break
2322 if all(info_dict.get(key) is False for key in live_keys):
2323 live_status = 'not_live'
2324 if live_status:
2325 info_dict['live_status'] = live_status
2326 for key in live_keys:
2327 if info_dict.get(key) is None:
2328 info_dict[key] = (live_status == key)
2329
2330 # Auto generate title fields corresponding to the *_number fields when missing
2331 # in order to always have clean titles. This is very common for TV series.
2332 for field in ('chapter', 'season', 'episode'):
2333 if info_dict.get('%s_number' % field) is not None and not info_dict.get(field):
2334 info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field])
2335
415f8d51 2336 def _raise_pending_errors(self, info):
2337 err = info.pop('__pending_error', None)
2338 if err:
2339 self.report_error(err, tb=False)
2340
dd82ffea
JMF
2341 def process_video_result(self, info_dict, download=True):
2342 assert info_dict.get('_type', 'video') == 'video'
9c906919 2343 self._num_videos += 1
dd82ffea 2344
bec1fad2 2345 if 'id' not in info_dict:
fc08bdd6 2346 raise ExtractorError('Missing "id" field in extractor result', ie=info_dict['extractor'])
2347 elif not info_dict.get('id'):
2348 raise ExtractorError('Extractor failed to obtain "id"', ie=info_dict['extractor'])
455a15e2 2349
c9969434
S
2350 def report_force_conversion(field, field_not, conversion):
2351 self.report_warning(
2352 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
2353 % (field, field_not, conversion))
2354
2355 def sanitize_string_field(info, string_field):
2356 field = info.get(string_field)
2357 if field is None or isinstance(field, compat_str):
2358 return
2359 report_force_conversion(string_field, 'a string', 'string')
2360 info[string_field] = compat_str(field)
2361
2362 def sanitize_numeric_fields(info):
2363 for numeric_field in self._NUMERIC_FIELDS:
2364 field = info.get(numeric_field)
f9934b96 2365 if field is None or isinstance(field, (int, float)):
c9969434
S
2366 continue
2367 report_force_conversion(numeric_field, 'numeric', 'int')
2368 info[numeric_field] = int_or_none(field)
2369
2370 sanitize_string_field(info_dict, 'id')
2371 sanitize_numeric_fields(info_dict)
4c3f8c3f 2372 if (info_dict.get('duration') or 0) <= 0 and info_dict.pop('duration', None):
50e93e03 2373 self.report_warning('"duration" field is negative, there is an error in extractor')
be6217b2 2374
dd82ffea
JMF
2375 if 'playlist' not in info_dict:
2376 # It isn't part of a playlist
2377 info_dict['playlist'] = None
2378 info_dict['playlist_index'] = None
2379
bc516a3f 2380 self._sanitize_thumbnails(info_dict)
d5519808 2381
536a55da 2382 thumbnail = info_dict.get('thumbnail')
bc516a3f 2383 thumbnails = info_dict.get('thumbnails')
536a55da
S
2384 if thumbnail:
2385 info_dict['thumbnail'] = sanitize_url(thumbnail)
2386 elif thumbnails:
d5519808
PH
2387 info_dict['thumbnail'] = thumbnails[-1]['url']
2388
ae30b840 2389 if info_dict.get('display_id') is None and 'id' in info_dict:
0afef30b
PH
2390 info_dict['display_id'] = info_dict['id']
2391
03f83004 2392 self._fill_common_fields(info_dict)
33d2fc2f 2393
05108a49
S
2394 for cc_kind in ('subtitles', 'automatic_captions'):
2395 cc = info_dict.get(cc_kind)
2396 if cc:
2397 for _, subtitle in cc.items():
2398 for subtitle_format in subtitle:
2399 if subtitle_format.get('url'):
2400 subtitle_format['url'] = sanitize_url(subtitle_format['url'])
2401 if subtitle_format.get('ext') is None:
2402 subtitle_format['ext'] = determine_ext(subtitle_format['url']).lower()
2403
2404 automatic_captions = info_dict.get('automatic_captions')
4bba3716 2405 subtitles = info_dict.get('subtitles')
4bba3716 2406
360e1ca5 2407 info_dict['requested_subtitles'] = self.process_subtitles(
05108a49 2408 info_dict['id'], subtitles, automatic_captions)
a504ced0 2409
dd82ffea
JMF
2410 if info_dict.get('formats') is None:
2411 # There's only one format available
2412 formats = [info_dict]
2413 else:
2414 formats = info_dict['formats']
2415
0a5a191a 2416 # or None ensures --clean-infojson removes it
2417 info_dict['_has_drm'] = any(f.get('has_drm') for f in formats) or None
88acdbc2 2418 if not self.params.get('allow_unplayable_formats'):
2419 formats = [f for f in formats if not f.get('has_drm')]
0a5a191a 2420 if info_dict['_has_drm'] and all(
c0b6e5c7 2421 f.get('acodec') == f.get('vcodec') == 'none' for f in formats):
2422 self.report_warning(
2423 'This video is DRM protected and only images are available for download. '
2424 'Use --list-formats to see them')
88acdbc2 2425
319b6059 2426 get_from_start = not info_dict.get('is_live') or bool(self.params.get('live_from_start'))
2427 if not get_from_start:
2428 info_dict['title'] += ' ' + datetime.datetime.now().strftime('%Y-%m-%d %H:%M')
2429 if info_dict.get('is_live') and formats:
adbc4ec4 2430 formats = [f for f in formats if bool(f.get('is_from_start')) == get_from_start]
319b6059 2431 if get_from_start and not formats:
a44ca5a4 2432 self.raise_no_formats(info_dict, msg=(
2433 '--live-from-start is passed, but there are no formats that can be downloaded from the start. '
2434 'If you want to download from the current time, use --no-live-from-start'))
adbc4ec4 2435
db95dc13 2436 if not formats:
1151c407 2437 self.raise_no_formats(info_dict)
db95dc13 2438
73af5cc8
S
2439 def is_wellformed(f):
2440 url = f.get('url')
a5ac0c47 2441 if not url:
73af5cc8
S
2442 self.report_warning(
2443 '"url" field is missing or empty - skipping format, '
2444 'there is an error in extractor')
a5ac0c47
S
2445 return False
2446 if isinstance(url, bytes):
2447 sanitize_string_field(f, 'url')
2448 return True
73af5cc8
S
2449
2450 # Filter out malformed formats for better extraction robustness
2451 formats = list(filter(is_wellformed, formats))
2452
181c7053
S
2453 formats_dict = {}
2454
dd82ffea 2455 # We check that all the formats have the format and format_id fields
db95dc13 2456 for i, format in enumerate(formats):
c9969434
S
2457 sanitize_string_field(format, 'format_id')
2458 sanitize_numeric_fields(format)
dcf77cf1 2459 format['url'] = sanitize_url(format['url'])
e74e3b63 2460 if not format.get('format_id'):
8016c922 2461 format['format_id'] = compat_str(i)
e2effb08
S
2462 else:
2463 # Sanitize format_id from characters used in format selector expression
ec85ded8 2464 format['format_id'] = re.sub(r'[\s,/+\[\]()]', '_', format['format_id'])
181c7053
S
2465 format_id = format['format_id']
2466 if format_id not in formats_dict:
2467 formats_dict[format_id] = []
2468 formats_dict[format_id].append(format)
2469
2470 # Make sure all formats have unique format_id
03b4de72 2471 common_exts = set(itertools.chain(*self._format_selection_exts.values()))
181c7053 2472 for format_id, ambiguous_formats in formats_dict.items():
48ee10ee 2473 ambigious_id = len(ambiguous_formats) > 1
2474 for i, format in enumerate(ambiguous_formats):
2475 if ambigious_id:
181c7053 2476 format['format_id'] = '%s-%d' % (format_id, i)
48ee10ee 2477 if format.get('ext') is None:
2478 format['ext'] = determine_ext(format['url']).lower()
2479 # Ensure there is no conflict between id and ext in format selection
2480 # See https://github.com/yt-dlp/yt-dlp/issues/1282
2481 if format['format_id'] != format['ext'] and format['format_id'] in common_exts:
2482 format['format_id'] = 'f%s' % format['format_id']
181c7053
S
2483
2484 for i, format in enumerate(formats):
8c51aa65 2485 if format.get('format') is None:
6febd1c1 2486 format['format'] = '{id} - {res}{note}'.format(
8c51aa65
JMF
2487 id=format['format_id'],
2488 res=self.format_resolution(format),
b868936c 2489 note=format_field(format, 'format_note', ' (%s)'),
8c51aa65 2490 )
6f0be937 2491 if format.get('protocol') is None:
b5559424 2492 format['protocol'] = determine_protocol(format)
239df021 2493 if format.get('resolution') is None:
2494 format['resolution'] = self.format_resolution(format, default=None)
176f1866 2495 if format.get('dynamic_range') is None and format.get('vcodec') != 'none':
2496 format['dynamic_range'] = 'SDR'
f2fe69c7 2497 if (info_dict.get('duration') and format.get('tbr')
2498 and not format.get('filesize') and not format.get('filesize_approx')):
56ba69e4 2499 format['filesize_approx'] = int(info_dict['duration'] * format['tbr'] * (1024 / 8))
f2fe69c7 2500
e5660ee6
JMF
2501 # Add HTTP headers, so that external programs can use them from the
2502 # json output
2503 full_format_info = info_dict.copy()
2504 full_format_info.update(format)
2505 format['http_headers'] = self._calc_headers(full_format_info)
0016b84e
S
2506 # Remove private housekeeping stuff
2507 if '__x_forwarded_for_ip' in info_dict:
2508 del info_dict['__x_forwarded_for_ip']
dd82ffea 2509
9f1a1c36 2510 if self.params.get('check_formats') is True:
282f5709 2511 formats = LazyList(self._check_formats(formats[::-1]), reverse=True)
9f1a1c36 2512
88acdbc2 2513 if not formats or formats[0] is not info_dict:
