]> jfr.im git - yt-dlp.git/blame - yt_dlp/YoutubeDL.py
[DailyWire] Add extractors (#4084)
[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,
61 LINK_TEMPLATES,
62 NO_DEFAULT,
1d485a1a 63 NUMBER_RE,
f8271158 64 OUTTMPL_TYPES,
65 POSTPROCESS_WHEN,
66 STR_FORMAT_RE_TMPL,
67 STR_FORMAT_TYPES,
68 ContentTooShortError,
69 DateRange,
70 DownloadCancelled,
71 DownloadError,
72 EntryNotInPlaylist,
73 ExistingVideoReached,
74 ExtractorError,
75 GeoRestrictedError,
76 HEADRequest,
f8271158 77 ISO3166Utils,
78 LazyList,
79 MaxDownloadsReached,
19a03940 80 Namespace,
f8271158 81 PagedList,
82 PerRequestProxyHandler,
7e88d7d7 83 PlaylistEntries,
f8271158 84 Popen,
85 PostProcessingError,
86 ReExtractInfo,
87 RejectedVideoReached,
88 SameFileError,
89 UnavailableVideoError,
90 YoutubeDLCookieProcessor,
91 YoutubeDLHandler,
92 YoutubeDLRedirectHandler,
eedb7ba5
S
93 age_restricted,
94 args_to_str,
ce02ed60 95 date_from_str,
ce02ed60 96 determine_ext,
b5559424 97 determine_protocol,
c0384f22 98 encode_compat_str,
ce02ed60 99 encodeFilename,
a06916d9 100 error_to_compat_str,
590bc6f6 101 expand_path,
90137ca4 102 filter_dict,
e29663c6 103 float_or_none,
02dbf93f 104 format_bytes,
e0fd9573 105 format_decimal_suffix,
f8271158 106 format_field,
525ef922 107 formatSeconds,
0bb322b9 108 get_domain,
c9969434 109 int_or_none,
732044af 110 iri_to_uri,
34921b43 111 join_nonempty,
ce02ed60 112 locked_file,
0202b52a 113 make_dir,
dca08720 114 make_HTTPS_handler,
8b7539d2 115 merge_headers,
3158150c 116 network_exceptions,
ec11a9f4 117 number_of_digits,
cd6fc19e 118 orderedSet,
083c9df9 119 parse_filesize,
dca08720 120 platform_name,
ce02ed60 121 preferredencoding,
eedb7ba5 122 prepend_extension,
51fb4995 123 register_socks_protocols,
3efb96a6 124 remove_terminal_sequences,
cfb56d1a 125 render_table,
eedb7ba5 126 replace_extension,
ce02ed60 127 sanitize_filename,
1bb5c511 128 sanitize_path,
dcf77cf1 129 sanitize_url,
67dda517 130 sanitized_Request,
e5660ee6 131 std_headers,
1211bb6d 132 str_or_none,
e29663c6 133 strftime_or_none,
ce02ed60 134 subtitles_filename,
819e0531 135 supports_terminal_sequences,
f2ebc5c7 136 timetuple_from_msec,
732044af 137 to_high_limit_path,
324ad820 138 traverse_obj,
6033d980 139 try_get,
29eb5174 140 url_basename,
7d1eb38a 141 variadic,
58b1f00d 142 version_tuple,
53973b4d 143 windows_enable_vt_mode,
ce02ed60
PH
144 write_json_file,
145 write_string,
4f026faf 146)
f8271158 147from .version import RELEASE_GIT_HEAD, __version__
8222d8de 148
e9c0cdd3
YCH
149if compat_os_name == 'nt':
150 import ctypes
151
2459b6e1 152
86e5f3ed 153class YoutubeDL:
8222d8de
JMF
154 """YoutubeDL class.
155
156 YoutubeDL objects are the ones responsible of downloading the
157 actual video file and writing it to disk if the user has requested
158 it, among some other tasks. In most cases there should be one per
159 program. As, given a video URL, the downloader doesn't know how to
160 extract all the needed information, task that InfoExtractors do, it
161 has to pass the URL to one of them.
162
163 For this, YoutubeDL objects have a method that allows
164 InfoExtractors to be registered in a given order. When it is passed
165 a URL, the YoutubeDL object handles it to the first InfoExtractor it
166 finds that reports being able to handle it. The InfoExtractor extracts
167 all the information about the video or videos the URL refers to, and
168 YoutubeDL process the extracted information, possibly using a File
169 Downloader to download the video.
170
171 YoutubeDL objects accept a lot of parameters. In order not to saturate
172 the object constructor with arguments, it receives a dictionary of
173 options instead. These options are available through the params
174 attribute for the InfoExtractors to use. The YoutubeDL also
175 registers itself as the downloader in charge for the InfoExtractors
176 that are added to it, so this is a "mutual registration".
177
178 Available options:
179
180 username: Username for authentication purposes.
181 password: Password for authentication purposes.
180940e0 182 videopassword: Password for accessing a video.
1da50aa3
S
183 ap_mso: Adobe Pass multiple-system operator identifier.
184 ap_username: Multiple-system operator account username.
185 ap_password: Multiple-system operator account password.
8222d8de
JMF
186 usenetrc: Use netrc for authentication instead.
187 verbose: Print additional info to stdout.
188 quiet: Do not print messages to stdout.
ad8915b7 189 no_warnings: Do not print out anything for warnings.
bb66c247 190 forceprint: A dict with keys WHEN mapped to a list of templates to
191 print to stdout. The allowed keys are video or any of the
192 items in utils.POSTPROCESS_WHEN.
ca30f449 193 For compatibility, a single list is also accepted
bb66c247 194 print_to_file: A dict with keys WHEN (same as forceprint) mapped to
195 a list of tuples with (template, filename)
8694c600 196 forcejson: Force printing info_dict as JSON.
63e0be34
PH
197 dump_single_json: Force printing the info_dict of the whole playlist
198 (or video) as a single JSON line.
c25228e5 199 force_write_download_archive: Force writing download archive regardless
200 of 'skip_download' or 'simulate'.
b7b04c78 201 simulate: Do not download the video files. If unset (or None),
202 simulate only if listsubtitles, listformats or list_thumbnails is used
eb8a4433 203 format: Video format code. see "FORMAT SELECTION" for more details.
093a1710 204 You can also pass a function. The function takes 'ctx' as
205 argument and returns the formats to download.
206 See "build_format_selector" for an implementation
63ad4d43 207 allow_unplayable_formats: Allow unplayable formats to be extracted and downloaded.
b7da73eb 208 ignore_no_formats_error: Ignore "No video formats" error. Usefull for
209 extracting metadata even if the video is not actually
210 available for download (experimental)
0930b11f 211 format_sort: A list of fields by which to sort the video formats.
212 See "Sorting Formats" for more details.
c25228e5 213 format_sort_force: Force the given format_sort. see "Sorting Formats"
214 for more details.
08d30158 215 prefer_free_formats: Whether to prefer video formats with free containers
216 over non-free ones of same quality.
c25228e5 217 allow_multiple_video_streams: Allow multiple video streams to be merged
218 into a single file
219 allow_multiple_audio_streams: Allow multiple audio streams to be merged
220 into a single file
0ba692ac 221 check_formats Whether to test if the formats are downloadable.
9f1a1c36 222 Can be True (check all), False (check none),
223 'selected' (check selected formats),
0ba692ac 224 or None (check only if requested by extractor)
4524baf0 225 paths: Dictionary of output paths. The allowed keys are 'home'
226 'temp' and the keys of OUTTMPL_TYPES (in utils.py)
de6000d9 227 outtmpl: Dictionary of templates for output names. Allowed keys
4524baf0 228 are 'default' and the keys of OUTTMPL_TYPES (in utils.py).
34488702 229 For compatibility with youtube-dl, a single string can also be used
a820dc72
RA
230 outtmpl_na_placeholder: Placeholder for unavailable meta fields.
231 restrictfilenames: Do not allow "&" and spaces in file names
232 trim_file_name: Limit length of filename (extension excluded)
4524baf0 233 windowsfilenames: Force the filenames to be windows compatible
b1940459 234 ignoreerrors: Do not stop on download/postprocessing errors.
235 Can be 'only_download' to ignore only download errors.
236 Default is 'only_download' for CLI, but False for API
26e2805c 237 skip_playlist_after_errors: Number of allowed failures until the rest of
238 the playlist is skipped
d22dec74 239 force_generic_extractor: Force downloader to use the generic extractor
0c3d0f51 240 overwrites: Overwrite all video and metadata files if True,
241 overwrite only non-video files if None
242 and don't overwrite any file if False
34488702 243 For compatibility with youtube-dl,
244 "nooverwrites" may also be used instead
c14e88f0 245 playlist_items: Specific indices of playlist to download.
75822ca7 246 playlistrandom: Download playlist items in random order.
7e9a6125 247 lazy_playlist: Process playlist entries as they are received.
8222d8de
JMF
248 matchtitle: Download only matching titles.
249 rejecttitle: Reject downloads for matching titles.
8bf9319e 250 logger: Log messages to a logging.Logger instance.
8222d8de 251 logtostderr: Log messages to stderr instead of stdout.
819e0531 252 consoletitle: Display progress in console window's titlebar.
8222d8de
JMF
253 writedescription: Write the video description to a .description file
254 writeinfojson: Write the video description to a .info.json file
75d43ca0 255 clean_infojson: Remove private fields from the infojson
34488702 256 getcomments: Extract video comments. This will not be written to disk
06167fbb 257 unless writeinfojson is also given
1fb07d10 258 writeannotations: Write the video annotations to a .annotations.xml file
8222d8de 259 writethumbnail: Write the thumbnail image to a file
c25228e5 260 allow_playlist_files: Whether to write playlists' description, infojson etc
261 also to disk when using the 'write*' options
ec82d85a 262 write_all_thumbnails: Write all thumbnail formats to files
732044af 263 writelink: Write an internet shortcut file, depending on the
264 current platform (.url/.webloc/.desktop)
265 writeurllink: Write a Windows internet shortcut file (.url)
266 writewebloclink: Write a macOS internet shortcut file (.webloc)
267 writedesktoplink: Write a Linux internet shortcut file (.desktop)
8222d8de 268 writesubtitles: Write the video subtitles to a file
741dd8ea 269 writeautomaticsub: Write the automatically generated subtitles to a file
8222d8de 270 listsubtitles: Lists all available subtitles for the video
a504ced0 271 subtitlesformat: The format code for subtitles
c32b0aab 272 subtitleslangs: List of languages of the subtitles to download (can be regex).
273 The list may contain "all" to refer to all the available
274 subtitles. The language can be prefixed with a "-" to
275 exclude it from the requested languages. Eg: ['all', '-live_chat']
8222d8de
JMF
276 keepvideo: Keep the video file after post-processing
277 daterange: A DateRange object, download only if the upload_date is in the range.
278 skip_download: Skip the actual download of the video file
c35f9e72 279 cachedir: Location of the cache files in the filesystem.
a0e07d31 280 False to disable filesystem cache.
47192f92 281 noplaylist: Download single video instead of a playlist if in doubt.
8dbe9899
PH
282 age_limit: An integer representing the user's age in years.
283 Unsuitable videos for the given age are skipped.
5fe18bdb
PH
284 min_views: An integer representing the minimum view count the video
285 must have in order to not be skipped.
286 Videos without view count information are always
287 downloaded. None for no limit.
288 max_views: An integer representing the maximum view count.
289 Videos that are more popular than that are not
290 downloaded.
291 Videos without view count information are always
292 downloaded. None for no limit.
293 download_archive: File name of a file where all downloads are recorded.
c1c9a79c
PH
294 Videos already present in the file are not downloaded
295 again.
8a51f564 296 break_on_existing: Stop the download process after attempting to download a
297 file that is in the archive.
298 break_on_reject: Stop the download process when encountering a video that
299 has been filtered out.
b222c271 300 break_per_url: Whether break_on_reject and break_on_existing
301 should act on each input URL as opposed to for the entire queue
d76fa1f3 302 cookiefile: File name or text stream from where cookies should be read and dumped to
f59f5ef8
MB
303 cookiesfrombrowser: A tuple containing the name of the browser, the profile
304 name/pathfrom where cookies are loaded, and the name of the
305 keyring. Eg: ('chrome', ) or ('vivaldi', 'default', 'BASICTEXT')
f81c62a6 306 legacyserverconnect: Explicitly allow HTTPS connection to servers that do not
307 support RFC 5746 secure renegotiation
f59f5ef8 308 nocheckcertificate: Do not verify SSL certificates
bb58c9ed 309 client_certificate: Path to client certificate file in PEM format. May include the private key
310 client_certificate_key: Path to private key file for client certificate
311 client_certificate_password: Password for client certificate private key, if encrypted.
312 If not provided and the key is encrypted, yt-dlp will ask interactively
7e8c0af0
PH
313 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
314 At the moment, this is only supported by YouTube.
8b7539d2 315 http_headers: A dictionary of custom headers to be used for all requests
a1ee09e8 316 proxy: URL of the proxy server to use
38cce791 317 geo_verification_proxy: URL of the proxy to use for IP address verification
504f20dd 318 on geo-restricted sites.
e344693b 319 socket_timeout: Time to wait for unresponsive hosts, in seconds
0783b09b
PH
320 bidi_workaround: Work around buggy terminals without bidirectional text
321 support, using fridibi
a0ddb8a2 322 debug_printtraffic:Print out sent and received HTTP traffic
04b4d394
PH
323 default_search: Prepend this string if an input url is not valid.
324 'auto' for elaborate guessing
62fec3b2 325 encoding: Use this encoding instead of the system-specified.
e8ee972c 326 extract_flat: Do not resolve URLs, return the immediate result.
057a5206
PH
327 Pass in 'in_playlist' to only show this behavior for
328 playlist items.
f2ebc5c7 329 wait_for_video: If given, wait for scheduled streams to become available.
330 The value should be a tuple containing the range
331 (min_secs, max_secs) to wait between retries
4f026faf 332 postprocessors: A list of dictionaries, each with an entry
71b640cc 333 * key: The name of the postprocessor. See
7a5c1cfe 334 yt_dlp/postprocessor/__init__.py for a list.
bb66c247 335 * when: When to run the postprocessor. Allowed values are
336 the entries of utils.POSTPROCESS_WHEN
56d868db 337 Assumed to be 'post_process' if not given
71b640cc
PH
338 progress_hooks: A list of functions that get called on download
339 progress, with a dictionary with the entries
5cda4eda 340 * status: One of "downloading", "error", or "finished".
ee69b99a 341 Check this first and ignore unknown values.
3ba7740d 342 * info_dict: The extracted info_dict
71b640cc 343
5cda4eda 344 If status is one of "downloading", or "finished", the
ee69b99a
PH
345 following properties may also be present:
346 * filename: The final filename (always present)
5cda4eda 347 * tmpfilename: The filename we're currently writing to
71b640cc
PH
348 * downloaded_bytes: Bytes on disk
349 * total_bytes: Size of the whole file, None if unknown
5cda4eda
PH
350 * total_bytes_estimate: Guess of the eventual file size,
351 None if unavailable.
352 * elapsed: The number of seconds since download started.
71b640cc
PH
353 * eta: The estimated time in seconds, None if unknown
354 * speed: The download speed in bytes/second, None if
355 unknown
5cda4eda
PH
356 * fragment_index: The counter of the currently
357 downloaded video fragment.
358 * fragment_count: The number of fragments (= individual
359 files that will be merged)
71b640cc
PH
360
361 Progress hooks are guaranteed to be called at least once
362 (with status "finished") if the download is successful.
819e0531 363 postprocessor_hooks: A list of functions that get called on postprocessing
364 progress, with a dictionary with the entries
365 * status: One of "started", "processing", or "finished".
366 Check this first and ignore unknown values.
367 * postprocessor: Name of the postprocessor
368 * info_dict: The extracted info_dict
369
370 Progress hooks are guaranteed to be called at least twice
371 (with status "started" and "finished") if the processing is successful.
45598f15 372 merge_output_format: Extension to use when merging formats.
6b591b29 373 final_ext: Expected final extension; used to detect when the file was
59a7a13e 374 already downloaded and converted
6271f1ca
PH
375 fixup: Automatically correct known faults of the file.
376 One of:
377 - "never": do nothing
378 - "warn": only emit a warning
379 - "detect_or_warn": check whether we can do anything
62cd676c 380 about it, warn otherwise (default)
504f20dd 381 source_address: Client-side IP address to bind to.
1cf376f5 382 sleep_interval_requests: Number of seconds to sleep between requests
383 during extraction
7aa589a5
S
384 sleep_interval: Number of seconds to sleep before each download when
385 used alone or a lower bound of a range for randomized
386 sleep before each download (minimum possible number
387 of seconds to sleep) when used along with
388 max_sleep_interval.
389 max_sleep_interval:Upper bound of a range for randomized sleep before each
390 download (maximum possible number of seconds to sleep).
391 Must only be used along with sleep_interval.
392 Actual sleep time will be a random float from range
393 [sleep_interval; max_sleep_interval].
1cf376f5 394 sleep_interval_subtitles: Number of seconds to sleep before each subtitle download
cfb56d1a
PH
395 listformats: Print an overview of available video formats and exit.
396 list_thumbnails: Print a table of all thumbnails and exit.
0a41f331 397 match_filter: A function that gets called for every video with the signature
398 (info_dict, *, incomplete: bool) -> Optional[str]
399 For backward compatibility with youtube-dl, the signature
400 (info_dict) -> Optional[str] is also allowed.
401 - If it returns a message, the video is ignored.
402 - If it returns None, the video is downloaded.
403 - If it returns utils.NO_DEFAULT, the user is interactively
404 asked whether to download the video.
347de493 405 match_filter_func in utils.py is one example for this.
7e5db8c9 406 no_color: Do not emit color codes in output.
0a840f58 407 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
504f20dd 408 HTTP header
0a840f58 409 geo_bypass_country:
773f291d
S
410 Two-letter ISO 3166-2 country code that will be used for
411 explicit geographic restriction bypassing via faking
504f20dd 412 X-Forwarded-For HTTP header
5f95927a
S
413 geo_bypass_ip_block:
414 IP range in CIDR notation that will be used similarly to
504f20dd 415 geo_bypass_country
52a8a1e1 416 external_downloader: A dictionary of protocol keys and the executable of the
417 external downloader to use for it. The allowed protocols
418 are default|http|ftp|m3u8|dash|rtsp|rtmp|mms.
419 Set the value to 'native' to use the native downloader
53ed7066 420 compat_opts: Compatibility options. See "Differences in default behavior".
3acf6d38 421 The following options do not work when used through the API:
b5ae35ee 422 filename, abort-on-error, multistreams, no-live-chat, format-sort
dac5df5a 423 no-clean-infojson, no-playlist-metafiles, no-keep-subs, no-attach-info-json.
e4f02757 424 Refer __init__.py for their implementation
819e0531 425 progress_template: Dictionary of templates for progress outputs.
426 Allowed keys are 'download', 'postprocess',
427 'download-title' (console title) and 'postprocess-title'.
428 The template is mapped on a dictionary with keys 'progress' and 'info'
23326151 429 retry_sleep_functions: Dictionary of functions that takes the number of attempts
430 as argument and returns the time to sleep in seconds.
431 Allowed keys are 'http', 'fragment', 'file_access'
5ec1b6b7 432 download_ranges: A function that gets called for every video with the signature
433 (info_dict, *, ydl) -> Iterable[Section].
434 Only the returned sections will be downloaded. Each Section contains:
435 * start_time: Start time of the section in seconds
436 * end_time: End time of the section in seconds
437 * title: Section title (Optional)
438 * index: Section number (Optional)
fe7e0c98 439
8222d8de 440 The following parameters are not used by YoutubeDL itself, they are used by
7a5c1cfe 441 the downloader (see yt_dlp/downloader/common.py):
51d9739f 442 nopart, updatetime, buffersize, ratelimit, throttledratelimit, min_filesize,
205a0654
EH
443 max_filesize, test, noresizebuffer, retries, file_access_retries, fragment_retries,
444 continuedl, noprogress, xattr_set_filesize, hls_use_mpegts, http_chunk_size,
59a7a13e 445 external_downloader_args, concurrent_fragment_downloads.
76b1bd67
JMF
446
447 The following options are used by the post processors:
c0b7d117
S
448 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
449 to the binary or its containing directory.
43820c03 450 postprocessor_args: A dictionary of postprocessor/executable keys (in lower case)
34488702 451 and a list of additional command-line arguments for the
452 postprocessor/executable. The dict can also have "PP+EXE" keys
453 which are used when the given exe is used by the given PP.
454 Use 'default' as the name for arguments to passed to all PP
455 For compatibility with youtube-dl, a single list of args
456 can also be used
e409895f 457
458 The following options are used by the extractors:
62bff2c1 459 extractor_retries: Number of times to retry for known errors
460 dynamic_mpd: Whether to process dynamic DASH manifests (default: True)
e409895f 461 hls_split_discontinuity: Split HLS playlists to different formats at
62bff2c1 462 discontinuities such as ad breaks (default: False)
5d3a0e79 463 extractor_args: A dictionary of arguments to be passed to the extractors.
464 See "EXTRACTOR ARGUMENTS" for details.
465 Eg: {'youtube': {'skip': ['dash', 'hls']}}
88f23a18 466 mark_watched: Mark videos watched (even with --simulate). Only for YouTube
1890fc63 467
468 The following options are deprecated and may be removed in the future:
469
7e9a6125 470 playliststart: - Use playlist_items
471 Playlist item to start at.
472 playlistend: - Use playlist_items
473 Playlist item to end at.
474 playlistreverse: - Use playlist_items
475 Download playlist items in reverse order.
1890fc63 476 forceurl: - Use forceprint
477 Force printing final URL.
478 forcetitle: - Use forceprint
479 Force printing title.
480 forceid: - Use forceprint
481 Force printing ID.
482 forcethumbnail: - Use forceprint
483 Force printing thumbnail URL.
484 forcedescription: - Use forceprint
485 Force printing description.
486 forcefilename: - Use forceprint
487 Force printing final filename.
488 forceduration: - Use forceprint
489 Force printing duration.
490 allsubtitles: - Use subtitleslangs = ['all']
491 Downloads all the subtitles of the video
492 (requires writesubtitles or writeautomaticsub)
493 include_ads: - Doesn't work
494 Download ads as well
495 call_home: - Not implemented
496 Boolean, true iff we are allowed to contact the
497 yt-dlp servers for debugging.
498 post_hooks: - Register a custom postprocessor
499 A list of functions that get called as the final step
500 for each video file, after all postprocessors have been
501 called. The filename will be passed as the only argument.
502 hls_prefer_native: - Use external_downloader = {'m3u8': 'native'} or {'m3u8': 'ffmpeg'}.
503 Use the native HLS downloader instead of ffmpeg/avconv
504 if True, otherwise use ffmpeg/avconv if False, otherwise
505 use downloader suggested by extractor if None.
506 prefer_ffmpeg: - avconv support is deprecated
507 If False, use avconv instead of ffmpeg if both are available,
508 otherwise prefer ffmpeg.
509 youtube_include_dash_manifest: - Use extractor_args
5d3a0e79 510 If True (default), DASH manifests and related
62bff2c1 511 data will be downloaded and processed by extractor.
512 You can reduce network I/O by disabling it if you don't
513 care about DASH. (only for youtube)
1890fc63 514 youtube_include_hls_manifest: - Use extractor_args
5d3a0e79 515 If True (default), HLS manifests and related
62bff2c1 516 data will be downloaded and processed by extractor.
