]> jfr.im git - yt-dlp.git/blame - yt_dlp/YoutubeDL.py
[cleanup] Misc
[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 (
003c69a8 30 compat_get_terminal_size,
e9c0cdd3 31 compat_os_name,
7d1eb38a 32 compat_shlex_quote,
ce02ed60
PH
33 compat_str,
34 compat_urllib_error,
35 compat_urllib_request,
8c25f81b 36)
982ee69a 37from .cookies import load_cookies
f8271158 38from .downloader import FFmpegFD, get_suitable_downloader, shorten_protocol_name
39from .downloader.rtmp import rtmpdump_version
40from .extractor import _LAZY_LOADER
41from .extractor import _PLUGIN_CLASSES as plugin_extractors
42from .extractor import gen_extractor_classes, get_info_extractor
43from .extractor.openload import PhantomJSwrapper
44from .minicurses import format_text
45from .postprocessor import _PLUGIN_CLASSES as plugin_postprocessors
46from .postprocessor import (
47 EmbedThumbnailPP,
48 FFmpegFixupDuplicateMoovPP,
49 FFmpegFixupDurationPP,
50 FFmpegFixupM3u8PP,
51 FFmpegFixupM4aPP,
52 FFmpegFixupStretchedPP,
53 FFmpegFixupTimestampPP,
54 FFmpegMergerPP,
55 FFmpegPostProcessor,
56 MoveFilesAfterDownloadPP,
57 get_postprocessor,
58)
59from .update import detect_variant
8c25f81b 60from .utils import (
f8271158 61 DEFAULT_OUTTMPL,
62 LINK_TEMPLATES,
63 NO_DEFAULT,
1d485a1a 64 NUMBER_RE,
f8271158 65 OUTTMPL_TYPES,
66 POSTPROCESS_WHEN,
67 STR_FORMAT_RE_TMPL,
68 STR_FORMAT_TYPES,
69 ContentTooShortError,
70 DateRange,
71 DownloadCancelled,
72 DownloadError,
73 EntryNotInPlaylist,
74 ExistingVideoReached,
75 ExtractorError,
76 GeoRestrictedError,
77 HEADRequest,
78 InAdvancePagedList,
79 ISO3166Utils,
80 LazyList,
81 MaxDownloadsReached,
19a03940 82 Namespace,
f8271158 83 PagedList,
84 PerRequestProxyHandler,
85 Popen,
86 PostProcessingError,
87 ReExtractInfo,
88 RejectedVideoReached,
89 SameFileError,
90 UnavailableVideoError,
91 YoutubeDLCookieProcessor,
92 YoutubeDLHandler,
93 YoutubeDLRedirectHandler,
eedb7ba5
S
94 age_restricted,
95 args_to_str,
ce02ed60 96 date_from_str,
ce02ed60 97 determine_ext,
b5559424 98 determine_protocol,
c0384f22 99 encode_compat_str,
ce02ed60 100 encodeFilename,
a06916d9 101 error_to_compat_str,
590bc6f6 102 expand_path,
90137ca4 103 filter_dict,
e29663c6 104 float_or_none,
02dbf93f 105 format_bytes,
e0fd9573 106 format_decimal_suffix,
f8271158 107 format_field,
525ef922 108 formatSeconds,
0bb322b9 109 get_domain,
c9969434 110 int_or_none,
732044af 111 iri_to_uri,
34921b43 112 join_nonempty,
ce02ed60 113 locked_file,
0202b52a 114 make_dir,
dca08720 115 make_HTTPS_handler,
8b7539d2 116 merge_headers,
3158150c 117 network_exceptions,
ec11a9f4 118 number_of_digits,
cd6fc19e 119 orderedSet,
083c9df9 120 parse_filesize,
dca08720 121 platform_name,
ce02ed60 122 preferredencoding,
eedb7ba5 123 prepend_extension,
51fb4995 124 register_socks_protocols,
3efb96a6 125 remove_terminal_sequences,
cfb56d1a 126 render_table,
eedb7ba5 127 replace_extension,
ce02ed60 128 sanitize_filename,
1bb5c511 129 sanitize_path,
dcf77cf1 130 sanitize_url,
67dda517 131 sanitized_Request,
e5660ee6 132 std_headers,
1211bb6d 133 str_or_none,
e29663c6 134 strftime_or_none,
ce02ed60 135 subtitles_filename,
819e0531 136 supports_terminal_sequences,
f2ebc5c7 137 timetuple_from_msec,
732044af 138 to_high_limit_path,
324ad820 139 traverse_obj,
6033d980 140 try_get,
29eb5174 141 url_basename,
7d1eb38a 142 variadic,
58b1f00d 143 version_tuple,
53973b4d 144 windows_enable_vt_mode,
ce02ed60
PH
145 write_json_file,
146 write_string,
4f026faf 147)
f8271158 148from .version import RELEASE_GIT_HEAD, __version__
8222d8de 149
e9c0cdd3
YCH
150if compat_os_name == 'nt':
151 import ctypes
152
2459b6e1 153
86e5f3ed 154class YoutubeDL:
8222d8de
JMF
155 """YoutubeDL class.
156
157 YoutubeDL objects are the ones responsible of downloading the
158 actual video file and writing it to disk if the user has requested
159 it, among some other tasks. In most cases there should be one per
160 program. As, given a video URL, the downloader doesn't know how to
161 extract all the needed information, task that InfoExtractors do, it
162 has to pass the URL to one of them.
163
164 For this, YoutubeDL objects have a method that allows
165 InfoExtractors to be registered in a given order. When it is passed
166 a URL, the YoutubeDL object handles it to the first InfoExtractor it
167 finds that reports being able to handle it. The InfoExtractor extracts
168 all the information about the video or videos the URL refers to, and
169 YoutubeDL process the extracted information, possibly using a File
170 Downloader to download the video.
171
172 YoutubeDL objects accept a lot of parameters. In order not to saturate
173 the object constructor with arguments, it receives a dictionary of
174 options instead. These options are available through the params
175 attribute for the InfoExtractors to use. The YoutubeDL also
176 registers itself as the downloader in charge for the InfoExtractors
177 that are added to it, so this is a "mutual registration".
178
179 Available options:
180
181 username: Username for authentication purposes.
182 password: Password for authentication purposes.
180940e0 183 videopassword: Password for accessing a video.
1da50aa3
S
184 ap_mso: Adobe Pass multiple-system operator identifier.
185 ap_username: Multiple-system operator account username.
186 ap_password: Multiple-system operator account password.
8222d8de
JMF
187 usenetrc: Use netrc for authentication instead.
188 verbose: Print additional info to stdout.
189 quiet: Do not print messages to stdout.
ad8915b7 190 no_warnings: Do not print out anything for warnings.
bb66c247 191 forceprint: A dict with keys WHEN mapped to a list of templates to
192 print to stdout. The allowed keys are video or any of the
193 items in utils.POSTPROCESS_WHEN.
ca30f449 194 For compatibility, a single list is also accepted
bb66c247 195 print_to_file: A dict with keys WHEN (same as forceprint) mapped to
196 a list of tuples with (template, filename)
53c18592 197 forceurl: Force printing final URL. (Deprecated)
198 forcetitle: Force printing title. (Deprecated)
199 forceid: Force printing ID. (Deprecated)
200 forcethumbnail: Force printing thumbnail URL. (Deprecated)
201 forcedescription: Force printing description. (Deprecated)
202 forcefilename: Force printing final filename. (Deprecated)
203 forceduration: Force printing duration. (Deprecated)
8694c600 204 forcejson: Force printing info_dict as JSON.
63e0be34
PH
205 dump_single_json: Force printing the info_dict of the whole playlist
206 (or video) as a single JSON line.
c25228e5 207 force_write_download_archive: Force writing download archive regardless
208 of 'skip_download' or 'simulate'.
b7b04c78 209 simulate: Do not download the video files. If unset (or None),
210 simulate only if listsubtitles, listformats or list_thumbnails is used
eb8a4433 211 format: Video format code. see "FORMAT SELECTION" for more details.
093a1710 212 You can also pass a function. The function takes 'ctx' as
213 argument and returns the formats to download.
214 See "build_format_selector" for an implementation
63ad4d43 215 allow_unplayable_formats: Allow unplayable formats to be extracted and downloaded.
b7da73eb 216 ignore_no_formats_error: Ignore "No video formats" error. Usefull for
217 extracting metadata even if the video is not actually
218 available for download (experimental)
0930b11f 219 format_sort: A list of fields by which to sort the video formats.
220 See "Sorting Formats" for more details.
c25228e5 221 format_sort_force: Force the given format_sort. see "Sorting Formats"
222 for more details.
08d30158 223 prefer_free_formats: Whether to prefer video formats with free containers
224 over non-free ones of same quality.
c25228e5 225 allow_multiple_video_streams: Allow multiple video streams to be merged
226 into a single file
227 allow_multiple_audio_streams: Allow multiple audio streams to be merged
228 into a single file
0ba692ac 229 check_formats Whether to test if the formats are downloadable.
9f1a1c36 230 Can be True (check all), False (check none),
231 'selected' (check selected formats),
0ba692ac 232 or None (check only if requested by extractor)
4524baf0 233 paths: Dictionary of output paths. The allowed keys are 'home'
234 'temp' and the keys of OUTTMPL_TYPES (in utils.py)
de6000d9 235 outtmpl: Dictionary of templates for output names. Allowed keys
4524baf0 236 are 'default' and the keys of OUTTMPL_TYPES (in utils.py).
34488702 237 For compatibility with youtube-dl, a single string can also be used
a820dc72
RA
238 outtmpl_na_placeholder: Placeholder for unavailable meta fields.
239 restrictfilenames: Do not allow "&" and spaces in file names
240 trim_file_name: Limit length of filename (extension excluded)
4524baf0 241 windowsfilenames: Force the filenames to be windows compatible
b1940459 242 ignoreerrors: Do not stop on download/postprocessing errors.