b3d9ef88
JMF
2514 # only set the 'formats' fields if the original info_dict list them
2515 # otherwise we end up with a circular reference, the first (and unique)
f89197d7 2516 # element in the 'formats' field in info_dict is info_dict itself,
dfb1b146 2517 # which can't be exported to json
b3d9ef88 2518 info_dict['formats'] = formats
4ec82a72 2519
2520 info_dict, _ = self.pre_process(info_dict)
2521
6db9c4d5 2522 if self._match_entry(info_dict, incomplete=self._format_fields) is not None:
09b49e1f 2523 return info_dict
2524
2525 self.post_extract(info_dict)
2526 info_dict, _ = self.pre_process(info_dict, 'after_filter')
2527
093a1710 2528 # The pre-processors may have modified the formats
2529 formats = info_dict.get('formats', [info_dict])
2530
fa9f30b8 2531 list_only = self.params.get('simulate') is None and (
2532 self.params.get('list_thumbnails') or self.params.get('listformats') or self.params.get('listsubtitles'))
2533 interactive_format_selection = not list_only and self.format_selector == '-'
b7b04c78 2534 if self.params.get('list_thumbnails'):
2535 self.list_thumbnails(info_dict)
b7b04c78 2536 if self.params.get('listsubtitles'):
2537 if 'automatic_captions' in info_dict:
2538 self.list_subtitles(
2539 info_dict['id'], automatic_captions, 'automatic captions')
2540 self.list_subtitles(info_dict['id'], subtitles, 'subtitles')
fa9f30b8 2541 if self.params.get('listformats') or interactive_format_selection:
b69fd25c 2542 self.list_formats(info_dict)
169dbde9 2543 if list_only:
b7b04c78 2544 # Without this printing, -F --print-json will not work
169dbde9 2545 self.__forced_printings(info_dict, self.prepare_filename(info_dict), incomplete=True)
c487cf00 2546 return info_dict
bfaae0a7 2547
187986a8 2548 format_selector = self.format_selector
2549 if format_selector is None:
0017d9ad 2550 req_format = self._default_format_spec(info_dict, download=download)
0760b0a7 2551 self.write_debug('Default format spec: %s' % req_format)
187986a8 2552 format_selector = self.build_format_selector(req_format)
317f7ab6 2553
fa9f30b8 2554 while True:
2555 if interactive_format_selection:
2556 req_format = input(
2557 self._format_screen('\nEnter format selector: ', self.Styles.EMPHASIS))
2558 try:
2559 format_selector = self.build_format_selector(req_format)
2560 except SyntaxError as err:
2561 self.report_error(err, tb=False, is_error=False)
2562 continue
2563
85e801a9 2564 formats_to_download = list(format_selector({
fa9f30b8 2565 'formats': formats,
85e801a9 2566 'has_merged_format': any('none' not in (f.get('acodec'), f.get('vcodec')) for f in formats),
2567 'incomplete_formats': (
2568 # All formats are video-only or
2569 all(f.get('vcodec') != 'none' and f.get('acodec') == 'none' for f in formats)
2570 # all formats are audio-only
2571 or all(f.get('vcodec') == 'none' and f.get('acodec') != 'none' for f in formats)),
2572 }))
fa9f30b8 2573 if interactive_format_selection and not formats_to_download:
2574 self.report_error('Requested format is not available', tb=False, is_error=False)
2575 continue
2576 break
317f7ab6 2577
dd82ffea 2578 if not formats_to_download:
b7da73eb 2579 if not self.params.get('ignore_no_formats_error'):
c0b6e5c7 2580 raise ExtractorError(
2581 'Requested format is not available. Use --list-formats for a list of available formats',
2582 expected=True, video_id=info_dict['id'], ie=info_dict['extractor'])
b62fa6d7 2583 self.report_warning('Requested format is not available')
2584 # Process what we can, even without any available formats.
2585 formats_to_download = [{}]
a13e6848 2586
5ec1b6b7 2587 requested_ranges = self.params.get('download_ranges')
2588 if requested_ranges:
2589 requested_ranges = tuple(requested_ranges(info_dict, self))
2590
2591 best_format, downloaded_formats = formats_to_download[-1], []
b62fa6d7 2592 if download:
2593 if best_format:
5ec1b6b7 2594 def to_screen(*msg):
2595 self.to_screen(f'[info] {info_dict["id"]}: {" ".join(", ".join(variadic(m)) for m in msg)}')
2596
2597 to_screen(f'Downloading {len(formats_to_download)} format(s):',
2598 (f['format_id'] for f in formats_to_download))
2599 if requested_ranges:
2600 to_screen(f'Downloading {len(requested_ranges)} time ranges:',
2601 (f'{int(c["start_time"])}-{int(c["end_time"])}' for c in requested_ranges))
a13e6848 2602 max_downloads_reached = False
5ec1b6b7 2603
2604 for fmt, chapter in itertools.product(formats_to_download, requested_ranges or [{}]):
2605 new_info = self._copy_infodict(info_dict)
b7da73eb 2606 new_info.update(fmt)
5ec1b6b7 2607 if chapter:
2608 new_info.update({
2609 'section_start': chapter.get('start_time'),
2610 'section_end': chapter.get('end_time', 0),
2611 'section_title': chapter.get('title'),
2612 'section_number': chapter.get('index'),
2613 })
2614 downloaded_formats.append(new_info)
a13e6848 2615 try:
2616 self.process_info(new_info)
2617 except MaxDownloadsReached:
2618 max_downloads_reached = True
415f8d51 2619 self._raise_pending_errors(new_info)
f46e2f9d 2620 # Remove copied info
2621 for key, val in tuple(new_info.items()):
2622 if info_dict.get(key) == val:
2623 new_info.pop(key)
a13e6848 2624 if max_downloads_reached:
2625 break
ebed8b37 2626
5ec1b6b7 2627 write_archive = {f.get('__write_download_archive', False) for f in downloaded_formats}
a13e6848 2628 assert write_archive.issubset({True, False, 'ignore'})
2629 if True in write_archive and False not in write_archive:
2630 self.record_download_archive(info_dict)
be72c624 2631
5ec1b6b7 2632 info_dict['requested_downloads'] = downloaded_formats
ed5835b4 2633 info_dict = self.run_all_pps('after_video', info_dict)
a13e6848 2634 if max_downloads_reached:
2635 raise MaxDownloadsReached()
ebed8b37 2636
49a57e70 2637 # We update the info dict with the selected best quality format (backwards compatibility)
be72c624 2638 info_dict.update(best_format)
dd82ffea
JMF
2639 return info_dict
2640
98c70d6f 2641 def process_subtitles(self, video_id, normal_subtitles, automatic_captions):
a504ced0 2642 """Select the requested subtitles and their format"""
d8a58ddc 2643 available_subs, normal_sub_langs = {}, []
98c70d6f
JMF
2644 if normal_subtitles and self.params.get('writesubtitles'):
2645 available_subs.update(normal_subtitles)
d8a58ddc 2646 normal_sub_langs = tuple(normal_subtitles.keys())
98c70d6f
JMF
2647 if automatic_captions and self.params.get('writeautomaticsub'):
2648 for lang, cap_info in automatic_captions.items():
360e1ca5
JMF
2649 if lang not in available_subs:
2650 available_subs[lang] = cap_info
2651
4d171848
JMF
2652 if (not self.params.get('writesubtitles') and not
2653 self.params.get('writeautomaticsub') or not
2654 available_subs):
2655 return None
a504ced0 2656
d8a58ddc 2657 all_sub_langs = tuple(available_subs.keys())
a504ced0 2658 if self.params.get('allsubtitles', False):
c32b0aab 2659 requested_langs = all_sub_langs
2660 elif self.params.get('subtitleslangs', False):
77c4a9ef 2661 # A list is used so that the order of languages will be the same as
2662 # given in subtitleslangs. See https://github.com/yt-dlp/yt-dlp/issues/1041
2663 requested_langs = []
2664 for lang_re in self.params.get('subtitleslangs'):
77c4a9ef 2665 discard = lang_re[0] == '-'
c32b0aab 2666 if discard:
77c4a9ef 2667 lang_re = lang_re[1:]
3aa91540 2668 if lang_re == 'all':
2669 if discard:
2670 requested_langs = []
2671 else:
2672 requested_langs.extend(all_sub_langs)
2673 continue
77c4a9ef 2674 current_langs = filter(re.compile(lang_re + '$').match, all_sub_langs)
c32b0aab 2675 if discard:
2676 for lang in current_langs:
77c4a9ef 2677 while lang in requested_langs:
2678 requested_langs.remove(lang)
c32b0aab 2679 else:
77c4a9ef 2680 requested_langs.extend(current_langs)
2681 requested_langs = orderedSet(requested_langs)
d8a58ddc 2682 elif normal_sub_langs:
2683 requested_langs = ['en'] if 'en' in normal_sub_langs else normal_sub_langs[:1]
a504ced0 2684 else:
d8a58ddc 2685 requested_langs = ['en'] if 'en' in all_sub_langs else all_sub_langs[:1]
ad3dc496 2686 if requested_langs:
2687 self.write_debug('Downloading subtitles: %s' % ', '.join(requested_langs))
a504ced0
JMF
2688
2689 formats_query = self.params.get('subtitlesformat', 'best')
2690 formats_preference = formats_query.split('/') if formats_query else []
2691 subs = {}
2692 for lang in requested_langs:
2693 formats = available_subs.get(lang)
2694 if formats is None:
86e5f3ed 2695 self.report_warning(f'{lang} subtitles not available for {video_id}')
a504ced0 2696 continue
a504ced0
JMF
2697 for ext in formats_preference:
2698 if ext == 'best':
2699 f = formats[-1]
2700 break
2701 matches = list(filter(lambda f: f['ext'] == ext, formats))
2702 if matches:
2703 f = matches[-1]
2704 break
2705 else:
2706 f = formats[-1]
2707 self.report_warning(
2708 'No subtitle format found matching "%s" for language %s, '
2709 'using %s' % (formats_query, lang, f['ext']))