517 You can reduce network I/O by disabling it if you don't
518 care about HLS. (only for youtube)
8222d8de
JMF
519 """
520
86e5f3ed 521 _NUMERIC_FIELDS = {
c9969434 522 'width', 'height', 'tbr', 'abr', 'asr', 'vbr', 'fps', 'filesize', 'filesize_approx',
e6f21b3d 523 'timestamp', 'release_timestamp',
c9969434
S
524 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
525 'average_rating', 'comment_count', 'age_limit',
526 'start_time', 'end_time',
527 'chapter_number', 'season_number', 'episode_number',
528 'track_number', 'disc_number', 'release_year',
86e5f3ed 529 }
c9969434 530
6db9c4d5 531 _format_fields = {
532 # NB: Keep in sync with the docstring of extractor/common.py
a44ca5a4 533 'url', 'manifest_url', 'manifest_stream_number', 'ext', 'format', 'format_id', 'format_note',
6db9c4d5 534 'width', 'height', 'resolution', 'dynamic_range', 'tbr', 'abr', 'acodec', 'asr',
535 'vbr', 'fps', 'vcodec', 'container', 'filesize', 'filesize_approx',
536 'player_url', 'protocol', 'fragment_base_url', 'fragments', 'is_from_start',
537 'preference', 'language', 'language_preference', 'quality', 'source_preference',
538 'http_headers', 'stretched_ratio', 'no_resume', 'has_drm', 'downloader_options',
539 'page_url', 'app', 'play_path', 'tc_url', 'flash_version', 'rtmp_live', 'rtmp_conn', 'rtmp_protocol', 'rtmp_real_time'
540 }
48ee10ee 541 _format_selection_exts = {
542 'audio': {'m4a', 'mp3', 'ogg', 'aac'},
543 'video': {'mp4', 'flv', 'webm', '3gp'},
544 'storyboards': {'mhtml'},
545 }
546
3511266b 547 def __init__(self, params=None, auto_init=True):
883d4b1e 548 """Create a FileDownloader object with the given options.
549 @param auto_init Whether to load the default extractors and print header (if verbose).
49a57e70 550 Set to 'no_verbose_header' to not print the header
883d4b1e 551 """
e9f9a10f
JMF
552 if params is None:
553 params = {}
592b7485 554 self.params = params
8b7491c8 555 self._ies = {}
56c73665 556 self._ies_instances = {}
1e43a6f7 557 self._pps = {k: [] for k in POSTPROCESS_WHEN}
b35496d8 558 self._printed_messages = set()
1cf376f5 559 self._first_webpage_request = True
ab8e5e51 560 self._post_hooks = []
933605d7 561 self._progress_hooks = []
819e0531 562 self._postprocessor_hooks = []
8222d8de
JMF
563 self._download_retcode = 0
564 self._num_downloads = 0
9c906919 565 self._num_videos = 0
592b7485 566 self._playlist_level = 0
567 self._playlist_urls = set()
a0e07d31 568 self.cache = Cache(self)
34308b30 569
819e0531 570 windows_enable_vt_mode()
591bb9d3 571 stdout = sys.stderr if self.params.get('logtostderr') else sys.stdout
572 self._out_files = Namespace(
573 out=stdout,
574 error=sys.stderr,
575 screen=sys.stderr if self.params.get('quiet') else stdout,
576 console=None if compat_os_name == 'nt' else next(
cf4f42cb 577 filter(supports_terminal_sequences, (sys.stderr, sys.stdout)), None)
591bb9d3 578 )
579 self._allow_colors = Namespace(**{
580 type_: not self.params.get('no_color') and supports_terminal_sequences(stream)
64fa820c 581 for type_, stream in self._out_files.items_ if type_ != 'console'
591bb9d3 582 })
819e0531 583
a61f4b28 584 if sys.version_info < (3, 6):
585 self.report_warning(
0181adef 586 'Python version %d.%d is not supported! Please update to Python 3.6 or above' % sys.version_info[:2])
a61f4b28 587
88acdbc2 588 if self.params.get('allow_unplayable_formats'):
589 self.report_warning(
ec11a9f4 590 f'You have asked for {self._format_err("UNPLAYABLE", self.Styles.EMPHASIS)} formats to be listed/downloaded. '
819e0531 591 'This is a developer option intended for debugging. \n'
592 ' If you experience any issues while using this option, '
ec11a9f4 593 f'{self._format_err("DO NOT", self.Styles.ERROR)} open a bug report')
88acdbc2 594
be5df5ee
S
595 def check_deprecated(param, option, suggestion):
596 if self.params.get(param) is not None:
86e5f3ed 597 self.report_warning(f'{option} is deprecated. Use {suggestion} instead')
be5df5ee
S
598 return True
599 return False
600
601 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
38cce791
YCH
602 if self.params.get('geo_verification_proxy') is None:
603 self.params['geo_verification_proxy'] = self.params['cn_verification_proxy']
604
0d1bb027 605 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
606 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
53ed7066 607 check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"')
0d1bb027 608
49a57e70 609 for msg in self.params.get('_warnings', []):
0d1bb027 610 self.report_warning(msg)
ee8dd27a 611 for msg in self.params.get('_deprecation_warnings', []):
612 self.deprecation_warning(msg)
0d1bb027 613
8a82af35 614 self.params['compat_opts'] = set(self.params.get('compat_opts', ()))
615 if not compat_has_legacy:
616 self.params['compat_opts'].add('no-compat-legacy')
617 if 'list-formats' in self.params['compat_opts']:
ec11a9f4 618 self.params['listformats_table'] = False
619
b5ae35ee 620 if 'overwrites' not in self.params and self.params.get('nooverwrites') is not None:
b868936c 621 # nooverwrites was unnecessarily changed to overwrites
622 # in 0c3d0f51778b153f65c21906031c2e091fcfb641
623 # This ensures compatibility with both keys
624 self.params['overwrites'] = not self.params['nooverwrites']
b5ae35ee 625 elif self.params.get('overwrites') is None:
626 self.params.pop('overwrites', None)
b868936c 627 else:
628 self.params['nooverwrites'] = not self.params['overwrites']
b9d973be 629
455a15e2 630 self.params.setdefault('forceprint', {})
631 self.params.setdefault('print_to_file', {})
bb66c247 632
633 # Compatibility with older syntax
ca30f449 634 if not isinstance(params['forceprint'], dict):
455a15e2 635 self.params['forceprint'] = {'video': params['forceprint']}
ca30f449 636
455a15e2 637 if self.params.get('bidi_workaround', False):
1c088fa8
PH
638 try:
639 import pty
640 master, slave = pty.openpty()
003c69a8 641 width = compat_get_terminal_size().columns
591bb9d3 642 width_args = [] if width is None else ['-w', str(width)]
643 sp_kwargs = {'stdin': subprocess.PIPE, 'stdout': slave, 'stderr': self._out_files.error}
5d681e96 644 try:
d3c93ec2 645 self._output_process = Popen(['bidiv'] + width_args, **sp_kwargs)
5d681e96 646 except OSError:
d3c93ec2 647 self._output_process = Popen(['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
5d681e96 648 self._output_channel = os.fdopen(master, 'rb')
1c088fa8 649 except OSError as ose:
66e7ace1 650 if ose.errno == errno.ENOENT:
49a57e70 651 self.report_warning(
652 'Could not find fribidi executable, ignoring --bidi-workaround. '
653 'Make sure that fribidi is an executable file in one of the directories in your $PATH.')
1c088fa8
PH
654 else:
655 raise
0783b09b 656
97ec5bc5 657 if auto_init:
658 if auto_init != 'no_verbose_header':
659 self.print_debug_header()
660 self.add_default_info_extractors()
661
3089bc74
S
662 if (sys.platform != 'win32'
663 and sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
455a15e2 664 and not self.params.get('restrictfilenames', False)):
e9137224 665 # Unicode filesystem API will throw errors (#1474, #13027)
34308b30 666 self.report_warning(
6febd1c1 667 'Assuming --restrict-filenames since file system encoding '
1b725173 668 'cannot encode all characters. '
6febd1c1 669 'Set the LC_ALL environment variable to fix this.')
4a98cdbf 670 self.params['restrictfilenames'] = True
34308b30 671
bf1824b3 672 self._parse_outtmpl()
486dd09e 673
187986a8 674 # Creating format selector here allows us to catch syntax errors before the extraction
675 self.format_selector = (
fa9f30b8 676 self.params.get('format') if self.params.get('format') in (None, '-')
093a1710 677 else self.params['format'] if callable(self.params['format'])
187986a8 678 else self.build_format_selector(self.params['format']))
679
8b7539d2 680 # Set http_headers defaults according to std_headers
681 self.params['http_headers'] = merge_headers(std_headers, self.params.get('http_headers', {}))
682
013b50b7 683 hooks = {
684 'post_hooks': self.add_post_hook,
685 'progress_hooks': self.add_progress_hook,
686 'postprocessor_hooks': self.add_postprocessor_hook,
687 }
688 for opt, fn in hooks.items():
689 for ph in self.params.get(opt, []):
690 fn(ph)
71b640cc 691
5bfc8bee 692 for pp_def_raw in self.params.get('postprocessors', []):
693 pp_def = dict(pp_def_raw)
694 when = pp_def.pop('when', 'post_process')
695 self.add_post_processor(
f9934b96 696 get_postprocessor(pp_def.pop('key'))(self, **pp_def),
5bfc8bee 697 when=when)
698
97ec5bc5 699 self._setup_opener()
51fb4995
YCH
700 register_socks_protocols()
701
ed39cac5 702 def preload_download_archive(fn):
703 """Preload the archive, if any is specified"""
704 if fn is None:
705 return False
49a57e70 706 self.write_debug(f'Loading archive file {fn!r}')
ed39cac5 707 try:
708 with locked_file(fn, 'r', encoding='utf-8') as archive_file:
709 for line in archive_file:
710 self.archive.add(line.strip())
86e5f3ed 711 except OSError as ioe:
ed39cac5 712 if ioe.errno != errno.ENOENT:
713 raise
714 return False
715 return True
716
717 self.archive = set()
718 preload_download_archive(self.params.get('download_archive'))
719
7d4111ed
PH
720 def warn_if_short_id(self, argv):
721 # short YouTube ID starting with dash?
722 idxs = [
723 i for i, a in enumerate(argv)
724 if re.match(r'^-[0-9A-Za-z_-]{10}$', a)]
725 if idxs:
726 correct_argv = (
7a5c1cfe 727 ['yt-dlp']
3089bc74
S
728 + [a for i, a in enumerate(argv) if i not in idxs]
729 + ['--'] + [argv[i] for i in idxs]
7d4111ed
PH
730 )
731 self.report_warning(
732 'Long argument string detected. '
49a57e70 733 'Use -- to separate parameters and URLs, like this:\n%s' %
7d4111ed
PH
734 args_to_str(correct_argv))
735
8222d8de
JMF
736 def add_info_extractor(self, ie):
737 """Add an InfoExtractor object to the end of the list."""
8b7491c8 738 ie_key = ie.ie_key()
739 self._ies[ie_key] = ie
e52d7f85 740 if not isinstance(ie, type):
8b7491c8 741 self._ies_instances[ie_key] = ie
e52d7f85 742 ie.set_downloader(self)
8222d8de 743
8b7491c8 744 def _get_info_extractor_class(self, ie_key):
745 ie = self._ies.get(ie_key)
746 if ie is None:
747 ie = get_info_extractor(ie_key)
748 self.add_info_extractor(ie)
749 return ie
750
56c73665
JMF
751 def get_info_extractor(self, ie_key):
752 """
753 Get an instance of an IE with name ie_key, it will try to get one from
754 the _ies list, if there's no instance it will create a new one and add
755 it to the extractor list.
756 """
757 ie = self._ies_instances.get(ie_key)
758 if ie is None:
759 ie = get_info_extractor(ie_key)()
760 self.add_info_extractor(ie)
761 return ie
762
023fa8c4
JMF
763 def add_default_info_extractors(self):
764 """
765 Add the InfoExtractors returned by gen_extractors to the end of the list
766 """
e52d7f85 767 for ie in gen_extractor_classes():
023fa8c4
JMF
768 self.add_info_extractor(ie)
769
56d868db 770 def add_post_processor(self, pp, when='post_process'):
8222d8de 771 """Add a PostProcessor object to the end of the chain."""
5bfa4862 772 self._pps[when].append(pp)
8222d8de
JMF
773 pp.set_downloader(self)
774
ab8e5e51
AM
775 def add_post_hook(self, ph):
776 """Add the post hook"""
777 self._post_hooks.append(ph)
778
933605d7 779 def add_progress_hook(self, ph):
819e0531 780 """Add the download progress hook"""
933605d7 781 self._progress_hooks.append(ph)
8ab470f1 782
819e0531 783 def add_postprocessor_hook(self, ph):
784 """Add the postprocessing progress hook"""
785 self._postprocessor_hooks.append(ph)
5bfc8bee 786 for pps in self._pps.values():
787 for pp in pps:
788 pp.add_progress_hook(ph)
819e0531 789
1c088fa8 790 def _bidi_workaround(self, message):
5d681e96 791 if not hasattr(self, '_output_channel'):
1c088fa8
PH
792 return message
793
5d681e96 794 assert hasattr(self, '_output_process')
11b85ce6 795 assert isinstance(message, compat_str)
6febd1c1 796 line_count = message.count('\n') + 1
0f06bcd7 797 self._output_process.stdin.write((message + '\n').encode())
5d681e96 798 self._output_process.stdin.flush()
0f06bcd7 799 res = ''.join(self._output_channel.readline().decode()
9e1a5b84 800 for _ in range(line_count))
6febd1c1 801 return res[:-len('\n')]
1c088fa8 802
b35496d8 803 def _write_string(self, message, out=None, only_once=False):
804 if only_once:
805 if message in self._printed_messages:
806 return
807 self._printed_messages.add(message)
808 write_string(message, out=out, encoding=self.params.get('encoding'))
734f90bb 809
cf4f42cb 810 def to_stdout(self, message, skip_eol=False, quiet=None):
0760b0a7 811 """Print message to stdout"""
cf4f42cb 812 if quiet is not None:
ae6a1b95 813 self.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument quiet. Use "YoutubeDL.to_screen" instead')
8a82af35 814 if skip_eol is not False:
815 self.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument skip_eol. Use "YoutubeDL.to_screen" instead')
0bf9dc1e 816 self._write_string(f'{self._bidi_workaround(message)}\n', self._out_files.out)
cf4f42cb 817
818 def to_screen(self, message, skip_eol=False, quiet=None):
819 """Print message to screen if not in quiet mode"""
8bf9319e 820 if self.params.get('logger'):
43afe285 821 self.params['logger'].debug(message)
cf4f42cb 822 return
823 if (self.params.get('quiet') if quiet is None else quiet) and not self.params.get('verbose'):
824 return
825 self._write_string(
826 '%s%s' % (self._bidi_workaround(message), ('' if skip_eol else '\n')),
591bb9d3 827 self._out_files.screen)
8222d8de 828
b35496d8 829 def to_stderr(self, message, only_once=False):
0760b0a7 830 """Print message to stderr"""
11b85ce6 831 assert isinstance(message, compat_str)
8bf9319e 832 if self.params.get('logger'):
43afe285
IB
833 self.params['logger'].error(message)
834 else:
5792c950 835 self._write_string(f'{self._bidi_workaround(message)}\n', self._out_files.error, only_once=only_once)
cf4f42cb 836
837 def _send_console_code(self, code):
591bb9d3 838 if compat_os_name == 'nt' or not self._out_files.console:
cf4f42cb 839 return
591bb9d3 840 self._write_string(code, self._out_files.console)
8222d8de 841
1e5b9a95
PH
842 def to_console_title(self, message):
843 if not self.params.get('consoletitle', False):
844 return
3efb96a6 845 message = remove_terminal_sequences(message)
4bede0d8
C
846 if compat_os_name == 'nt':
847 if ctypes.windll.kernel32.GetConsoleWindow():
848 # c_wchar_p() might not be necessary if `message` is
849 # already of type unicode()
850 ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
cf4f42cb 851 else:
852 self._send_console_code(f'\033]0;{message}\007')
1e5b9a95 853
bdde425c 854 def save_console_title(self):
cf4f42cb 855 if not self.params.get('consoletitle') or self.params.get('simulate'):
bdde425c 856 return
592b7485 857 self._send_console_code('\033[22;0t') # Save the title on stack
bdde425c
PH
858
859 def restore_console_title(self):
cf4f42cb 860 if not self.params.get('consoletitle') or self.params.get('simulate'):
bdde425c 861 return
592b7485 862 self._send_console_code('\033[23;0t') # Restore the title from stack
bdde425c
PH
863
864 def __enter__(self):
865 self.save_console_title()
866 return self
867
868 def __exit__(self, *args):
869 self.restore_console_title()
f89197d7 870
dca08720 871 if self.params.get('cookiefile') is not None:
1bab3437 872 self.cookiejar.save(ignore_discard=True, ignore_expires=True)
bdde425c 873
fa9f30b8 874 def trouble(self, message=None, tb=None, is_error=True):
8222d8de
JMF
875 """Determine action to take when a download problem appears.
876
877 Depending on if the downloader has been configured to ignore
878 download errors or not, this method may throw an exception or
879 not when errors are found, after printing the message.
880
fa9f30b8 881 @param tb If given, is additional traceback information
882 @param is_error Whether to raise error according to ignorerrors
8222d8de
JMF
883 """
884 if message is not None:
885 self.to_stderr(message)
886 if self.params.get('verbose'):
887 if tb is None:
888 if sys.exc_info()[0]: # if .trouble has been called from an except block
6febd1c1 889 tb = ''
8222d8de 890 if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
6febd1c1 891 tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
c0384f22 892 tb += encode_compat_str(traceback.format_exc())
8222d8de
JMF
893 else:
894 tb_data = traceback.format_list(traceback.extract_stack())
6febd1c1 895 tb = ''.join(tb_data)
c19bc311 896 if tb:
897 self.to_stderr(tb)
fa9f30b8 898 if not is_error:
899 return
b1940459 900 if not self.params.get('ignoreerrors'):
8222d8de
JMF
901 if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
902 exc_info = sys.exc_info()[1].exc_info
903 else:
904 exc_info = sys.exc_info()
905 raise DownloadError(message, exc_info)
906 self._download_retcode = 1
907
19a03940 908 Styles = Namespace(
909 HEADERS='yellow',
910 EMPHASIS='light blue',
492272fe 911 FILENAME='green',
19a03940 912 ID='green',
913 DELIM='blue',
914 ERROR='red',
915 WARNING='yellow',
916 SUPPRESS='light black',
917 )
ec11a9f4 918
7578d77d 919 def _format_text(self, handle, allow_colors, text, f, fallback=None, *, test_encoding=False):
e5a998f3 920 text = str(text)
ec11a9f4 921 if test_encoding:
922 original_text = text
5c104538 923 # handle.encoding can be None. See https://github.com/yt-dlp/yt-dlp/issues/2711
924 encoding = self.params.get('encoding') or getattr(handle, 'encoding', None) or 'ascii'
ec11a9f4 925 text = text.encode(encoding, 'ignore').decode(encoding)
926 if fallback is not None and text != original_text:
927 text = fallback
7578d77d 928 return format_text(text, f) if allow_colors else text if fallback is None else fallback
ec11a9f4 929
591bb9d3 930 def _format_out(self, *args, **kwargs):
931 return self._format_text(self._out_files.out, self._allow_colors.out, *args, **kwargs)
932
ec11a9f4 933 def _format_screen(self, *args, **kwargs):
591bb9d3 934 return self._format_text(self._out_files.screen, self._allow_colors.screen, *args, **kwargs)
ec11a9f4 935
936 def _format_err(self, *args, **kwargs):
591bb9d3 937 return self._format_text(self._out_files.error, self._allow_colors.error, *args, **kwargs)
819e0531 938
c84aeac6 939 def report_warning(self, message, only_once=False):
8222d8de
JMF
940 '''
941 Print the message to stderr, it will be prefixed with 'WARNING:'
942 If stderr is a tty file the 'WARNING:' will be colored
943 '''
6d07ce01
JMF
944 if self.params.get('logger') is not None:
945 self.params['logger'].warning(message)
8222d8de 946 else:
ad8915b7
PH
947 if self.params.get('no_warnings'):
948 return
ec11a9f4 949 self.to_stderr(f'{self._format_err("WARNING:", self.Styles.WARNING)} {message}', only_once)
8222d8de 950
ee8dd27a 951 def deprecation_warning(self, message):
952 if self.params.get('logger') is not None:
a44ca5a4 953 self.params['logger'].warning(f'DeprecationWarning: {message}')
ee8dd27a 954 else:
955 self.to_stderr(f'{self._format_err("DeprecationWarning:", self.Styles.ERROR)} {message}', True)
956
fa9f30b8 957 def report_error(self, message, *args, **kwargs):
8222d8de
JMF
958 '''
959 Do the same as trouble, but prefixes the message with 'ERROR:', colored
960 in red if stderr is a tty file.
961 '''
fa9f30b8 962 self.trouble(f'{self._format_err("ERROR:", self.Styles.ERROR)} {message}', *args, **kwargs)
8222d8de 963
b35496d8 964 def write_debug(self, message, only_once=False):
0760b0a7 965 '''Log debug message or Print message to stderr'''
966 if not self.params.get('verbose', False):
967 return
8a82af35 968 message = f'[debug] {message}'
0760b0a7 969 if self.params.get('logger'):
970 self.params['logger'].debug(message)
971 else:
b35496d8 972 self.to_stderr(message, only_once)
0760b0a7 973
8222d8de
JMF
974 def report_file_already_downloaded(self, file_name):
975 """Report file has already been fully downloaded."""
976 try:
6febd1c1 977 self.to_screen('[download] %s has already been downloaded' % file_name)
ce02ed60 978 except UnicodeEncodeError:
6febd1c1 979 self.to_screen('[download] The file has already been downloaded')
8222d8de 980
0c3d0f51 981 def report_file_delete(self, file_name):
982 """Report that existing file will be deleted."""
983 try:
c25228e5 984 self.to_screen('Deleting existing file %s' % file_name)
0c3d0f51 985 except UnicodeEncodeError:
c25228e5 986 self.to_screen('Deleting existing file')
0c3d0f51 987
319b6059 988 def raise_no_formats(self, info, forced=False, *, msg=None):
0a5a191a 989 has_drm = info.get('_has_drm')
319b6059 990 ignored, expected = self.params.get('ignore_no_formats_error'), bool(msg)
991 msg = msg or has_drm and 'This video is DRM protected' or 'No video formats found!'
992 if forced or not ignored:
1151c407 993 raise ExtractorError(msg, video_id=info['id'], ie=info['extractor'],
319b6059 994 expected=has_drm or ignored or expected)
88acdbc2 995 else:
996 self.report_warning(msg)
997
de6000d9 998 def parse_outtmpl(self):
bf1824b3 999 self.deprecation_warning('"YoutubeDL.parse_outtmpl" is deprecated and may be removed in a future version')
1000 self._parse_outtmpl()
1001 return self.params['outtmpl']
1002
1003 def _parse_outtmpl(self):
1004 sanitize = lambda x: x
1005 if self.params.get('restrictfilenames'): # Remove spaces in the default template
71ce444a 1006 sanitize = lambda x: x.replace(' - ', ' ').replace(' ', '-')
bf1824b3 1007
1008 outtmpl = self.params.setdefault('outtmpl', {})
1009 if not isinstance(outtmpl, dict):
1010 self.params['outtmpl'] = outtmpl = {'default': outtmpl}
1011 outtmpl.update({k: sanitize(v) for k, v in DEFAULT_OUTTMPL.items() if outtmpl.get(k) is None})
de6000d9 1012
21cd8fae 1013 def get_output_path(self, dir_type='', filename=None):
1014 paths = self.params.get('paths', {})
1015 assert isinstance(paths, dict)
1016 path = os.path.join(
1017 expand_path(paths.get('home', '').strip()),
1018 expand_path(paths.get(dir_type, '').strip()) if dir_type else '',
1019 filename or '')
21cd8fae 1020 return sanitize_path(path, force=self.params.get('windowsfilenames'))
1021
76a264ac 1022 @staticmethod
901130bb 1023 def _outtmpl_expandpath(outtmpl):
1024 # expand_path translates '%%' into '%' and '$$' into '$'
1025 # correspondingly that is not what we want since we need to keep
1026 # '%%' intact for template dict substitution step. Working around
1027 # with boundary-alike separator hack.