243 Can be 'only_download' to ignore only download errors.
244 Default is 'only_download' for CLI, but False for API
26e2805c 245 skip_playlist_after_errors: Number of allowed failures until the rest of
246 the playlist is skipped
d22dec74 247 force_generic_extractor: Force downloader to use the generic extractor
0c3d0f51 248 overwrites: Overwrite all video and metadata files if True,
249 overwrite only non-video files if None
250 and don't overwrite any file if False
34488702 251 For compatibility with youtube-dl,
252 "nooverwrites" may also be used instead
8222d8de
JMF
253 playliststart: Playlist item to start at.
254 playlistend: Playlist item to end at.
c14e88f0 255 playlist_items: Specific indices of playlist to download.
ff815fe6 256 playlistreverse: Download playlist items in reverse order.
75822ca7 257 playlistrandom: Download playlist items in random order.
8222d8de
JMF
258 matchtitle: Download only matching titles.
259 rejecttitle: Reject downloads for matching titles.
8bf9319e 260 logger: Log messages to a logging.Logger instance.
8222d8de 261 logtostderr: Log messages to stderr instead of stdout.
819e0531 262 consoletitle: Display progress in console window's titlebar.
8222d8de
JMF
263 writedescription: Write the video description to a .description file
264 writeinfojson: Write the video description to a .info.json file
75d43ca0 265 clean_infojson: Remove private fields from the infojson
34488702 266 getcomments: Extract video comments. This will not be written to disk
06167fbb 267 unless writeinfojson is also given
1fb07d10 268 writeannotations: Write the video annotations to a .annotations.xml file
8222d8de 269 writethumbnail: Write the thumbnail image to a file
c25228e5 270 allow_playlist_files: Whether to write playlists' description, infojson etc
271 also to disk when using the 'write*' options
ec82d85a 272 write_all_thumbnails: Write all thumbnail formats to files
732044af 273 writelink: Write an internet shortcut file, depending on the
274 current platform (.url/.webloc/.desktop)
275 writeurllink: Write a Windows internet shortcut file (.url)
276 writewebloclink: Write a macOS internet shortcut file (.webloc)
277 writedesktoplink: Write a Linux internet shortcut file (.desktop)
8222d8de 278 writesubtitles: Write the video subtitles to a file
741dd8ea 279 writeautomaticsub: Write the automatically generated subtitles to a file
245524e6 280 allsubtitles: Deprecated - Use subtitleslangs = ['all']
c32b0aab 281 Downloads all the subtitles of the video
0b7f3118 282 (requires writesubtitles or writeautomaticsub)
8222d8de 283 listsubtitles: Lists all available subtitles for the video
a504ced0 284 subtitlesformat: The format code for subtitles
c32b0aab 285 subtitleslangs: List of languages of the subtitles to download (can be regex).
286 The list may contain "all" to refer to all the available
287 subtitles. The language can be prefixed with a "-" to
288 exclude it from the requested languages. Eg: ['all', '-live_chat']
8222d8de
JMF
289 keepvideo: Keep the video file after post-processing
290 daterange: A DateRange object, download only if the upload_date is in the range.
291 skip_download: Skip the actual download of the video file
c35f9e72 292 cachedir: Location of the cache files in the filesystem.
a0e07d31 293 False to disable filesystem cache.
47192f92 294 noplaylist: Download single video instead of a playlist if in doubt.
8dbe9899
PH
295 age_limit: An integer representing the user's age in years.
296 Unsuitable videos for the given age are skipped.
5fe18bdb
PH
297 min_views: An integer representing the minimum view count the video
298 must have in order to not be skipped.
299 Videos without view count information are always
300 downloaded. None for no limit.
301 max_views: An integer representing the maximum view count.
302 Videos that are more popular than that are not
303 downloaded.
304 Videos without view count information are always
305 downloaded. None for no limit.
306 download_archive: File name of a file where all downloads are recorded.
c1c9a79c
PH
307 Videos already present in the file are not downloaded
308 again.
8a51f564 309 break_on_existing: Stop the download process after attempting to download a
310 file that is in the archive.
311 break_on_reject: Stop the download process when encountering a video that
312 has been filtered out.
b222c271 313 break_per_url: Whether break_on_reject and break_on_existing
314 should act on each input URL as opposed to for the entire queue
d76fa1f3 315 cookiefile: File name or text stream from where cookies should be read and dumped to
f59f5ef8
MB
316 cookiesfrombrowser: A tuple containing the name of the browser, the profile
317 name/pathfrom where cookies are loaded, and the name of the
318 keyring. Eg: ('chrome', ) or ('vivaldi', 'default', 'BASICTEXT')
f81c62a6 319 legacyserverconnect: Explicitly allow HTTPS connection to servers that do not
320 support RFC 5746 secure renegotiation
f59f5ef8 321 nocheckcertificate: Do not verify SSL certificates
bb58c9ed 322 client_certificate: Path to client certificate file in PEM format. May include the private key
323 client_certificate_key: Path to private key file for client certificate
324 client_certificate_password: Password for client certificate private key, if encrypted.
325 If not provided and the key is encrypted, yt-dlp will ask interactively
7e8c0af0
PH
326 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
327 At the moment, this is only supported by YouTube.
8b7539d2 328 http_headers: A dictionary of custom headers to be used for all requests
a1ee09e8 329 proxy: URL of the proxy server to use
38cce791 330 geo_verification_proxy: URL of the proxy to use for IP address verification
504f20dd 331 on geo-restricted sites.
e344693b 332 socket_timeout: Time to wait for unresponsive hosts, in seconds
0783b09b
PH
333 bidi_workaround: Work around buggy terminals without bidirectional text
334 support, using fridibi
a0ddb8a2 335 debug_printtraffic:Print out sent and received HTTP traffic
91f071af 336 include_ads: Download ads as well (deprecated)
04b4d394
PH
337 default_search: Prepend this string if an input url is not valid.
338 'auto' for elaborate guessing
62fec3b2 339 encoding: Use this encoding instead of the system-specified.
e8ee972c 340 extract_flat: Do not resolve URLs, return the immediate result.
057a5206
PH
341 Pass in 'in_playlist' to only show this behavior for
342 playlist items.
f2ebc5c7 343 wait_for_video: If given, wait for scheduled streams to become available.
344 The value should be a tuple containing the range
345 (min_secs, max_secs) to wait between retries
4f026faf 346 postprocessors: A list of dictionaries, each with an entry
71b640cc 347 * key: The name of the postprocessor. See
7a5c1cfe 348 yt_dlp/postprocessor/__init__.py for a list.
bb66c247 349 * when: When to run the postprocessor. Allowed values are
350 the entries of utils.POSTPROCESS_WHEN
56d868db 351 Assumed to be 'post_process' if not given
b5ae35ee 352 post_hooks: Deprecated - Register a custom postprocessor instead
353 A list of functions that get called as the final step
ab8e5e51
AM
354 for each video file, after all postprocessors have been
355 called. The filename will be passed as the only argument.
71b640cc
PH
356 progress_hooks: A list of functions that get called on download
357 progress, with a dictionary with the entries
5cda4eda 358 * status: One of "downloading", "error", or "finished".
ee69b99a 359 Check this first and ignore unknown values.
3ba7740d 360 * info_dict: The extracted info_dict
71b640cc 361
5cda4eda 362 If status is one of "downloading", or "finished", the
ee69b99a
PH
363 following properties may also be present:
364 * filename: The final filename (always present)
5cda4eda 365 * tmpfilename: The filename we're currently writing to
71b640cc
PH
366 * downloaded_bytes: Bytes on disk
367 * total_bytes: Size of the whole file, None if unknown
5cda4eda
PH
368 * total_bytes_estimate: Guess of the eventual file size,
369 None if unavailable.
370 * elapsed: The number of seconds since download started.
71b640cc
PH
371 * eta: The estimated time in seconds, None if unknown
372 * speed: The download speed in bytes/second, None if
373 unknown
5cda4eda
PH
374 * fragment_index: The counter of the currently
375 downloaded video fragment.
376 * fragment_count: The number of fragments (= individual
377 files that will be merged)
71b640cc
PH
378
379 Progress hooks are guaranteed to be called at least once
380 (with status "finished") if the download is successful.
819e0531 381 postprocessor_hooks: A list of functions that get called on postprocessing
382 progress, with a dictionary with the entries
383 * status: One of "started", "processing", or "finished".
384 Check this first and ignore unknown values.
385 * postprocessor: Name of the postprocessor
386 * info_dict: The extracted info_dict
387
388 Progress hooks are guaranteed to be called at least twice
389 (with status "started" and "finished") if the processing is successful.
45598f15 390 merge_output_format: Extension to use when merging formats.
6b591b29 391 final_ext: Expected final extension; used to detect when the file was
59a7a13e 392 already downloaded and converted
6271f1ca
PH
393 fixup: Automatically correct known faults of the file.
394 One of:
395 - "never": do nothing
396 - "warn": only emit a warning
397 - "detect_or_warn": check whether we can do anything
62cd676c 398 about it, warn otherwise (default)
504f20dd 399 source_address: Client-side IP address to bind to.
6ec6cb4e 400 call_home: Boolean, true iff we are allowed to contact the
7a5c1cfe 401 yt-dlp servers for debugging. (BROKEN)
1cf376f5 402 sleep_interval_requests: Number of seconds to sleep between requests
403 during extraction
7aa589a5
S
404 sleep_interval: Number of seconds to sleep before each download when
405 used alone or a lower bound of a range for randomized
406 sleep before each download (minimum possible number
407 of seconds to sleep) when used along with
408 max_sleep_interval.
409 max_sleep_interval:Upper bound of a range for randomized sleep before each
410 download (maximum possible number of seconds to sleep).
411 Must only be used along with sleep_interval.
412 Actual sleep time will be a random float from range
413 [sleep_interval; max_sleep_interval].
1cf376f5 414 sleep_interval_subtitles: Number of seconds to sleep before each subtitle download
cfb56d1a
PH
415 listformats: Print an overview of available video formats and exit.
416 list_thumbnails: Print a table of all thumbnails and exit.
0a41f331 417 match_filter: A function that gets called for every video with the signature
418 (info_dict, *, incomplete: bool) -> Optional[str]
419 For backward compatibility with youtube-dl, the signature
420 (info_dict) -> Optional[str] is also allowed.
421 - If it returns a message, the video is ignored.
422 - If it returns None, the video is downloaded.