2710 subs[lang] = f
2711 return subs
2712
bb66c247 2713 def _forceprint(self, key, info_dict):
2714 if info_dict is None:
2715 return
2716 info_copy = info_dict.copy()
2717 info_copy['formats_table'] = self.render_formats_table(info_dict)
2718 info_copy['thumbnails_table'] = self.render_thumbnails_table(info_dict)
2719 info_copy['subtitles_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('subtitles'))
2720 info_copy['automatic_captions_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('automatic_captions'))
2721
2722 def format_tmpl(tmpl):
2723 mobj = re.match(r'\w+(=?)$', tmpl)
2724 if mobj and mobj.group(1):
2725 return f'{tmpl[:-1]} = %({tmpl[:-1]})r'
2726 elif mobj:
2727 return f'%({tmpl})s'
2728 return tmpl
8130779d 2729
bb66c247 2730 for tmpl in self.params['forceprint'].get(key, []):
2731 self.to_stdout(self.evaluate_outtmpl(format_tmpl(tmpl), info_copy))
2732
2733 for tmpl, file_tmpl in self.params['print_to_file'].get(key, []):
5127e92a 2734 filename = self.prepare_filename(info_dict, outtmpl=file_tmpl)
bb66c247 2735 tmpl = format_tmpl(tmpl)
2736 self.to_screen(f'[info] Writing {tmpl!r} to: {filename}')
8d93e69d 2737 if self._ensure_dir_exists(filename):
86e5f3ed 2738 with open(filename, 'a', encoding='utf-8') as f:
8d93e69d 2739 f.write(self.evaluate_outtmpl(tmpl, info_copy) + '\n')
ca30f449 2740
d06daf23 2741 def __forced_printings(self, info_dict, filename, incomplete):
53c18592 2742 def print_mandatory(field, actual_field=None):
2743 if actual_field is None:
2744 actual_field = field
d06daf23 2745 if (self.params.get('force%s' % field, False)
53c18592 2746 and (not incomplete or info_dict.get(actual_field) is not None)):
2747 self.to_stdout(info_dict[actual_field])
d06daf23
S
2748
2749 def print_optional(field):
2750 if (self.params.get('force%s' % field, False)
2751 and info_dict.get(field) is not None):
2752 self.to_stdout(info_dict[field])
2753
53c18592 2754 info_dict = info_dict.copy()
2755 if filename is not None:
2756 info_dict['filename'] = filename
2757 if info_dict.get('requested_formats') is not None:
2758 # For RTMP URLs, also include the playpath
2759 info_dict['urls'] = '\n'.join(f['url'] + f.get('play_path', '') for f in info_dict['requested_formats'])
10331a26 2760 elif info_dict.get('url'):
53c18592 2761 info_dict['urls'] = info_dict['url'] + info_dict.get('play_path', '')
2762
bb66c247 2763 if (self.params.get('forcejson')
2764 or self.params['forceprint'].get('video')
2765 or self.params['print_to_file'].get('video')):
2b8a2973 2766 self.post_extract(info_dict)
bb66c247 2767 self._forceprint('video', info_dict)
53c18592 2768
d06daf23
S
2769 print_mandatory('title')
2770 print_mandatory('id')
53c18592 2771 print_mandatory('url', 'urls')
d06daf23
S
2772 print_optional('thumbnail')
2773 print_optional('description')
53c18592 2774 print_optional('filename')
b868936c 2775 if self.params.get('forceduration') and info_dict.get('duration') is not None:
d06daf23
S
2776 self.to_stdout(formatSeconds(info_dict['duration']))
2777 print_mandatory('format')
53c18592 2778
2b8a2973 2779 if self.params.get('forcejson'):
6e84b215 2780 self.to_stdout(json.dumps(self.sanitize_info(info_dict)))
d06daf23 2781
e8e73840 2782 def dl(self, name, info, subtitle=False, test=False):
88acdbc2 2783 if not info.get('url'):
1151c407 2784 self.raise_no_formats(info, True)
e8e73840 2785
2786 if test:
2787 verbose = self.params.get('verbose')
2788 params = {
2789 'test': True,
a169858f 2790 'quiet': self.params.get('quiet') or not verbose,
e8e73840 2791 'verbose': verbose,
2792 'noprogress': not verbose,
2793 'nopart': True,
2794 'skip_unavailable_fragments': False,
2795 'keep_fragments': False,
2796 'overwrites': True,
2797 '_no_ytdl_file': True,
2798 }
2799 else:
2800 params = self.params
96fccc10 2801 fd = get_suitable_downloader(info, params, to_stdout=(name == '-'))(self, params)
e8e73840 2802 if not test:
2803 for ph in self._progress_hooks:
2804 fd.add_progress_hook(ph)
42676437
M
2805 urls = '", "'.join(
2806 (f['url'].split(',')[0] + ',<data>' if f['url'].startswith('data:') else f['url'])
2807 for f in info.get('requested_formats', []) or [info])
3a408f9d 2808 self.write_debug(f'Invoking {fd.FD_NAME} downloader on "{urls}"')
03b4de72 2809
adbc4ec4
THD
2810 # Note: Ideally info should be a deep-copied so that hooks cannot modify it.
2811 # But it may contain objects that are not deep-copyable
2812 new_info = self._copy_infodict(info)
e8e73840 2813 if new_info.get('http_headers') is None:
2814 new_info['http_headers'] = self._calc_headers(new_info)
2815 return fd.download(name, new_info, subtitle)
2816
e04938ab 2817 def existing_file(self, filepaths, *, default_overwrite=True):
2818 existing_files = list(filter(os.path.exists, orderedSet(filepaths)))
2819 if existing_files and not self.params.get('overwrites', default_overwrite):
2820 return existing_files[0]
2821
2822 for file in existing_files:
2823 self.report_file_delete(file)
2824 os.remove(file)
2825 return None
2826
8222d8de 2827 def process_info(self, info_dict):
09b49e1f 2828 """Process a single resolved IE result. (Modifies it in-place)"""
8222d8de
JMF
2829
2830 assert info_dict.get('_type', 'video') == 'video'
f46e2f9d 2831 original_infodict = info_dict
fd288278 2832
4513a41a 2833 if 'format' not in info_dict and 'ext' in info_dict:
8222d8de
JMF
2834 info_dict['format'] = info_dict['ext']
2835
09b49e1f 2836 # This is mostly just for backward compatibility of process_info
2837 # As a side-effect, this allows for format-specific filters
c77495e3 2838 if self._match_entry(info_dict) is not None:
9e907ebd 2839 info_dict['__write_download_archive'] = 'ignore'
8222d8de
JMF
2840 return
2841
09b49e1f 2842 # Does nothing under normal operation - for backward compatibility of process_info
277d6ff5 2843 self.post_extract(info_dict)
0c14d66a 2844 self._num_downloads += 1
8222d8de 2845
dcf64d43 2846 # info_dict['_filename'] needs to be set for backward compatibility
de6000d9 2847 info_dict['_filename'] = full_filename = self.prepare_filename(info_dict, warn=True)
2848 temp_filename = self.prepare_filename(info_dict, 'temp')
0202b52a 2849 files_to_move = {}
8222d8de
JMF
2850
2851 # Forced printings
4513a41a 2852 self.__forced_printings(info_dict, full_filename, incomplete=('format' not in info_dict))
8222d8de 2853
ca6d59d2 2854 def check_max_downloads():
2855 if self._num_downloads >= float(self.params.get('max_downloads') or 'inf'):
2856 raise MaxDownloadsReached()
2857
b7b04c78 2858 if self.params.get('simulate'):
9e907ebd 2859 info_dict['__write_download_archive'] = self.params.get('force_write_download_archive')
ca6d59d2 2860 check_max_downloads()
8222d8de
JMF
2861 return
2862
de6000d9 2863 if full_filename is None:
8222d8de 2864 return
e92caff5 2865 if not self._ensure_dir_exists(encodeFilename(full_filename)):
0202b52a 2866 return
e92caff5 2867 if not self._ensure_dir_exists(encodeFilename(temp_filename)):
8222d8de
JMF
2868 return
2869
80c03fa9 2870 if self._write_description('video', info_dict,
2871 self.prepare_filename(info_dict, 'description')) is None:
2872 return
2873
2874 sub_files = self._write_subtitles(info_dict, temp_filename)
2875 if sub_files is None:
2876 return
2877 files_to_move.update(dict(sub_files))
2878
2879 thumb_files = self._write_thumbnails(
2880 'video', info_dict, temp_filename, self.prepare_filename(info_dict, 'thumbnail'))
2881 if thumb_files is None:
2882 return
2883 files_to_move.update(dict(thumb_files))
8222d8de 2884
80c03fa9 2885 infofn = self.prepare_filename(info_dict, 'infojson')
2886 _infojson_written = self._write_info_json('video', info_dict, infofn)
2887 if _infojson_written:
dac5df5a 2888 info_dict['infojson_filename'] = infofn
e75bb0d6 2889 # For backward compatibility, even though it was a private field
80c03fa9 2890 info_dict['__infojson_filename'] = infofn
2891 elif _infojson_written is None:
2892 return
2893
2894 # Note: Annotations are deprecated
2895 annofn = None
1fb07d10 2896 if self.params.get('writeannotations', False):
de6000d9 2897 annofn = self.prepare_filename(info_dict, 'annotation')
80c03fa9 2898 if annofn:
e92caff5 2899 if not self._ensure_dir_exists(encodeFilename(annofn)):
0202b52a 2900 return
0c3d0f51 2901 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(annofn)):
6febd1c1 2902 self.to_screen('[info] Video annotations are already present')
ffddb112
RA
2903 elif not info_dict.get('annotations'):
2904 self.report_warning('There are no annotations to write.')
7b6fefc9
PH
2905 else:
2906 try:
6febd1c1 2907 self.to_screen('[info] Writing video annotations to: ' + annofn)
86e5f3ed 2908 with open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
7b6fefc9
PH
2909 annofile.write(info_dict['annotations'])
2910 except (KeyError, TypeError):
6febd1c1 2911 self.report_warning('There are no annotations to write.')