1028 sep = ''.join([random.choice(ascii_letters) for _ in range(32)])
86e5f3ed 1029 outtmpl = outtmpl.replace('%%', f'%{sep}%').replace('$$', f'${sep}$')
901130bb 1030
1031 # outtmpl should be expand_path'ed before template dict substitution
1032 # because meta fields may contain env variables we don't want to
1033 # be expanded. For example, for outtmpl "%(title)s.%(ext)s" and
1034 # title "Hello $PATH", we don't want `$PATH` to be expanded.
1035 return expand_path(outtmpl).replace(sep, '')
1036
1037 @staticmethod
1038 def escape_outtmpl(outtmpl):
1039 ''' Escape any remaining strings like %s, %abc% etc. '''
1040 return re.sub(
1041 STR_FORMAT_RE_TMPL.format('', '(?![%(\0])'),
1042 lambda mobj: ('' if mobj.group('has_key') else '%') + mobj.group(0),
1043 outtmpl)
1044
1045 @classmethod
1046 def validate_outtmpl(cls, outtmpl):
76a264ac 1047 ''' @return None or Exception object '''
7d1eb38a 1048 outtmpl = re.sub(
37893bb0 1049 STR_FORMAT_RE_TMPL.format('[^)]*', '[ljqBUDS]'),
7d1eb38a 1050 lambda mobj: f'{mobj.group(0)[:-1]}s',
1051 cls._outtmpl_expandpath(outtmpl))
76a264ac 1052 try:
7d1eb38a 1053 cls.escape_outtmpl(outtmpl) % collections.defaultdict(int)
76a264ac 1054 return None
1055 except ValueError as err:
1056 return err
1057
03b4de72 1058 @staticmethod
1059 def _copy_infodict(info_dict):
1060 info_dict = dict(info_dict)
09b49e1f 1061 info_dict.pop('__postprocessors', None)
415f8d51 1062 info_dict.pop('__pending_error', None)
03b4de72 1063 return info_dict
1064
e0fd9573 1065 def prepare_outtmpl(self, outtmpl, info_dict, sanitize=False):
1066 """ Make the outtmpl and info_dict suitable for substitution: ydl.escape_outtmpl(outtmpl) % info_dict
1067 @param sanitize Whether to sanitize the output as a filename.
1068 For backward compatibility, a function can also be passed
1069 """
1070
6e84b215 1071 info_dict.setdefault('epoch', int(time.time())) # keep epoch consistent once set
143db31d 1072
03b4de72 1073 info_dict = self._copy_infodict(info_dict)
752cda38 1074 info_dict['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
53c18592 1075 formatSeconds(info_dict['duration'], '-' if sanitize else ':')
143db31d 1076 if info_dict.get('duration', None) is not None
1077 else None)
1d485a1a 1078 info_dict['autonumber'] = int(self.params.get('autonumber_start', 1) - 1 + self._num_downloads)
9c906919 1079 info_dict['video_autonumber'] = self._num_videos
752cda38 1080 if info_dict.get('resolution') is None:
1081 info_dict['resolution'] = self.format_resolution(info_dict, default=None)
143db31d 1082
e6f21b3d 1083 # For fields playlist_index, playlist_autonumber and autonumber convert all occurrences
143db31d 1084 # of %(field)s to %(field)0Nd for backward compatibility
1085 field_size_compat_map = {
0a5a191a 1086 'playlist_index': number_of_digits(info_dict.get('__last_playlist_index') or 0),
ec11a9f4 1087 'playlist_autonumber': number_of_digits(info_dict.get('n_entries') or 0),
752cda38 1088 'autonumber': self.params.get('autonumber_size') or 5,
143db31d 1089 }
752cda38 1090
385a27fa 1091 TMPL_DICT = {}
37893bb0 1092 EXTERNAL_FORMAT_RE = re.compile(STR_FORMAT_RE_TMPL.format('[^)]*', f'[{STR_FORMAT_TYPES}ljqBUDS]'))
385a27fa 1093 MATH_FUNCTIONS = {
1094 '+': float.__add__,
1095 '-': float.__sub__,
1096 }
e625be0d 1097 # Field is of the form key1.key2...
1098 # where keys (except first) can be string, int or slice
2b8a2973 1099 FIELD_RE = r'\w*(?:\.(?:\w+|{num}|{num}?(?::{num}?){{1,2}}))*'.format(num=r'(?:-?\d+)')
1d485a1a 1100 MATH_FIELD_RE = rf'(?:{FIELD_RE}|-?{NUMBER_RE})'
385a27fa 1101 MATH_OPERATORS_RE = r'(?:%s)' % '|'.join(map(re.escape, MATH_FUNCTIONS.keys()))
1d485a1a 1102 INTERNAL_FORMAT_RE = re.compile(rf'''(?x)
e625be0d 1103 (?P<negate>-)?
1d485a1a 1104 (?P<fields>{FIELD_RE})
1105 (?P<maths>(?:{MATH_OPERATORS_RE}{MATH_FIELD_RE})*)
e625be0d 1106 (?:>(?P<strf_format>.+?))?
34baa9fd 1107 (?P<remaining>
1108 (?P<alternate>(?<!\\),[^|&)]+)?
1109 (?:&(?P<replacement>.*?))?
1110 (?:\|(?P<default>.*?))?
1d485a1a 1111 )$''')
752cda38 1112
2b8a2973 1113 def _traverse_infodict(k):
1114 k = k.split('.')
1115 if k[0] == '':
1116 k.pop(0)
1117 return traverse_obj(info_dict, k, is_user_input=True, traverse_string=True)
76a264ac 1118
752cda38 1119 def get_value(mdict):
1120 # Object traversal
2b8a2973 1121 value = _traverse_infodict(mdict['fields'])
752cda38 1122 # Negative
1123 if mdict['negate']:
1124 value = float_or_none(value)
1125 if value is not None:
1126 value *= -1
1127 # Do maths
385a27fa 1128 offset_key = mdict['maths']
1129 if offset_key:
752cda38 1130 value = float_or_none(value)
1131 operator = None
385a27fa 1132 while offset_key:
1133 item = re.match(
1134 MATH_FIELD_RE if operator else MATH_OPERATORS_RE,
1135 offset_key).group(0)
1136 offset_key = offset_key[len(item):]
1137 if operator is None:
752cda38 1138 operator = MATH_FUNCTIONS[item]
385a27fa 1139 continue
1140 item, multiplier = (item[1:], -1) if item[0] == '-' else (item, 1)
1141 offset = float_or_none(item)
1142 if offset is None:
2b8a2973 1143 offset = float_or_none(_traverse_infodict(item))
385a27fa 1144 try:
1145 value = operator(value, multiplier * offset)
1146 except (TypeError, ZeroDivisionError):
1147 return None
1148 operator = None
752cda38 1149 # Datetime formatting
1150 if mdict['strf_format']:
7c37ff97 1151 value = strftime_or_none(value, mdict['strf_format'].replace('\\,', ','))
752cda38 1152
1153 return value
1154
b868936c 1155 na = self.params.get('outtmpl_na_placeholder', 'NA')
1156
e0fd9573 1157 def filename_sanitizer(key, value, restricted=self.params.get('restrictfilenames')):
5c3895ff 1158 return sanitize_filename(str(value), restricted=restricted, is_id=(
1159 bool(re.search(r'(^|[_.])id(\.|$)', key))
8a82af35 1160 if 'filename-sanitization' in self.params['compat_opts']
5c3895ff 1161 else NO_DEFAULT))
e0fd9573 1162
1163 sanitizer = sanitize if callable(sanitize) else filename_sanitizer
1164 sanitize = bool(sanitize)
1165
6e84b215 1166 def _dumpjson_default(obj):
1167 if isinstance(obj, (set, LazyList)):
1168 return list(obj)
adbc4ec4 1169 return repr(obj)
6e84b215 1170
752cda38 1171 def create_key(outer_mobj):
1172 if not outer_mobj.group('has_key'):
b836dc94 1173 return outer_mobj.group(0)
752cda38 1174 key = outer_mobj.group('key')
752cda38 1175 mobj = re.match(INTERNAL_FORMAT_RE, key)
e0fd9573 1176 initial_field = mobj.group('fields') if mobj else ''
e978789f 1177 value, replacement, default = None, None, na
7c37ff97 1178 while mobj:
e625be0d 1179 mobj = mobj.groupdict()
7c37ff97 1180 default = mobj['default'] if mobj['default'] is not None else default
752cda38 1181 value = get_value(mobj)
e978789f 1182 replacement = mobj['replacement']
7c37ff97 1183 if value is None and mobj['alternate']:
34baa9fd 1184 mobj = re.match(INTERNAL_FORMAT_RE, mobj['remaining'][1:])
7c37ff97 1185 else:
1186 break
752cda38 1187
b868936c 1188 fmt = outer_mobj.group('format')
752cda38 1189 if fmt == 's' and value is not None and key in field_size_compat_map.keys():
86e5f3ed 1190 fmt = f'0{field_size_compat_map[key]:d}d'
752cda38 1191
e978789f 1192 value = default if value is None else value if replacement is None else replacement
752cda38 1193
4476d2c7 1194 flags = outer_mobj.group('conversion') or ''
7d1eb38a 1195 str_fmt = f'{fmt[:-1]}s'
524e2e4f 1196 if fmt[-1] == 'l': # list
4476d2c7 1197 delim = '\n' if '#' in flags else ', '
9e907ebd 1198 value, fmt = delim.join(map(str, variadic(value, allowed_types=(str, bytes)))), str_fmt
524e2e4f 1199 elif fmt[-1] == 'j': # json
4476d2c7 1200 value, fmt = json.dumps(value, default=_dumpjson_default, indent=4 if '#' in flags else None), str_fmt
524e2e4f 1201 elif fmt[-1] == 'q': # quoted
4476d2c7 1202 value = map(str, variadic(value) if '#' in flags else [value])
1203 value, fmt = ' '.join(map(compat_shlex_quote, value)), str_fmt
524e2e4f 1204 elif fmt[-1] == 'B': # bytes
0f06bcd7 1205 value = f'%{str_fmt}'.encode() % str(value).encode()
f5aa5cfb 1206 value, fmt = value.decode('utf-8', 'ignore'), 's'
524e2e4f 1207 elif fmt[-1] == 'U': # unicode normalized
524e2e4f 1208 value, fmt = unicodedata.normalize(
1209 # "+" = compatibility equivalence, "#" = NFD
4476d2c7 1210 'NF%s%s' % ('K' if '+' in flags else '', 'D' if '#' in flags else 'C'),
524e2e4f 1211 value), str_fmt
e0fd9573 1212 elif fmt[-1] == 'D': # decimal suffix
abbeeebc 1213 num_fmt, fmt = fmt[:-1].replace('#', ''), 's'
1214 value = format_decimal_suffix(value, f'%{num_fmt}f%s' if num_fmt else '%d%s',
1215 factor=1024 if '#' in flags else 1000)
37893bb0 1216 elif fmt[-1] == 'S': # filename sanitization
e0fd9573 1217 value, fmt = filename_sanitizer(initial_field, value, restricted='#' in flags), str_fmt
7d1eb38a 1218 elif fmt[-1] == 'c':
524e2e4f 1219 if value:
1220 value = str(value)[0]
76a264ac 1221 else:
524e2e4f 1222 fmt = str_fmt
76a264ac 1223 elif fmt[-1] not in 'rs': # numeric
a439a3a4 1224 value = float_or_none(value)
752cda38 1225 if value is None:
1226 value, fmt = default, 's'
901130bb 1227
752cda38 1228 if sanitize:
1229 if fmt[-1] == 'r':
1230 # If value is an object, sanitize might convert it to a string
1231 # So we convert it to repr first
7d1eb38a 1232 value, fmt = repr(value), str_fmt
639f1cea 1233 if fmt[-1] in 'csr':
e0fd9573 1234 value = sanitizer(initial_field, value)
901130bb 1235
b868936c 1236 key = '%s\0%s' % (key.replace('%', '%\0'), outer_mobj.group('format'))
385a27fa 1237 TMPL_DICT[key] = value
b868936c 1238 return '{prefix}%({key}){fmt}'.format(key=key, fmt=fmt, prefix=outer_mobj.group('prefix'))
752cda38 1239
385a27fa 1240 return EXTERNAL_FORMAT_RE.sub(create_key, outtmpl), TMPL_DICT
143db31d 1241
819e0531 1242 def evaluate_outtmpl(self, outtmpl, info_dict, *args, **kwargs):
1243 outtmpl, info_dict = self.prepare_outtmpl(outtmpl, info_dict, *args, **kwargs)
1244 return self.escape_outtmpl(outtmpl) % info_dict
1245
5127e92a 1246 def _prepare_filename(self, info_dict, *, outtmpl=None, tmpl_type=None):
1247 assert None in (outtmpl, tmpl_type), 'outtmpl and tmpl_type are mutually exclusive'
1248 if outtmpl is None:
bf1824b3 1249 outtmpl = self.params['outtmpl'].get(tmpl_type or 'default', self.params['outtmpl']['default'])
8222d8de 1250 try:
5127e92a 1251 outtmpl = self._outtmpl_expandpath(outtmpl)
e0fd9573 1252 filename = self.evaluate_outtmpl(outtmpl, info_dict, True)
6a0546e3 1253 if not filename:
1254 return None
15da37c7 1255
5127e92a 1256 if tmpl_type in ('', 'temp'):
6a0546e3 1257 final_ext, ext = self.params.get('final_ext'), info_dict.get('ext')
1258 if final_ext and ext and final_ext != ext and filename.endswith(f'.{final_ext}'):
1259 filename = replace_extension(filename, ext, final_ext)
5127e92a 1260 elif tmpl_type:
6a0546e3 1261 force_ext = OUTTMPL_TYPES[tmpl_type]
1262 if force_ext:
1263 filename = replace_extension(filename, force_ext, info_dict.get('ext'))
de6000d9 1264
bdc3fd2f
U
1265 # https://github.com/blackjack4494/youtube-dlc/issues/85
1266 trim_file_name = self.params.get('trim_file_name', False)
1267 if trim_file_name:
5c22c63d 1268 no_ext, *ext = filename.rsplit('.', 2)
1269 filename = join_nonempty(no_ext[:trim_file_name], *ext, delim='.')
bdc3fd2f 1270
0202b52a 1271 return filename
8222d8de 1272 except ValueError as err:
6febd1c1 1273 self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')')
8222d8de
JMF
1274 return None
1275
5127e92a 1276 def prepare_filename(self, info_dict, dir_type='', *, outtmpl=None, warn=False):
1277 """Generate the output filename"""
1278 if outtmpl:
1279 assert not dir_type, 'outtmpl and dir_type are mutually exclusive'
1280 dir_type = None
1281 filename = self._prepare_filename(info_dict, tmpl_type=dir_type, outtmpl=outtmpl)
80c03fa9 1282 if not filename and dir_type not in ('', 'temp'):
1283 return ''
de6000d9 1284
c84aeac6 1285 if warn:
21cd8fae 1286 if not self.params.get('paths'):
de6000d9 1287 pass
1288 elif filename == '-':
c84aeac6 1289 self.report_warning('--paths is ignored when an outputting to stdout', only_once=True)
de6000d9 1290 elif os.path.isabs(filename):
c84aeac6 1291 self.report_warning('--paths is ignored since an absolute path is given in output template', only_once=True)
de6000d9 1292 if filename == '-' or not filename:
1293 return filename
1294
21cd8fae 1295 return self.get_output_path(dir_type, filename)
0202b52a 1296
120fe513 1297 def _match_entry(self, info_dict, incomplete=False, silent=False):
ecdec191 1298 """ Returns None if the file should be downloaded """
8222d8de 1299
c77495e3 1300 video_title = info_dict.get('title', info_dict.get('id', 'video'))
1301
8b0d7497 1302 def check_filter():
8b0d7497 1303 if 'title' in info_dict:
1304 # This can happen when we're just evaluating the playlist
1305 title = info_dict['title']
1306 matchtitle = self.params.get('matchtitle', False)
1307 if matchtitle:
1308 if not re.search(matchtitle, title, re.IGNORECASE):
1309 return '"' + title + '" title did not match pattern "' + matchtitle + '"'
1310 rejecttitle = self.params.get('rejecttitle', False)
1311 if rejecttitle:
1312 if re.search(rejecttitle, title, re.IGNORECASE):
1313 return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
1314 date = info_dict.get('upload_date')
1315 if date is not None:
1316 dateRange = self.params.get('daterange', DateRange())
1317 if date not in dateRange:
86e5f3ed 1318 return f'{date_from_str(date).isoformat()} upload date is not in range {dateRange}'
8b0d7497 1319 view_count = info_dict.get('view_count')
1320 if view_count is not None:
1321 min_views = self.params.get('min_views')
1322 if min_views is not None and view_count < min_views:
1323 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
1324 max_views = self.params.get('max_views')
1325 if max_views is not None and view_count > max_views:
1326 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
1327 if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')):
1328 return 'Skipping "%s" because it is age restricted' % video_title
8b0d7497 1329
8f18aca8 1330 match_filter = self.params.get('match_filter')
1331 if match_filter is not None:
1332 try:
1333 ret = match_filter(info_dict, incomplete=incomplete)
1334 except TypeError:
1335 # For backward compatibility
1336 ret = None if incomplete else match_filter(info_dict)
492272fe 1337 if ret is NO_DEFAULT:
1338 while True:
1339 filename = self._format_screen(self.prepare_filename(info_dict), self.Styles.FILENAME)
1340 reply = input(self._format_screen(
1341 f'Download "{filename}"? (Y/n): ', self.Styles.EMPHASIS)).lower().strip()
1342 if reply in {'y', ''}:
1343 return None
1344 elif reply == 'n':
1345 return f'Skipping {video_title}'
492272fe 1346 elif ret is not None:
8f18aca8 1347 return ret
8b0d7497 1348 return None
1349
c77495e3 1350 if self.in_download_archive(info_dict):
1351 reason = '%s has already been recorded in the archive' % video_title
1352 break_opt, break_err = 'break_on_existing', ExistingVideoReached
1353 else:
1354 reason = check_filter()
1355 break_opt, break_err = 'break_on_reject', RejectedVideoReached
8b0d7497 1356 if reason is not None:
120fe513 1357 if not silent:
1358 self.to_screen('[download] ' + reason)
c77495e3 1359 if self.params.get(break_opt, False):
1360 raise break_err()
8b0d7497 1361 return reason
fe7e0c98 1362
b6c45014
JMF
1363 @staticmethod
1364 def add_extra_info(info_dict, extra_info):
1365 '''Set the keys from extra_info in info dict if they are missing'''
1366 for key, value in extra_info.items():
1367 info_dict.setdefault(key, value)
1368
409e1828 1369 def extract_info(self, url, download=True, ie_key=None, extra_info=None,
61aa5ba3 1370 process=True, force_generic_extractor=False):
41d1cca3 1371 """
1372 Return a list with a dictionary for each video extracted.
1373
1374 Arguments:
1375 url -- URL to extract
1376
1377 Keyword arguments:
1378 download -- whether to download videos during extraction
1379 ie_key -- extractor key hint
1380 extra_info -- dictionary containing the extra values to add to each result
1381 process -- whether to resolve all unresolved references (URLs, playlist items),
1382 must be True for download to work.
1383 force_generic_extractor -- force using the generic extractor
1384 """
fe7e0c98 1385
409e1828 1386 if extra_info is None:
1387 extra_info = {}
1388
61aa5ba3 1389 if not ie_key and force_generic_extractor:
d22dec74
S
1390 ie_key = 'Generic'
1391
8222d8de 1392 if ie_key:
8b7491c8 1393 ies = {ie_key: self._get_info_extractor_class(ie_key)}
8222d8de
JMF
1394 else:
1395 ies = self._ies
1396
8b7491c8 1397 for ie_key, ie in ies.items():
8222d8de
JMF
1398 if not ie.suitable(url):
1399 continue
1400
1401 if not ie.working():
6febd1c1
PH
1402 self.report_warning('The program functionality for this site has been marked as broken, '
1403 'and will probably not work.')
8222d8de 1404
1151c407 1405 temp_id = ie.get_temp_id(url)
a0566bbf 1406 if temp_id is not None and self.in_download_archive({'id': temp_id, 'ie_key': ie_key}):
5e5be0c0 1407 self.to_screen(f'[{ie_key}] {temp_id}: has already been recorded in the archive')
1408 if self.params.get('break_on_existing', False):
1409 raise ExistingVideoReached()
a0566bbf 1410 break
8b7491c8 1411 return self.__extract_info(url, self.get_info_extractor(ie_key), download, extra_info, process)
a0566bbf 1412 else:
1413 self.report_error('no suitable InfoExtractor for URL %s' % url)
1414
7e88d7d7 1415 def _handle_extraction_exceptions(func):
b5ae35ee 1416 @functools.wraps(func)
a0566bbf 1417 def wrapper(self, *args, **kwargs):
6da22e7d 1418 while True:
1419 try:
1420 return func(self, *args, **kwargs)
1421 except (DownloadCancelled, LazyList.IndexError, PagedList.IndexError):
8222d8de 1422 raise
6da22e7d 1423 except ReExtractInfo as e:
1424 if e.expected:
1425 self.to_screen(f'{e}; Re-extracting data')
1426 else:
1427 self.to_stderr('\r')
1428 self.report_warning(f'{e}; Re-extracting data')
1429 continue
1430 except GeoRestrictedError as e:
1431 msg = e.msg
1432 if e.countries:
1433 msg += '\nThis video is available in %s.' % ', '.join(
1434 map(ISO3166Utils.short2full, e.countries))
1435 msg += '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1436 self.report_error(msg)
1437 except ExtractorError as e: # An error we somewhat expected
1438 self.report_error(str(e), e.format_traceback())
1439 except Exception as e:
1440 if self.params.get('ignoreerrors'):
1441 self.report_error(str(e), tb=encode_compat_str(traceback.format_exc()))
1442 else:
1443 raise
1444 break
a0566bbf 1445 return wrapper
1446
f2ebc5c7 1447 def _wait_for_video(self, ie_result):
1448 if (not self.params.get('wait_for_video')
1449 or ie_result.get('_type', 'video') != 'video'
1450 or ie_result.get('formats') or ie_result.get('url')):
1451 return
1452
1453 format_dur = lambda dur: '%02d:%02d:%02d' % timetuple_from_msec(dur * 1000)[:-1]
1454 last_msg = ''
1455
1456 def progress(msg):
1457 nonlocal last_msg
1458 self.to_screen(msg + ' ' * (len(last_msg) - len(msg)) + '\r', skip_eol=True)
1459 last_msg = msg
1460
1461 min_wait, max_wait = self.params.get('wait_for_video')
1462 diff = try_get(ie_result, lambda x: x['release_timestamp'] - time.time())
1463 if diff is None and ie_result.get('live_status') == 'is_upcoming':
16c620bc 1464 diff = round(random.uniform(min_wait, max_wait) if (max_wait and min_wait) else (max_wait or min_wait), 0)
f2ebc5c7 1465 self.report_warning('Release time of video is not known')
1466 elif (diff or 0) <= 0:
1467 self.report_warning('Video should already be available according to extracted info')
38d79fd1 1468 diff = min(max(diff or 0, min_wait or 0), max_wait or float('inf'))
f2ebc5c7 1469 self.to_screen(f'[wait] Waiting for {format_dur(diff)} - Press Ctrl+C to try now')
1470
1471 wait_till = time.time() + diff
1472 try:
1473 while True:
1474 diff = wait_till - time.time()
1475 if diff <= 0:
1476 progress('')
1477 raise ReExtractInfo('[wait] Wait period ended', expected=True)
1478 progress(f'[wait] Remaining time until next attempt: {self._format_screen(format_dur(diff), self.Styles.EMPHASIS)}')
1479 time.sleep(1)
1480 except KeyboardInterrupt:
1481 progress('')
1482 raise ReExtractInfo('[wait] Interrupted by user', expected=True)
1483 except BaseException as e:
1484 if not isinstance(e, ReExtractInfo):
1485 self.to_screen('')
1486 raise
1487
7e88d7d7 1488 @_handle_extraction_exceptions
58f197b7 1489 def __extract_info(self, url, ie, download, extra_info, process):
a0566bbf 1490 ie_result = ie.extract(url)
1491 if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
1492 return
1493 if isinstance(ie_result, list):
1494 # Backwards compatibility: old IE result format
1495 ie_result = {
1496 '_type': 'compat_list',
1497 'entries': ie_result,
1498 }
e37d0efb 1499 if extra_info.get('original_url'):
1500 ie_result.setdefault('original_url', extra_info['original_url'])
a0566bbf 1501 self.add_default_extra_info(ie_result, ie, url)
1502 if process:
f2ebc5c7 1503 self._wait_for_video(ie_result)
a0566bbf 1504 return self.process_ie_result(ie_result, download, extra_info)
8222d8de 1505 else:
a0566bbf 1506 return ie_result
fe7e0c98 1507
ea38e55f 1508 def add_default_extra_info(self, ie_result, ie, url):
6033d980 1509 if url is not None:
1510 self.add_extra_info(ie_result, {
1511 'webpage_url': url,
1512 'original_url': url,
57ebfca3 1513 })
1514 webpage_url = ie_result.get('webpage_url')
1515 if webpage_url:
1516 self.add_extra_info(ie_result, {
1517 'webpage_url_basename': url_basename(webpage_url),
1518 'webpage_url_domain': get_domain(webpage_url),
6033d980 1519 })
1520 if ie is not None:
1521 self.add_extra_info(ie_result, {
1522 'extractor': ie.IE_NAME,
1523 'extractor_key': ie.ie_key(),
1524 })
ea38e55f 1525
58adec46 1526 def process_ie_result(self, ie_result, download=True, extra_info=None):
8222d8de
JMF
1527 """
1528 Take the result of the ie(may be modified) and resolve all unresolved
1529 references (URLs, playlist items).