423 - If it returns utils.NO_DEFAULT, the user is interactively
424 asked whether to download the video.
347de493 425 match_filter_func in utils.py is one example for this.
7e5db8c9 426 no_color: Do not emit color codes in output.
0a840f58 427 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
504f20dd 428 HTTP header
0a840f58 429 geo_bypass_country:
773f291d
S
430 Two-letter ISO 3166-2 country code that will be used for
431 explicit geographic restriction bypassing via faking
504f20dd 432 X-Forwarded-For HTTP header
5f95927a
S
433 geo_bypass_ip_block:
434 IP range in CIDR notation that will be used similarly to
504f20dd 435 geo_bypass_country
71b640cc 436
85729c51 437 The following options determine which downloader is picked:
52a8a1e1 438 external_downloader: A dictionary of protocol keys and the executable of the
439 external downloader to use for it. The allowed protocols
440 are default|http|ftp|m3u8|dash|rtsp|rtmp|mms.
441 Set the value to 'native' to use the native downloader
442 hls_prefer_native: Deprecated - Use external_downloader = {'m3u8': 'native'}
443 or {'m3u8': 'ffmpeg'} instead.
444 Use the native HLS downloader instead of ffmpeg/avconv
bf09af3a
S
445 if True, otherwise use ffmpeg/avconv if False, otherwise
446 use downloader suggested by extractor if None.
53ed7066 447 compat_opts: Compatibility options. See "Differences in default behavior".
3acf6d38 448 The following options do not work when used through the API:
b5ae35ee 449 filename, abort-on-error, multistreams, no-live-chat, format-sort
dac5df5a 450 no-clean-infojson, no-playlist-metafiles, no-keep-subs, no-attach-info-json.
e4f02757 451 Refer __init__.py for their implementation
819e0531 452 progress_template: Dictionary of templates for progress outputs.
453 Allowed keys are 'download', 'postprocess',
454 'download-title' (console title) and 'postprocess-title'.
455 The template is mapped on a dictionary with keys 'progress' and 'info'
23326151 456 retry_sleep_functions: Dictionary of functions that takes the number of attempts
457 as argument and returns the time to sleep in seconds.
458 Allowed keys are 'http', 'fragment', 'file_access'
fe7e0c98 459
8222d8de 460 The following parameters are not used by YoutubeDL itself, they are used by
7a5c1cfe 461 the downloader (see yt_dlp/downloader/common.py):
51d9739f 462 nopart, updatetime, buffersize, ratelimit, throttledratelimit, min_filesize,
205a0654
EH
463 max_filesize, test, noresizebuffer, retries, file_access_retries, fragment_retries,
464 continuedl, noprogress, xattr_set_filesize, hls_use_mpegts, http_chunk_size,
59a7a13e 465 external_downloader_args, concurrent_fragment_downloads.
76b1bd67
JMF
466
467 The following options are used by the post processors:
d4a24f40 468 prefer_ffmpeg: If False, use avconv instead of ffmpeg if both are available,
e4172ac9 469 otherwise prefer ffmpeg. (avconv support is deprecated)
c0b7d117
S
470 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
471 to the binary or its containing directory.
43820c03 472 postprocessor_args: A dictionary of postprocessor/executable keys (in lower case)
34488702 473 and a list of additional command-line arguments for the
474 postprocessor/executable. The dict can also have "PP+EXE" keys
475 which are used when the given exe is used by the given PP.
476 Use 'default' as the name for arguments to passed to all PP
477 For compatibility with youtube-dl, a single list of args
478 can also be used
e409895f 479
480 The following options are used by the extractors:
62bff2c1 481 extractor_retries: Number of times to retry for known errors
482 dynamic_mpd: Whether to process dynamic DASH manifests (default: True)
e409895f 483 hls_split_discontinuity: Split HLS playlists to different formats at
62bff2c1 484 discontinuities such as ad breaks (default: False)
5d3a0e79 485 extractor_args: A dictionary of arguments to be passed to the extractors.
486 See "EXTRACTOR ARGUMENTS" for details.
487 Eg: {'youtube': {'skip': ['dash', 'hls']}}
88f23a18 488 mark_watched: Mark videos watched (even with --simulate). Only for YouTube
5d3a0e79 489 youtube_include_dash_manifest: Deprecated - Use extractor_args instead.
490 If True (default), DASH manifests and related
62bff2c1 491 data will be downloaded and processed by extractor.
492 You can reduce network I/O by disabling it if you don't
493 care about DASH. (only for youtube)
5d3a0e79 494 youtube_include_hls_manifest: Deprecated - Use extractor_args instead.
495 If True (default), HLS manifests and related
62bff2c1 496 data will be downloaded and processed by extractor.
497 You can reduce network I/O by disabling it if you don't
498 care about HLS. (only for youtube)
8222d8de
JMF
499 """
500
86e5f3ed 501 _NUMERIC_FIELDS = {
c9969434 502 'width', 'height', 'tbr', 'abr', 'asr', 'vbr', 'fps', 'filesize', 'filesize_approx',
e6f21b3d 503 'timestamp', 'release_timestamp',
c9969434
S
504 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
505 'average_rating', 'comment_count', 'age_limit',
506 'start_time', 'end_time',
507 'chapter_number', 'season_number', 'episode_number',
508 'track_number', 'disc_number', 'release_year',
86e5f3ed 509 }
c9969434 510
6db9c4d5 511 _format_fields = {
512 # NB: Keep in sync with the docstring of extractor/common.py
a44ca5a4 513 'url', 'manifest_url', 'manifest_stream_number', 'ext', 'format', 'format_id', 'format_note',
6db9c4d5 514 'width', 'height', 'resolution', 'dynamic_range', 'tbr', 'abr', 'acodec', 'asr',
515 'vbr', 'fps', 'vcodec', 'container', 'filesize', 'filesize_approx',
516 'player_url', 'protocol', 'fragment_base_url', 'fragments', 'is_from_start',
517 'preference', 'language', 'language_preference', 'quality', 'source_preference',
518 'http_headers', 'stretched_ratio', 'no_resume', 'has_drm', 'downloader_options',
519 'page_url', 'app', 'play_path', 'tc_url', 'flash_version', 'rtmp_live', 'rtmp_conn', 'rtmp_protocol', 'rtmp_real_time'
520 }
48ee10ee 521 _format_selection_exts = {
522 'audio': {'m4a', 'mp3', 'ogg', 'aac'},
523 'video': {'mp4', 'flv', 'webm', '3gp'},
524 'storyboards': {'mhtml'},
525 }
526
3511266b 527 def __init__(self, params=None, auto_init=True):
883d4b1e 528 """Create a FileDownloader object with the given options.
529 @param auto_init Whether to load the default extractors and print header (if verbose).
49a57e70 530 Set to 'no_verbose_header' to not print the header
883d4b1e 531 """
e9f9a10f
JMF
532 if params is None:
533 params = {}
592b7485 534 self.params = params
8b7491c8 535 self._ies = {}
56c73665 536 self._ies_instances = {}
1e43a6f7 537 self._pps = {k: [] for k in POSTPROCESS_WHEN}
b35496d8 538 self._printed_messages = set()
1cf376f5 539 self._first_webpage_request = True
ab8e5e51 540 self._post_hooks = []
933605d7 541 self._progress_hooks = []
819e0531 542 self._postprocessor_hooks = []
8222d8de
JMF
543 self._download_retcode = 0
544 self._num_downloads = 0
9c906919 545 self._num_videos = 0
592b7485 546 self._playlist_level = 0
547 self._playlist_urls = set()
a0e07d31 548 self.cache = Cache(self)
34308b30 549
819e0531 550 windows_enable_vt_mode()
591bb9d3 551 stdout = sys.stderr if self.params.get('logtostderr') else sys.stdout
552 self._out_files = Namespace(
553 out=stdout,
554 error=sys.stderr,
555 screen=sys.stderr if self.params.get('quiet') else stdout,
556 console=None if compat_os_name == 'nt' else next(
cf4f42cb 557 filter(supports_terminal_sequences, (sys.stderr, sys.stdout)), None)
591bb9d3 558 )
559 self._allow_colors = Namespace(**{
560 type_: not self.params.get('no_color') and supports_terminal_sequences(stream)
7896214c 561 for type_, stream in self._out_files if type_ != 'console'
591bb9d3 562 })
819e0531 563
a61f4b28 564 if sys.version_info < (3, 6):
565 self.report_warning(
0181adef 566 'Python version %d.%d is not supported! Please update to Python 3.6 or above' % sys.version_info[:2])
a61f4b28 567
88acdbc2 568 if self.params.get('allow_unplayable_formats'):
569 self.report_warning(
ec11a9f4 570 f'You have asked for {self._format_err("UNPLAYABLE", self.Styles.EMPHASIS)} formats to be listed/downloaded. '
819e0531 571 'This is a developer option intended for debugging. \n'
572 ' If you experience any issues while using this option, '
ec11a9f4 573 f'{self._format_err("DO NOT", self.Styles.ERROR)} open a bug report')
88acdbc2 574
be5df5ee
S
575 def check_deprecated(param, option, suggestion):
576 if self.params.get(param) is not None:
86e5f3ed 577 self.report_warning(f'{option} is deprecated. Use {suggestion} instead')
be5df5ee
S
578 return True
579 return False
580
581 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
38cce791
YCH
582 if self.params.get('geo_verification_proxy') is None:
583 self.params['geo_verification_proxy'] = self.params['cn_verification_proxy']
584
0d1bb027 585 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
586 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
53ed7066 587 check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"')
0d1bb027 588
49a57e70 589 for msg in self.params.get('_warnings', []):
0d1bb027 590 self.report_warning(msg)
ee8dd27a 591 for msg in self.params.get('_deprecation_warnings', []):
592 self.deprecation_warning(msg)
0d1bb027 593
ec11a9f4 594 if 'list-formats' in self.params.get('compat_opts', []):
595 self.params['listformats_table'] = False
596
b5ae35ee 597 if 'overwrites' not in self.params and self.params.get('nooverwrites') is not None:
b868936c 598 # nooverwrites was unnecessarily changed to overwrites
599 # in 0c3d0f51778b153f65c21906031c2e091fcfb641
600 # This ensures compatibility with both keys
601 self.params['overwrites'] = not self.params['nooverwrites']
b5ae35ee 602 elif self.params.get('overwrites') is None:
603 self.params.pop('overwrites', None)
b868936c 604 else:
605 self.params['nooverwrites'] = not self.params['overwrites']
b9d973be 606
455a15e2 607 self.params.setdefault('forceprint', {})
608 self.params.setdefault('print_to_file', {})
bb66c247 609
610 # Compatibility with older syntax
ca30f449 611 if not isinstance(params['forceprint'], dict):
455a15e2 612 self.params['forceprint'] = {'video': params['forceprint']}
ca30f449 613
455a15e2 614 if self.params.get('bidi_workaround', False):
1c088fa8
PH
615 try:
616 import pty
617 master, slave = pty.openpty()
003c69a8 618 width = compat_get_terminal_size().columns
591bb9d3 619 width_args = [] if width is None else ['-w', str(width)]
620 sp_kwargs = {'stdin': subprocess.PIPE, 'stdout': slave, 'stderr': self._out_files.error}
5d681e96 621 try:
d3c93ec2 622 self._output_process = Popen(['bidiv'] + width_args, **sp_kwargs)
5d681e96 623 except OSError:
d3c93ec2 624 self._output_process = Popen(['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
5d681e96 625 self._output_channel = os.fdopen(master, 'rb')
1c088fa8 626 except OSError as ose:
66e7ace1 627 if ose.errno == errno.ENOENT:
49a57e70 628 self.report_warning(
629 'Could not find fribidi executable, ignoring --bidi-workaround. '
630 'Make sure that fribidi is an executable file in one of the directories in your $PATH.')