86e5f3ed 2912 except OSError:
6febd1c1 2913 self.report_error('Cannot write annotations file: ' + annofn)
7b6fefc9 2914 return
1fb07d10 2915
732044af 2916 # Write internet shortcut files
08438d2c 2917 def _write_link_file(link_type):
60f3e995 2918 url = try_get(info_dict['webpage_url'], iri_to_uri)
2919 if not url:
2920 self.report_warning(
2921 f'Cannot write internet shortcut file because the actual URL of "{info_dict["webpage_url"]}" is unknown')
2922 return True
08438d2c 2923 linkfn = replace_extension(self.prepare_filename(info_dict, 'link'), link_type, info_dict.get('ext'))
0e6b018a
Z
2924 if not self._ensure_dir_exists(encodeFilename(linkfn)):
2925 return False
10e3742e 2926 if self.params.get('overwrites', True) and os.path.exists(encodeFilename(linkfn)):
08438d2c 2927 self.to_screen(f'[info] Internet shortcut (.{link_type}) is already present')
2928 return True
2929 try:
2930 self.to_screen(f'[info] Writing internet shortcut (.{link_type}) to: {linkfn}')
86e5f3ed 2931 with open(encodeFilename(to_high_limit_path(linkfn)), 'w', encoding='utf-8',
2932 newline='\r\n' if link_type == 'url' else '\n') as linkfile:
60f3e995 2933 template_vars = {'url': url}
08438d2c 2934 if link_type == 'desktop':
2935 template_vars['filename'] = linkfn[:-(len(link_type) + 1)]
2936 linkfile.write(LINK_TEMPLATES[link_type] % template_vars)
86e5f3ed 2937 except OSError:
08438d2c 2938 self.report_error(f'Cannot write internet shortcut {linkfn}')
2939 return False
732044af 2940 return True
2941
08438d2c 2942 write_links = {
2943 'url': self.params.get('writeurllink'),
2944 'webloc': self.params.get('writewebloclink'),
2945 'desktop': self.params.get('writedesktoplink'),
2946 }
2947 if self.params.get('writelink'):
2948 link_type = ('webloc' if sys.platform == 'darwin'
2949 else 'desktop' if sys.platform.startswith('linux')
2950 else 'url')
2951 write_links[link_type] = True
2952
2953 if any(should_write and not _write_link_file(link_type)
2954 for link_type, should_write in write_links.items()):
2955 return
732044af 2956
f46e2f9d 2957 def replace_info_dict(new_info):
2958 nonlocal info_dict
2959 if new_info == info_dict:
2960 return
2961 info_dict.clear()
2962 info_dict.update(new_info)
2963
415f8d51 2964 new_info, files_to_move = self.pre_process(info_dict, 'before_dl', files_to_move)
2965 replace_info_dict(new_info)
56d868db 2966
a13e6848 2967 if self.params.get('skip_download'):
56d868db 2968 info_dict['filepath'] = temp_filename
2969 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
2970 info_dict['__files_to_move'] = files_to_move
f46e2f9d 2971 replace_info_dict(self.run_pp(MoveFilesAfterDownloadPP(self, False), info_dict))
9e907ebd 2972 info_dict['__write_download_archive'] = self.params.get('force_write_download_archive')
56d868db 2973 else:
2974 # Download
b868936c 2975 info_dict.setdefault('__postprocessors', [])
4340deca 2976 try:
0202b52a 2977
e04938ab 2978 def existing_video_file(*filepaths):
6b591b29 2979 ext = info_dict.get('ext')
e04938ab 2980 converted = lambda file: replace_extension(file, self.params.get('final_ext') or ext, ext)
2981 file = self.existing_file(itertools.chain(*zip(map(converted, filepaths), filepaths)),
2982 default_overwrite=False)
2983 if file:
2984 info_dict['ext'] = os.path.splitext(file)[1][1:]
2985 return file
0202b52a 2986
7b2c3f47 2987 fd, success = None, True
fccf90e7 2988 if info_dict.get('protocol') or info_dict.get('url'):
56ba69e4 2989 fd = get_suitable_downloader(info_dict, self.params, to_stdout=temp_filename == '-')
2990 if fd is not FFmpegFD and (
2991 info_dict.get('section_start') or info_dict.get('section_end')):
7b2c3f47 2992 msg = ('This format cannot be partially downloaded' if FFmpegFD.available()
56ba69e4 2993 else 'You have requested downloading the video partially, but ffmpeg is not installed')
2994 self.report_error(f'{msg}. Aborting')
5ec1b6b7 2995 return
5ec1b6b7 2996
4340deca 2997 if info_dict.get('requested_formats') is not None:
81cd954a
S
2998
2999 def compatible_formats(formats):
d03cfdce 3000 # TODO: some formats actually allow this (mkv, webm, ogg, mp4), but not all of them.
3001 video_formats = [format for format in formats if format.get('vcodec') != 'none']
3002 audio_formats = [format for format in formats if format.get('acodec') != 'none']
3003 if len(video_formats) > 2 or len(audio_formats) > 2:
3004 return False
3005
81cd954a 3006 # Check extension
86e5f3ed 3007 exts = {format.get('ext') for format in formats}
d03cfdce 3008 COMPATIBLE_EXTS = (
86e5f3ed 3009 {'mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma'},
3010 {'webm'},
d03cfdce 3011 )
3012 for ext_sets in COMPATIBLE_EXTS:
3013 if ext_sets.issuperset(exts):
3014 return True
81cd954a
S
3015 # TODO: Check acodec/vcodec
3016 return False
3017
3018 requested_formats = info_dict['requested_formats']
0202b52a 3019 old_ext = info_dict['ext']
4e3b637d 3020 if self.params.get('merge_output_format') is None:
3021 if not compatible_formats(requested_formats):
3022 info_dict['ext'] = 'mkv'
3023 self.report_warning(
3024 'Requested formats are incompatible for merge and will be merged into mkv')
3025 if (info_dict['ext'] == 'webm'
3026 and info_dict.get('thumbnails')
3027 # check with type instead of pp_key, __name__, or isinstance
3028 # since we dont want any custom PPs to trigger this
c487cf00 3029 and any(type(pp) == EmbedThumbnailPP for pp in self._pps['post_process'])): # noqa: E721
4e3b637d 3030 info_dict['ext'] = 'mkv'
3031 self.report_warning(
3032 'webm doesn\'t support embedding a thumbnail, mkv will be used')
124bc071 3033 new_ext = info_dict['ext']
0202b52a 3034
124bc071 3035 def correct_ext(filename, ext=new_ext):
96fccc10 3036 if filename == '-':
3037 return filename
0202b52a 3038 filename_real_ext = os.path.splitext(filename)[1][1:]
3039 filename_wo_ext = (
3040 os.path.splitext(filename)[0]
124bc071 3041 if filename_real_ext in (old_ext, new_ext)
0202b52a 3042 else filename)
86e5f3ed 3043 return f'{filename_wo_ext}.{ext}'
0202b52a 3044
38c6902b 3045 # Ensure filename always has a correct extension for successful merge
0202b52a 3046 full_filename = correct_ext(full_filename)
3047 temp_filename = correct_ext(temp_filename)
e04938ab 3048 dl_filename = existing_video_file(full_filename, temp_filename)
1ea24129 3049 info_dict['__real_download'] = False
18e674b4 3050
7b2c3f47 3051 merger = FFmpegMergerPP(self)
adbc4ec4 3052 downloaded = []
dbf5416a 3053 if dl_filename is not None:
6c7274ec 3054 self.report_file_already_downloaded(dl_filename)
adbc4ec4
THD
3055 elif fd:
3056 for f in requested_formats if fd != FFmpegFD else []:
3057 f['filepath'] = fname = prepend_extension(
3058 correct_ext(temp_filename, info_dict['ext']),
3059 'f%s' % f['format_id'], info_dict['ext'])
3060 downloaded.append(fname)
dbf5416a 3061 info_dict['url'] = '\n'.join(f['url'] for f in requested_formats)
3062 success, real_download = self.dl(temp_filename, info_dict)
3063 info_dict['__real_download'] = real_download
18e674b4 3064 else:
18e674b4 3065 if self.params.get('allow_unplayable_formats'):
3066 self.report_warning(
3067 'You have requested merging of multiple formats '
3068 'while also allowing unplayable formats to be downloaded. '
3069 'The formats won\'t be merged to prevent data corruption.')
3070 elif not merger.available:
e8969bda 3071 msg = 'You have requested merging of multiple formats but ffmpeg is not installed'
3072 if not self.params.get('ignoreerrors'):
3073 self.report_error(f'{msg}. Aborting due to --abort-on-error')
3074 return
3075 self.report_warning(f'{msg}. The formats won\'t be merged')
18e674b4 3076
96fccc10 3077 if temp_filename == '-':
adbc4ec4 3078 reason = ('using a downloader other than ffmpeg' if FFmpegFD.can_merge_formats(info_dict, self.params)
96fccc10 3079 else 'but the formats are incompatible for simultaneous download' if merger.available
3080 else 'but ffmpeg is not installed')
3081 self.report_warning(
3082 f'You have requested downloading multiple formats to stdout {reason}. '
3083 'The formats will be streamed one after the other')
3084 fname = temp_filename
dbf5416a 3085 for f in requested_formats:
3086 new_info = dict(info_dict)
3087 del new_info['requested_formats']
3088 new_info.update(f)
96fccc10 3089 if temp_filename != '-':
124bc071 3090 fname = prepend_extension(
3091 correct_ext(temp_filename, new_info['ext']),
3092 'f%s' % f['format_id'], new_info['ext'])
96fccc10 3093 if not self._ensure_dir_exists(fname):
3094 return
a21e0ab1 3095 f['filepath'] = fname
96fccc10 3096 downloaded.append(fname)
dbf5416a 3097 partial_success, real_download = self.dl(fname, new_info)
3098 info_dict['__real_download'] = info_dict['__real_download'] or real_download
3099 success = success and partial_success
adbc4ec4
THD
3100
3101 if downloaded and merger.available and not self.params.get('allow_unplayable_formats'):
3102 info_dict['__postprocessors'].append(merger)
3103 info_dict['__files_to_merge'] = downloaded
3104 # Even if there were no downloads, it is being merged only now
3105 info_dict['__real_download'] = True
3106 else:
3107 for file in downloaded:
3108 files_to_move[file] = None
4340deca
P
3109 else:
3110 # Just a single file
e04938ab 3111 dl_filename = existing_video_file(full_filename, temp_filename)
6c7274ec 3112 if dl_filename is None or dl_filename == temp_filename:
3113 # dl_filename == temp_filename could mean that the file was partially downloaded with --no-part.