1530
1531 It will also download the videos if 'download'.
1532 Returns the resolved ie_result.
1533 """
58adec46 1534 if extra_info is None:
1535 extra_info = {}
e8ee972c
PH
1536 result_type = ie_result.get('_type', 'video')
1537
057a5206 1538 if result_type in ('url', 'url_transparent'):
134c6ea8 1539 ie_result['url'] = sanitize_url(ie_result['url'])
e37d0efb 1540 if ie_result.get('original_url'):
1541 extra_info.setdefault('original_url', ie_result['original_url'])
1542
057a5206 1543 extract_flat = self.params.get('extract_flat', False)
3089bc74
S
1544 if ((extract_flat == 'in_playlist' and 'playlist' in extra_info)
1545 or extract_flat is True):
ecb54191 1546 info_copy = ie_result.copy()
6033d980 1547 ie = try_get(ie_result.get('ie_key'), self.get_info_extractor)
360167b9 1548 if ie and not ie_result.get('id'):
4614bc22 1549 info_copy['id'] = ie.get_temp_id(ie_result['url'])
6033d980 1550 self.add_default_extra_info(info_copy, ie, ie_result['url'])
4614bc22 1551 self.add_extra_info(info_copy, extra_info)
b5475f11 1552 info_copy, _ = self.pre_process(info_copy)
ecb54191 1553 self.__forced_printings(info_copy, self.prepare_filename(info_copy), incomplete=True)
415f8d51 1554 self._raise_pending_errors(info_copy)
4614bc22 1555 if self.params.get('force_write_download_archive', False):
1556 self.record_download_archive(info_copy)
e8ee972c
PH
1557 return ie_result
1558
8222d8de 1559 if result_type == 'video':
b6c45014 1560 self.add_extra_info(ie_result, extra_info)
9c2b75b5 1561 ie_result = self.process_video_result(ie_result, download=download)
415f8d51 1562 self._raise_pending_errors(ie_result)
28b0eb0f 1563 additional_urls = (ie_result or {}).get('additional_urls')
9c2b75b5 1564 if additional_urls:
e9f4ccd1 1565 # TODO: Improve MetadataParserPP to allow setting a list
9c2b75b5 1566 if isinstance(additional_urls, compat_str):
1567 additional_urls = [additional_urls]
1568 self.to_screen(
1569 '[info] %s: %d additional URL(s) requested' % (ie_result['id'], len(additional_urls)))
1570 self.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls))
1571 ie_result['additional_entries'] = [
1572 self.extract_info(
b69fd25c 1573 url, download, extra_info=extra_info,
9c2b75b5 1574 force_generic_extractor=self.params.get('force_generic_extractor'))
1575 for url in additional_urls
1576 ]
1577 return ie_result
8222d8de
JMF
1578 elif result_type == 'url':
1579 # We have to add extra_info to the results because it may be
1580 # contained in a playlist
07cce701 1581 return self.extract_info(
1582 ie_result['url'], download,
1583 ie_key=ie_result.get('ie_key'),
1584 extra_info=extra_info)
7fc3fa05
PH
1585 elif result_type == 'url_transparent':
1586 # Use the information from the embedding page
1587 info = self.extract_info(
1588 ie_result['url'], ie_key=ie_result.get('ie_key'),
1589 extra_info=extra_info, download=False, process=False)
1590
1640eb09
S
1591 # extract_info may return None when ignoreerrors is enabled and
1592 # extraction failed with an error, don't crash and return early
1593 # in this case
1594 if not info:
1595 return info
1596
412c617d 1597 new_result = info.copy()
90137ca4 1598 new_result.update(filter_dict(ie_result, lambda k, v: (
1599 v is not None and k not in {'_type', 'url', 'id', 'extractor', 'extractor_key', 'ie_key'})))
7fc3fa05 1600
0563f7ac
S
1601 # Extracted info may not be a video result (i.e.
1602 # info.get('_type', 'video') != video) but rather an url or
1603 # url_transparent. In such cases outer metadata (from ie_result)
1604 # should be propagated to inner one (info). For this to happen
1605 # _type of info should be overridden with url_transparent. This
067aa17e 1606 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
0563f7ac
S
1607 if new_result.get('_type') == 'url':
1608 new_result['_type'] = 'url_transparent'
7fc3fa05
PH
1609
1610 return self.process_ie_result(
1611 new_result, download=download, extra_info=extra_info)
40fcba5e 1612 elif result_type in ('playlist', 'multi_video'):
30a074c2 1613 # Protect from infinite recursion due to recursively nested playlists
1614 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
1615 webpage_url = ie_result['webpage_url']
1616 if webpage_url in self._playlist_urls:
7e85e872 1617 self.to_screen(
30a074c2 1618 '[download] Skipping already downloaded playlist: %s'
1619 % ie_result.get('title') or ie_result.get('id'))
1620 return
7e85e872 1621
30a074c2 1622 self._playlist_level += 1
1623 self._playlist_urls.add(webpage_url)
03f83004 1624 self._fill_common_fields(ie_result, False)
bc516a3f 1625 self._sanitize_thumbnails(ie_result)
30a074c2 1626 try:
1627 return self.__process_playlist(ie_result, download)
1628 finally:
1629 self._playlist_level -= 1
1630 if not self._playlist_level:
1631 self._playlist_urls.clear()
8222d8de 1632 elif result_type == 'compat_list':
c9bf4114
PH
1633 self.report_warning(
1634 'Extractor %s returned a compat_list result. '
1635 'It needs to be updated.' % ie_result.get('extractor'))
5f6a1245 1636
8222d8de 1637 def _fixup(r):
b868936c 1638 self.add_extra_info(r, {
1639 'extractor': ie_result['extractor'],
1640 'webpage_url': ie_result['webpage_url'],
1641 'webpage_url_basename': url_basename(ie_result['webpage_url']),
0bb322b9 1642 'webpage_url_domain': get_domain(ie_result['webpage_url']),
b868936c 1643 'extractor_key': ie_result['extractor_key'],
1644 })
8222d8de
JMF
1645 return r
1646 ie_result['entries'] = [
b6c45014 1647 self.process_ie_result(_fixup(r), download, extra_info)
8222d8de
JMF
1648 for r in ie_result['entries']
1649 ]
1650 return ie_result
1651 else:
1652 raise Exception('Invalid result type: %s' % result_type)
1653
e92caff5 1654 def _ensure_dir_exists(self, path):
1655 return make_dir(path, self.report_error)
1656
3b603dbd 1657 @staticmethod
1658 def _playlist_infodict(ie_result, **kwargs):
1659 return {
1660 **ie_result,
1661 'playlist': ie_result.get('title') or ie_result.get('id'),
1662 'playlist_id': ie_result.get('id'),
1663 'playlist_title': ie_result.get('title'),
1664 'playlist_uploader': ie_result.get('uploader'),
1665 'playlist_uploader_id': ie_result.get('uploader_id'),
1666 'playlist_index': 0,
1667 **kwargs,
1668 }
1669
30a074c2 1670 def __process_playlist(self, ie_result, download):
7e88d7d7 1671 """Process each entry in the playlist"""
1672 title = ie_result.get('title') or ie_result.get('id') or '<Untitled>'
1673 self.to_screen(f'[download] Downloading playlist: {title}')
f0d785d3 1674
7e88d7d7 1675 all_entries = PlaylistEntries(self, ie_result)
7e9a6125 1676 entries = orderedSet(all_entries.get_requested_items(), lazy=True)
1677
1678 lazy = self.params.get('lazy_playlist')
1679 if lazy:
1680 resolved_entries, n_entries = [], 'N/A'
1681 ie_result['requested_entries'], ie_result['entries'] = None, None
1682 else:
1683 entries = resolved_entries = list(entries)
1684 n_entries = len(resolved_entries)
1685 ie_result['requested_entries'], ie_result['entries'] = tuple(zip(*resolved_entries)) or ([], [])
1686 if not ie_result.get('playlist_count'):
1687 # Better to do this after potentially exhausting entries
1688 ie_result['playlist_count'] = all_entries.get_full_count()
498f5606 1689
e08a85d8 1690 _infojson_written = False
0bfc53d0 1691 write_playlist_files = self.params.get('allow_playlist_files', True)
1692 if write_playlist_files and self.params.get('list_thumbnails'):
1693 self.list_thumbnails(ie_result)
1694 if write_playlist_files and not self.params.get('simulate'):
7e9a6125 1695 ie_copy = self._playlist_infodict(ie_result, n_entries=int_or_none(n_entries))
e08a85d8 1696 _infojson_written = self._write_info_json(
1697 'playlist', ie_result, self.prepare_filename(ie_copy, 'pl_infojson'))
1698 if _infojson_written is None:
80c03fa9 1699 return
1700 if self._write_description('playlist', ie_result,
1701 self.prepare_filename(ie_copy, 'pl_description')) is None:
1702 return
681de68e 1703 # TODO: This should be passed to ThumbnailsConvertor if necessary
80c03fa9 1704 self._write_thumbnails('playlist', ie_copy, self.prepare_filename(ie_copy, 'pl_thumbnail'))
30a074c2 1705
7e9a6125 1706 if lazy:
1707 if self.params.get('playlistreverse') or self.params.get('playlistrandom'):
1708 self.report_warning('playlistreverse and playlistrandom are not supported with lazy_playlist', only_once=True)
1709 elif self.params.get('playlistreverse'):
1710 entries.reverse()
1711 elif self.params.get('playlistrandom'):
30a074c2 1712 random.shuffle(entries)
1713
7e88d7d7 1714 self.to_screen(f'[{ie_result["extractor"]}] Playlist {title}: Downloading {n_entries} videos'
1715 f'{format_field(ie_result, "playlist_count", " of %s")}')
30a074c2 1716
26e2805c 1717 failures = 0
1718 max_failures = self.params.get('skip_playlist_after_errors') or float('inf')
7e9a6125 1719 for i, (playlist_index, entry) in enumerate(entries):
1720 if lazy:
1721 resolved_entries.append((playlist_index, entry))
1722
7e88d7d7 1723 # TODO: Add auto-generated fields
1724 if self._match_entry(entry, incomplete=True) is not None:
1725 continue
1726
19a03940 1727 self.to_screen('[download] Downloading video %s of %s' % (
7e9a6125 1728 self._format_screen(i + 1, self.Styles.ID), self._format_screen(n_entries, self.Styles.EMPHASIS)))
7e88d7d7 1729
1730 entry['__x_forwarded_for_ip'] = ie_result.get('__x_forwarded_for_ip')
7e9a6125 1731 if not lazy and 'playlist-index' in self.params.get('compat_opts', []):
1732 playlist_index = ie_result['requested_entries'][i]
1733
7e88d7d7 1734 entry_result = self.__process_iterable_entry(entry, download, {
7e9a6125 1735 'n_entries': int_or_none(n_entries),
1736 '__last_playlist_index': max(ie_result['requested_entries'] or (0, 0)),
f0d785d3 1737 'playlist_count': ie_result.get('playlist_count'),
71729754 1738 'playlist_index': playlist_index,
7e9a6125 1739 'playlist_autonumber': i + 1,
7e88d7d7 1740 'playlist': title,
30a074c2 1741 'playlist_id': ie_result.get('id'),
1742 'playlist_title': ie_result.get('title'),
1743 'playlist_uploader': ie_result.get('uploader'),
1744 'playlist_uploader_id': ie_result.get('uploader_id'),
30a074c2 1745 'extractor': ie_result['extractor'],
1746 'webpage_url': ie_result['webpage_url'],
1747 'webpage_url_basename': url_basename(ie_result['webpage_url']),
0bb322b9 1748 'webpage_url_domain': get_domain(ie_result['webpage_url']),
30a074c2 1749 'extractor_key': ie_result['extractor_key'],
7e88d7d7 1750 })
26e2805c 1751 if not entry_result:
1752 failures += 1
1753 if failures >= max_failures:
1754 self.report_error(
7e88d7d7 1755 f'Skipping the remaining entries in playlist "{title}" since {failures} items failed extraction')
26e2805c 1756 break
7e9a6125 1757 resolved_entries[i] = (playlist_index, entry_result)
7e88d7d7 1758
1759 # Update with processed data
7e9a6125 1760 ie_result['requested_entries'], ie_result['entries'] = tuple(zip(*resolved_entries)) or ([], [])
e08a85d8 1761
1762 # Write the updated info to json
cb96c5be 1763 if _infojson_written is True and self._write_info_json(
e08a85d8 1764 'updated playlist', ie_result,
1765 self.prepare_filename(ie_copy, 'pl_infojson'), overwrite=True) is None:
1766 return
ca30f449 1767
ed5835b4 1768 ie_result = self.run_all_pps('playlist', ie_result)
7e88d7d7 1769 self.to_screen(f'[download] Finished downloading playlist: {title}')
30a074c2 1770 return ie_result
1771
7e88d7d7 1772 @_handle_extraction_exceptions
a0566bbf 1773 def __process_iterable_entry(self, entry, download, extra_info):
1774 return self.process_ie_result(
1775 entry, download=download, extra_info=extra_info)
1776
67134eab
JMF
1777 def _build_format_filter(self, filter_spec):
1778 " Returns a function to filter the formats according to the filter_spec "
083c9df9
PH
1779
1780 OPERATORS = {
1781 '<': operator.lt,
1782 '<=': operator.le,
1783 '>': operator.gt,
1784 '>=': operator.ge,
1785 '=': operator.eq,
1786 '!=': operator.ne,
1787 }
67134eab 1788 operator_rex = re.compile(r'''(?x)\s*
187986a8 1789 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)\s*
1790 (?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1791 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)\s*
083c9df9 1792 ''' % '|'.join(map(re.escape, OPERATORS.keys())))
187986a8 1793 m = operator_rex.fullmatch(filter_spec)
9ddb6925
S
1794 if m:
1795 try:
1796 comparison_value = int(m.group('value'))
1797 except ValueError:
1798 comparison_value = parse_filesize(m.group('value'))
1799 if comparison_value is None:
1800 comparison_value = parse_filesize(m.group('value') + 'B')
1801 if comparison_value is None:
1802 raise ValueError(
1803 'Invalid value %r in format specification %r' % (
67134eab 1804 m.group('value'), filter_spec))
9ddb6925
S
1805 op = OPERATORS[m.group('op')]
1806
083c9df9 1807 if not m:
9ddb6925
S
1808 STR_OPERATORS = {
1809 '=': operator.eq,
10d33b34
YCH
1810 '^=': lambda attr, value: attr.startswith(value),
1811 '$=': lambda attr, value: attr.endswith(value),
1812 '*=': lambda attr, value: value in attr,
1ce9a3cb 1813 '~=': lambda attr, value: value.search(attr) is not None
9ddb6925 1814 }
187986a8 1815 str_operator_rex = re.compile(r'''(?x)\s*
1816 (?P<key>[a-zA-Z0-9._-]+)\s*
1ce9a3cb
LF
1817 (?P<negation>!\s*)?(?P<op>%s)\s*(?P<none_inclusive>\?\s*)?
1818 (?P<quote>["'])?
1819 (?P<value>(?(quote)(?:(?!(?P=quote))[^\\]|\\.)+|[\w.-]+))
1820 (?(quote)(?P=quote))\s*
9ddb6925 1821 ''' % '|'.join(map(re.escape, STR_OPERATORS.keys())))
187986a8 1822 m = str_operator_rex.fullmatch(filter_spec)
9ddb6925 1823 if m:
1ce9a3cb
LF
1824 if m.group('op') == '~=':
1825 comparison_value = re.compile(m.group('value'))
1826 else:
1827 comparison_value = re.sub(r'''\\([\\"'])''', r'\1', m.group('value'))
2cc779f4
S
1828 str_op = STR_OPERATORS[m.group('op')]
1829 if m.group('negation'):
e118a879 1830 op = lambda attr, value: not str_op(attr, value)
2cc779f4
S
1831 else:
1832 op = str_op
083c9df9 1833
9ddb6925 1834 if not m:
187986a8 1835 raise SyntaxError('Invalid filter specification %r' % filter_spec)
083c9df9
PH
1836
1837 def _filter(f):
1838 actual_value = f.get(m.group('key'))
1839 if actual_value is None:
1840 return m.group('none_inclusive')
1841 return op(actual_value, comparison_value)
67134eab
JMF
1842 return _filter
1843
9f1a1c36 1844 def _check_formats(self, formats):
1845 for f in formats:
1846 self.to_screen('[info] Testing format %s' % f['format_id'])
75689fe5 1847 path = self.get_output_path('temp')
1848 if not self._ensure_dir_exists(f'{path}/'):
1849 continue
1850 temp_file = tempfile.NamedTemporaryFile(suffix='.tmp', delete=False, dir=path or None)
9f1a1c36 1851 temp_file.close()
1852 try:
1853 success, _ = self.dl(temp_file.name, f, test=True)
8a82af35 1854 except (DownloadError, OSError, ValueError) + network_exceptions:
9f1a1c36 1855 success = False
1856 finally:
1857 if os.path.exists(temp_file.name):
1858 try:
1859 os.remove(temp_file.name)
1860 except OSError:
1861 self.report_warning('Unable to delete temporary file "%s"' % temp_file.name)
1862 if success:
1863 yield f
1864 else:
1865 self.to_screen('[info] Unable to download format %s. Skipping...' % f['format_id'])
1866
0017d9ad 1867 def _default_format_spec(self, info_dict, download=True):
0017d9ad 1868
af0f7428
S
1869 def can_merge():
1870 merger = FFmpegMergerPP(self)
1871 return merger.available and merger.can_merge()
1872
91ebc640 1873 prefer_best = (
b7b04c78 1874 not self.params.get('simulate')
91ebc640 1875 and download
1876 and (
1877 not can_merge()
21633673 1878 or info_dict.get('is_live') and not self.params.get('live_from_start')
bf1824b3 1879 or self.params['outtmpl']['default'] == '-'))
53ed7066 1880 compat = (
1881 prefer_best
1882 or self.params.get('allow_multiple_audio_streams', False)
8a82af35 1883 or 'format-spec' in self.params['compat_opts'])
91ebc640 1884
1885 return (
53ed7066 1886 'best/bestvideo+bestaudio' if prefer_best
1887 else 'bestvideo*+bestaudio/best' if not compat
91ebc640 1888 else 'bestvideo+bestaudio/best')
0017d9ad 1889
67134eab
JMF
1890 def build_format_selector(self, format_spec):
1891 def syntax_error(note, start):
1892 message = (
1893 'Invalid format specification: '
86e5f3ed 1894 '{}\n\t{}\n\t{}^'.format(note, format_spec, ' ' * start[1]))
67134eab
JMF
1895 return SyntaxError(message)
1896
1897 PICKFIRST = 'PICKFIRST'
1898 MERGE = 'MERGE'
1899 SINGLE = 'SINGLE'
0130afb7 1900 GROUP = 'GROUP'
67134eab
JMF
1901 FormatSelector = collections.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
1902
91ebc640 1903 allow_multiple_streams = {'audio': self.params.get('allow_multiple_audio_streams', False),
1904 'video': self.params.get('allow_multiple_video_streams', False)}
909d24dd 1905
9f1a1c36 1906 check_formats = self.params.get('check_formats') == 'selected'
e8e73840 1907
67134eab
JMF
1908 def _parse_filter(tokens):
1909 filter_parts = []
1910 for type, string, start, _, _ in tokens:
1911 if type == tokenize.OP and string == ']':
1912 return ''.join(filter_parts)
1913 else:
1914 filter_parts.append(string)
1915
232541df 1916 def _remove_unused_ops(tokens):
17cc1534 1917 # Remove operators that we don't use and join them with the surrounding strings
232541df
JMF
1918 # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
1919 ALLOWED_OPS = ('/', '+', ',', '(', ')')
1920 last_string, last_start, last_end, last_line = None, None, None, None
1921 for type, string, start, end, line in tokens:
1922 if type == tokenize.OP and string == '[':
1923 if last_string:
1924 yield tokenize.NAME, last_string, last_start, last_end, last_line
1925 last_string = None
1926 yield type, string, start, end, line
1927 # everything inside brackets will be handled by _parse_filter
1928 for type, string, start, end, line in tokens:
1929 yield type, string, start, end, line
1930 if type == tokenize.OP and string == ']':
1931 break
1932 elif type == tokenize.OP and string in ALLOWED_OPS:
1933 if last_string:
1934 yield tokenize.NAME, last_string, last_start, last_end, last_line
1935 last_string = None
1936 yield type, string, start, end, line
1937 elif type in [tokenize.NAME, tokenize.NUMBER, tokenize.OP]:
1938 if not last_string:
1939 last_string = string
1940 last_start = start
1941 last_end = end
1942 else:
1943 last_string += string
1944 if last_string:
1945 yield tokenize.NAME, last_string, last_start, last_end, last_line
1946
cf2ac6df 1947 def _parse_format_selection(tokens, inside_merge=False, inside_choice=False, inside_group=False):
67134eab
JMF
1948 selectors = []
1949 current_selector = None
1950 for type, string, start, _, _ in tokens:
1951 # ENCODING is only defined in python 3.x
1952 if type == getattr(tokenize, 'ENCODING', None):
1953 continue
1954 elif type in [tokenize.NAME, tokenize.NUMBER]:
1955 current_selector = FormatSelector(SINGLE, string, [])
1956 elif type == tokenize.OP:
cf2ac6df
JMF
1957 if string == ')':
1958 if not inside_group:
1959 # ')' will be handled by the parentheses group
1960 tokens.restore_last_token()
67134eab 1961 break
cf2ac6df 1962 elif inside_merge and string in ['/', ',']:
0130afb7
JMF
1963 tokens.restore_last_token()
1964 break
cf2ac6df
JMF
1965 elif inside_choice and string == ',':
1966 tokens.restore_last_token()
1967 break
1968 elif string == ',':
0a31a350
JMF
1969 if not current_selector:
1970 raise syntax_error('"," must follow a format selector', start)
67134eab
JMF
1971 selectors.append(current_selector)
1972 current_selector = None
1973 elif string == '/':
d96d604e
JMF
1974 if not current_selector:
1975 raise syntax_error('"/" must follow a format selector', start)
67134eab 1976 first_choice = current_selector
cf2ac6df 1977 second_choice = _parse_format_selection(tokens, inside_choice=True)
f5f4a27a 1978 current_selector = FormatSelector(PICKFIRST, (first_choice, second_choice), [])
67134eab
JMF
1979 elif string == '[':
1980 if not current_selector:
1981 current_selector = FormatSelector(SINGLE, 'best', [])
1982 format_filter = _parse_filter(tokens)
1983 current_selector.filters.append(format_filter)
0130afb7
JMF
1984 elif string == '(':
1985 if current_selector:
1986 raise syntax_error('Unexpected "("', start)
cf2ac6df
JMF
1987 group = _parse_format_selection(tokens, inside_group=True)
1988 current_selector = FormatSelector(GROUP, group, [])
67134eab 1989 elif string == '+':
d03cfdce 1990 if not current_selector:
1991 raise syntax_error('Unexpected "+"', start)
1992 selector_1 = current_selector
1993 selector_2 = _parse_format_selection(tokens, inside_merge=True)
1994 if not selector_2:
1995 raise syntax_error('Expected a selector', start)
1996 current_selector = FormatSelector(MERGE, (selector_1, selector_2), [])
67134eab 1997 else:
86e5f3ed 1998 raise syntax_error(f'Operator not recognized: "{string}"', start)
67134eab
JMF
1999 elif type == tokenize.ENDMARKER:
2000 break
2001 if current_selector:
2002 selectors.append(current_selector)
2003 return selectors
2004
f8d4ad9a 2005 def _merge(formats_pair):
2006 format_1, format_2 = formats_pair
2007
2008 formats_info = []
2009 formats_info.extend(format_1.get('requested_formats', (format_1,)))
2010 formats_info.extend(format_2.get('requested_formats', (format_2,)))
2011
2012 if not allow_multiple_streams['video'] or not allow_multiple_streams['audio']:
551f9388 2013 get_no_more = {'video': False, 'audio': False}
f8d4ad9a 2014 for (i, fmt_info) in enumerate(formats_info):
551f9388 2015 if fmt_info.get('acodec') == fmt_info.get('vcodec') == 'none':
2016 formats_info.pop(i)
2017 continue
2018 for aud_vid in ['audio', 'video']:
f8d4ad9a 2019 if not allow_multiple_streams[aud_vid] and fmt_info.get(aud_vid[0] + 'codec') != 'none':
2020 if get_no_more[aud_vid]:
2021 formats_info.pop(i)
f5510afe 2022 break
f8d4ad9a 2023 get_no_more[aud_vid] = True
2024
2025 if len(formats_info) == 1:
2026 return formats_info[0]
2027
2028 video_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('vcodec') != 'none']
2029 audio_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('acodec') != 'none']
2030
2031 the_only_video = video_fmts[0] if len(video_fmts) == 1 else None
2032 the_only_audio = audio_fmts[0] if len(audio_fmts) == 1 else None
2033
2034 output_ext = self.params.get('merge_output_format')
2035 if not output_ext:
2036 if the_only_video:
2037 output_ext = the_only_video['ext']
2038 elif the_only_audio and not video_fmts:
2039 output_ext = the_only_audio['ext']
2040 else:
2041 output_ext = 'mkv'
2042
975a0d0d 2043 filtered = lambda *keys: filter(None, (traverse_obj(fmt, *keys) for fmt in formats_info))
2044
f8d4ad9a 2045 new_dict = {
2046 'requested_formats': formats_info,
975a0d0d 2047 'format': '+'.join(filtered('format')),
2048 'format_id': '+'.join(filtered('format_id')),
f8d4ad9a 2049 'ext': output_ext,
975a0d0d 2050 'protocol': '+'.join(map(determine_protocol, formats_info)),
093a1710 2051 'language': '+'.join(orderedSet(filtered('language'))) or None,
2052 'format_note': '+'.join(orderedSet(filtered('format_note'))) or None,
2053 'filesize_approx': sum(filtered('filesize', 'filesize_approx')) or None,
975a0d0d 2054 'tbr': sum(filtered('tbr', 'vbr', 'abr')),
f8d4ad9a 2055 }
2056
2057 if the_only_video:
2058 new_dict.update({
2059 'width': the_only_video.get('width'),