1c088fa8
PH
631 else:
632 raise
0783b09b 633
97ec5bc5 634 if auto_init:
635 if auto_init != 'no_verbose_header':
636 self.print_debug_header()
637 self.add_default_info_extractors()
638
3089bc74
S
639 if (sys.platform != 'win32'
640 and sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
455a15e2 641 and not self.params.get('restrictfilenames', False)):
e9137224 642 # Unicode filesystem API will throw errors (#1474, #13027)
34308b30 643 self.report_warning(
6febd1c1 644 'Assuming --restrict-filenames since file system encoding '
1b725173 645 'cannot encode all characters. '
6febd1c1 646 'Set the LC_ALL environment variable to fix this.')
4a98cdbf 647 self.params['restrictfilenames'] = True
34308b30 648
de6000d9 649 self.outtmpl_dict = self.parse_outtmpl()
486dd09e 650
187986a8 651 # Creating format selector here allows us to catch syntax errors before the extraction
652 self.format_selector = (
fa9f30b8 653 self.params.get('format') if self.params.get('format') in (None, '-')
093a1710 654 else self.params['format'] if callable(self.params['format'])
187986a8 655 else self.build_format_selector(self.params['format']))
656
8b7539d2 657 # Set http_headers defaults according to std_headers
658 self.params['http_headers'] = merge_headers(std_headers, self.params.get('http_headers', {}))
659
013b50b7 660 hooks = {
661 'post_hooks': self.add_post_hook,
662 'progress_hooks': self.add_progress_hook,
663 'postprocessor_hooks': self.add_postprocessor_hook,
664 }
665 for opt, fn in hooks.items():
666 for ph in self.params.get(opt, []):
667 fn(ph)
71b640cc 668
5bfc8bee 669 for pp_def_raw in self.params.get('postprocessors', []):
670 pp_def = dict(pp_def_raw)
671 when = pp_def.pop('when', 'post_process')
672 self.add_post_processor(
f9934b96 673 get_postprocessor(pp_def.pop('key'))(self, **pp_def),
5bfc8bee 674 when=when)
675
97ec5bc5 676 self._setup_opener()
51fb4995
YCH
677 register_socks_protocols()
678
ed39cac5 679 def preload_download_archive(fn):
680 """Preload the archive, if any is specified"""
681 if fn is None:
682 return False
49a57e70 683 self.write_debug(f'Loading archive file {fn!r}')
ed39cac5 684 try:
685 with locked_file(fn, 'r', encoding='utf-8') as archive_file:
686 for line in archive_file:
687 self.archive.add(line.strip())
86e5f3ed 688 except OSError as ioe:
ed39cac5 689 if ioe.errno != errno.ENOENT:
690 raise
691 return False
692 return True
693
694 self.archive = set()
695 preload_download_archive(self.params.get('download_archive'))
696
7d4111ed
PH
697 def warn_if_short_id(self, argv):
698 # short YouTube ID starting with dash?
699 idxs = [
700 i for i, a in enumerate(argv)
701 if re.match(r'^-[0-9A-Za-z_-]{10}$', a)]
702 if idxs:
703 correct_argv = (
7a5c1cfe 704 ['yt-dlp']
3089bc74
S
705 + [a for i, a in enumerate(argv) if i not in idxs]
706 + ['--'] + [argv[i] for i in idxs]
7d4111ed
PH
707 )
708 self.report_warning(
709 'Long argument string detected. '
49a57e70 710 'Use -- to separate parameters and URLs, like this:\n%s' %
7d4111ed
PH
711 args_to_str(correct_argv))
712
8222d8de
JMF
713 def add_info_extractor(self, ie):
714 """Add an InfoExtractor object to the end of the list."""
8b7491c8 715 ie_key = ie.ie_key()
716 self._ies[ie_key] = ie
e52d7f85 717 if not isinstance(ie, type):
8b7491c8 718 self._ies_instances[ie_key] = ie
e52d7f85 719 ie.set_downloader(self)
8222d8de 720
8b7491c8 721 def _get_info_extractor_class(self, ie_key):
722 ie = self._ies.get(ie_key)
723 if ie is None:
724 ie = get_info_extractor(ie_key)
725 self.add_info_extractor(ie)
726 return ie
727
56c73665
JMF
728 def get_info_extractor(self, ie_key):
729 """
730 Get an instance of an IE with name ie_key, it will try to get one from
731 the _ies list, if there's no instance it will create a new one and add
732 it to the extractor list.
733 """
734 ie = self._ies_instances.get(ie_key)
735 if ie is None:
736 ie = get_info_extractor(ie_key)()
737 self.add_info_extractor(ie)
738 return ie
739
023fa8c4
JMF
740 def add_default_info_extractors(self):
741 """
742 Add the InfoExtractors returned by gen_extractors to the end of the list
743 """
e52d7f85 744 for ie in gen_extractor_classes():
023fa8c4
JMF
745 self.add_info_extractor(ie)
746
56d868db 747 def add_post_processor(self, pp, when='post_process'):
8222d8de 748 """Add a PostProcessor object to the end of the chain."""
5bfa4862 749 self._pps[when].append(pp)
8222d8de
JMF
750 pp.set_downloader(self)
751
ab8e5e51
AM
752 def add_post_hook(self, ph):
753 """Add the post hook"""
754 self._post_hooks.append(ph)
755
933605d7 756 def add_progress_hook(self, ph):
819e0531 757 """Add the download progress hook"""
933605d7 758 self._progress_hooks.append(ph)
8ab470f1 759
819e0531 760 def add_postprocessor_hook(self, ph):
761 """Add the postprocessing progress hook"""
762 self._postprocessor_hooks.append(ph)
5bfc8bee 763 for pps in self._pps.values():
764 for pp in pps:
765 pp.add_progress_hook(ph)
819e0531 766
1c088fa8 767 def _bidi_workaround(self, message):
5d681e96 768 if not hasattr(self, '_output_channel'):
1c088fa8
PH
769 return message
770
5d681e96 771 assert hasattr(self, '_output_process')
11b85ce6 772 assert isinstance(message, compat_str)
6febd1c1 773 line_count = message.count('\n') + 1
0f06bcd7 774 self._output_process.stdin.write((message + '\n').encode())
5d681e96 775 self._output_process.stdin.flush()
0f06bcd7 776 res = ''.join(self._output_channel.readline().decode()
9e1a5b84 777 for _ in range(line_count))
6febd1c1 778 return res[:-len('\n')]
1c088fa8 779
b35496d8 780 def _write_string(self, message, out=None, only_once=False):
781 if only_once:
782 if message in self._printed_messages:
783 return
784 self._printed_messages.add(message)
785 write_string(message, out=out, encoding=self.params.get('encoding'))
734f90bb 786
cf4f42cb 787 def to_stdout(self, message, skip_eol=False, quiet=None):
0760b0a7 788 """Print message to stdout"""
cf4f42cb 789 if quiet is not None:
ae6a1b95 790 self.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument quiet. Use "YoutubeDL.to_screen" instead')
cf4f42cb 791 self._write_string(
792 '%s%s' % (self._bidi_workaround(message), ('' if skip_eol else '\n')),
591bb9d3 793 self._out_files.out)
cf4f42cb 794
795 def to_screen(self, message, skip_eol=False, quiet=None):
796 """Print message to screen if not in quiet mode"""
8bf9319e 797 if self.params.get('logger'):
43afe285 798 self.params['logger'].debug(message)
cf4f42cb 799 return
800 if (self.params.get('quiet') if quiet is None else quiet) and not self.params.get('verbose'):
801 return
802 self._write_string(
803 '%s%s' % (self._bidi_workaround(message), ('' if skip_eol else '\n')),
591bb9d3 804 self._out_files.screen)
8222d8de 805
b35496d8 806 def to_stderr(self, message, only_once=False):
0760b0a7 807 """Print message to stderr"""
11b85ce6 808 assert isinstance(message, compat_str)
8bf9319e 809 if self.params.get('logger'):
43afe285
IB
810 self.params['logger'].error(message)
811 else:
5792c950 812 self._write_string(f'{self._bidi_workaround(message)}\n', self._out_files.error, only_once=only_once)
cf4f42cb 813
814 def _send_console_code(self, code):
591bb9d3 815 if compat_os_name == 'nt' or not self._out_files.console:
cf4f42cb 816 return
591bb9d3 817 self._write_string(code, self._out_files.console)
8222d8de 818
1e5b9a95
PH
819 def to_console_title(self, message):
820 if not self.params.get('consoletitle', False):
821 return
3efb96a6 822 message = remove_terminal_sequences(message)
4bede0d8
C
823 if compat_os_name == 'nt':
824 if ctypes.windll.kernel32.GetConsoleWindow():
825 # c_wchar_p() might not be necessary if `message` is
826 # already of type unicode()
827 ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
cf4f42cb 828 else:
829 self._send_console_code(f'\033]0;{message}\007')
1e5b9a95 830
bdde425c 831 def save_console_title(self):
cf4f42cb 832 if not self.params.get('consoletitle') or self.params.get('simulate'):
bdde425c 833 return
592b7485 834 self._send_console_code('\033[22;0t') # Save the title on stack
bdde425c
PH
835
836 def restore_console_title(self):
cf4f42cb 837 if not self.params.get('consoletitle') or self.params.get('simulate'):
bdde425c 838 return
592b7485 839 self._send_console_code('\033[23;0t') # Restore the title from stack
bdde425c
PH
840
841 def __enter__(self):
842 self.save_console_title()
843 return self
844
845 def __exit__(self, *args):
846 self.restore_console_title()
f89197d7 847
dca08720 848 if self.params.get('cookiefile') is not None:
1bab3437 849 self.cookiejar.save(ignore_discard=True, ignore_expires=True)
bdde425c 850
fa9f30b8 851 def trouble(self, message=None, tb=None, is_error=True):
8222d8de
JMF
852 """Determine action to take when a download problem appears.