3114 # So we should try to resume the download
e8e73840 3115 success, real_download = self.dl(temp_filename, info_dict)
0202b52a 3116 info_dict['__real_download'] = real_download
6c7274ec 3117 else:
3118 self.report_file_already_downloaded(dl_filename)
0202b52a 3119
0202b52a 3120 dl_filename = dl_filename or temp_filename
c571435f 3121 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
0202b52a 3122
3158150c 3123 except network_exceptions as err:
7960b056 3124 self.report_error('unable to download video data: %s' % error_to_compat_str(err))
4340deca 3125 return
86e5f3ed 3126 except OSError as err:
4340deca
P
3127 raise UnavailableVideoError(err)
3128 except (ContentTooShortError, ) as err:
86e5f3ed 3129 self.report_error(f'content too short (expected {err.expected} bytes and served {err.downloaded})')
4340deca 3130 return
8222d8de 3131
415f8d51 3132 self._raise_pending_errors(info_dict)
de6000d9 3133 if success and full_filename != '-':
f17f8651 3134
fd7cfb64 3135 def fixup():
3136 do_fixup = True
3137 fixup_policy = self.params.get('fixup')
3138 vid = info_dict['id']
3139
3140 if fixup_policy in ('ignore', 'never'):
3141 return
3142 elif fixup_policy == 'warn':
3fe75fdc 3143 do_fixup = 'warn'
f89b3e2d 3144 elif fixup_policy != 'force':
3145 assert fixup_policy in ('detect_or_warn', None)
3146 if not info_dict.get('__real_download'):
3147 do_fixup = False
fd7cfb64 3148
3149 def ffmpeg_fixup(cndn, msg, cls):
3fe75fdc 3150 if not (do_fixup and cndn):
fd7cfb64 3151 return
3fe75fdc 3152 elif do_fixup == 'warn':
fd7cfb64 3153 self.report_warning(f'{vid}: {msg}')
3154 return
3155 pp = cls(self)
3156 if pp.available:
3157 info_dict['__postprocessors'].append(pp)
3158 else:
3159 self.report_warning(f'{vid}: {msg}. Install ffmpeg to fix this automatically')
3160
3161 stretched_ratio = info_dict.get('stretched_ratio')
3162 ffmpeg_fixup(
3163 stretched_ratio not in (1, None),
3164 f'Non-uniform pixel ratio {stretched_ratio}',
3165 FFmpegFixupStretchedPP)
3166
3167 ffmpeg_fixup(
3168 (info_dict.get('requested_formats') is None
3169 and info_dict.get('container') == 'm4a_dash'
3170 and info_dict.get('ext') == 'm4a'),
3171 'writing DASH m4a. Only some players support this container',
3172 FFmpegFixupM4aPP)
3173
993191c0 3174 downloader = get_suitable_downloader(info_dict, self.params) if 'protocol' in info_dict else None
24146491 3175 downloader = downloader.FD_NAME if downloader else None
adbc4ec4
THD
3176
3177 if info_dict.get('requested_formats') is None: # Not necessary if doing merger
24146491 3178 ffmpeg_fixup(downloader == 'hlsnative' and not self.params.get('hls_use_mpegts')
494f5230 3179 or info_dict.get('is_live') and self.params.get('hls_use_mpegts') is None,
adbc4ec4
THD
3180 'Possible MPEG-TS in MP4 container or malformed AAC timestamps',
3181 FFmpegFixupM3u8PP)
3182 ffmpeg_fixup(info_dict.get('is_live') and downloader == 'DashSegmentsFD',
3183 'Possible duplicate MOOV atoms', FFmpegFixupDuplicateMoovPP)
3184
24146491 3185 ffmpeg_fixup(downloader == 'web_socket_fragment', 'Malformed timestamps detected', FFmpegFixupTimestampPP)
3186 ffmpeg_fixup(downloader == 'web_socket_fragment', 'Malformed duration detected', FFmpegFixupDurationPP)
fd7cfb64 3187
3188 fixup()
8222d8de 3189 try:
f46e2f9d 3190 replace_info_dict(self.post_process(dl_filename, info_dict, files_to_move))
af819c21 3191 except PostProcessingError as err:
3192 self.report_error('Postprocessing: %s' % str(err))
8222d8de 3193 return
ab8e5e51
AM
3194 try:
3195 for ph in self._post_hooks:
23c1a667 3196 ph(info_dict['filepath'])
ab8e5e51
AM
3197 except Exception as err:
3198 self.report_error('post hooks: %s' % str(err))
3199 return
9e907ebd 3200 info_dict['__write_download_archive'] = True
2d30509f 3201
c487cf00 3202 assert info_dict is original_infodict # Make sure the info_dict was modified in-place
a13e6848 3203 if self.params.get('force_write_download_archive'):
9e907ebd 3204 info_dict['__write_download_archive'] = True
ca6d59d2 3205 check_max_downloads()
8222d8de 3206
aa9369a2 3207 def __download_wrapper(self, func):
3208 @functools.wraps(func)
3209 def wrapper(*args, **kwargs):
3210 try:
3211 res = func(*args, **kwargs)
3212 except UnavailableVideoError as e:
3213 self.report_error(e)
b222c271 3214 except DownloadCancelled as e:
3215 self.to_screen(f'[info] {e}')
3216 if not self.params.get('break_per_url'):
3217 raise
aa9369a2 3218 else:
3219 if self.params.get('dump_single_json', False):
3220 self.post_extract(res)
3221 self.to_stdout(json.dumps(self.sanitize_info(res)))
3222 return wrapper
3223
8222d8de
JMF
3224 def download(self, url_list):
3225 """Download a given list of URLs."""
aa9369a2 3226 url_list = variadic(url_list) # Passing a single URL is a common mistake
bf1824b3 3227 outtmpl = self.params['outtmpl']['default']
3089bc74
S
3228 if (len(url_list) > 1
3229 and outtmpl != '-'
3230 and '%' not in outtmpl
3231 and self.params.get('max_downloads') != 1):
acd69589 3232 raise SameFileError(outtmpl)
8222d8de
JMF
3233
3234 for url in url_list:
aa9369a2 3235 self.__download_wrapper(self.extract_info)(
3236 url, force_generic_extractor=self.params.get('force_generic_extractor', False))
8222d8de
JMF
3237
3238 return self._download_retcode
3239
1dcc4c0c 3240 def download_with_info_file(self, info_filename):
31bd3925
JMF
3241 with contextlib.closing(fileinput.FileInput(
3242 [info_filename], mode='r',
3243 openhook=fileinput.hook_encoded('utf-8'))) as f:
3244 # FileInput doesn't have a read method, we can't call json.load
8012d892 3245 info = self.sanitize_info(json.loads('\n'.join(f)), self.params.get('clean_infojson', True))
d4943898 3246 try:
aa9369a2 3247 self.__download_wrapper(self.process_ie_result)(info, download=True)
f2ebc5c7 3248 except (DownloadError, EntryNotInPlaylist, ReExtractInfo) as e:
bf5f605e 3249 if not isinstance(e, EntryNotInPlaylist):
3250 self.to_stderr('\r')
d4943898
JMF
3251 webpage_url = info.get('webpage_url')
3252 if webpage_url is not None:
aa9369a2 3253 self.report_warning(f'The info failed to download: {e}; trying with URL {webpage_url}')
d4943898
JMF
3254 return self.download([webpage_url])
3255 else:
3256 raise
3257 return self._download_retcode
1dcc4c0c 3258
cb202fd2 3259 @staticmethod
8012d892 3260 def sanitize_info(info_dict, remove_private_keys=False):
3261 ''' Sanitize the infodict for converting to json '''
3ad56b42 3262 if info_dict is None:
3263 return info_dict
6e84b215 3264 info_dict.setdefault('epoch', int(time.time()))
6a5a30f9 3265 info_dict.setdefault('_type', 'video')
09b49e1f 3266
8012d892 3267 if remove_private_keys:
0a5a191a 3268 reject = lambda k, v: v is None or k.startswith('__') or k in {
f46e2f9d 3269 'requested_downloads', 'requested_formats', 'requested_subtitles', 'requested_entries',
0a5a191a 3270 'entries', 'filepath', '_filename', 'infojson_filename', 'original_url', 'playlist_autonumber',
6e84b215 3271 }
ae8f99e6 3272 else:
09b49e1f 3273 reject = lambda k, v: False
adbc4ec4
THD
3274
3275 def filter_fn(obj):
3276 if isinstance(obj, dict):
3277 return {k: filter_fn(v) for k, v in obj.items() if not reject(k, v)}
3278 elif isinstance(obj, (list, tuple, set, LazyList)):
3279 return list(map(filter_fn, obj))
3280 elif obj is None or isinstance(obj, (str, int, float, bool)):
3281 return obj
3282 else:
3283 return repr(obj)
3284
5226731e 3285 return filter_fn(info_dict)
cb202fd2 3286
8012d892 3287 @staticmethod
3288 def filter_requested_info(info_dict, actually_filter=True):
3289 ''' Alias of sanitize_info for backward compatibility '''
3290 return YoutubeDL.sanitize_info(info_dict, actually_filter)
3291
43d7f5a5 3292 def _delete_downloaded_files(self, *files_to_delete, info={}, msg=None):
3293 for filename in set(filter(None, files_to_delete)):
3294 if msg:
3295 self.to_screen(msg % filename)
3296 try:
3297 os.remove(filename)
3298 except OSError:
3299 self.report_warning(f'Unable to delete file {filename}')
3300 if filename in info.get('__files_to_move', []): # NB: Delete even if None
3301 del info['__files_to_move'][filename]
3302
ed5835b4 3303 @staticmethod
3304 def post_extract(info_dict):
3305 def actual_post_extract(info_dict):
3306 if info_dict.get('_type') in ('playlist', 'multi_video'):
3307 for video_dict in info_dict.get('entries', {}):
3308 actual_post_extract(video_dict or {})
3309 return
3310
09b49e1f 3311 post_extractor = info_dict.pop('__post_extractor', None) or (lambda: {})
3312 info_dict.update(post_extractor())
ed5835b4 3313
3314 actual_post_extract(info_dict or {})
3315
dcf64d43 3316 def run_pp(self, pp, infodict):
5bfa4862 3317 files_to_delete = []
dcf64d43 3318 if '__files_to_move' not in infodict:
3319 infodict['__files_to_move'] = {}
b1940459 3320 try:
3321 files_to_delete, infodict = pp.run(infodict)
3322 except PostProcessingError as e:
3323 # Must be True and not 'only_download'
3324 if self.params.get('ignoreerrors') is True:
3325 self.report_error(e)
3326 return infodict
3327 raise
3328
5bfa4862 3329 if not files_to_delete:
dcf64d43 3330 return infodict
5bfa4862 3331 if self.params.get('keepvideo', False):
3332 for f in files_to_delete:
dcf64d43 3333 infodict['__files_to_move'].setdefault(f, '')
5bfa4862 3334 else:
43d7f5a5 3335 self._delete_downloaded_files(
3336 *files_to_delete, info=infodict, msg='Deleting original file %s (pass -k to keep)')
dcf64d43 3337 return infodict
5bfa4862 3338
ed5835b4 3339 def run_all_pps(self, key, info, *, additional_pps=None):
bb66c247 3340 self._forceprint(key, info)
ed5835b4 3341 for pp in (additional_pps or []) + self._pps[key]:
dc5f409c 3342 info = self.run_pp(pp, info)
ed5835b4 3343 return info
277d6ff5 3344
56d868db 3345 def pre_process(self, ie_info, key='pre_process', files_to_move=None):
5bfa4862 3346 info = dict(ie_info)
56d868db 3347 info['__files_to_move'] = files_to_move or {}
415f8d51 3348 try:
3349 info = self.run_all_pps(key, info)
3350 except PostProcessingError as err:
3351 msg = f'Preprocessing: {err}'
3352 info.setdefault('__pending_error', msg)
3353 self.report_error(msg, is_error=False)
56d868db 3354 return info, info.pop('__files_to_move', None)
5bfa4862 3355
f46e2f9d 3356 def post_process(self, filename, info, files_to_move=None):
8222d8de 3357 """Run all the postprocessors on the given file."""