2060 'height': the_only_video.get('height'),
2061 'resolution': the_only_video.get('resolution') or self.format_resolution(the_only_video),
2062 'fps': the_only_video.get('fps'),
49a57e70 2063 'dynamic_range': the_only_video.get('dynamic_range'),
f8d4ad9a 2064 'vcodec': the_only_video.get('vcodec'),
2065 'vbr': the_only_video.get('vbr'),
2066 'stretched_ratio': the_only_video.get('stretched_ratio'),
2067 })
2068
2069 if the_only_audio:
2070 new_dict.update({
2071 'acodec': the_only_audio.get('acodec'),
2072 'abr': the_only_audio.get('abr'),
975a0d0d 2073 'asr': the_only_audio.get('asr'),
f8d4ad9a 2074 })
2075
2076 return new_dict
2077
e8e73840 2078 def _check_formats(formats):
981052c9 2079 if not check_formats:
2080 yield from formats
b5ac45b1 2081 return
9f1a1c36 2082 yield from self._check_formats(formats)
e8e73840 2083
67134eab 2084 def _build_selector_function(selector):
909d24dd 2085 if isinstance(selector, list): # ,
67134eab
JMF
2086 fs = [_build_selector_function(s) for s in selector]
2087
317f7ab6 2088 def selector_function(ctx):
67134eab 2089 for f in fs:
981052c9 2090 yield from f(ctx)
67134eab 2091 return selector_function
909d24dd 2092
2093 elif selector.type == GROUP: # ()
0130afb7 2094 selector_function = _build_selector_function(selector.selector)
909d24dd 2095
2096 elif selector.type == PICKFIRST: # /
67134eab
JMF
2097 fs = [_build_selector_function(s) for s in selector.selector]
2098
317f7ab6 2099 def selector_function(ctx):
67134eab 2100 for f in fs:
317f7ab6 2101 picked_formats = list(f(ctx))
67134eab
JMF
2102 if picked_formats:
2103 return picked_formats
2104 return []
67134eab 2105
981052c9 2106 elif selector.type == MERGE: # +
2107 selector_1, selector_2 = map(_build_selector_function, selector.selector)
2108
2109 def selector_function(ctx):
adbc4ec4 2110 for pair in itertools.product(selector_1(ctx), selector_2(ctx)):
981052c9 2111 yield _merge(pair)
2112
909d24dd 2113 elif selector.type == SINGLE: # atom
598d185d 2114 format_spec = selector.selector or 'best'
909d24dd 2115
f8d4ad9a 2116 # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
909d24dd 2117 if format_spec == 'all':
2118 def selector_function(ctx):
9222c381 2119 yield from _check_formats(ctx['formats'][::-1])
f8d4ad9a 2120 elif format_spec == 'mergeall':
2121 def selector_function(ctx):
316f2650 2122 formats = list(_check_formats(
2123 f for f in ctx['formats'] if f.get('vcodec') != 'none' or f.get('acodec') != 'none'))
e01d6aa4 2124 if not formats:
2125 return
921b76ca 2126 merged_format = formats[-1]
2127 for f in formats[-2::-1]:
f8d4ad9a 2128 merged_format = _merge((merged_format, f))
2129 yield merged_format
909d24dd 2130
2131 else:
85e801a9 2132 format_fallback, seperate_fallback, format_reverse, format_idx = False, None, True, 1
eff63539 2133 mobj = re.match(
2134 r'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
2135 format_spec)
2136 if mobj is not None:
2137 format_idx = int_or_none(mobj.group('n'), default=1)
e8e73840 2138 format_reverse = mobj.group('bw')[0] == 'b'
eff63539 2139 format_type = (mobj.group('type') or [None])[0]
2140 not_format_type = {'v': 'a', 'a': 'v'}.get(format_type)
2141 format_modified = mobj.group('mod') is not None
909d24dd 2142
2143 format_fallback = not format_type and not format_modified # for b, w
8326b00a 2144 _filter_f = (
eff63539 2145 (lambda f: f.get('%scodec' % format_type) != 'none')
2146 if format_type and format_modified # bv*, ba*, wv*, wa*
2147 else (lambda f: f.get('%scodec' % not_format_type) == 'none')
2148 if format_type # bv, ba, wv, wa
2149 else (lambda f: f.get('vcodec') != 'none' and f.get('acodec') != 'none')
2150 if not format_modified # b, w
8326b00a 2151 else lambda f: True) # b*, w*
2152 filter_f = lambda f: _filter_f(f) and (
2153 f.get('vcodec') != 'none' or f.get('acodec') != 'none')
67134eab 2154 else:
48ee10ee 2155 if format_spec in self._format_selection_exts['audio']:
b11c04a8 2156 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none'
48ee10ee 2157 elif format_spec in self._format_selection_exts['video']:
b11c04a8 2158 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none' and f.get('vcodec') != 'none'
85e801a9 2159 seperate_fallback = lambda f: f.get('ext') == format_spec and f.get('vcodec') != 'none'
48ee10ee 2160 elif format_spec in self._format_selection_exts['storyboards']:
b11c04a8 2161 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') == 'none' and f.get('vcodec') == 'none'
2162 else:
b5ae35ee 2163 filter_f = lambda f: f.get('format_id') == format_spec # id
909d24dd 2164
2165 def selector_function(ctx):
2166 formats = list(ctx['formats'])
909d24dd 2167 matches = list(filter(filter_f, formats)) if filter_f is not None else formats
85e801a9 2168 if not matches:
2169 if format_fallback and ctx['incomplete_formats']:
2170 # for extractors with incomplete formats (audio only (soundcloud)
2171 # or video only (imgur)) best/worst will fallback to
2172 # best/worst {video,audio}-only format
2173 matches = formats
2174 elif seperate_fallback and not ctx['has_merged_format']:
2175 # for compatibility with youtube-dl when there is no pre-merged format
2176 matches = list(filter(seperate_fallback, formats))
981052c9 2177 matches = LazyList(_check_formats(matches[::-1 if format_reverse else 1]))
2178 try:
e8e73840 2179 yield matches[format_idx - 1]
4abea8ca 2180 except LazyList.IndexError:
981052c9 2181 return
083c9df9 2182
67134eab 2183 filters = [self._build_format_filter(f) for f in selector.filters]
083c9df9 2184
317f7ab6 2185 def final_selector(ctx):
adbc4ec4 2186 ctx_copy = dict(ctx)
67134eab 2187 for _filter in filters:
317f7ab6
S
2188 ctx_copy['formats'] = list(filter(_filter, ctx_copy['formats']))
2189 return selector_function(ctx_copy)
67134eab 2190 return final_selector
083c9df9 2191
0f06bcd7 2192 stream = io.BytesIO(format_spec.encode())
0130afb7 2193 try:
f9934b96 2194 tokens = list(_remove_unused_ops(tokenize.tokenize(stream.readline)))
0130afb7
JMF
2195 except tokenize.TokenError:
2196 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec)))
2197
86e5f3ed 2198 class TokenIterator:
0130afb7
JMF
2199 def __init__(self, tokens):
2200 self.tokens = tokens
2201 self.counter = 0
2202
2203 def __iter__(self):
2204 return self
2205
2206 def __next__(self):
2207 if self.counter >= len(self.tokens):
2208 raise StopIteration()
2209 value = self.tokens[self.counter]
2210 self.counter += 1
2211 return value
2212
2213 next = __next__
2214
2215 def restore_last_token(self):
2216 self.counter -= 1
2217
2218 parsed_selector = _parse_format_selection(iter(TokenIterator(tokens)))
67134eab 2219 return _build_selector_function(parsed_selector)
a9c58ad9 2220
e5660ee6 2221 def _calc_headers(self, info_dict):
8b7539d2 2222 res = merge_headers(self.params['http_headers'], info_dict.get('http_headers') or {})
e5660ee6 2223
c487cf00 2224 cookies = self._calc_cookies(info_dict['url'])
e5660ee6
JMF
2225 if cookies:
2226 res['Cookie'] = cookies
2227
0016b84e
S
2228 if 'X-Forwarded-For' not in res:
2229 x_forwarded_for_ip = info_dict.get('__x_forwarded_for_ip')
2230 if x_forwarded_for_ip:
2231 res['X-Forwarded-For'] = x_forwarded_for_ip
2232
e5660ee6
JMF
2233 return res
2234
c487cf00 2235 def _calc_cookies(self, url):
2236 pr = sanitized_Request(url)
e5660ee6 2237 self.cookiejar.add_cookie_header(pr)
662435f7 2238 return pr.get_header('Cookie')
e5660ee6 2239
9f1a1c36 2240 def _sort_thumbnails(self, thumbnails):
2241 thumbnails.sort(key=lambda t: (
2242 t.get('preference') if t.get('preference') is not None else -1,
2243 t.get('width') if t.get('width') is not None else -1,
2244 t.get('height') if t.get('height') is not None else -1,
2245 t.get('id') if t.get('id') is not None else '',
2246 t.get('url')))
2247
b0249bca 2248 def _sanitize_thumbnails(self, info_dict):
bc516a3f 2249 thumbnails = info_dict.get('thumbnails')
2250 if thumbnails is None:
2251 thumbnail = info_dict.get('thumbnail')
2252 if thumbnail:
2253 info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}]
9f1a1c36 2254 if not thumbnails:
2255 return
2256
2257 def check_thumbnails(thumbnails):
2258 for t in thumbnails:
2259 self.to_screen(f'[info] Testing thumbnail {t["id"]}')
2260 try:
2261 self.urlopen(HEADRequest(t['url']))
2262 except network_exceptions as err:
2263 self.to_screen(f'[info] Unable to connect to thumbnail {t["id"]} URL {t["url"]!r} - {err}. Skipping...')
2264 continue
2265 yield t
2266
2267 self._sort_thumbnails(thumbnails)
2268 for i, t in enumerate(thumbnails):
2269 if t.get('id') is None:
2270 t['id'] = '%d' % i
2271 if t.get('width') and t.get('height'):
2272 t['resolution'] = '%dx%d' % (t['width'], t['height'])
2273 t['url'] = sanitize_url(t['url'])
2274
2275 if self.params.get('check_formats') is True:
282f5709 2276 info_dict['thumbnails'] = LazyList(check_thumbnails(thumbnails[::-1]), reverse=True)
9f1a1c36 2277 else:
2278 info_dict['thumbnails'] = thumbnails
bc516a3f 2279
03f83004
LNO
2280 def _fill_common_fields(self, info_dict, is_video=True):
2281 # TODO: move sanitization here
2282 if is_video:
2283 # playlists are allowed to lack "title"
d4736fdb 2284 title = info_dict.get('title', NO_DEFAULT)
2285 if title is NO_DEFAULT:
03f83004
LNO
2286 raise ExtractorError('Missing "title" field in extractor result',
2287 video_id=info_dict['id'], ie=info_dict['extractor'])
d4736fdb 2288 info_dict['fulltitle'] = title
2289 if not title:
2290 if title == '':
2291 self.write_debug('Extractor gave empty title. Creating a generic title')
2292 else:
2293 self.report_warning('Extractor failed to obtain "title". Creating a generic title instead')
1d485a1a 2294 info_dict['title'] = f'{info_dict["extractor"].replace(":", "-")} video #{info_dict["id"]}'
03f83004
LNO
2295
2296 if info_dict.get('duration') is not None:
2297 info_dict['duration_string'] = formatSeconds(info_dict['duration'])
2298
2299 for ts_key, date_key in (
2300 ('timestamp', 'upload_date'),
2301 ('release_timestamp', 'release_date'),
2302 ('modified_timestamp', 'modified_date'),
2303 ):
2304 if info_dict.get(date_key) is None and info_dict.get(ts_key) is not None:
2305 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
2306 # see http://bugs.python.org/issue1646728)
19a03940 2307 with contextlib.suppress(ValueError, OverflowError, OSError):
03f83004
LNO
2308 upload_date = datetime.datetime.utcfromtimestamp(info_dict[ts_key])
2309 info_dict[date_key] = upload_date.strftime('%Y%m%d')
03f83004
LNO
2310
2311 live_keys = ('is_live', 'was_live')
2312 live_status = info_dict.get('live_status')
2313 if live_status is None:
2314 for key in live_keys:
2315 if info_dict.get(key) is False:
2316 continue
2317 if info_dict.get(key):
2318 live_status = key
2319 break
2320 if all(info_dict.get(key) is False for key in live_keys):
2321 live_status = 'not_live'
2322 if live_status:
2323 info_dict['live_status'] = live_status
2324 for key in live_keys:
2325 if info_dict.get(key) is None:
2326 info_dict[key] = (live_status == key)
2327
2328 # Auto generate title fields corresponding to the *_number fields when missing
2329 # in order to always have clean titles. This is very common for TV series.
2330 for field in ('chapter', 'season', 'episode'):
2331 if info_dict.get('%s_number' % field) is not None and not info_dict.get(field):
2332 info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field])
2333
415f8d51 2334 def _raise_pending_errors(self, info):
2335 err = info.pop('__pending_error', None)
2336 if err:
2337 self.report_error(err, tb=False)
2338
dd82ffea
JMF
2339 def process_video_result(self, info_dict, download=True):
2340 assert info_dict.get('_type', 'video') == 'video'
9c906919 2341 self._num_videos += 1
dd82ffea 2342
bec1fad2 2343 if 'id' not in info_dict:
fc08bdd6 2344 raise ExtractorError('Missing "id" field in extractor result', ie=info_dict['extractor'])
2345 elif not info_dict.get('id'):
2346 raise ExtractorError('Extractor failed to obtain "id"', ie=info_dict['extractor'])
455a15e2 2347
c9969434
S
2348 def report_force_conversion(field, field_not, conversion):
2349 self.report_warning(
2350 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
2351 % (field, field_not, conversion))
2352
2353 def sanitize_string_field(info, string_field):
2354 field = info.get(string_field)
2355 if field is None or isinstance(field, compat_str):
2356 return
2357 report_force_conversion(string_field, 'a string', 'string')
2358 info[string_field] = compat_str(field)
2359
2360 def sanitize_numeric_fields(info):
2361 for numeric_field in self._NUMERIC_FIELDS:
2362 field = info.get(numeric_field)
f9934b96 2363 if field is None or isinstance(field, (int, float)):
c9969434
S
2364 continue
2365 report_force_conversion(numeric_field, 'numeric', 'int')
2366 info[numeric_field] = int_or_none(field)
2367
2368 sanitize_string_field(info_dict, 'id')
2369 sanitize_numeric_fields(info_dict)
4c3f8c3f 2370 if (info_dict.get('duration') or 0) <= 0 and info_dict.pop('duration', None):
50e93e03 2371 self.report_warning('"duration" field is negative, there is an error in extractor')
be6217b2 2372
dd82ffea
JMF
2373 if 'playlist' not in info_dict:
2374 # It isn't part of a playlist
2375 info_dict['playlist'] = None
2376 info_dict['playlist_index'] = None
2377
bc516a3f 2378 self._sanitize_thumbnails(info_dict)
d5519808 2379
536a55da 2380 thumbnail = info_dict.get('thumbnail')
bc516a3f 2381 thumbnails = info_dict.get('thumbnails')
536a55da
S
2382 if thumbnail:
2383 info_dict['thumbnail'] = sanitize_url(thumbnail)
2384 elif thumbnails:
d5519808
PH
2385 info_dict['thumbnail'] = thumbnails[-1]['url']
2386
ae30b840 2387 if info_dict.get('display_id') is None and 'id' in info_dict:
0afef30b
PH
2388 info_dict['display_id'] = info_dict['id']
2389
03f83004 2390 self._fill_common_fields(info_dict)
33d2fc2f 2391
05108a49
S
2392 for cc_kind in ('subtitles', 'automatic_captions'):
2393 cc = info_dict.get(cc_kind)
2394 if cc:
2395 for _, subtitle in cc.items():
2396 for subtitle_format in subtitle:
2397 if subtitle_format.get('url'):
2398 subtitle_format['url'] = sanitize_url(subtitle_format['url'])
2399 if subtitle_format.get('ext') is None:
2400 subtitle_format['ext'] = determine_ext(subtitle_format['url']).lower()
2401
2402 automatic_captions = info_dict.get('automatic_captions')
4bba3716 2403 subtitles = info_dict.get('subtitles')
4bba3716 2404
360e1ca5 2405 info_dict['requested_subtitles'] = self.process_subtitles(
05108a49 2406 info_dict['id'], subtitles, automatic_captions)
a504ced0 2407
dd82ffea
JMF
2408 if info_dict.get('formats') is None:
2409 # There's only one format available
2410 formats = [info_dict]
2411 else:
2412 formats = info_dict['formats']
2413
0a5a191a 2414 # or None ensures --clean-infojson removes it
2415 info_dict['_has_drm'] = any(f.get('has_drm') for f in formats) or None
88acdbc2 2416 if not self.params.get('allow_unplayable_formats'):
2417 formats = [f for f in formats if not f.get('has_drm')]
0a5a191a 2418 if info_dict['_has_drm'] and all(
c0b6e5c7 2419 f.get('acodec') == f.get('vcodec') == 'none' for f in formats):
2420 self.report_warning(
2421 'This video is DRM protected and only images are available for download. '
2422 'Use --list-formats to see them')
88acdbc2 2423
319b6059 2424 get_from_start = not info_dict.get('is_live') or bool(self.params.get('live_from_start'))
2425 if not get_from_start:
2426 info_dict['title'] += ' ' + datetime.datetime.now().strftime('%Y-%m-%d %H:%M')
2427 if info_dict.get('is_live') and formats:
adbc4ec4 2428 formats = [f for f in formats if bool(f.get('is_from_start')) == get_from_start]
319b6059 2429 if get_from_start and not formats:
a44ca5a4 2430 self.raise_no_formats(info_dict, msg=(
2431 '--live-from-start is passed, but there are no formats that can be downloaded from the start. '
2432 'If you want to download from the current time, use --no-live-from-start'))
adbc4ec4 2433
db95dc13 2434 if not formats:
1151c407 2435 self.raise_no_formats(info_dict)
db95dc13 2436
73af5cc8
S
2437 def is_wellformed(f):
2438 url = f.get('url')
a5ac0c47 2439 if not url:
73af5cc8
S
2440 self.report_warning(
2441 '"url" field is missing or empty - skipping format, '
2442 'there is an error in extractor')
a5ac0c47
S
2443 return False
2444 if isinstance(url, bytes):
2445 sanitize_string_field(f, 'url')
2446 return True
73af5cc8
S
2447
2448 # Filter out malformed formats for better extraction robustness
2449 formats = list(filter(is_wellformed, formats))
2450
181c7053
S
2451 formats_dict = {}
2452
dd82ffea 2453 # We check that all the formats have the format and format_id fields
db95dc13 2454 for i, format in enumerate(formats):
c9969434
S
2455 sanitize_string_field(format, 'format_id')
2456 sanitize_numeric_fields(format)
dcf77cf1 2457 format['url'] = sanitize_url(format['url'])
e74e3b63 2458 if not format.get('format_id'):
8016c922 2459 format['format_id'] = compat_str(i)
e2effb08
S
2460 else:
2461 # Sanitize format_id from characters used in format selector expression
ec85ded8 2462 format['format_id'] = re.sub(r'[\s,/+\[\]()]', '_', format['format_id'])
181c7053
S
2463 format_id = format['format_id']
2464 if format_id not in formats_dict:
2465 formats_dict[format_id] = []
2466 formats_dict[format_id].append(format)
2467
2468 # Make sure all formats have unique format_id
03b4de72 2469 common_exts = set(itertools.chain(*self._format_selection_exts.values()))
181c7053 2470 for format_id, ambiguous_formats in formats_dict.items():
48ee10ee 2471 ambigious_id = len(ambiguous_formats) > 1
2472 for i, format in enumerate(ambiguous_formats):
2473 if ambigious_id:
181c7053 2474 format['format_id'] = '%s-%d' % (format_id, i)
48ee10ee 2475 if format.get('ext') is None:
2476 format['ext'] = determine_ext(format['url']).lower()
2477 # Ensure there is no conflict between id and ext in format selection
2478 # See https://github.com/yt-dlp/yt-dlp/issues/1282
2479 if format['format_id'] != format['ext'] and format['format_id'] in common_exts:
2480 format['format_id'] = 'f%s' % format['format_id']
181c7053
S
2481
2482 for i, format in enumerate(formats):
8c51aa65 2483 if format.get('format') is None:
6febd1c1 2484 format['format'] = '{id} - {res}{note}'.format(
8c51aa65
JMF
2485 id=format['format_id'],
2486 res=self.format_resolution(format),
b868936c 2487 note=format_field(format, 'format_note', ' (%s)'),
8c51aa65 2488 )
6f0be937 2489 if format.get('protocol') is None:
b5559424 2490 format['protocol'] = determine_protocol(format)
239df021 2491 if format.get('resolution') is None:
2492 format['resolution'] = self.format_resolution(format, default=None)
176f1866 2493 if format.get('dynamic_range') is None and format.get('vcodec') != 'none':
2494 format['dynamic_range'] = 'SDR'
f2fe69c7 2495 if (info_dict.get('duration') and format.get('tbr')
2496 and not format.get('filesize') and not format.get('filesize_approx')):
56ba69e4 2497 format['filesize_approx'] = int(info_dict['duration'] * format['tbr'] * (1024 / 8))
f2fe69c7 2498
e5660ee6
JMF
2499 # Add HTTP headers, so that external programs can use them from the
2500 # json output
2501 full_format_info = info_dict.copy()
2502 full_format_info.update(format)
2503 format['http_headers'] = self._calc_headers(full_format_info)
0016b84e
S
2504 # Remove private housekeeping stuff
2505 if '__x_forwarded_for_ip' in info_dict:
2506 del info_dict['__x_forwarded_for_ip']
dd82ffea 2507
9f1a1c36 2508 if self.params.get('check_formats') is True:
282f5709 2509 formats = LazyList(self._check_formats(formats[::-1]), reverse=True)
9f1a1c36 2510
88acdbc2 2511 if not formats or formats[0] is not info_dict:
b3d9ef88
JMF
2512 # only set the 'formats' fields if the original info_dict list them
2513 # otherwise we end up with a circular reference, the first (and unique)
f89197d7 2514 # element in the 'formats' field in info_dict is info_dict itself,
dfb1b146 2515 # which can't be exported to json
b3d9ef88 2516 info_dict['formats'] = formats
4ec82a72 2517
2518 info_dict, _ = self.pre_process(info_dict)
2519
6db9c4d5 2520 if self._match_entry(info_dict, incomplete=self._format_fields) is not None:
09b49e1f 2521 return info_dict
2522
2523 self.post_extract(info_dict)
2524 info_dict, _ = self.pre_process(info_dict, 'after_filter')
2525
093a1710 2526 # The pre-processors may have modified the formats
2527 formats = info_dict.get('formats', [info_dict])
2528
fa9f30b8 2529 list_only = self.params.get('simulate') is None and (
2530 self.params.get('list_thumbnails') or self.params.get('listformats') or self.params.get('listsubtitles'))
2531 interactive_format_selection = not list_only and self.format_selector == '-'
b7b04c78 2532 if self.params.get('list_thumbnails'):
2533 self.list_thumbnails(info_dict)
b7b04c78 2534 if self.params.get('listsubtitles'):
2535 if 'automatic_captions' in info_dict:
2536 self.list_subtitles(
2537 info_dict['id'], automatic_captions, 'automatic captions')
2538 self.list_subtitles(info_dict['id'], subtitles, 'subtitles')
fa9f30b8 2539 if self.params.get('listformats') or interactive_format_selection:
b69fd25c 2540 self.list_formats(info_dict)
169dbde9 2541 if list_only:
b7b04c78 2542 # Without this printing, -F --print-json will not work
169dbde9 2543 self.__forced_printings(info_dict, self.prepare_filename(info_dict), incomplete=True)
c487cf00 2544 return info_dict
bfaae0a7 2545
187986a8 2546 format_selector = self.format_selector
2547 if format_selector is None:
0017d9ad 2548 req_format = self._default_format_spec(info_dict, download=download)
0760b0a7 2549 self.write_debug('Default format spec: %s' % req_format)
187986a8 2550 format_selector = self.build_format_selector(req_format)
317f7ab6 2551
fa9f30b8 2552 while True:
2553 if interactive_format_selection:
2554 req_format = input(
2555 self._format_screen('\nEnter format selector: ', self.Styles.EMPHASIS))
2556 try:
2557 format_selector = self.build_format_selector(req_format)
2558 except SyntaxError as err:
2559 self.report_error(err, tb=False, is_error=False)
2560 continue
2561
85e801a9 2562 formats_to_download = list(format_selector({
fa9f30b8 2563 'formats': formats,
85e801a9 2564 'has_merged_format': any('none' not in (f.get('acodec'), f.get('vcodec')) for f in formats),
2565 'incomplete_formats': (
2566 # All formats are video-only or
2567 all(f.get('vcodec') != 'none' and f.get('acodec') == 'none' for f in formats)
2568 # all formats are audio-only
2569 or all(f.get('vcodec') == 'none' and f.get('acodec') != 'none' for f in formats)),
2570 }))
fa9f30b8 2571 if interactive_format_selection and not formats_to_download:
2572 self.report_error('Requested format is not available', tb=False, is_error=False)
2573 continue
2574 break
317f7ab6 2575
dd82ffea 2576 if not formats_to_download:
b7da73eb 2577 if not self.params.get('ignore_no_formats_error'):
c0b6e5c7 2578 raise ExtractorError(
2579 'Requested format is not available. Use --list-formats for a list of available formats',
2580 expected=True, video_id=info_dict['id'], ie=info_dict['extractor'])
b62fa6d7 2581 self.report_warning('Requested format is not available')
2582 # Process what we can, even without any available formats.