853
854 Depending on if the downloader has been configured to ignore
855 download errors or not, this method may throw an exception or
856 not when errors are found, after printing the message.
857
fa9f30b8 858 @param tb If given, is additional traceback information
859 @param is_error Whether to raise error according to ignorerrors
8222d8de
JMF
860 """
861 if message is not None:
862 self.to_stderr(message)
863 if self.params.get('verbose'):
864 if tb is None:
865 if sys.exc_info()[0]: # if .trouble has been called from an except block
6febd1c1 866 tb = ''
8222d8de 867 if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
6febd1c1 868 tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
c0384f22 869 tb += encode_compat_str(traceback.format_exc())
8222d8de
JMF
870 else:
871 tb_data = traceback.format_list(traceback.extract_stack())
6febd1c1 872 tb = ''.join(tb_data)
c19bc311 873 if tb:
874 self.to_stderr(tb)
fa9f30b8 875 if not is_error:
876 return
b1940459 877 if not self.params.get('ignoreerrors'):
8222d8de
JMF
878 if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
879 exc_info = sys.exc_info()[1].exc_info
880 else:
881 exc_info = sys.exc_info()
882 raise DownloadError(message, exc_info)
883 self._download_retcode = 1
884
19a03940 885 Styles = Namespace(
886 HEADERS='yellow',
887 EMPHASIS='light blue',
492272fe 888 FILENAME='green',
19a03940 889 ID='green',
890 DELIM='blue',
891 ERROR='red',
892 WARNING='yellow',
893 SUPPRESS='light black',
894 )
ec11a9f4 895
7578d77d 896 def _format_text(self, handle, allow_colors, text, f, fallback=None, *, test_encoding=False):
e5a998f3 897 text = str(text)
ec11a9f4 898 if test_encoding:
899 original_text = text
5c104538 900 # handle.encoding can be None. See https://github.com/yt-dlp/yt-dlp/issues/2711
901 encoding = self.params.get('encoding') or getattr(handle, 'encoding', None) or 'ascii'
ec11a9f4 902 text = text.encode(encoding, 'ignore').decode(encoding)
903 if fallback is not None and text != original_text:
904 text = fallback
7578d77d 905 return format_text(text, f) if allow_colors else text if fallback is None else fallback
ec11a9f4 906
591bb9d3 907 def _format_out(self, *args, **kwargs):
908 return self._format_text(self._out_files.out, self._allow_colors.out, *args, **kwargs)
909
ec11a9f4 910 def _format_screen(self, *args, **kwargs):
591bb9d3 911 return self._format_text(self._out_files.screen, self._allow_colors.screen, *args, **kwargs)
ec11a9f4 912
913 def _format_err(self, *args, **kwargs):
591bb9d3 914 return self._format_text(self._out_files.error, self._allow_colors.error, *args, **kwargs)
819e0531 915
c84aeac6 916 def report_warning(self, message, only_once=False):
8222d8de
JMF
917 '''
918 Print the message to stderr, it will be prefixed with 'WARNING:'
919 If stderr is a tty file the 'WARNING:' will be colored
920 '''
6d07ce01
JMF
921 if self.params.get('logger') is not None:
922 self.params['logger'].warning(message)
8222d8de 923 else:
ad8915b7
PH
924 if self.params.get('no_warnings'):
925 return
ec11a9f4 926 self.to_stderr(f'{self._format_err("WARNING:", self.Styles.WARNING)} {message}', only_once)
8222d8de 927
ee8dd27a 928 def deprecation_warning(self, message):
929 if self.params.get('logger') is not None:
a44ca5a4 930 self.params['logger'].warning(f'DeprecationWarning: {message}')
ee8dd27a 931 else:
932 self.to_stderr(f'{self._format_err("DeprecationWarning:", self.Styles.ERROR)} {message}', True)
933
fa9f30b8 934 def report_error(self, message, *args, **kwargs):
8222d8de
JMF
935 '''
936 Do the same as trouble, but prefixes the message with 'ERROR:', colored
937 in red if stderr is a tty file.
938 '''
fa9f30b8 939 self.trouble(f'{self._format_err("ERROR:", self.Styles.ERROR)} {message}', *args, **kwargs)
8222d8de 940
b35496d8 941 def write_debug(self, message, only_once=False):
0760b0a7 942 '''Log debug message or Print message to stderr'''
943 if not self.params.get('verbose', False):
944 return
945 message = '[debug] %s' % message
946 if self.params.get('logger'):
947 self.params['logger'].debug(message)
948 else:
b35496d8 949 self.to_stderr(message, only_once)
0760b0a7 950
8222d8de
JMF
951 def report_file_already_downloaded(self, file_name):
952 """Report file has already been fully downloaded."""
953 try:
6febd1c1 954 self.to_screen('[download] %s has already been downloaded' % file_name)
ce02ed60 955 except UnicodeEncodeError:
6febd1c1 956 self.to_screen('[download] The file has already been downloaded')
8222d8de 957
0c3d0f51 958 def report_file_delete(self, file_name):
959 """Report that existing file will be deleted."""
960 try:
c25228e5 961 self.to_screen('Deleting existing file %s' % file_name)
0c3d0f51 962 except UnicodeEncodeError:
c25228e5 963 self.to_screen('Deleting existing file')
0c3d0f51 964
319b6059 965 def raise_no_formats(self, info, forced=False, *, msg=None):
0a5a191a 966 has_drm = info.get('_has_drm')
319b6059 967 ignored, expected = self.params.get('ignore_no_formats_error'), bool(msg)
968 msg = msg or has_drm and 'This video is DRM protected' or 'No video formats found!'
969 if forced or not ignored:
1151c407 970 raise ExtractorError(msg, video_id=info['id'], ie=info['extractor'],
319b6059 971 expected=has_drm or ignored or expected)
88acdbc2 972 else:
973 self.report_warning(msg)
974
de6000d9 975 def parse_outtmpl(self):
976 outtmpl_dict = self.params.get('outtmpl', {})
977 if not isinstance(outtmpl_dict, dict):
978 outtmpl_dict = {'default': outtmpl_dict}
71ce444a 979 # Remove spaces in the default template
980 if self.params.get('restrictfilenames'):
981 sanitize = lambda x: x.replace(' - ', ' ').replace(' ', '-')
982 else:
983 sanitize = lambda x: x
de6000d9 984 outtmpl_dict.update({
71ce444a 985 k: sanitize(v) for k, v in DEFAULT_OUTTMPL.items()
80c03fa9 986 if outtmpl_dict.get(k) is None})
86e5f3ed 987 for _, val in outtmpl_dict.items():
de6000d9 988 if isinstance(val, bytes):
86e5f3ed 989 self.report_warning('Parameter outtmpl is bytes, but should be a unicode string')
de6000d9 990 return outtmpl_dict
991
21cd8fae 992 def get_output_path(self, dir_type='', filename=None):
993 paths = self.params.get('paths', {})
994 assert isinstance(paths, dict)
995 path = os.path.join(
996 expand_path(paths.get('home', '').strip()),
997 expand_path(paths.get(dir_type, '').strip()) if dir_type else '',
998 filename or '')
21cd8fae 999 return sanitize_path(path, force=self.params.get('windowsfilenames'))
1000
76a264ac 1001 @staticmethod
901130bb 1002 def _outtmpl_expandpath(outtmpl):
1003 # expand_path translates '%%' into '%' and '$$' into '$'
1004 # correspondingly that is not what we want since we need to keep
1005 # '%%' intact for template dict substitution step. Working around
1006 # with boundary-alike separator hack.
1007 sep = ''.join([random.choice(ascii_letters) for _ in range(32)])
86e5f3ed 1008 outtmpl = outtmpl.replace('%%', f'%{sep}%').replace('$$', f'${sep}$')
901130bb 1009
1010 # outtmpl should be expand_path'ed before template dict substitution
1011 # because meta fields may contain env variables we don't want to
1012 # be expanded. For example, for outtmpl "%(title)s.%(ext)s" and
1013 # title "Hello $PATH", we don't want `$PATH` to be expanded.