8222d8de 3358 info['filepath'] = filename
dcf64d43 3359 info['__files_to_move'] = files_to_move or {}
ed5835b4 3360 info = self.run_all_pps('post_process', info, additional_pps=info.get('__postprocessors'))
dcf64d43 3361 info = self.run_pp(MoveFilesAfterDownloadPP(self), info)
3362 del info['__files_to_move']
ed5835b4 3363 return self.run_all_pps('after_move', info)
c1c9a79c 3364
5db07df6 3365 def _make_archive_id(self, info_dict):
e9fef7ee
S
3366 video_id = info_dict.get('id')
3367 if not video_id:
3368 return
5db07df6
PH
3369 # Future-proof against any change in case
3370 # and backwards compatibility with prior versions
e9fef7ee 3371 extractor = info_dict.get('extractor_key') or info_dict.get('ie_key') # key in a playlist
7012b23c 3372 if extractor is None:
1211bb6d
S
3373 url = str_or_none(info_dict.get('url'))
3374 if not url:
3375 return
e9fef7ee 3376 # Try to find matching extractor for the URL and take its ie_key
8b7491c8 3377 for ie_key, ie in self._ies.items():
1211bb6d 3378 if ie.suitable(url):
8b7491c8 3379 extractor = ie_key
e9fef7ee
S
3380 break
3381 else:
3382 return
86e5f3ed 3383 return f'{extractor.lower()} {video_id}'
5db07df6
PH
3384
3385 def in_download_archive(self, info_dict):
3386 fn = self.params.get('download_archive')
3387 if fn is None:
3388 return False
3389
3390 vid_id = self._make_archive_id(info_dict)
e9fef7ee 3391 if not vid_id:
7012b23c 3392 return False # Incomplete video information
5db07df6 3393
a45e8619 3394 return vid_id in self.archive
c1c9a79c
PH
3395
3396 def record_download_archive(self, info_dict):
3397 fn = self.params.get('download_archive')
3398 if fn is None:
3399 return
5db07df6
PH
3400 vid_id = self._make_archive_id(info_dict)
3401 assert vid_id
a13e6848 3402 self.write_debug(f'Adding to archive: {vid_id}')
c1c9a79c 3403 with locked_file(fn, 'a', encoding='utf-8') as archive_file:
6febd1c1 3404 archive_file.write(vid_id + '\n')
a45e8619 3405 self.archive.add(vid_id)
dd82ffea 3406
8c51aa65 3407 @staticmethod
8abeeb94 3408 def format_resolution(format, default='unknown'):
9359f3d4 3409 if format.get('vcodec') == 'none' and format.get('acodec') != 'none':
fb04e403 3410 return 'audio only'
f49d89ee
PH
3411 if format.get('resolution') is not None:
3412 return format['resolution']
35615307 3413 if format.get('width') and format.get('height'):
ff51ed58 3414 return '%dx%d' % (format['width'], format['height'])
35615307 3415 elif format.get('height'):
ff51ed58 3416 return '%sp' % format['height']
35615307 3417 elif format.get('width'):
ff51ed58 3418 return '%dx?' % format['width']
3419 return default
8c51aa65 3420
8130779d 3421 def _list_format_headers(self, *headers):
3422 if self.params.get('listformats_table', True) is not False:
591bb9d3 3423 return [self._format_out(header, self.Styles.HEADERS) for header in headers]
8130779d 3424 return headers
3425
c57f7757
PH
3426 def _format_note(self, fdict):
3427 res = ''
3428 if fdict.get('ext') in ['f4f', 'f4m']:
f304da8a 3429 res += '(unsupported)'
32f90364
PH
3430 if fdict.get('language'):
3431 if res:
3432 res += ' '
f304da8a 3433 res += '[%s]' % fdict['language']
c57f7757 3434 if fdict.get('format_note') is not None:
f304da8a 3435 if res:
3436 res += ' '
3437 res += fdict['format_note']
c57f7757 3438 if fdict.get('tbr') is not None:
f304da8a 3439 if res:
3440 res += ', '
3441 res += '%4dk' % fdict['tbr']
c57f7757
PH
3442 if fdict.get('container') is not None:
3443 if res:
3444 res += ', '
3445 res += '%s container' % fdict['container']
3089bc74
S
3446 if (fdict.get('vcodec') is not None
3447 and fdict.get('vcodec') != 'none'):
c57f7757
PH
3448 if res:
3449 res += ', '
3450 res += fdict['vcodec']
91c7271a 3451 if fdict.get('vbr') is not None:
c57f7757
PH
3452 res += '@'
3453 elif fdict.get('vbr') is not None and fdict.get('abr') is not None:
3454 res += 'video@'
3455 if fdict.get('vbr') is not None:
3456 res += '%4dk' % fdict['vbr']
fbb21cf5 3457 if fdict.get('fps') is not None:
5d583bdf
S
3458 if res:
3459 res += ', '
3460 res += '%sfps' % fdict['fps']
c57f7757
PH
3461 if fdict.get('acodec') is not None:
3462 if res:
3463 res += ', '
3464 if fdict['acodec'] == 'none':
3465 res += 'video only'
3466 else:
3467 res += '%-5s' % fdict['acodec']
3468 elif fdict.get('abr') is not None:
3469 if res:
3470 res += ', '
3471 res += 'audio'
3472 if fdict.get('abr') is not None:
3473 res += '@%3dk' % fdict['abr']
3474 if fdict.get('asr') is not None:
3475 res += ' (%5dHz)' % fdict['asr']
3476 if fdict.get('filesize') is not None:
3477 if res:
3478 res += ', '
3479 res += format_bytes(fdict['filesize'])
9732d77e
PH
3480 elif fdict.get('filesize_approx') is not None:
3481 if res:
3482 res += ', '
3483 res += '~' + format_bytes(fdict['filesize_approx'])
c57f7757 3484 return res
91c7271a 3485
8130779d 3486 def render_formats_table(self, info_dict):
b69fd25c 3487 if not info_dict.get('formats') and not info_dict.get('url'):
8130779d 3488 return None
b69fd25c 3489
94badb25 3490 formats = info_dict.get('formats', [info_dict])
8130779d 3491 if not self.params.get('listformats_table', True) is not False:
76d321f6 3492 table = [
3493 [
3494 format_field(f, 'format_id'),
3495 format_field(f, 'ext'),
3496 self.format_resolution(f),
8130779d 3497 self._format_note(f)
3498 ] for f in formats if f.get('preference') is None or f['preference'] >= -1000]
3499 return render_table(['format code', 'extension', 'resolution', 'note'], table, extra_gap=1)
3500
591bb9d3 3501 delim = self._format_out('\u2502', self.Styles.DELIM, '|', test_encoding=True)
8130779d 3502 table = [
3503 [
591bb9d3 3504 self._format_out(format_field(f, 'format_id'), self.Styles.ID),
8130779d 3505 format_field(f, 'ext'),
3506 format_field(f, func=self.format_resolution, ignore=('audio only', 'images')),
3507 format_field(f, 'fps', '\t%d'),
3508 format_field(f, 'dynamic_range', '%s', ignore=(None, 'SDR')).replace('HDR', ''),
3509 delim,
3510 format_field(f, 'filesize', ' \t%s', func=format_bytes) + format_field(f, 'filesize_approx', '~\t%s', func=format_bytes),
3511 format_field(f, 'tbr', '\t%dk'),
3512 shorten_protocol_name(f.get('protocol', '')),
3513 delim,
3514 format_field(f, 'vcodec', default='unknown').replace(
3515 'none', 'images' if f.get('acodec') == 'none'
591bb9d3 3516 else self._format_out('audio only', self.Styles.SUPPRESS)),
8130779d 3517 format_field(f, 'vbr', '\t%dk'),
3518 format_field(f, 'acodec', default='unknown').replace(
3519 'none', '' if f.get('vcodec') == 'none'
591bb9d3 3520 else self._format_out('video only', self.Styles.SUPPRESS)),
8130779d 3521 format_field(f, 'abr', '\t%dk'),
3522 format_field(f, 'asr', '\t%dHz'),
3523 join_nonempty(
591bb9d3 3524 self._format_out('UNSUPPORTED', 'light red') if f.get('ext') in ('f4f', 'f4m') else None,
8130779d 3525 format_field(f, 'language', '[%s]'),
3526 join_nonempty(format_field(f, 'format_note'),
3527 format_field(f, 'container', ignore=(None, f.get('ext'))),
3528 delim=', '),
3529 delim=' '),
3530 ] for f in formats if f.get('preference') is None or f['preference'] >= -1000]
3531 header_line = self._list_format_headers(
3532 'ID', 'EXT', 'RESOLUTION', '\tFPS', 'HDR', delim, '\tFILESIZE', '\tTBR', 'PROTO',
3533 delim, 'VCODEC', '\tVBR', 'ACODEC', '\tABR', '\tASR', 'MORE INFO')
3534
3535 return render_table(
3536 header_line, table, hide_empty=True,
591bb9d3 3537 delim=self._format_out('\u2500', self.Styles.DELIM, '-', test_encoding=True))
8130779d 3538
3539 def render_thumbnails_table(self, info_dict):
88f23a18 3540 thumbnails = list(info_dict.get('thumbnails') or [])
cfb56d1a 3541 if not thumbnails:
8130779d 3542 return None
3543 return render_table(
ec11a9f4 3544 self._list_format_headers('ID', 'Width', 'Height', 'URL'),
6970b600 3545 [[t.get('id'), t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails])
2412044c 3546
8130779d 3547 def render_subtitles_table(self, video_id, subtitles):
2412044c 3548 def _row(lang, formats):
49c258e1 3549 exts, names = zip(*((f['ext'], f.get('name') or 'unknown') for f in reversed(formats)))
2412044c 3550 if len(set(names)) == 1:
7aee40c1 3551 names = [] if names[0] == 'unknown' else names[:1]
2412044c 3552 return [lang, ', '.join(names), ', '.join(exts)]
3553
8130779d 3554 if not subtitles:
3555 return None
3556 return render_table(
ec11a9f4 3557 self._list_format_headers('Language', 'Name', 'Formats'),
2412044c 3558 [_row(lang, formats) for lang, formats in subtitles.items()],
8130779d 3559 hide_empty=True)
3560
3561 def __list_table(self, video_id, name, func, *args):
3562 table = func(*args)
3563 if not table:
3564 self.to_screen(f'{video_id} has no {name}')
3565 return