2583 formats_to_download = [{}]
a13e6848 2584
5ec1b6b7 2585 requested_ranges = self.params.get('download_ranges')
2586 if requested_ranges:
2587 requested_ranges = tuple(requested_ranges(info_dict, self))
2588
2589 best_format, downloaded_formats = formats_to_download[-1], []
b62fa6d7 2590 if download:
2591 if best_format:
5ec1b6b7 2592 def to_screen(*msg):
2593 self.to_screen(f'[info] {info_dict["id"]}: {" ".join(", ".join(variadic(m)) for m in msg)}')
2594
2595 to_screen(f'Downloading {len(formats_to_download)} format(s):',
2596 (f['format_id'] for f in formats_to_download))
2597 if requested_ranges:
2598 to_screen(f'Downloading {len(requested_ranges)} time ranges:',
2599 (f'{int(c["start_time"])}-{int(c["end_time"])}' for c in requested_ranges))
a13e6848 2600 max_downloads_reached = False
5ec1b6b7 2601
2602 for fmt, chapter in itertools.product(formats_to_download, requested_ranges or [{}]):
2603 new_info = self._copy_infodict(info_dict)
b7da73eb 2604 new_info.update(fmt)
5ec1b6b7 2605 if chapter:
2606 new_info.update({
2607 'section_start': chapter.get('start_time'),
2608 'section_end': chapter.get('end_time', 0),
2609 'section_title': chapter.get('title'),
2610 'section_number': chapter.get('index'),
2611 })
2612 downloaded_formats.append(new_info)
a13e6848 2613 try:
2614 self.process_info(new_info)
2615 except MaxDownloadsReached:
2616 max_downloads_reached = True
415f8d51 2617 self._raise_pending_errors(new_info)
f46e2f9d 2618 # Remove copied info
2619 for key, val in tuple(new_info.items()):
2620 if info_dict.get(key) == val:
2621 new_info.pop(key)
a13e6848 2622 if max_downloads_reached:
2623 break
ebed8b37 2624
5ec1b6b7 2625 write_archive = {f.get('__write_download_archive', False) for f in downloaded_formats}
a13e6848 2626 assert write_archive.issubset({True, False, 'ignore'})
2627 if True in write_archive and False not in write_archive:
2628 self.record_download_archive(info_dict)
be72c624 2629
5ec1b6b7 2630 info_dict['requested_downloads'] = downloaded_formats
ed5835b4 2631 info_dict = self.run_all_pps('after_video', info_dict)
a13e6848 2632 if max_downloads_reached:
2633 raise MaxDownloadsReached()
ebed8b37 2634
49a57e70 2635 # We update the info dict with the selected best quality format (backwards compatibility)
be72c624 2636 info_dict.update(best_format)
dd82ffea
JMF
2637 return info_dict
2638
98c70d6f 2639 def process_subtitles(self, video_id, normal_subtitles, automatic_captions):
a504ced0 2640 """Select the requested subtitles and their format"""
d8a58ddc 2641 available_subs, normal_sub_langs = {}, []
98c70d6f
JMF
2642 if normal_subtitles and self.params.get('writesubtitles'):
2643 available_subs.update(normal_subtitles)
d8a58ddc 2644 normal_sub_langs = tuple(normal_subtitles.keys())
98c70d6f
JMF
2645 if automatic_captions and self.params.get('writeautomaticsub'):
2646 for lang, cap_info in automatic_captions.items():
360e1ca5
JMF
2647 if lang not in available_subs:
2648 available_subs[lang] = cap_info
2649
4d171848
JMF
2650 if (not self.params.get('writesubtitles') and not
2651 self.params.get('writeautomaticsub') or not
2652 available_subs):
2653 return None
a504ced0 2654
d8a58ddc 2655 all_sub_langs = tuple(available_subs.keys())
a504ced0 2656 if self.params.get('allsubtitles', False):
c32b0aab 2657 requested_langs = all_sub_langs
2658 elif self.params.get('subtitleslangs', False):
77c4a9ef 2659 # A list is used so that the order of languages will be the same as
2660 # given in subtitleslangs. See https://github.com/yt-dlp/yt-dlp/issues/1041
2661 requested_langs = []
2662 for lang_re in self.params.get('subtitleslangs'):
77c4a9ef 2663 discard = lang_re[0] == '-'
c32b0aab 2664 if discard:
77c4a9ef 2665 lang_re = lang_re[1:]
3aa91540 2666 if lang_re == 'all':
2667 if discard:
2668 requested_langs = []
2669 else:
2670 requested_langs.extend(all_sub_langs)
2671 continue
77c4a9ef 2672 current_langs = filter(re.compile(lang_re + '$').match, all_sub_langs)
c32b0aab 2673 if discard:
2674 for lang in current_langs:
77c4a9ef 2675 while lang in requested_langs:
2676 requested_langs.remove(lang)
c32b0aab 2677 else:
77c4a9ef 2678 requested_langs.extend(current_langs)
2679 requested_langs = orderedSet(requested_langs)
d8a58ddc 2680 elif normal_sub_langs:
2681 requested_langs = ['en'] if 'en' in normal_sub_langs else normal_sub_langs[:1]
a504ced0 2682 else:
d8a58ddc 2683 requested_langs = ['en'] if 'en' in all_sub_langs else all_sub_langs[:1]
ad3dc496 2684 if requested_langs:
2685 self.write_debug('Downloading subtitles: %s' % ', '.join(requested_langs))
a504ced0
JMF
2686
2687 formats_query = self.params.get('subtitlesformat', 'best')
2688 formats_preference = formats_query.split('/') if formats_query else []
2689 subs = {}
2690 for lang in requested_langs:
2691 formats = available_subs.get(lang)
2692 if formats is None:
86e5f3ed 2693 self.report_warning(f'{lang} subtitles not available for {video_id}')
a504ced0 2694 continue
a504ced0
JMF
2695 for ext in formats_preference:
2696 if ext == 'best':
2697 f = formats[-1]
2698 break
2699 matches = list(filter(lambda f: f['ext'] == ext, formats))
2700 if matches:
2701 f = matches[-1]
2702 break
2703 else:
2704 f = formats[-1]
2705 self.report_warning(
2706 'No subtitle format found matching "%s" for language %s, '
2707 'using %s' % (formats_query, lang, f['ext']))
2708 subs[lang] = f
2709 return subs
2710
bb66c247 2711 def _forceprint(self, key, info_dict):
2712 if info_dict is None:
2713 return
2714 info_copy = info_dict.copy()
2715 info_copy['formats_table'] = self.render_formats_table(info_dict)
2716 info_copy['thumbnails_table'] = self.render_thumbnails_table(info_dict)
2717 info_copy['subtitles_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('subtitles'))
2718 info_copy['automatic_captions_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('automatic_captions'))
2719
2720 def format_tmpl(tmpl):
2721 mobj = re.match(r'\w+(=?)$', tmpl)
2722 if mobj and mobj.group(1):
2723 return f'{tmpl[:-1]} = %({tmpl[:-1]})r'
2724 elif mobj:
2725 return f'%({tmpl})s'
2726 return tmpl
8130779d 2727
bb66c247 2728 for tmpl in self.params['forceprint'].get(key, []):
2729 self.to_stdout(self.evaluate_outtmpl(format_tmpl(tmpl), info_copy))
2730
2731 for tmpl, file_tmpl in self.params['print_to_file'].get(key, []):
5127e92a 2732 filename = self.prepare_filename(info_dict, outtmpl=file_tmpl)
bb66c247 2733 tmpl = format_tmpl(tmpl)
2734 self.to_screen(f'[info] Writing {tmpl!r} to: {filename}')
8d93e69d 2735 if self._ensure_dir_exists(filename):
86e5f3ed 2736 with open(filename, 'a', encoding='utf-8') as f:
8d93e69d 2737 f.write(self.evaluate_outtmpl(tmpl, info_copy) + '\n')
ca30f449 2738
d06daf23 2739 def __forced_printings(self, info_dict, filename, incomplete):
53c18592 2740 def print_mandatory(field, actual_field=None):
2741 if actual_field is None:
2742 actual_field = field
d06daf23 2743 if (self.params.get('force%s' % field, False)
53c18592 2744 and (not incomplete or info_dict.get(actual_field) is not None)):
2745 self.to_stdout(info_dict[actual_field])
d06daf23
S
2746
2747 def print_optional(field):
2748 if (self.params.get('force%s' % field, False)
2749 and info_dict.get(field) is not None):
2750 self.to_stdout(info_dict[field])
2751
53c18592 2752 info_dict = info_dict.copy()
2753 if filename is not None:
2754 info_dict['filename'] = filename
2755 if info_dict.get('requested_formats') is not None:
2756 # For RTMP URLs, also include the playpath
2757 info_dict['urls'] = '\n'.join(f['url'] + f.get('play_path', '') for f in info_dict['requested_formats'])
10331a26 2758 elif info_dict.get('url'):
53c18592 2759 info_dict['urls'] = info_dict['url'] + info_dict.get('play_path', '')
2760
bb66c247 2761 if (self.params.get('forcejson')
2762 or self.params['forceprint'].get('video')
2763 or self.params['print_to_file'].get('video')):
2b8a2973 2764 self.post_extract(info_dict)
bb66c247 2765 self._forceprint('video', info_dict)
53c18592 2766
d06daf23
S
2767 print_mandatory('title')
2768 print_mandatory('id')
53c18592 2769 print_mandatory('url', 'urls')
d06daf23
S
2770 print_optional('thumbnail')
2771 print_optional('description')
53c18592 2772 print_optional('filename')
b868936c 2773 if self.params.get('forceduration') and info_dict.get('duration') is not None:
d06daf23
S
2774 self.to_stdout(formatSeconds(info_dict['duration']))
2775 print_mandatory('format')
53c18592 2776
2b8a2973 2777 if self.params.get('forcejson'):
6e84b215 2778 self.to_stdout(json.dumps(self.sanitize_info(info_dict)))
d06daf23 2779
e8e73840 2780 def dl(self, name, info, subtitle=False, test=False):
88acdbc2 2781 if not info.get('url'):
1151c407 2782 self.raise_no_formats(info, True)
e8e73840 2783
2784 if test:
2785 verbose = self.params.get('verbose')
2786 params = {
2787 'test': True,
a169858f 2788 'quiet': self.params.get('quiet') or not verbose,
e8e73840 2789 'verbose': verbose,
2790 'noprogress': not verbose,
2791 'nopart': True,
2792 'skip_unavailable_fragments': False,
2793 'keep_fragments': False,
2794 'overwrites': True,
2795 '_no_ytdl_file': True,
2796 }
2797 else:
2798 params = self.params
96fccc10 2799 fd = get_suitable_downloader(info, params, to_stdout=(name == '-'))(self, params)
e8e73840 2800 if not test:
2801 for ph in self._progress_hooks:
2802 fd.add_progress_hook(ph)
42676437
M
2803 urls = '", "'.join(
2804 (f['url'].split(',')[0] + ',<data>' if f['url'].startswith('data:') else f['url'])
2805 for f in info.get('requested_formats', []) or [info])
3a408f9d 2806 self.write_debug(f'Invoking {fd.FD_NAME} downloader on "{urls}"')
03b4de72 2807
adbc4ec4
THD
2808 # Note: Ideally info should be a deep-copied so that hooks cannot modify it.
2809 # But it may contain objects that are not deep-copyable
2810 new_info = self._copy_infodict(info)
e8e73840 2811 if new_info.get('http_headers') is None:
2812 new_info['http_headers'] = self._calc_headers(new_info)
2813 return fd.download(name, new_info, subtitle)
2814
e04938ab 2815 def existing_file(self, filepaths, *, default_overwrite=True):
2816 existing_files = list(filter(os.path.exists, orderedSet(filepaths)))
2817 if existing_files and not self.params.get('overwrites', default_overwrite):
2818 return existing_files[0]
2819
2820 for file in existing_files:
2821 self.report_file_delete(file)
2822 os.remove(file)
2823 return None
2824
8222d8de 2825 def process_info(self, info_dict):
09b49e1f 2826 """Process a single resolved IE result. (Modifies it in-place)"""
8222d8de
JMF
2827
2828 assert info_dict.get('_type', 'video') == 'video'
f46e2f9d 2829 original_infodict = info_dict
fd288278 2830
4513a41a 2831 if 'format' not in info_dict and 'ext' in info_dict:
8222d8de
JMF
2832 info_dict['format'] = info_dict['ext']
2833
09b49e1f 2834 # This is mostly just for backward compatibility of process_info
2835 # As a side-effect, this allows for format-specific filters
c77495e3 2836 if self._match_entry(info_dict) is not None:
9e907ebd 2837 info_dict['__write_download_archive'] = 'ignore'
8222d8de
JMF
2838 return
2839
09b49e1f 2840 # Does nothing under normal operation - for backward compatibility of process_info
277d6ff5 2841 self.post_extract(info_dict)
0c14d66a 2842 self._num_downloads += 1
8222d8de 2843
dcf64d43 2844 # info_dict['_filename'] needs to be set for backward compatibility
de6000d9 2845 info_dict['_filename'] = full_filename = self.prepare_filename(info_dict, warn=True)
2846 temp_filename = self.prepare_filename(info_dict, 'temp')
0202b52a 2847 files_to_move = {}
8222d8de
JMF
2848
2849 # Forced printings
4513a41a 2850 self.__forced_printings(info_dict, full_filename, incomplete=('format' not in info_dict))
8222d8de 2851
ca6d59d2 2852 def check_max_downloads():
2853 if self._num_downloads >= float(self.params.get('max_downloads') or 'inf'):
2854 raise MaxDownloadsReached()
2855
b7b04c78 2856 if self.params.get('simulate'):
9e907ebd 2857 info_dict['__write_download_archive'] = self.params.get('force_write_download_archive')
ca6d59d2 2858 check_max_downloads()
8222d8de
JMF
2859 return
2860
de6000d9 2861 if full_filename is None:
8222d8de 2862 return
e92caff5 2863 if not self._ensure_dir_exists(encodeFilename(full_filename)):
0202b52a 2864 return
e92caff5 2865 if not self._ensure_dir_exists(encodeFilename(temp_filename)):
8222d8de
JMF
2866 return
2867
80c03fa9 2868 if self._write_description('video', info_dict,
2869 self.prepare_filename(info_dict, 'description')) is None:
2870 return
2871
2872 sub_files = self._write_subtitles(info_dict, temp_filename)
2873 if sub_files is None:
2874 return
2875 files_to_move.update(dict(sub_files))
2876
2877 thumb_files = self._write_thumbnails(
2878 'video', info_dict, temp_filename, self.prepare_filename(info_dict, 'thumbnail'))
2879 if thumb_files is None:
2880 return
2881 files_to_move.update(dict(thumb_files))
8222d8de 2882
80c03fa9 2883 infofn = self.prepare_filename(info_dict, 'infojson')
2884 _infojson_written = self._write_info_json('video', info_dict, infofn)
2885 if _infojson_written:
dac5df5a 2886 info_dict['infojson_filename'] = infofn
e75bb0d6 2887 # For backward compatibility, even though it was a private field
80c03fa9 2888 info_dict['__infojson_filename'] = infofn
2889 elif _infojson_written is None:
2890 return
2891
2892 # Note: Annotations are deprecated
2893 annofn = None
1fb07d10 2894 if self.params.get('writeannotations', False):
de6000d9 2895 annofn = self.prepare_filename(info_dict, 'annotation')
80c03fa9 2896 if annofn:
e92caff5 2897 if not self._ensure_dir_exists(encodeFilename(annofn)):
0202b52a 2898 return
0c3d0f51 2899 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(annofn)):
6febd1c1 2900 self.to_screen('[info] Video annotations are already present')
ffddb112
RA
2901 elif not info_dict.get('annotations'):
2902 self.report_warning('There are no annotations to write.')
7b6fefc9
PH
2903 else:
2904 try:
6febd1c1 2905 self.to_screen('[info] Writing video annotations to: ' + annofn)
86e5f3ed 2906 with open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
7b6fefc9
PH
2907 annofile.write(info_dict['annotations'])
2908 except (KeyError, TypeError):
6febd1c1 2909 self.report_warning('There are no annotations to write.')
86e5f3ed 2910 except OSError:
6febd1c1 2911 self.report_error('Cannot write annotations file: ' + annofn)
7b6fefc9 2912 return
1fb07d10 2913
732044af 2914 # Write internet shortcut files
08438d2c 2915 def _write_link_file(link_type):
60f3e995 2916 url = try_get(info_dict['webpage_url'], iri_to_uri)
2917 if not url:
2918 self.report_warning(
2919 f'Cannot write internet shortcut file because the actual URL of "{info_dict["webpage_url"]}" is unknown')
2920 return True
08438d2c 2921 linkfn = replace_extension(self.prepare_filename(info_dict, 'link'), link_type, info_dict.get('ext'))
0e6b018a
Z
2922 if not self._ensure_dir_exists(encodeFilename(linkfn)):
2923 return False
10e3742e 2924 if self.params.get('overwrites', True) and os.path.exists(encodeFilename(linkfn)):
08438d2c 2925 self.to_screen(f'[info] Internet shortcut (.{link_type}) is already present')
2926 return True
2927 try:
2928 self.to_screen(f'[info] Writing internet shortcut (.{link_type}) to: {linkfn}')
86e5f3ed 2929 with open(encodeFilename(to_high_limit_path(linkfn)), 'w', encoding='utf-8',
2930 newline='\r\n' if link_type == 'url' else '\n') as linkfile:
60f3e995 2931 template_vars = {'url': url}
08438d2c 2932 if link_type == 'desktop':
2933 template_vars['filename'] = linkfn[:-(len(link_type) + 1)]
2934 linkfile.write(LINK_TEMPLATES[link_type] % template_vars)
86e5f3ed 2935 except OSError:
08438d2c 2936 self.report_error(f'Cannot write internet shortcut {linkfn}')
2937 return False
732044af 2938 return True
2939
08438d2c 2940 write_links = {
2941 'url': self.params.get('writeurllink'),
2942 'webloc': self.params.get('writewebloclink'),
2943 'desktop': self.params.get('writedesktoplink'),
2944 }
2945 if self.params.get('writelink'):
2946 link_type = ('webloc' if sys.platform == 'darwin'
2947 else 'desktop' if sys.platform.startswith('linux')
2948 else 'url')
2949 write_links[link_type] = True
2950
2951 if any(should_write and not _write_link_file(link_type)
2952 for link_type, should_write in write_links.items()):
2953 return
732044af 2954
f46e2f9d 2955 def replace_info_dict(new_info):
2956 nonlocal info_dict
2957 if new_info == info_dict:
2958 return
2959 info_dict.clear()
2960 info_dict.update(new_info)
2961
415f8d51 2962 new_info, files_to_move = self.pre_process(info_dict, 'before_dl', files_to_move)
2963 replace_info_dict(new_info)
56d868db 2964
a13e6848 2965 if self.params.get('skip_download'):
56d868db 2966 info_dict['filepath'] = temp_filename
2967 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
2968 info_dict['__files_to_move'] = files_to_move
f46e2f9d 2969 replace_info_dict(self.run_pp(MoveFilesAfterDownloadPP(self, False), info_dict))
9e907ebd 2970 info_dict['__write_download_archive'] = self.params.get('force_write_download_archive')
56d868db 2971 else:
2972 # Download
b868936c 2973 info_dict.setdefault('__postprocessors', [])
4340deca 2974 try:
0202b52a 2975
e04938ab 2976 def existing_video_file(*filepaths):
6b591b29 2977 ext = info_dict.get('ext')
e04938ab 2978 converted = lambda file: replace_extension(file, self.params.get('final_ext') or ext, ext)
2979 file = self.existing_file(itertools.chain(*zip(map(converted, filepaths), filepaths)),
2980 default_overwrite=False)
2981 if file:
2982 info_dict['ext'] = os.path.splitext(file)[1][1:]
2983 return file
0202b52a 2984
2985 success = True
56ba69e4 2986 merger, fd = FFmpegMergerPP(self), None
fccf90e7 2987 if info_dict.get('protocol') or info_dict.get('url'):
56ba69e4 2988 fd = get_suitable_downloader(info_dict, self.params, to_stdout=temp_filename == '-')
2989 if fd is not FFmpegFD and (
2990 info_dict.get('section_start') or info_dict.get('section_end')):
2991 msg = ('This format cannot be partially downloaded' if merger.available
2992 else 'You have requested downloading the video partially, but ffmpeg is not installed')
2993 self.report_error(f'{msg}. Aborting')
5ec1b6b7 2994 return
5ec1b6b7 2995
4340deca 2996 if info_dict.get('requested_formats') is not None:
81cd954a
S
2997
2998 def compatible_formats(formats):
d03cfdce 2999 # TODO: some formats actually allow this (mkv, webm, ogg, mp4), but not all of them.