1014 return expand_path(outtmpl).replace(sep, '')
1015
1016 @staticmethod
1017 def escape_outtmpl(outtmpl):
1018 ''' Escape any remaining strings like %s, %abc% etc. '''
1019 return re.sub(
1020 STR_FORMAT_RE_TMPL.format('', '(?![%(\0])'),
1021 lambda mobj: ('' if mobj.group('has_key') else '%') + mobj.group(0),
1022 outtmpl)
1023
1024 @classmethod
1025 def validate_outtmpl(cls, outtmpl):
76a264ac 1026 ''' @return None or Exception object '''
7d1eb38a 1027 outtmpl = re.sub(
37893bb0 1028 STR_FORMAT_RE_TMPL.format('[^)]*', '[ljqBUDS]'),
7d1eb38a 1029 lambda mobj: f'{mobj.group(0)[:-1]}s',
1030 cls._outtmpl_expandpath(outtmpl))
76a264ac 1031 try:
7d1eb38a 1032 cls.escape_outtmpl(outtmpl) % collections.defaultdict(int)
76a264ac 1033 return None
1034 except ValueError as err:
1035 return err
1036
03b4de72 1037 @staticmethod
1038 def _copy_infodict(info_dict):
1039 info_dict = dict(info_dict)
09b49e1f 1040 info_dict.pop('__postprocessors', None)
415f8d51 1041 info_dict.pop('__pending_error', None)
03b4de72 1042 return info_dict
1043
e0fd9573 1044 def prepare_outtmpl(self, outtmpl, info_dict, sanitize=False):
1045 """ Make the outtmpl and info_dict suitable for substitution: ydl.escape_outtmpl(outtmpl) % info_dict
1046 @param sanitize Whether to sanitize the output as a filename.
1047 For backward compatibility, a function can also be passed
1048 """
1049
6e84b215 1050 info_dict.setdefault('epoch', int(time.time())) # keep epoch consistent once set
143db31d 1051
03b4de72 1052 info_dict = self._copy_infodict(info_dict)
752cda38 1053 info_dict['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
53c18592 1054 formatSeconds(info_dict['duration'], '-' if sanitize else ':')
143db31d 1055 if info_dict.get('duration', None) is not None
1056 else None)
1d485a1a 1057 info_dict['autonumber'] = int(self.params.get('autonumber_start', 1) - 1 + self._num_downloads)
9c906919 1058 info_dict['video_autonumber'] = self._num_videos
752cda38 1059 if info_dict.get('resolution') is None:
1060 info_dict['resolution'] = self.format_resolution(info_dict, default=None)
143db31d 1061
e6f21b3d 1062 # For fields playlist_index, playlist_autonumber and autonumber convert all occurrences
143db31d 1063 # of %(field)s to %(field)0Nd for backward compatibility
1064 field_size_compat_map = {
0a5a191a 1065 'playlist_index': number_of_digits(info_dict.get('__last_playlist_index') or 0),
ec11a9f4 1066 'playlist_autonumber': number_of_digits(info_dict.get('n_entries') or 0),
752cda38 1067 'autonumber': self.params.get('autonumber_size') or 5,
143db31d 1068 }
752cda38 1069
385a27fa 1070 TMPL_DICT = {}
37893bb0 1071 EXTERNAL_FORMAT_RE = re.compile(STR_FORMAT_RE_TMPL.format('[^)]*', f'[{STR_FORMAT_TYPES}ljqBUDS]'))
385a27fa 1072 MATH_FUNCTIONS = {
1073 '+': float.__add__,
1074 '-': float.__sub__,
1075 }
e625be0d 1076 # Field is of the form key1.key2...
1077 # where keys (except first) can be string, int or slice
2b8a2973 1078 FIELD_RE = r'\w*(?:\.(?:\w+|{num}|{num}?(?::{num}?){{1,2}}))*'.format(num=r'(?:-?\d+)')
1d485a1a 1079 MATH_FIELD_RE = rf'(?:{FIELD_RE}|-?{NUMBER_RE})'
385a27fa 1080 MATH_OPERATORS_RE = r'(?:%s)' % '|'.join(map(re.escape, MATH_FUNCTIONS.keys()))
1d485a1a 1081 INTERNAL_FORMAT_RE = re.compile(rf'''(?x)
e625be0d 1082 (?P<negate>-)?
1d485a1a 1083 (?P<fields>{FIELD_RE})
1084 (?P<maths>(?:{MATH_OPERATORS_RE}{MATH_FIELD_RE})*)
e625be0d 1085 (?:>(?P<strf_format>.+?))?
34baa9fd 1086 (?P<remaining>
1087 (?P<alternate>(?<!\\),[^|&)]+)?
1088 (?:&(?P<replacement>.*?))?
1089 (?:\|(?P<default>.*?))?
1d485a1a 1090 )$''')
752cda38 1091
2b8a2973 1092 def _traverse_infodict(k):
1093 k = k.split('.')
1094 if k[0] == '':
1095 k.pop(0)
1096 return traverse_obj(info_dict, k, is_user_input=True, traverse_string=True)
76a264ac 1097
752cda38 1098 def get_value(mdict):
1099 # Object traversal
2b8a2973 1100 value = _traverse_infodict(mdict['fields'])
752cda38 1101 # Negative
1102 if mdict['negate']:
1103 value = float_or_none(value)
1104 if value is not None:
1105 value *= -1
1106 # Do maths
385a27fa 1107 offset_key = mdict['maths']
1108 if offset_key:
752cda38 1109 value = float_or_none(value)
1110 operator = None
385a27fa 1111 while offset_key:
1112 item = re.match(
1113 MATH_FIELD_RE if operator else MATH_OPERATORS_RE,
1114 offset_key).group(0)
1115 offset_key = offset_key[len(item):]
1116 if operator is None:
752cda38 1117 operator = MATH_FUNCTIONS[item]
385a27fa 1118 continue
1119 item, multiplier = (item[1:], -1) if item[0] == '-' else (item, 1)
1120 offset = float_or_none(item)
1121 if offset is None:
2b8a2973 1122 offset = float_or_none(_traverse_infodict(item))
385a27fa 1123 try:
1124 value = operator(value, multiplier * offset)
1125 except (TypeError, ZeroDivisionError):
1126 return None
1127 operator = None
752cda38 1128 # Datetime formatting
1129 if mdict['strf_format']:
7c37ff97 1130 value = strftime_or_none(value, mdict['strf_format'].replace('\\,', ','))
752cda38 1131
1132 return value
1133
b868936c 1134 na = self.params.get('outtmpl_na_placeholder', 'NA')
1135
e0fd9573 1136 def filename_sanitizer(key, value, restricted=self.params.get('restrictfilenames')):
5c3895ff 1137 return sanitize_filename(str(value), restricted=restricted, is_id=(
1138 bool(re.search(r'(^|[_.])id(\.|$)', key))
1139 if 'filename-sanitization' in self.params.get('compat_opts', [])
1140 else NO_DEFAULT))
e0fd9573 1141
1142 sanitizer = sanitize if callable(sanitize) else filename_sanitizer
1143 sanitize = bool(sanitize)
1144
6e84b215 1145 def _dumpjson_default(obj):
1146 if isinstance(obj, (set, LazyList)):
1147 return list(obj)
adbc4ec4 1148 return repr(obj)
6e84b215 1149
752cda38 1150 def create_key(outer_mobj):
1151 if not outer_mobj.group('has_key'):
b836dc94 1152 return outer_mobj.group(0)
752cda38 1153 key = outer_mobj.group('key')
752cda38 1154 mobj = re.match(INTERNAL_FORMAT_RE, key)
e0fd9573 1155 initial_field = mobj.group('fields') if mobj else ''
e978789f 1156 value, replacement, default = None, None, na
7c37ff97 1157 while mobj:
e625be0d 1158 mobj = mobj.groupdict()
7c37ff97 1159 default = mobj['default'] if mobj['default'] is not None else default
752cda38 1160 value = get_value(mobj)
e978789f 1161 replacement = mobj['replacement']
7c37ff97 1162 if value is None and mobj['alternate']:
34baa9fd 1163 mobj = re.match(INTERNAL_FORMAT_RE, mobj['remaining'][1:])
7c37ff97 1164 else:
1165 break
752cda38 1166
b868936c 1167 fmt = outer_mobj.group('format')
752cda38 1168 if fmt == 's' and value is not None and key in field_size_compat_map.keys():
86e5f3ed 1169 fmt = f'0{field_size_compat_map[key]:d}d'
752cda38 1170
e978789f 1171 value = default if value is None else value if replacement is None else replacement
752cda38 1172
4476d2c7 1173 flags = outer_mobj.group('conversion') or ''
7d1eb38a 1174 str_fmt = f'{fmt[:-1]}s'
524e2e4f 1175 if fmt[-1] == 'l': # list
4476d2c7 1176 delim = '\n' if '#' in flags else ', '
9e907ebd 1177 value, fmt = delim.join(map(str, variadic(value, allowed_types=(str, bytes)))), str_fmt
524e2e4f 1178 elif fmt[-1] == 'j': # json
4476d2c7 1179 value, fmt = json.dumps(value, default=_dumpjson_default, indent=4 if '#' in flags else None), str_fmt
524e2e4f 1180 elif fmt[-1] == 'q': # quoted
4476d2c7 1181 value = map(str, variadic(value) if '#' in flags else [value])
1182 value, fmt = ' '.join(map(compat_shlex_quote, value)), str_fmt
524e2e4f 1183 elif fmt[-1] == 'B': # bytes
0f06bcd7 1184 value = f'%{str_fmt}'.encode() % str(value).encode()
f5aa5cfb 1185 value, fmt = value.decode('utf-8', 'ignore'), 's'
524e2e4f 1186 elif fmt[-1] == 'U': # unicode normalized
524e2e4f 1187 value, fmt = unicodedata.normalize(
1188 # "+" = compatibility equivalence, "#" = NFD
4476d2c7 1189 'NF%s%s' % ('K' if '+' in flags else '', 'D' if '#' in flags else 'C'),
524e2e4f 1190 value), str_fmt
e0fd9573 1191 elif fmt[-1] == 'D': # decimal suffix
abbeeebc 1192 num_fmt, fmt = fmt[:-1].replace('#', ''), 's'
1193 value = format_decimal_suffix(value, f'%{num_fmt}f%s' if num_fmt else '%d%s',
1194 factor=1024 if '#' in flags else 1000)
37893bb0 1195 elif fmt[-1] == 'S': # filename sanitization
e0fd9573 1196 value, fmt = filename_sanitizer(initial_field, value, restricted='#' in flags), str_fmt
7d1eb38a 1197 elif fmt[-1] == 'c':
524e2e4f 1198 if value:
1199 value = str(value)[0]
76a264ac 1200 else:
524e2e4f 1201 fmt = str_fmt
76a264ac 1202 elif fmt[-1] not in 'rs': # numeric
a439a3a4 1203 value = float_or_none(value)
752cda38 1204 if value is None:
1205 value, fmt = default, 's'
901130bb 1206
752cda38 1207 if sanitize:
1208 if fmt[-1] == 'r':
1209 # If value is an object, sanitize might convert it to a string
1210 # So we convert it to repr first
7d1eb38a 1211 value, fmt = repr(value), str_fmt
639f1cea 1212 if fmt[-1] in 'csr':
e0fd9573 1213 value = sanitizer(initial_field, value)
901130bb 1214
b868936c 1215 key = '%s\0%s' % (key.replace('%', '%\0'), outer_mobj.group('format'))
385a27fa 1216 TMPL_DICT[key] = value
b868936c 1217 return '{prefix}%({key}){fmt}'.format(key=key, fmt=fmt, prefix=outer_mobj.group('prefix'))
752cda38 1218
385a27fa 1219 return EXTERNAL_FORMAT_RE.sub(create_key, outtmpl), TMPL_DICT
143db31d 1220
819e0531 1221 def evaluate_outtmpl(self, outtmpl, info_dict, *args, **kwargs):
1222 outtmpl, info_dict = self.prepare_outtmpl(outtmpl, info_dict, *args, **kwargs)
1223 return self.escape_outtmpl(outtmpl) % info_dict
1224
5127e92a 1225 def _prepare_filename(self, info_dict, *, outtmpl=None, tmpl_type=None):
1226 assert None in (outtmpl, tmpl_type), 'outtmpl and tmpl_type are mutually exclusive'
1227 if outtmpl is None:
1228 outtmpl = self.outtmpl_dict.get(tmpl_type or 'default', self.outtmpl_dict['default'])
8222d8de 1229 try:
5127e92a 1230 outtmpl = self._outtmpl_expandpath(outtmpl)
e0fd9573 1231 filename = self.evaluate_outtmpl(outtmpl, info_dict, True)
6a0546e3 1232 if not filename:
1233 return None
15da37c7 1234
5127e92a 1235 if tmpl_type in ('', 'temp'):
6a0546e3 1236 final_ext, ext = self.params.get('final_ext'), info_dict.get('ext')
1237 if final_ext and ext and final_ext != ext and filename.endswith(f'.{final_ext}'):
1238 filename = replace_extension(filename, ext, final_ext)
5127e92a 1239 elif tmpl_type:
6a0546e3 1240 force_ext = OUTTMPL_TYPES[tmpl_type]
1241 if force_ext:
1242 filename = replace_extension(filename, force_ext, info_dict.get('ext'))
de6000d9 1243
bdc3fd2f
U
1244 # https://github.com/blackjack4494/youtube-dlc/issues/85
1245 trim_file_name = self.params.get('trim_file_name', False)
1246 if trim_file_name:
5c22c63d 1247 no_ext, *ext = filename.rsplit('.', 2)
1248 filename = join_nonempty(no_ext[:trim_file_name], *ext, delim='.')