3566 self.to_screen(f'[info] Available {name} for {video_id}:')
3567 self.to_stdout(table)
3568
3569 def list_formats(self, info_dict):
3570 self.__list_table(info_dict['id'], 'formats', self.render_formats_table, info_dict)
3571
3572 def list_thumbnails(self, info_dict):
3573 self.__list_table(info_dict['id'], 'thumbnails', self.render_thumbnails_table, info_dict)
3574
3575 def list_subtitles(self, video_id, subtitles, name='subtitles'):
3576 self.__list_table(video_id, name, self.render_subtitles_table, video_id, subtitles)
a504ced0 3577
dca08720
PH
3578 def urlopen(self, req):
3579 """ Start an HTTP download """
f9934b96 3580 if isinstance(req, str):
67dda517 3581 req = sanitized_Request(req)
19a41fc6 3582 return self._opener.open(req, timeout=self._socket_timeout)
dca08720
PH
3583
3584 def print_debug_header(self):
3585 if not self.params.get('verbose'):
3586 return
49a57e70 3587
560738f3 3588 # These imports can be slow. So import them only as needed
3589 from .extractor.extractors import _LAZY_LOADER
3590 from .extractor.extractors import _PLUGIN_CLASSES as plugin_extractors
3591
49a57e70 3592 def get_encoding(stream):
2a938746 3593 ret = str(getattr(stream, 'encoding', 'missing (%s)' % type(stream).__name__))
49a57e70 3594 if not supports_terminal_sequences(stream):
53973b4d 3595 from .utils import WINDOWS_VT_MODE # Must be imported locally
e3c7d495 3596 ret += ' (No VT)' if WINDOWS_VT_MODE is False else ' (No ANSI)'
49a57e70 3597 return ret
3598
591bb9d3 3599 encoding_str = 'Encodings: locale %s, fs %s, pref %s, %s' % (
49a57e70 3600 locale.getpreferredencoding(),
3601 sys.getfilesystemencoding(),
591bb9d3 3602 self.get_encoding(),
3603 ', '.join(
64fa820c 3604 f'{key} {get_encoding(stream)}' for key, stream in self._out_files.items_
591bb9d3 3605 if stream is not None and key != 'console')
3606 )
883d4b1e 3607
3608 logger = self.params.get('logger')
3609 if logger:
3610 write_debug = lambda msg: logger.debug(f'[debug] {msg}')
3611 write_debug(encoding_str)
3612 else:
96565c7e 3613 write_string(f'[debug] {encoding_str}\n', encoding=None)
49a57e70 3614 write_debug = lambda msg: self._write_string(f'[debug] {msg}\n')
734f90bb 3615
4c88ff87 3616 source = detect_variant()
36eaf303 3617 write_debug(join_nonempty(
3618 'yt-dlp version', __version__,
3619 f'[{RELEASE_GIT_HEAD}]' if RELEASE_GIT_HEAD else '',
3620 '' if source == 'unknown' else f'({source})',
3621 delim=' '))
6e21fdd2 3622 if not _LAZY_LOADER:
3623 if os.environ.get('YTDLP_NO_LAZY_EXTRACTORS'):
49a57e70 3624 write_debug('Lazy loading extractors is forcibly disabled')
6e21fdd2 3625 else:
49a57e70 3626 write_debug('Lazy loading extractors is disabled')
3ae5e797 3627 if plugin_extractors or plugin_postprocessors:
49a57e70 3628 write_debug('Plugins: %s' % [
3ae5e797 3629 '%s%s' % (klass.__name__, '' if klass.__name__ == name else f' as {name}')
3630 for name, klass in itertools.chain(plugin_extractors.items(), plugin_postprocessors.items())])
8a82af35 3631 if self.params['compat_opts']:
3632 write_debug('Compatibility options: %s' % ', '.join(self.params['compat_opts']))
36eaf303 3633
3634 if source == 'source':
dca08720 3635 try:
f0c9fb96 3636 stdout, _, _ = Popen.run(
36eaf303 3637 ['git', 'rev-parse', '--short', 'HEAD'],
f0c9fb96 3638 text=True, cwd=os.path.dirname(os.path.abspath(__file__)),
3639 stdout=subprocess.PIPE, stderr=subprocess.PIPE)
3640 if re.fullmatch('[0-9a-f]+', stdout.strip()):
3641 write_debug(f'Git HEAD: {stdout.strip()}')
70a1165b 3642 except Exception:
19a03940 3643 with contextlib.suppress(Exception):
36eaf303 3644 sys.exc_clear()
b300cda4
S
3645
3646 def python_implementation():
3647 impl_name = platform.python_implementation()
3648 if impl_name == 'PyPy' and hasattr(sys, 'pypy_version_info'):
3649 return impl_name + ' version %d.%d.%d' % sys.pypy_version_info[:3]
3650 return impl_name
3651
49a57e70 3652 write_debug('Python version %s (%s %s) - %s' % (
e5813e53 3653 platform.python_version(),
3654 python_implementation(),
3655 platform.architecture()[0],
b300cda4 3656 platform_name()))
d28b5171 3657
8913ef74 3658 exe_versions, ffmpeg_features = FFmpegPostProcessor.get_versions_and_features(self)
3659 ffmpeg_features = {key for key, val in ffmpeg_features.items() if val}
3660 if ffmpeg_features:
19a03940 3661 exe_versions['ffmpeg'] += ' (%s)' % ','.join(sorted(ffmpeg_features))
8913ef74 3662
4c83c967 3663 exe_versions['rtmpdump'] = rtmpdump_version()
feee8d32 3664 exe_versions['phantomjs'] = PhantomJSwrapper._version()
d28b5171 3665 exe_str = ', '.join(
2831b468 3666 f'{exe} {v}' for exe, v in sorted(exe_versions.items()) if v
3667 ) or 'none'
49a57e70 3668 write_debug('exe versions: %s' % exe_str)
dca08720 3669
1d485a1a 3670 from .compat.compat_utils import get_package_info
9b8ee23b 3671 from .dependencies import available_dependencies
3672
3673 write_debug('Optional libraries: %s' % (', '.join(sorted({
1d485a1a 3674 join_nonempty(*get_package_info(m)) for m in available_dependencies.values()
9b8ee23b 3675 })) or 'none'))
2831b468 3676
97ec5bc5 3677 self._setup_opener()
dca08720
PH
3678 proxy_map = {}
3679 for handler in self._opener.handlers:
3680 if hasattr(handler, 'proxies'):
3681 proxy_map.update(handler.proxies)
49a57e70 3682 write_debug(f'Proxy map: {proxy_map}')
dca08720 3683
49a57e70 3684 # Not implemented
3685 if False and self.params.get('call_home'):
0f06bcd7 3686 ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode()
49a57e70 3687 write_debug('Public IP address: %s' % ipaddr)
58b1f00d 3688 latest_version = self.urlopen(
0f06bcd7 3689 'https://yt-dl.org/latest/version').read().decode()
58b1f00d
PH
3690 if version_tuple(latest_version) > version_tuple(__version__):
3691 self.report_warning(
3692 'You are using an outdated version (newest version: %s)! '
3693 'See https://yt-dl.org/update if you need help updating.' %
3694 latest_version)
3695
e344693b 3696 def _setup_opener(self):
97ec5bc5 3697 if hasattr(self, '_opener'):
3698 return
6ad14cab 3699 timeout_val = self.params.get('socket_timeout')
17bddf3e 3700 self._socket_timeout = 20 if timeout_val is None else float(timeout_val)
6ad14cab 3701
982ee69a 3702 opts_cookiesfrombrowser = self.params.get('cookiesfrombrowser')
dca08720
PH
3703 opts_cookiefile = self.params.get('cookiefile')
3704 opts_proxy = self.params.get('proxy')
3705
982ee69a 3706 self.cookiejar = load_cookies(opts_cookiefile, opts_cookiesfrombrowser, self)
dca08720 3707
6a3f4c3f 3708 cookie_processor = YoutubeDLCookieProcessor(self.cookiejar)
dca08720
PH
3709 if opts_proxy is not None:
3710 if opts_proxy == '':
3711 proxies = {}
3712 else:
3713 proxies = {'http': opts_proxy, 'https': opts_proxy}
3714 else:
3715 proxies = compat_urllib_request.getproxies()
067aa17e 3716 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
dca08720
PH
3717 if 'http' in proxies and 'https' not in proxies:
3718 proxies['https'] = proxies['http']
91410c9b 3719 proxy_handler = PerRequestProxyHandler(proxies)
a0ddb8a2
PH
3720
3721 debuglevel = 1 if self.params.get('debug_printtraffic') else 0
be4a824d
PH
3722 https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel)
3723 ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel)
fca6dba8 3724 redirect_handler = YoutubeDLRedirectHandler()
f9934b96 3725 data_handler = urllib.request.DataHandler()
6240b0a2
JMF
3726
3727 # When passing our own FileHandler instance, build_opener won't add the
3728 # default FileHandler and allows us to disable the file protocol, which
3729 # can be used for malicious purposes (see
067aa17e 3730 # https://github.com/ytdl-org/youtube-dl/issues/8227)
6240b0a2
JMF
3731 file_handler = compat_urllib_request.FileHandler()
3732
3733 def file_open(*args, **kwargs):
7a5c1cfe 3734 raise compat_urllib_error.URLError('file:// scheme is explicitly disabled in yt-dlp for security reasons')
6240b0a2
JMF
3735 file_handler.file_open = file_open
3736
3737 opener = compat_urllib_request.build_opener(
fca6dba8 3738 proxy_handler, https_handler, cookie_processor, ydlh, redirect_handler, data_handler, file_handler)
2461f79d 3739
dca08720
PH
3740 # Delete the default user-agent header, which would otherwise apply in
3741 # cases where our custom HTTP handler doesn't come into play
067aa17e 3742 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
dca08720
PH
3743 opener.addheaders = []
3744 self._opener = opener
62fec3b2
PH
3745
3746 def encode(self, s):
3747 if isinstance(s, bytes):
3748 return s # Already encoded
3749
3750 try:
3751 return s.encode(self.get_encoding())
3752 except UnicodeEncodeError as err:
3753 err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.'