3000 video_formats = [format for format in formats if format.get('vcodec') != 'none']
3001 audio_formats = [format for format in formats if format.get('acodec') != 'none']
3002 if len(video_formats) > 2 or len(audio_formats) > 2:
3003 return False
3004
81cd954a 3005 # Check extension
86e5f3ed 3006 exts = {format.get('ext') for format in formats}
d03cfdce 3007 COMPATIBLE_EXTS = (
86e5f3ed 3008 {'mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma'},
3009 {'webm'},
d03cfdce 3010 )
3011 for ext_sets in COMPATIBLE_EXTS:
3012 if ext_sets.issuperset(exts):
3013 return True
81cd954a
S
3014 # TODO: Check acodec/vcodec
3015 return False
3016
3017 requested_formats = info_dict['requested_formats']
0202b52a 3018 old_ext = info_dict['ext']
4e3b637d 3019 if self.params.get('merge_output_format') is None:
3020 if not compatible_formats(requested_formats):
3021 info_dict['ext'] = 'mkv'
3022 self.report_warning(
3023 'Requested formats are incompatible for merge and will be merged into mkv')
3024 if (info_dict['ext'] == 'webm'
3025 and info_dict.get('thumbnails')
3026 # check with type instead of pp_key, __name__, or isinstance
3027 # since we dont want any custom PPs to trigger this
c487cf00 3028 and any(type(pp) == EmbedThumbnailPP for pp in self._pps['post_process'])): # noqa: E721
4e3b637d 3029 info_dict['ext'] = 'mkv'
3030 self.report_warning(
3031 'webm doesn\'t support embedding a thumbnail, mkv will be used')
124bc071 3032 new_ext = info_dict['ext']
0202b52a 3033
124bc071 3034 def correct_ext(filename, ext=new_ext):
96fccc10 3035 if filename == '-':
3036 return filename
0202b52a 3037 filename_real_ext = os.path.splitext(filename)[1][1:]
3038 filename_wo_ext = (
3039 os.path.splitext(filename)[0]
124bc071 3040 if filename_real_ext in (old_ext, new_ext)
0202b52a 3041 else filename)
86e5f3ed 3042 return f'{filename_wo_ext}.{ext}'
0202b52a 3043
38c6902b 3044 # Ensure filename always has a correct extension for successful merge
0202b52a 3045 full_filename = correct_ext(full_filename)
3046 temp_filename = correct_ext(temp_filename)
e04938ab 3047 dl_filename = existing_video_file(full_filename, temp_filename)
1ea24129 3048 info_dict['__real_download'] = False
18e674b4 3049
adbc4ec4 3050 downloaded = []
dbf5416a 3051 if dl_filename is not None:
6c7274ec 3052 self.report_file_already_downloaded(dl_filename)
adbc4ec4
THD
3053 elif fd:
3054 for f in requested_formats if fd != FFmpegFD else []:
3055 f['filepath'] = fname = prepend_extension(
3056 correct_ext(temp_filename, info_dict['ext']),
3057 'f%s' % f['format_id'], info_dict['ext'])
3058 downloaded.append(fname)
dbf5416a 3059 info_dict['url'] = '\n'.join(f['url'] for f in requested_formats)
3060 success, real_download = self.dl(temp_filename, info_dict)
3061 info_dict['__real_download'] = real_download
18e674b4 3062 else:
18e674b4 3063 if self.params.get('allow_unplayable_formats'):
3064 self.report_warning(
3065 'You have requested merging of multiple formats '
3066 'while also allowing unplayable formats to be downloaded. '
3067 'The formats won\'t be merged to prevent data corruption.')
3068 elif not merger.available:
e8969bda 3069 msg = 'You have requested merging of multiple formats but ffmpeg is not installed'
3070 if not self.params.get('ignoreerrors'):
3071 self.report_error(f'{msg}. Aborting due to --abort-on-error')
3072 return
3073 self.report_warning(f'{msg}. The formats won\'t be merged')
18e674b4 3074
96fccc10 3075 if temp_filename == '-':
adbc4ec4 3076 reason = ('using a downloader other than ffmpeg' if FFmpegFD.can_merge_formats(info_dict, self.params)
96fccc10 3077 else 'but the formats are incompatible for simultaneous download' if merger.available
3078 else 'but ffmpeg is not installed')
3079 self.report_warning(
3080 f'You have requested downloading multiple formats to stdout {reason}. '
3081 'The formats will be streamed one after the other')
3082 fname = temp_filename
dbf5416a 3083 for f in requested_formats:
3084 new_info = dict(info_dict)
3085 del new_info['requested_formats']
3086 new_info.update(f)
96fccc10 3087 if temp_filename != '-':
124bc071 3088 fname = prepend_extension(
3089 correct_ext(temp_filename, new_info['ext']),
3090 'f%s' % f['format_id'], new_info['ext'])
96fccc10 3091 if not self._ensure_dir_exists(fname):
3092 return
a21e0ab1 3093 f['filepath'] = fname
96fccc10 3094 downloaded.append(fname)
dbf5416a 3095 partial_success, real_download = self.dl(fname, new_info)
3096 info_dict['__real_download'] = info_dict['__real_download'] or real_download
3097 success = success and partial_success
adbc4ec4
THD
3098
3099 if downloaded and merger.available and not self.params.get('allow_unplayable_formats'):
3100 info_dict['__postprocessors'].append(merger)
3101 info_dict['__files_to_merge'] = downloaded
3102 # Even if there were no downloads, it is being merged only now
3103 info_dict['__real_download'] = True
3104 else:
3105 for file in downloaded:
3106 files_to_move[file] = None
4340deca
P
3107 else:
3108 # Just a single file
e04938ab 3109 dl_filename = existing_video_file(full_filename, temp_filename)
6c7274ec 3110 if dl_filename is None or dl_filename == temp_filename:
3111 # dl_filename == temp_filename could mean that the file was partially downloaded with --no-part.
3112 # So we should try to resume the download
e8e73840 3113 success, real_download = self.dl(temp_filename, info_dict)
0202b52a 3114 info_dict['__real_download'] = real_download
6c7274ec 3115 else:
3116 self.report_file_already_downloaded(dl_filename)
0202b52a 3117
0202b52a 3118 dl_filename = dl_filename or temp_filename
c571435f 3119 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
0202b52a 3120
3158150c 3121 except network_exceptions as err:
7960b056 3122 self.report_error('unable to download video data: %s' % error_to_compat_str(err))
4340deca 3123 return
86e5f3ed 3124 except OSError as err:
4340deca
P
3125 raise UnavailableVideoError(err)
3126 except (ContentTooShortError, ) as err:
86e5f3ed 3127 self.report_error(f'content too short (expected {err.expected} bytes and served {err.downloaded})')
4340deca 3128 return
8222d8de 3129
415f8d51 3130 self._raise_pending_errors(info_dict)
de6000d9 3131 if success and full_filename != '-':
f17f8651 3132
fd7cfb64 3133 def fixup():
3134 do_fixup = True
3135 fixup_policy = self.params.get('fixup')
3136 vid = info_dict['id']
3137
3138 if fixup_policy in ('ignore', 'never'):
3139 return
3140 elif fixup_policy == 'warn':
3fe75fdc 3141 do_fixup = 'warn'
f89b3e2d 3142 elif fixup_policy != 'force':
3143 assert fixup_policy in ('detect_or_warn', None)
3144 if not info_dict.get('__real_download'):
3145 do_fixup = False
fd7cfb64 3146
3147 def ffmpeg_fixup(cndn, msg, cls):
3fe75fdc 3148 if not (do_fixup and cndn):
fd7cfb64 3149 return
3fe75fdc 3150 elif do_fixup == 'warn':
fd7cfb64 3151 self.report_warning(f'{vid}: {msg}')
3152 return
3153 pp = cls(self)
3154 if pp.available:
3155 info_dict['__postprocessors'].append(pp)
3156 else:
3157 self.report_warning(f'{vid}: {msg}. Install ffmpeg to fix this automatically')
3158
3159 stretched_ratio = info_dict.get('stretched_ratio')
3160 ffmpeg_fixup(
3161 stretched_ratio not in (1, None),
3162 f'Non-uniform pixel ratio {stretched_ratio}',
3163 FFmpegFixupStretchedPP)
3164
3165 ffmpeg_fixup(
3166 (info_dict.get('requested_formats') is None
3167 and info_dict.get('container') == 'm4a_dash'
3168 and info_dict.get('ext') == 'm4a'),
3169 'writing DASH m4a. Only some players support this container',
3170 FFmpegFixupM4aPP)
3171
993191c0 3172 downloader = get_suitable_downloader(info_dict, self.params) if 'protocol' in info_dict else None
24146491 3173 downloader = downloader.FD_NAME if downloader else None
adbc4ec4
THD
3174
3175 if info_dict.get('requested_formats') is None: # Not necessary if doing merger
24146491 3176 ffmpeg_fixup(downloader == 'hlsnative' and not self.params.get('hls_use_mpegts')
494f5230 3177 or info_dict.get('is_live') and self.params.get('hls_use_mpegts') is None,
adbc4ec4
THD
3178 'Possible MPEG-TS in MP4 container or malformed AAC timestamps',
3179 FFmpegFixupM3u8PP)
3180 ffmpeg_fixup(info_dict.get('is_live') and downloader == 'DashSegmentsFD',
3181 'Possible duplicate MOOV atoms', FFmpegFixupDuplicateMoovPP)
3182
24146491 3183 ffmpeg_fixup(downloader == 'web_socket_fragment', 'Malformed timestamps detected', FFmpegFixupTimestampPP)
3184 ffmpeg_fixup(downloader == 'web_socket_fragment', 'Malformed duration detected', FFmpegFixupDurationPP)
fd7cfb64 3185
3186 fixup()
8222d8de 3187 try:
f46e2f9d 3188 replace_info_dict(self.post_process(dl_filename, info_dict, files_to_move))
af819c21 3189 except PostProcessingError as err:
3190 self.report_error('Postprocessing: %s' % str(err))
8222d8de 3191 return
ab8e5e51
AM
3192 try:
3193 for ph in self._post_hooks:
23c1a667 3194 ph(info_dict['filepath'])
ab8e5e51
AM
3195 except Exception as err:
3196 self.report_error('post hooks: %s' % str(err))
3197 return
9e907ebd 3198 info_dict['__write_download_archive'] = True
2d30509f 3199
c487cf00 3200 assert info_dict is original_infodict # Make sure the info_dict was modified in-place
a13e6848 3201 if self.params.get('force_write_download_archive'):
9e907ebd 3202 info_dict['__write_download_archive'] = True
ca6d59d2 3203 check_max_downloads()
8222d8de 3204
aa9369a2 3205 def __download_wrapper(self, func):
3206 @functools.wraps(func)
3207 def wrapper(*args, **kwargs):
3208 try:
3209 res = func(*args, **kwargs)
3210 except UnavailableVideoError as e:
3211 self.report_error(e)
b222c271 3212 except DownloadCancelled as e:
3213 self.to_screen(f'[info] {e}')
3214 if not self.params.get('break_per_url'):
3215 raise
aa9369a2 3216 else:
3217 if self.params.get('dump_single_json', False):
3218 self.post_extract(res)
3219 self.to_stdout(json.dumps(self.sanitize_info(res)))
3220 return wrapper
3221
8222d8de
JMF
3222 def download(self, url_list):
3223 """Download a given list of URLs."""
aa9369a2 3224 url_list = variadic(url_list) # Passing a single URL is a common mistake
bf1824b3 3225 outtmpl = self.params['outtmpl']['default']
3089bc74
S
3226 if (len(url_list) > 1
3227 and outtmpl != '-'
3228 and '%' not in outtmpl
3229 and self.params.get('max_downloads') != 1):
acd69589 3230 raise SameFileError(outtmpl)
8222d8de
JMF
3231
3232 for url in url_list:
aa9369a2 3233 self.__download_wrapper(self.extract_info)(
3234 url, force_generic_extractor=self.params.get('force_generic_extractor', False))
8222d8de
JMF
3235
3236 return self._download_retcode
3237
1dcc4c0c 3238 def download_with_info_file(self, info_filename):
31bd3925
JMF
3239 with contextlib.closing(fileinput.FileInput(
3240 [info_filename], mode='r',
3241 openhook=fileinput.hook_encoded('utf-8'))) as f:
3242 # FileInput doesn't have a read method, we can't call json.load
8012d892 3243 info = self.sanitize_info(json.loads('\n'.join(f)), self.params.get('clean_infojson', True))
d4943898 3244 try:
aa9369a2 3245 self.__download_wrapper(self.process_ie_result)(info, download=True)
f2ebc5c7 3246 except (DownloadError, EntryNotInPlaylist, ReExtractInfo) as e:
bf5f605e 3247 if not isinstance(e, EntryNotInPlaylist):
3248 self.to_stderr('\r')
d4943898
JMF
3249 webpage_url = info.get('webpage_url')
3250 if webpage_url is not None:
aa9369a2 3251 self.report_warning(f'The info failed to download: {e}; trying with URL {webpage_url}')
d4943898
JMF
3252 return self.download([webpage_url])
3253 else:
3254 raise
3255 return self._download_retcode
1dcc4c0c 3256
cb202fd2 3257 @staticmethod
8012d892 3258 def sanitize_info(info_dict, remove_private_keys=False):
3259 ''' Sanitize the infodict for converting to json '''
3ad56b42 3260 if info_dict is None:
3261 return info_dict
6e84b215 3262 info_dict.setdefault('epoch', int(time.time()))
6a5a30f9 3263 info_dict.setdefault('_type', 'video')
09b49e1f 3264
8012d892 3265 if remove_private_keys:
0a5a191a 3266 reject = lambda k, v: v is None or k.startswith('__') or k in {
f46e2f9d 3267 'requested_downloads', 'requested_formats', 'requested_subtitles', 'requested_entries',
0a5a191a 3268 'entries', 'filepath', '_filename', 'infojson_filename', 'original_url', 'playlist_autonumber',
6e84b215 3269 }
ae8f99e6 3270 else:
09b49e1f 3271 reject = lambda k, v: False
adbc4ec4
THD
3272
3273 def filter_fn(obj):
3274 if isinstance(obj, dict):
3275 return {k: filter_fn(v) for k, v in obj.items() if not reject(k, v)}
3276 elif isinstance(obj, (list, tuple, set, LazyList)):
3277 return list(map(filter_fn, obj))
3278 elif obj is None or isinstance(obj, (str, int, float, bool)):
3279 return obj
3280 else:
3281 return repr(obj)
3282
5226731e 3283 return filter_fn(info_dict)
cb202fd2 3284
8012d892 3285 @staticmethod
3286 def filter_requested_info(info_dict, actually_filter=True):
3287 ''' Alias of sanitize_info for backward compatibility '''
3288 return YoutubeDL.sanitize_info(info_dict, actually_filter)
3289
43d7f5a5 3290 def _delete_downloaded_files(self, *files_to_delete, info={}, msg=None):
3291 for filename in set(filter(None, files_to_delete)):
3292 if msg:
3293 self.to_screen(msg % filename)
3294 try:
3295 os.remove(filename)
3296 except OSError:
3297 self.report_warning(f'Unable to delete file {filename}')
3298 if filename in info.get('__files_to_move', []): # NB: Delete even if None
3299 del info['__files_to_move'][filename]
3300
ed5835b4 3301 @staticmethod
3302 def post_extract(info_dict):
3303 def actual_post_extract(info_dict):
3304 if info_dict.get('_type') in ('playlist', 'multi_video'):
3305 for video_dict in info_dict.get('entries', {}):
3306 actual_post_extract(video_dict or {})
3307 return
3308
09b49e1f 3309 post_extractor = info_dict.pop('__post_extractor', None) or (lambda: {})
3310 info_dict.update(post_extractor())
ed5835b4 3311
3312 actual_post_extract(info_dict or {})
3313
dcf64d43 3314 def run_pp(self, pp, infodict):
5bfa4862 3315 files_to_delete = []
dcf64d43 3316 if '__files_to_move' not in infodict:
3317 infodict['__files_to_move'] = {}
b1940459 3318 try:
3319 files_to_delete, infodict = pp.run(infodict)
3320 except PostProcessingError as e:
3321 # Must be True and not 'only_download'
3322 if self.params.get('ignoreerrors') is True:
3323 self.report_error(e)
3324 return infodict
3325 raise
3326
5bfa4862 3327 if not files_to_delete:
dcf64d43 3328 return infodict
5bfa4862 3329 if self.params.get('keepvideo', False):
3330 for f in files_to_delete:
dcf64d43 3331 infodict['__files_to_move'].setdefault(f, '')
5bfa4862 3332 else:
43d7f5a5 3333 self._delete_downloaded_files(
3334 *files_to_delete, info=infodict, msg='Deleting original file %s (pass -k to keep)')
dcf64d43 3335 return infodict
5bfa4862 3336
ed5835b4 3337 def run_all_pps(self, key, info, *, additional_pps=None):
bb66c247 3338 self._forceprint(key, info)
ed5835b4 3339 for pp in (additional_pps or []) + self._pps[key]:
dc5f409c 3340 info = self.run_pp(pp, info)
ed5835b4 3341 return info
277d6ff5 3342
56d868db 3343 def pre_process(self, ie_info, key='pre_process', files_to_move=None):
5bfa4862 3344 info = dict(ie_info)
56d868db 3345 info['__files_to_move'] = files_to_move or {}
415f8d51 3346 try:
3347 info = self.run_all_pps(key, info)
3348 except PostProcessingError as err:
3349 msg = f'Preprocessing: {err}'
3350 info.setdefault('__pending_error', msg)
3351 self.report_error(msg, is_error=False)
56d868db 3352 return info, info.pop('__files_to_move', None)
5bfa4862 3353
f46e2f9d 3354 def post_process(self, filename, info, files_to_move=None):
8222d8de 3355 """Run all the postprocessors on the given file."""