bdc3fd2f 1249
0202b52a 1250 return filename
8222d8de 1251 except ValueError as err:
6febd1c1 1252 self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')')
8222d8de
JMF
1253 return None
1254
5127e92a 1255 def prepare_filename(self, info_dict, dir_type='', *, outtmpl=None, warn=False):
1256 """Generate the output filename"""
1257 if outtmpl:
1258 assert not dir_type, 'outtmpl and dir_type are mutually exclusive'
1259 dir_type = None
1260 filename = self._prepare_filename(info_dict, tmpl_type=dir_type, outtmpl=outtmpl)
80c03fa9 1261 if not filename and dir_type not in ('', 'temp'):
1262 return ''
de6000d9 1263
c84aeac6 1264 if warn:
21cd8fae 1265 if not self.params.get('paths'):
de6000d9 1266 pass
1267 elif filename == '-':
c84aeac6 1268 self.report_warning('--paths is ignored when an outputting to stdout', only_once=True)
de6000d9 1269 elif os.path.isabs(filename):
c84aeac6 1270 self.report_warning('--paths is ignored since an absolute path is given in output template', only_once=True)
de6000d9 1271 if filename == '-' or not filename:
1272 return filename
1273
21cd8fae 1274 return self.get_output_path(dir_type, filename)
0202b52a 1275
120fe513 1276 def _match_entry(self, info_dict, incomplete=False, silent=False):
ecdec191 1277 """ Returns None if the file should be downloaded """
8222d8de 1278
c77495e3 1279 video_title = info_dict.get('title', info_dict.get('id', 'video'))
1280
8b0d7497 1281 def check_filter():
8b0d7497 1282 if 'title' in info_dict:
1283 # This can happen when we're just evaluating the playlist
1284 title = info_dict['title']
1285 matchtitle = self.params.get('matchtitle', False)
1286 if matchtitle:
1287 if not re.search(matchtitle, title, re.IGNORECASE):
1288 return '"' + title + '" title did not match pattern "' + matchtitle + '"'
1289 rejecttitle = self.params.get('rejecttitle', False)
1290 if rejecttitle:
1291 if re.search(rejecttitle, title, re.IGNORECASE):
1292 return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
1293 date = info_dict.get('upload_date')
1294 if date is not None:
1295 dateRange = self.params.get('daterange', DateRange())
1296 if date not in dateRange:
86e5f3ed 1297 return f'{date_from_str(date).isoformat()} upload date is not in range {dateRange}'
8b0d7497 1298 view_count = info_dict.get('view_count')
1299 if view_count is not None:
1300 min_views = self.params.get('min_views')
1301 if min_views is not None and view_count < min_views:
1302 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
1303 max_views = self.params.get('max_views')
1304 if max_views is not None and view_count > max_views:
1305 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
1306 if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')):
1307 return 'Skipping "%s" because it is age restricted' % video_title
8b0d7497 1308
8f18aca8 1309 match_filter = self.params.get('match_filter')
1310 if match_filter is not None:
1311 try:
1312 ret = match_filter(info_dict, incomplete=incomplete)
1313 except TypeError:
1314 # For backward compatibility
1315 ret = None if incomplete else match_filter(info_dict)
492272fe 1316 if ret is NO_DEFAULT:
1317 while True:
1318 filename = self._format_screen(self.prepare_filename(info_dict), self.Styles.FILENAME)
1319 reply = input(self._format_screen(
1320 f'Download "{filename}"? (Y/n): ', self.Styles.EMPHASIS)).lower().strip()
1321 if reply in {'y', ''}:
1322 return None
1323 elif reply == 'n':
1324 return f'Skipping {video_title}'
492272fe 1325 elif ret is not None:
8f18aca8 1326 return ret
8b0d7497 1327 return None
1328
c77495e3 1329 if self.in_download_archive(info_dict):
1330 reason = '%s has already been recorded in the archive' % video_title
1331 break_opt, break_err = 'break_on_existing', ExistingVideoReached
1332 else:
1333 reason = check_filter()
1334 break_opt, break_err = 'break_on_reject', RejectedVideoReached
8b0d7497 1335 if reason is not None:
120fe513 1336 if not silent:
1337 self.to_screen('[download] ' + reason)
c77495e3 1338 if self.params.get(break_opt, False):
1339 raise break_err()
8b0d7497 1340 return reason
fe7e0c98 1341
b6c45014
JMF
1342 @staticmethod
1343 def add_extra_info(info_dict, extra_info):
1344 '''Set the keys from extra_info in info dict if they are missing'''
1345 for key, value in extra_info.items():
1346 info_dict.setdefault(key, value)
1347
409e1828 1348 def extract_info(self, url, download=True, ie_key=None, extra_info=None,
61aa5ba3 1349 process=True, force_generic_extractor=False):
41d1cca3 1350 """
1351 Return a list with a dictionary for each video extracted.
1352
1353 Arguments:
1354 url -- URL to extract
1355
1356 Keyword arguments:
1357 download -- whether to download videos during extraction
1358 ie_key -- extractor key hint
1359 extra_info -- dictionary containing the extra values to add to each result
1360 process -- whether to resolve all unresolved references (URLs, playlist items),
1361 must be True for download to work.
1362 force_generic_extractor -- force using the generic extractor
1363 """
fe7e0c98 1364
409e1828 1365 if extra_info is None:
1366 extra_info = {}
1367
61aa5ba3 1368 if not ie_key and force_generic_extractor:
d22dec74
S
1369 ie_key = 'Generic'
1370
8222d8de 1371 if ie_key:
8b7491c8 1372 ies = {ie_key: self._get_info_extractor_class(ie_key)}
8222d8de
JMF
1373 else:
1374 ies = self._ies
1375
8b7491c8 1376 for ie_key, ie in ies.items():
8222d8de
JMF
1377 if not ie.suitable(url):
1378 continue
1379
1380 if not ie.working():
6febd1c1
PH
1381 self.report_warning('The program functionality for this site has been marked as broken, '
1382 'and will probably not work.')