3754 raise
3755
3756 def get_encoding(self):
3757 encoding = self.params.get('encoding')
3758 if encoding is None:
3759 encoding = preferredencoding()
3760 return encoding
ec82d85a 3761
e08a85d8 3762 def _write_info_json(self, label, ie_result, infofn, overwrite=None):
cb96c5be 3763 ''' Write infojson and returns True = written, 'exists' = Already exists, False = skip, None = error '''
e08a85d8 3764 if overwrite is None:
3765 overwrite = self.params.get('overwrites', True)
80c03fa9 3766 if not self.params.get('writeinfojson'):
3767 return False
3768 elif not infofn:
3769 self.write_debug(f'Skipping writing {label} infojson')
3770 return False
3771 elif not self._ensure_dir_exists(infofn):
3772 return None
e08a85d8 3773 elif not overwrite and os.path.exists(infofn):
80c03fa9 3774 self.to_screen(f'[info] {label.title()} metadata is already present')
cb96c5be 3775 return 'exists'
3776
3777 self.to_screen(f'[info] Writing {label} metadata as JSON to: {infofn}')
3778 try:
3779 write_json_file(self.sanitize_info(ie_result, self.params.get('clean_infojson', True)), infofn)
3780 return True
86e5f3ed 3781 except OSError:
cb96c5be 3782 self.report_error(f'Cannot write {label} metadata to JSON file {infofn}')
3783 return None
80c03fa9 3784
3785 def _write_description(self, label, ie_result, descfn):
3786 ''' Write description and returns True = written, False = skip, None = error '''
3787 if not self.params.get('writedescription'):
3788 return False
3789 elif not descfn:
3790 self.write_debug(f'Skipping writing {label} description')
3791 return False
3792 elif not self._ensure_dir_exists(descfn):
3793 return None
3794 elif not self.params.get('overwrites', True) and os.path.exists(descfn):
3795 self.to_screen(f'[info] {label.title()} description is already present')
3796 elif ie_result.get('description') is None:
3797 self.report_warning(f'There\'s no {label} description to write')
3798 return False
3799 else:
3800 try:
3801 self.to_screen(f'[info] Writing {label} description to: {descfn}')
86e5f3ed 3802 with open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
80c03fa9 3803 descfile.write(ie_result['description'])
86e5f3ed 3804 except OSError:
80c03fa9 3805 self.report_error(f'Cannot write {label} description file {descfn}')
3806 return None
3807 return True
3808
3809 def _write_subtitles(self, info_dict, filename):
3810 ''' Write subtitles to file and return list of (sub_filename, final_sub_filename); or None if error'''
3811 ret = []
3812 subtitles = info_dict.get('requested_subtitles')
3813 if not subtitles or not (self.params.get('writesubtitles') or self.params.get('writeautomaticsub')):
3814 # subtitles download errors are already managed as troubles in relevant IE
3815 # that way it will silently go on when used with unsupporting IE
3816 return ret
3817
3818 sub_filename_base = self.prepare_filename(info_dict, 'subtitle')
3819 if not sub_filename_base:
3820 self.to_screen('[info] Skipping writing video subtitles')
3821 return ret
3822 for sub_lang, sub_info in subtitles.items():
3823 sub_format = sub_info['ext']
3824 sub_filename = subtitles_filename(filename, sub_lang, sub_format, info_dict.get('ext'))
3825 sub_filename_final = subtitles_filename(sub_filename_base, sub_lang, sub_format, info_dict.get('ext'))
e04938ab 3826 existing_sub = self.existing_file((sub_filename_final, sub_filename))
3827 if existing_sub:
80c03fa9 3828 self.to_screen(f'[info] Video subtitle {sub_lang}.{sub_format} is already present')
e04938ab 3829 sub_info['filepath'] = existing_sub
3830 ret.append((existing_sub, sub_filename_final))
80c03fa9 3831 continue
3832
3833 self.to_screen(f'[info] Writing video subtitles to: {sub_filename}')
3834 if sub_info.get('data') is not None:
3835 try:
3836 # Use newline='' to prevent conversion of newline characters
3837 # See https://github.com/ytdl-org/youtube-dl/issues/10268
86e5f3ed 3838 with open(sub_filename, 'w', encoding='utf-8', newline='') as subfile:
80c03fa9 3839 subfile.write(sub_info['data'])
3840 sub_info['filepath'] = sub_filename
3841 ret.append((sub_filename, sub_filename_final))
3842 continue
86e5f3ed 3843 except OSError:
80c03fa9 3844 self.report_error(f'Cannot write video subtitles file {sub_filename}')
3845 return None
3846
3847 try:
3848 sub_copy = sub_info.copy()
3849 sub_copy.setdefault('http_headers', info_dict.get('http_headers'))
3850 self.dl(sub_filename, sub_copy, subtitle=True)
3851 sub_info['filepath'] = sub_filename
3852 ret.append((sub_filename, sub_filename_final))
6020e05d 3853 except (DownloadError, ExtractorError, IOError, OSError, ValueError) + network_exceptions as err:
c70c418d 3854 msg = f'Unable to download video subtitles for {sub_lang!r}: {err}'
6020e05d 3855 if self.params.get('ignoreerrors') is not True: # False or 'only_download'
c70c418d 3856 if not self.params.get('ignoreerrors'):
3857 self.report_error(msg)
3858 raise DownloadError(msg)
3859 self.report_warning(msg)
519804a9 3860 return ret
80c03fa9 3861
3862 def _write_thumbnails(self, label, info_dict, filename, thumb_filename_base=None):
3863 ''' Write thumbnails to file and return list of (thumb_filename, final_thumb_filename) '''
6c4fd172 3864 write_all = self.params.get('write_all_thumbnails', False)
80c03fa9 3865 thumbnails, ret = [], []
6c4fd172 3866 if write_all or self.params.get('writethumbnail', False):
0202b52a 3867 thumbnails = info_dict.get('thumbnails') or []
6c4fd172 3868 multiple = write_all and len(thumbnails) > 1
ec82d85a 3869
80c03fa9 3870 if thumb_filename_base is None:
3871 thumb_filename_base = filename
3872 if thumbnails and not thumb_filename_base:
3873 self.write_debug(f'Skipping writing {label} thumbnail')
3874 return ret
3875
dd0228ce 3876 for idx, t in list(enumerate(thumbnails))[::-1]:
80c03fa9 3877 thumb_ext = (f'{t["id"]}.' if multiple else '') + determine_ext(t['url'], 'jpg')
aa9369a2 3878 thumb_display_id = f'{label} thumbnail {t["id"]}'
80c03fa9 3879 thumb_filename = replace_extension(filename, thumb_ext, info_dict.get('ext'))
3880 thumb_filename_final = replace_extension(thumb_filename_base, thumb_ext, info_dict.get('ext'))
ec82d85a 3881
e04938ab 3882 existing_thumb = self.existing_file((thumb_filename_final, thumb_filename))
3883 if existing_thumb:
aa9369a2 3884 self.to_screen('[info] %s is already present' % (
3885 thumb_display_id if multiple else f'{label} thumbnail').capitalize())
e04938ab 3886 t['filepath'] = existing_thumb
3887 ret.append((existing_thumb, thumb_filename_final))
ec82d85a 3888 else:
80c03fa9 3889 self.to_screen(f'[info] Downloading {thumb_display_id} ...')
ec82d85a 3890 try:
297e9952 3891 uf = self.urlopen(sanitized_Request(t['url'], headers=t.get('http_headers', {})))
80c03fa9 3892 self.to_screen(f'[info] Writing {thumb_display_id} to: {thumb_filename}')
d3d89c32 3893 with open(encodeFilename(thumb_filename), 'wb') as thumbf:
ec82d85a 3894 shutil.copyfileobj(uf, thumbf)
80c03fa9 3895 ret.append((thumb_filename, thumb_filename_final))
885cc0b7 3896 t['filepath'] = thumb_filename
3158150c 3897 except network_exceptions as err:
dd0228ce 3898 thumbnails.pop(idx)
80c03fa9 3899 self.report_warning(f'Unable to download {thumb_display_id}: {err}')
6c4fd172 3900 if ret and not write_all:
3901 break
0202b52a 3902 return ret