8222d8de 3356 info['filepath'] = filename
dcf64d43 3357 info['__files_to_move'] = files_to_move or {}
ed5835b4 3358 info = self.run_all_pps('post_process', info, additional_pps=info.get('__postprocessors'))
dcf64d43 3359 info = self.run_pp(MoveFilesAfterDownloadPP(self), info)
3360 del info['__files_to_move']
ed5835b4 3361 return self.run_all_pps('after_move', info)
c1c9a79c 3362
5db07df6 3363 def _make_archive_id(self, info_dict):
e9fef7ee
S
3364 video_id = info_dict.get('id')
3365 if not video_id:
3366 return
5db07df6
PH
3367 # Future-proof against any change in case
3368 # and backwards compatibility with prior versions
e9fef7ee 3369 extractor = info_dict.get('extractor_key') or info_dict.get('ie_key') # key in a playlist
7012b23c 3370 if extractor is None:
1211bb6d
S
3371 url = str_or_none(info_dict.get('url'))
3372 if not url:
3373 return
e9fef7ee 3374 # Try to find matching extractor for the URL and take its ie_key
8b7491c8 3375 for ie_key, ie in self._ies.items():
1211bb6d 3376 if ie.suitable(url):
8b7491c8 3377 extractor = ie_key
e9fef7ee
S
3378 break
3379 else:
3380 return
86e5f3ed 3381 return f'{extractor.lower()} {video_id}'
5db07df6
PH
3382
3383 def in_download_archive(self, info_dict):
3384 fn = self.params.get('download_archive')
3385 if fn is None:
3386 return False
3387
3388 vid_id = self._make_archive_id(info_dict)
e9fef7ee 3389 if not vid_id:
7012b23c 3390 return False # Incomplete video information
5db07df6 3391
a45e8619 3392 return vid_id in self.archive
c1c9a79c
PH
3393
3394 def record_download_archive(self, info_dict):
3395 fn = self.params.get('download_archive')
3396 if fn is None:
3397 return
5db07df6
PH
3398 vid_id = self._make_archive_id(info_dict)
3399 assert vid_id
a13e6848 3400 self.write_debug(f'Adding to archive: {vid_id}')
c1c9a79c 3401 with locked_file(fn, 'a', encoding='utf-8') as archive_file:
6febd1c1 3402 archive_file.write(vid_id + '\n')
a45e8619 3403 self.archive.add(vid_id)
dd82ffea 3404
8c51aa65 3405 @staticmethod
8abeeb94 3406 def format_resolution(format, default='unknown'):
9359f3d4 3407 if format.get('vcodec') == 'none' and format.get('acodec') != 'none':
fb04e403 3408 return 'audio only'
f49d89ee
PH
3409 if format.get('resolution') is not None:
3410 return format['resolution']
35615307 3411 if format.get('width') and format.get('height'):
ff51ed58 3412 return '%dx%d' % (format['width'], format['height'])
35615307 3413 elif format.get('height'):
ff51ed58 3414 return '%sp' % format['height']
35615307 3415 elif format.get('width'):
ff51ed58 3416 return '%dx?' % format['width']
3417 return default
8c51aa65 3418
8130779d 3419 def _list_format_headers(self, *headers):
3420 if self.params.get('listformats_table', True) is not False:
591bb9d3 3421 return [self._format_out(header, self.Styles.HEADERS) for header in headers]
8130779d 3422 return headers
3423
c57f7757
PH
3424 def _format_note(self, fdict):
3425 res = ''
3426 if fdict.get('ext') in ['f4f', 'f4m']:
f304da8a 3427 res += '(unsupported)'
32f90364
PH
3428 if fdict.get('language'):
3429 if res:
3430 res += ' '
f304da8a 3431 res += '[%s]' % fdict['language']
c57f7757 3432 if fdict.get('format_note') is not None:
f304da8a 3433 if res:
3434 res += ' '
3435 res += fdict['format_note']
c57f7757 3436 if fdict.get('tbr') is not None:
f304da8a 3437 if res:
3438 res += ', '
3439 res += '%4dk' % fdict['tbr']
c57f7757
PH
3440 if fdict.get('container') is not None:
3441 if res:
3442 res += ', '
3443 res += '%s container' % fdict['container']
3089bc74
S
3444 if (fdict.get('vcodec') is not None
3445 and fdict.get('vcodec') != 'none'):
c57f7757
PH
3446 if res:
3447 res += ', '
3448 res += fdict['vcodec']
91c7271a 3449 if fdict.get('vbr') is not None:
c57f7757
PH
3450 res += '@'
3451 elif fdict.get('vbr') is not None and fdict.get('abr') is not None:
3452 res += 'video@'
3453 if fdict.get('vbr') is not None:
3454 res += '%4dk' % fdict['vbr']
fbb21cf5 3455 if fdict.get('fps') is not None:
5d583bdf
S
3456 if res:
3457 res += ', '
3458 res += '%sfps' % fdict['fps']
c57f7757
PH
3459 if fdict.get('acodec') is not None:
3460 if res:
3461 res += ', '
3462 if fdict['acodec'] == 'none':
3463 res += 'video only'
3464 else:
3465 res += '%-5s' % fdict['acodec']
3466 elif fdict.get('abr') is not None:
3467 if res:
3468 res += ', '
3469 res += 'audio'
3470 if fdict.get('abr') is not None:
3471 res += '@%3dk' % fdict['abr']
3472 if fdict.get('asr') is not None:
3473 res += ' (%5dHz)' % fdict['asr']
3474 if fdict.get('filesize') is not None:
3475 if res:
3476 res += ', '
3477 res += format_bytes(fdict['filesize'])
9732d77e
PH
3478 elif fdict.get('filesize_approx') is not None:
3479 if res:
3480 res += ', '
3481 res += '~' + format_bytes(fdict['filesize_approx'])
c57f7757 3482 return res
91c7271a 3483
8130779d 3484 def render_formats_table(self, info_dict):
b69fd25c 3485 if not info_dict.get('formats') and not info_dict.get('url'):
8130779d 3486 return None
b69fd25c 3487
94badb25 3488 formats = info_dict.get('formats', [info_dict])
8130779d 3489 if not self.params.get('listformats_table', True) is not False:
76d321f6 3490 table = [
3491 [
3492 format_field(f, 'format_id'),
3493 format_field(f, 'ext'),
3494 self.format_resolution(f),
8130779d 3495 self._format_note(f)
3496 ] for f in formats if f.get('preference') is None or f['preference'] >= -1000]
3497 return render_table(['format code', 'extension', 'resolution', 'note'], table, extra_gap=1)
3498
591bb9d3 3499 delim = self._format_out('\u2502', self.Styles.DELIM, '|', test_encoding=True)
8130779d 3500 table = [
3501 [
591bb9d3 3502 self._format_out(format_field(f, 'format_id'), self.Styles.ID),
8130779d 3503 format_field(f, 'ext'),
3504 format_field(f, func=self.format_resolution, ignore=('audio only', 'images')),
3505 format_field(f, 'fps', '\t%d'),
3506 format_field(f, 'dynamic_range', '%s', ignore=(None, 'SDR')).replace('HDR', ''),
3507 delim,
3508 format_field(f, 'filesize', ' \t%s', func=format_bytes) + format_field(f, 'filesize_approx', '~\t%s', func=format_bytes),
3509 format_field(f, 'tbr', '\t%dk'),
3510 shorten_protocol_name(f.get('protocol', '')),
3511 delim,
3512 format_field(f, 'vcodec', default='unknown').replace(
3513 'none', 'images' if f.get('acodec') == 'none'
591bb9d3 3514 else self._format_out('audio only', self.Styles.SUPPRESS)),
8130779d 3515 format_field(f, 'vbr', '\t%dk'),
3516 format_field(f, 'acodec', default='unknown').replace(
3517 'none', '' if f.get('vcodec') == 'none'
591bb9d3 3518 else self._format_out('video only', self.Styles.SUPPRESS)),
8130779d 3519 format_field(f, 'abr', '\t%dk'),
3520 format_field(f, 'asr', '\t%dHz'),
3521 join_nonempty(
591bb9d3 3522 self._format_out('UNSUPPORTED', 'light red') if f.get('ext') in ('f4f', 'f4m') else None,
8130779d 3523 format_field(f, 'language', '[%s]'),
3524 join_nonempty(format_field(f, 'format_note'),
3525 format_field(f, 'container', ignore=(None, f.get('ext'))),
3526 delim=', '),
3527 delim=' '),
3528 ] for f in formats if f.get('preference') is None or f['preference'] >= -1000]
3529 header_line = self._list_format_headers(
3530 'ID', 'EXT', 'RESOLUTION', '\tFPS', 'HDR', delim, '\tFILESIZE', '\tTBR', 'PROTO',
3531 delim, 'VCODEC', '\tVBR', 'ACODEC', '\tABR', '\tASR', 'MORE INFO')
3532
3533 return render_table(
3534 header_line, table, hide_empty=True,
591bb9d3 3535 delim=self._format_out('\u2500', self.Styles.DELIM, '-', test_encoding=True))
8130779d 3536
3537 def render_thumbnails_table(self, info_dict):
88f23a18 3538 thumbnails = list(info_dict.get('thumbnails') or [])
cfb56d1a 3539 if not thumbnails:
8130779d 3540 return None
3541 return render_table(
ec11a9f4 3542 self._list_format_headers('ID', 'Width', 'Height', 'URL'),
6970b600 3543 [[t.get('id'), t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails])
2412044c 3544
8130779d 3545 def render_subtitles_table(self, video_id, subtitles):
2412044c 3546 def _row(lang, formats):
49c258e1 3547 exts, names = zip(*((f['ext'], f.get('name') or 'unknown') for f in reversed(formats)))
2412044c 3548 if len(set(names)) == 1:
7aee40c1 3549 names = [] if names[0] == 'unknown' else names[:1]
2412044c 3550 return [lang, ', '.join(names), ', '.join(exts)]
3551
8130779d 3552 if not subtitles:
3553 return None
3554 return render_table(
ec11a9f4 3555 self._list_format_headers('Language', 'Name', 'Formats'),
2412044c 3556 [_row(lang, formats) for lang, formats in subtitles.items()],
8130779d 3557 hide_empty=True)
3558
3559 def __list_table(self, video_id, name, func, *args):
3560 table = func(*args)
3561 if not table:
3562 self.to_screen(f'{video_id} has no {name}')
3563 return
3564 self.to_screen(f'[info] Available {name} for {video_id}:')
3565 self.to_stdout(table)
3566
3567 def list_formats(self, info_dict):
3568 self.__list_table(info_dict['id'], 'formats', self.render_formats_table, info_dict)
3569
3570 def list_thumbnails(self, info_dict):
3571 self.__list_table(info_dict['id'], 'thumbnails', self.render_thumbnails_table, info_dict)
3572
3573 def list_subtitles(self, video_id, subtitles, name='subtitles'):
3574 self.__list_table(video_id, name, self.render_subtitles_table, video_id, subtitles)
a504ced0 3575
dca08720
PH
3576 def urlopen(self, req):
3577 """ Start an HTTP download """
f9934b96 3578 if isinstance(req, str):
67dda517 3579 req = sanitized_Request(req)
19a41fc6 3580 return self._opener.open(req, timeout=self._socket_timeout)
dca08720
PH
3581
3582 def print_debug_header(self):
3583 if not self.params.get('verbose'):
3584 return
49a57e70 3585
560738f3 3586 # These imports can be slow. So import them only as needed
3587 from .extractor.extractors import _LAZY_LOADER
3588 from .extractor.extractors import _PLUGIN_CLASSES as plugin_extractors
3589
49a57e70 3590 def get_encoding(stream):
2a938746 3591 ret = str(getattr(stream, 'encoding', 'missing (%s)' % type(stream).__name__))
49a57e70 3592 if not supports_terminal_sequences(stream):
53973b4d 3593 from .utils import WINDOWS_VT_MODE # Must be imported locally
e3c7d495 3594 ret += ' (No VT)' if WINDOWS_VT_MODE is False else ' (No ANSI)'
49a57e70 3595 return ret
3596
591bb9d3 3597 encoding_str = 'Encodings: locale %s, fs %s, pref %s, %s' % (
49a57e70 3598 locale.getpreferredencoding(),
3599 sys.getfilesystemencoding(),
591bb9d3 3600 self.get_encoding(),
3601 ', '.join(
64fa820c 3602 f'{key} {get_encoding(stream)}' for key, stream in self._out_files.items_
591bb9d3 3603 if stream is not None and key != 'console')
3604 )
883d4b1e 3605
3606 logger = self.params.get('logger')
3607 if logger:
3608 write_debug = lambda msg: logger.debug(f'[debug] {msg}')
3609 write_debug(encoding_str)
3610 else:
96565c7e 3611 write_string(f'[debug] {encoding_str}\n', encoding=None)
49a57e70 3612 write_debug = lambda msg: self._write_string(f'[debug] {msg}\n')
734f90bb 3613
4c88ff87 3614 source = detect_variant()
36eaf303 3615 write_debug(join_nonempty(
3616 'yt-dlp version', __version__,
3617 f'[{RELEASE_GIT_HEAD}]' if RELEASE_GIT_HEAD else '',
3618 '' if source == 'unknown' else f'({source})',
3619 delim=' '))
6e21fdd2 3620 if not _LAZY_LOADER:
3621 if os.environ.get('YTDLP_NO_LAZY_EXTRACTORS'):
49a57e70 3622 write_debug('Lazy loading extractors is forcibly disabled')
6e21fdd2 3623 else:
49a57e70 3624 write_debug('Lazy loading extractors is disabled')
3ae5e797 3625 if plugin_extractors or plugin_postprocessors:
49a57e70 3626 write_debug('Plugins: %s' % [
3ae5e797 3627 '%s%s' % (klass.__name__, '' if klass.__name__ == name else f' as {name}')
3628 for name, klass in itertools.chain(plugin_extractors.items(), plugin_postprocessors.items())])
8a82af35 3629 if self.params['compat_opts']:
3630 write_debug('Compatibility options: %s' % ', '.join(self.params['compat_opts']))
36eaf303 3631
3632 if source == 'source':
dca08720 3633 try:
f0c9fb96 3634 stdout, _, _ = Popen.run(
36eaf303 3635 ['git', 'rev-parse', '--short', 'HEAD'],
f0c9fb96 3636 text=True, cwd=os.path.dirname(os.path.abspath(__file__)),
3637 stdout=subprocess.PIPE, stderr=subprocess.PIPE)
3638 if re.fullmatch('[0-9a-f]+', stdout.strip()):
3639 write_debug(f'Git HEAD: {stdout.strip()}')
70a1165b 3640 except Exception:
19a03940 3641 with contextlib.suppress(Exception):
36eaf303 3642 sys.exc_clear()
b300cda4
S
3643
3644 def python_implementation():
3645 impl_name = platform.python_implementation()
3646 if impl_name == 'PyPy' and hasattr(sys, 'pypy_version_info'):
3647 return impl_name + ' version %d.%d.%d' % sys.pypy_version_info[:3]
3648 return impl_name
3649
49a57e70 3650 write_debug('Python version %s (%s %s) - %s' % (
e5813e53 3651 platform.python_version(),
3652 python_implementation(),
3653 platform.architecture()[0],
b300cda4 3654 platform_name()))
d28b5171 3655
8913ef74 3656 exe_versions, ffmpeg_features = FFmpegPostProcessor.get_versions_and_features(self)
3657 ffmpeg_features = {key for key, val in ffmpeg_features.items() if val}
3658 if ffmpeg_features:
19a03940 3659 exe_versions['ffmpeg'] += ' (%s)' % ','.join(sorted(ffmpeg_features))
8913ef74 3660
4c83c967 3661 exe_versions['rtmpdump'] = rtmpdump_version()
feee8d32 3662 exe_versions['phantomjs'] = PhantomJSwrapper._version()
d28b5171 3663 exe_str = ', '.join(
2831b468 3664 f'{exe} {v}' for exe, v in sorted(exe_versions.items()) if v
3665 ) or 'none'
49a57e70 3666 write_debug('exe versions: %s' % exe_str)
dca08720 3667
1d485a1a 3668 from .compat.compat_utils import get_package_info
9b8ee23b 3669 from .dependencies import available_dependencies
3670
3671 write_debug('Optional libraries: %s' % (', '.join(sorted({
1d485a1a 3672 join_nonempty(*get_package_info(m)) for m in available_dependencies.values()
9b8ee23b 3673 })) or 'none'))
2831b468 3674
97ec5bc5 3675 self._setup_opener()
dca08720
PH
3676 proxy_map = {}
3677 for handler in self._opener.handlers:
3678 if hasattr(handler, 'proxies'):
3679 proxy_map.update(handler.proxies)
49a57e70 3680 write_debug(f'Proxy map: {proxy_map}')
dca08720 3681
49a57e70 3682 # Not implemented
3683 if False and self.params.get('call_home'):
0f06bcd7 3684 ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode()
49a57e70 3685 write_debug('Public IP address: %s' % ipaddr)
58b1f00d 3686 latest_version = self.urlopen(
0f06bcd7 3687 'https://yt-dl.org/latest/version').read().decode()
58b1f00d
PH
3688 if version_tuple(latest_version) > version_tuple(__version__):
3689 self.report_warning(
3690 'You are using an outdated version (newest version: %s)! '
3691 'See https://yt-dl.org/update if you need help updating.' %
3692 latest_version)
3693
e344693b 3694 def _setup_opener(self):
97ec5bc5 3695 if hasattr(self, '_opener'):
3696 return
6ad14cab 3697 timeout_val = self.params.get('socket_timeout')
17bddf3e 3698 self._socket_timeout = 20 if timeout_val is None else float(timeout_val)
6ad14cab 3699
982ee69a 3700 opts_cookiesfrombrowser = self.params.get('cookiesfrombrowser')
dca08720
PH
3701 opts_cookiefile = self.params.get('cookiefile')
3702 opts_proxy = self.params.get('proxy')
3703
982ee69a 3704 self.cookiejar = load_cookies(opts_cookiefile, opts_cookiesfrombrowser, self)
dca08720 3705
6a3f4c3f 3706 cookie_processor = YoutubeDLCookieProcessor(self.cookiejar)
dca08720
PH
3707 if opts_proxy is not None:
3708 if opts_proxy == '':
3709 proxies = {}
3710 else:
3711 proxies = {'http': opts_proxy, 'https': opts_proxy}
3712 else:
3713 proxies = compat_urllib_request.getproxies()
067aa17e 3714 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
dca08720
PH
3715 if 'http' in proxies and 'https' not in proxies:
3716 proxies['https'] = proxies['http']
91410c9b 3717 proxy_handler = PerRequestProxyHandler(proxies)
a0ddb8a2
PH
3718
3719 debuglevel = 1 if self.params.get('debug_printtraffic') else 0
be4a824d
PH
3720 https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel)
3721 ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel)
fca6dba8 3722 redirect_handler = YoutubeDLRedirectHandler()
f9934b96 3723 data_handler = urllib.request.DataHandler()
6240b0a2
JMF
3724
3725 # When passing our own FileHandler instance, build_opener won't add the
3726 # default FileHandler and allows us to disable the file protocol, which
3727 # can be used for malicious purposes (see
067aa17e 3728 # https://github.com/ytdl-org/youtube-dl/issues/8227)
6240b0a2
JMF
3729 file_handler = compat_urllib_request.FileHandler()
3730
3731 def file_open(*args, **kwargs):
7a5c1cfe 3732 raise compat_urllib_error.URLError('file:// scheme is explicitly disabled in yt-dlp for security reasons')
6240b0a2
JMF
3733 file_handler.file_open = file_open
3734
3735 opener = compat_urllib_request.build_opener(
fca6dba8 3736 proxy_handler, https_handler, cookie_processor, ydlh, redirect_handler, data_handler, file_handler)
2461f79d 3737
dca08720
PH
3738 # Delete the default user-agent header, which would otherwise apply in
3739 # cases where our custom HTTP handler doesn't come into play
067aa17e 3740 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
dca08720
PH
3741 opener.addheaders = []
3742 self._opener = opener
62fec3b2
PH
3743
3744 def encode(self, s):
3745 if isinstance(s, bytes):
3746 return s # Already encoded
3747
3748 try:
3749 return s.encode(self.get_encoding())
3750 except UnicodeEncodeError as err:
3751 err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.'
3752 raise
3753
3754 def get_encoding(self):
3755 encoding = self.params.get('encoding')
3756 if encoding is None:
3757 encoding = preferredencoding()
3758 return encoding
ec82d85a 3759
e08a85d8 3760 def _write_info_json(self, label, ie_result, infofn, overwrite=None):
cb96c5be 3761 ''' Write infojson and returns True = written, 'exists' = Already exists, False = skip, None = error '''
e08a85d8 3762 if overwrite is None:
3763 overwrite = self.params.get('overwrites', True)
80c03fa9 3764 if not self.params.get('writeinfojson'):
3765 return False
3766 elif not infofn:
3767 self.write_debug(f'Skipping writing {label} infojson')
3768 return False
3769 elif not self._ensure_dir_exists(infofn):
3770 return None
e08a85d8 3771 elif not overwrite and os.path.exists(infofn):
80c03fa9 3772 self.to_screen(f'[info] {label.title()} metadata is already present')
cb96c5be 3773 return 'exists'
3774
3775 self.to_screen(f'[info] Writing {label} metadata as JSON to: {infofn}')
3776 try:
3777 write_json_file(self.sanitize_info(ie_result, self.params.get('clean_infojson', True)), infofn)
3778 return True
86e5f3ed 3779 except OSError:
cb96c5be 3780 self.report_error(f'Cannot write {label} metadata to JSON file {infofn}')
3781 return None
80c03fa9 3782
3783 def _write_description(self, label, ie_result, descfn):
3784 ''' Write description and returns True = written, False = skip, None = error '''
3785 if not self.params.get('writedescription'):
3786 return False
3787 elif not descfn:
3788 self.write_debug(f'Skipping writing {label} description')
3789 return False
3790 elif not self._ensure_dir_exists(descfn):
3791 return None
3792 elif not self.params.get('overwrites', True) and os.path.exists(descfn):
3793 self.to_screen(f'[info] {label.title()} description is already present')
3794 elif ie_result.get('description') is None:
3795 self.report_warning(f'There\'s no {label} description to write')
3796 return False
3797 else:
3798 try:
3799 self.to_screen(f'[info] Writing {label} description to: {descfn}')
86e5f3ed 3800 with open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
80c03fa9 3801 descfile.write(ie_result['description'])
86e5f3ed 3802 except OSError:
80c03fa9 3803 self.report_error(f'Cannot write {label} description file {descfn}')
3804 return None
3805 return True
3806
3807 def _write_subtitles(self, info_dict, filename):
3808 ''' Write subtitles to file and return list of (sub_filename, final_sub_filename); or None if error'''
3809 ret = []
3810 subtitles = info_dict.get('requested_subtitles')
3811 if not subtitles or not (self.params.get('writesubtitles') or self.params.get('writeautomaticsub')):
3812 # subtitles download errors are already managed as troubles in relevant IE
3813 # that way it will silently go on when used with unsupporting IE
3814 return ret
3815
3816 sub_filename_base = self.prepare_filename(info_dict, 'subtitle')
3817 if not sub_filename_base:
3818 self.to_screen('[info] Skipping writing video subtitles')
3819 return ret
3820 for sub_lang, sub_info in subtitles.items():
3821 sub_format = sub_info['ext']
3822 sub_filename = subtitles_filename(filename, sub_lang, sub_format, info_dict.get('ext'))
3823 sub_filename_final = subtitles_filename(sub_filename_base, sub_lang, sub_format, info_dict.get('ext'))
e04938ab 3824 existing_sub = self.existing_file((sub_filename_final, sub_filename))
3825 if existing_sub:
80c03fa9 3826 self.to_screen(f'[info] Video subtitle {sub_lang}.{sub_format} is already present')
e04938ab 3827 sub_info['filepath'] = existing_sub
3828 ret.append((existing_sub, sub_filename_final))
80c03fa9 3829 continue
3830
3831 self.to_screen(f'[info] Writing video subtitles to: {sub_filename}')
3832 if sub_info.get('data') is not None:
3833 try:
3834 # Use newline='' to prevent conversion of newline characters
3835 # See https://github.com/ytdl-org/youtube-dl/issues/10268
86e5f3ed 3836 with open(sub_filename, 'w', encoding='utf-8', newline='') as subfile:
80c03fa9 3837 subfile.write(sub_info['data'])
3838 sub_info['filepath'] = sub_filename
3839 ret.append((sub_filename, sub_filename_final))
3840 continue
86e5f3ed 3841 except OSError:
80c03fa9 3842 self.report_error(f'Cannot write video subtitles file {sub_filename}')
3843 return None
3844
3845 try:
3846 sub_copy = sub_info.copy()
3847 sub_copy.setdefault('http_headers', info_dict.get('http_headers'))
3848 self.dl(sub_filename, sub_copy, subtitle=True)
3849 sub_info['filepath'] = sub_filename
3850 ret.append((sub_filename, sub_filename_final))
6020e05d 3851 except (DownloadError, ExtractorError, IOError, OSError, ValueError) + network_exceptions as err:
c70c418d 3852 msg = f'Unable to download video subtitles for {sub_lang!r}: {err}'
6020e05d 3853 if self.params.get('ignoreerrors') is not True: # False or 'only_download'
c70c418d 3854 if not self.params.get('ignoreerrors'):
3855 self.report_error(msg)
3856 raise DownloadError(msg)
3857 self.report_warning(msg)
519804a9 3858 return ret
80c03fa9 3859
3860 def _write_thumbnails(self, label, info_dict, filename, thumb_filename_base=None):
3861 ''' Write thumbnails to file and return list of (thumb_filename, final_thumb_filename) '''
6c4fd172 3862 write_all = self.params.get('write_all_thumbnails', False)
80c03fa9 3863 thumbnails, ret = [], []
6c4fd172 3864 if write_all or self.params.get('writethumbnail', False):
0202b52a 3865 thumbnails = info_dict.get('thumbnails') or []
6c4fd172 3866 multiple = write_all and len(thumbnails) > 1
ec82d85a 3867
80c03fa9 3868 if thumb_filename_base is None:
3869 thumb_filename_base = filename
3870 if thumbnails and not thumb_filename_base:
3871 self.write_debug(f'Skipping writing {label} thumbnail')
3872 return ret
3873
dd0228ce 3874 for idx, t in list(enumerate(thumbnails))[::-1]:
80c03fa9 3875 thumb_ext = (f'{t["id"]}.' if multiple else '') + determine_ext(t['url'], 'jpg')
aa9369a2 3876 thumb_display_id = f'{label} thumbnail {t["id"]}'
80c03fa9 3877 thumb_filename = replace_extension(filename, thumb_ext, info_dict.get('ext'))
3878 thumb_filename_final = replace_extension(thumb_filename_base, thumb_ext, info_dict.get('ext'))
ec82d85a 3879
e04938ab 3880 existing_thumb = self.existing_file((thumb_filename_final, thumb_filename))
3881 if existing_thumb:
aa9369a2 3882 self.to_screen('[info] %s is already present' % (
3883 thumb_display_id if multiple else f'{label} thumbnail').capitalize())
e04938ab 3884 t['filepath'] = existing_thumb
3885 ret.append((existing_thumb, thumb_filename_final))
ec82d85a 3886 else:
80c03fa9 3887 self.to_screen(f'[info] Downloading {thumb_display_id} ...')
ec82d85a 3888 try:
297e9952 3889 uf = self.urlopen(sanitized_Request(t['url'], headers=t.get('http_headers', {})))
80c03fa9 3890 self.to_screen(f'[info] Writing {thumb_display_id} to: {thumb_filename}')
d3d89c32 3891 with open(encodeFilename(thumb_filename), 'wb') as thumbf:
ec82d85a 3892 shutil.copyfileobj(uf, thumbf)
80c03fa9 3893 ret.append((thumb_filename, thumb_filename_final))
885cc0b7 3894 t['filepath'] = thumb_filename
3158150c 3895 except network_exceptions as err:
dd0228ce 3896 thumbnails.pop(idx)
80c03fa9 3897 self.report_warning(f'Unable to download {thumb_display_id}: {err}')
6c4fd172 3898 if ret and not write_all:
3899 break
0202b52a 3900 return ret