8222d8de 1383
1151c407 1384 temp_id = ie.get_temp_id(url)
a0566bbf 1385 if temp_id is not None and self.in_download_archive({'id': temp_id, 'ie_key': ie_key}):
5e5be0c0 1386 self.to_screen(f'[{ie_key}] {temp_id}: has already been recorded in the archive')
1387 if self.params.get('break_on_existing', False):
1388 raise ExistingVideoReached()
a0566bbf 1389 break
8b7491c8 1390 return self.__extract_info(url, self.get_info_extractor(ie_key), download, extra_info, process)
a0566bbf 1391 else:
1392 self.report_error('no suitable InfoExtractor for URL %s' % url)
1393
8e5fecc8 1394 def __handle_extraction_exceptions(func):
b5ae35ee 1395 @functools.wraps(func)
a0566bbf 1396 def wrapper(self, *args, **kwargs):
6da22e7d 1397 while True:
1398 try:
1399 return func(self, *args, **kwargs)
1400 except (DownloadCancelled, LazyList.IndexError, PagedList.IndexError):
8222d8de 1401 raise
6da22e7d 1402 except ReExtractInfo as e:
1403 if e.expected:
1404 self.to_screen(f'{e}; Re-extracting data')
1405 else:
1406 self.to_stderr('\r')
1407 self.report_warning(f'{e}; Re-extracting data')
1408 continue
1409 except GeoRestrictedError as e:
1410 msg = e.msg
1411 if e.countries:
1412 msg += '\nThis video is available in %s.' % ', '.join(
1413 map(ISO3166Utils.short2full, e.countries))
1414 msg += '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1415 self.report_error(msg)
1416 except ExtractorError as e: # An error we somewhat expected
1417 self.report_error(str(e), e.format_traceback())
1418 except Exception as e:
1419 if self.params.get('ignoreerrors'):
1420 self.report_error(str(e), tb=encode_compat_str(traceback.format_exc()))
1421 else:
1422 raise
1423 break
a0566bbf 1424 return wrapper
1425
f2ebc5c7 1426 def _wait_for_video(self, ie_result):
1427 if (not self.params.get('wait_for_video')
1428 or ie_result.get('_type', 'video') != 'video'
1429 or ie_result.get('formats') or ie_result.get('url')):
1430 return
1431
1432 format_dur = lambda dur: '%02d:%02d:%02d' % timetuple_from_msec(dur * 1000)[:-1]
1433 last_msg = ''
1434
1435 def progress(msg):
1436 nonlocal last_msg
1437 self.to_screen(msg + ' ' * (len(last_msg) - len(msg)) + '\r', skip_eol=True)
1438 last_msg = msg
1439
1440 min_wait, max_wait = self.params.get('wait_for_video')
1441 diff = try_get(ie_result, lambda x: x['release_timestamp'] - time.time())
1442 if diff is None and ie_result.get('live_status') == 'is_upcoming':
16c620bc 1443 diff = round(random.uniform(min_wait, max_wait) if (max_wait and min_wait) else (max_wait or min_wait), 0)
f2ebc5c7 1444 self.report_warning('Release time of video is not known')
1445 elif (diff or 0) <= 0:
1446 self.report_warning('Video should already be available according to extracted info')
38d79fd1 1447 diff = min(max(diff or 0, min_wait or 0), max_wait or float('inf'))
f2ebc5c7 1448 self.to_screen(f'[wait] Waiting for {format_dur(diff)} - Press Ctrl+C to try now')
1449
1450 wait_till = time.time() + diff
1451 try:
1452 while True:
1453 diff = wait_till - time.time()
1454 if diff <= 0:
1455 progress('')
1456 raise ReExtractInfo('[wait] Wait period ended', expected=True)
1457 progress(f'[wait] Remaining time until next attempt: {self._format_screen(format_dur(diff), self.Styles.EMPHASIS)}')
1458 time.sleep(1)
1459 except KeyboardInterrupt:
1460 progress('')
1461 raise ReExtractInfo('[wait] Interrupted by user', expected=True)
1462 except BaseException as e:
1463 if not isinstance(e, ReExtractInfo):
1464 self.to_screen('')
1465 raise
1466
a0566bbf 1467 @__handle_extraction_exceptions
58f197b7 1468 def __extract_info(self, url, ie, download, extra_info, process):
a0566bbf 1469 ie_result = ie.extract(url)
1470 if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
1471 return
1472 if isinstance(ie_result, list):
1473 # Backwards compatibility: old IE result format
1474 ie_result = {
1475 '_type': 'compat_list',
1476 'entries': ie_result,
1477 }
e37d0efb 1478 if extra_info.get('original_url'):
1479 ie_result.setdefault('original_url', extra_info['original_url'])
a0566bbf 1480 self.add_default_extra_info(ie_result, ie, url)
1481 if process:
f2ebc5c7 1482 self._wait_for_video(ie_result)
a0566bbf 1483 return self.process_ie_result(ie_result, download, extra_info)
8222d8de 1484 else:
a0566bbf 1485 return ie_result
fe7e0c98 1486
ea38e55f 1487 def add_default_extra_info(self, ie_result, ie, url):
6033d980 1488 if url is not None:
1489 self.add_extra_info(ie_result, {
1490 'webpage_url': url,
1491 'original_url': url,
57ebfca3 1492 })
1493 webpage_url = ie_result.get('webpage_url')
1494 if webpage_url:
1495 self.add_extra_info(ie_result, {
1496 'webpage_url_basename': url_basename(webpage_url),
1497 'webpage_url_domain': get_domain(webpage_url),
6033d980 1498 })
1499 if ie is not None:
1500 self.add_extra_info(ie_result, {
1501 'extractor': ie.IE_NAME,
1502 'extractor_key': ie.ie_key(),
1503 })
ea38e55f 1504
58adec46 1505 def process_ie_result(self, ie_result, download=True, extra_info=None):
8222d8de
JMF
1506 """
1507 Take the result of the ie(may be modified) and resolve all unresolved
1508 references (URLs, playlist items).
1509
1510 It will also download the videos if 'download'.
1511 Returns the resolved ie_result.
1512 """
58adec46 1513 if extra_info is None:
1514 extra_info = {}
e8ee972c
PH
1515 result_type = ie_result.get('_type', 'video')
1516
057a5206 1517 if result_type in ('url', 'url_transparent'):
134c6ea8 1518 ie_result['url'] = sanitize_url(ie_result['url'])
e37d0efb 1519 if ie_result.get('original_url'):
1520 extra_info.setdefault('original_url', ie_result['original_url'])
1521
057a5206 1522 extract_flat = self.params.get('extract_flat', False)
3089bc74
S
1523 if ((extract_flat == 'in_playlist' and 'playlist' in extra_info)
1524 or extract_flat is True):
ecb54191 1525 info_copy = ie_result.copy()
6033d980 1526 ie = try_get(ie_result.get('ie_key'), self.get_info_extractor)
360167b9 1527 if ie and not ie_result.get('id'):
4614bc22 1528 info_copy['id'] = ie.get_temp_id(ie_result['url'])
6033d980 1529 self.add_default_extra_info(info_copy, ie, ie_result['url'])
4614bc22 1530 self.add_extra_info(info_copy, extra_info)
b5475f11 1531 info_copy, _ = self.pre_process(info_copy)
ecb54191 1532 self.__forced_printings(info_copy, self.prepare_filename(info_copy), incomplete=True)
415f8d51 1533 self._raise_pending_errors(info_copy)
4614bc22 1534 if self.params.get('force_write_download_archive', False):
1535 self.record_download_archive(info_copy)
e8ee972c
PH
1536 return ie_result
1537
8222d8de 1538 if result_type == 'video':
b6c45014 1539 self.add_extra_info(ie_result, extra_info)
9c2b75b5 1540 ie_result = self.process_video_result(ie_result, download=download)
415f8d51 1541 self._raise_pending_errors(ie_result)
28b0eb0f 1542 additional_urls = (ie_result or {}).get('additional_urls')
9c2b75b5 1543 if additional_urls:
e9f4ccd1 1544 # TODO: Improve MetadataParserPP to allow setting a list
9c2b75b5 1545 if isinstance(additional_urls, compat_str):
1546 additional_urls = [additional_urls]
1547 self.to_screen(
1548 '[info] %s: %d additional URL(s) requested' % (ie_result['id'], len(additional_urls)))
1549 self.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls))
1550 ie_result['additional_entries'] = [
1551 self.extract_info(
b69fd25c 1552 url, download, extra_info=extra_info,
9c2b75b5 1553 force_generic_extractor=self.params.get('force_generic_extractor'))
1554 for url in additional_urls
1555 ]
1556 return ie_result
8222d8de
JMF
1557 elif result_type == 'url':
1558 # We have to add extra_info to the results because it may be
1559 # contained in a playlist
07cce701 1560 return self.extract_info(
1561 ie_result['url'], download,
1562 ie_key=ie_result.get('ie_key'),
1563 extra_info=extra_info)
7fc3fa05
PH
1564 elif result_type == 'url_transparent':
1565 # Use the information from the embedding page
1566 info = self.extract_info(
1567 ie_result['url'], ie_key=ie_result.get('ie_key'),
1568 extra_info=extra_info, download=False, process=False)
1569
1640eb09
S
1570 # extract_info may return None when ignoreerrors is enabled and
1571 # extraction failed with an error, don't crash and return early
1572 # in this case
1573 if not info:
1574 return info
1575
412c617d 1576 new_result = info.copy()
90137ca4 1577 new_result.update(filter_dict(ie_result, lambda k, v: (
1578 v is not None and k not in {'_type', 'url', 'id', 'extractor', 'extractor_key', 'ie_key'})))
7fc3fa05 1579
0563f7ac
S
1580 # Extracted info may not be a video result (i.e.
1581 # info.get('_type', 'video') != video) but rather an url or
1582 # url_transparent. In such cases outer metadata (from ie_result)
1583 # should be propagated to inner one (info). For this to happen
1584 # _type of info should be overridden with url_transparent. This
067aa17e 1585 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
0563f7ac
S
1586 if new_result.get('_type') == 'url':
1587 new_result['_type'] = 'url_transparent'
7fc3fa05
PH
1588
1589 return self.process_ie_result(
1590 new_result, download=download, extra_info=extra_info)
40fcba5e 1591 elif result_type in ('playlist', 'multi_video'):
30a074c2 1592 # Protect from infinite recursion due to recursively nested playlists
1593 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
1594 webpage_url = ie_result['webpage_url']
1595 if webpage_url in self._playlist_urls:
7e85e872 1596 self.to_screen(
30a074c2 1597 '[download] Skipping already downloaded playlist: %s'
1598 % ie_result.get('title') or ie_result.get('id'))
1599 return
7e85e872 1600
30a074c2 1601 self._playlist_level += 1
1602 self._playlist_urls.add(webpage_url)
03f83004 1603 self._fill_common_fields(ie_result, False)
bc516a3f 1604 self._sanitize_thumbnails(ie_result)
30a074c2 1605 try:
1606 return self.__process_playlist(ie_result, download)
1607 finally:
1608 self._playlist_level -= 1
1609 if not self._playlist_level:
1610 self._playlist_urls.clear()
8222d8de 1611 elif result_type == 'compat_list':
c9bf4114
PH
1612 self.report_warning(
1613 'Extractor %s returned a compat_list result. '
1614 'It needs to be updated.' % ie_result.get('extractor'))
5f6a1245 1615
8222d8de 1616 def _fixup(r):
b868936c 1617 self.add_extra_info(r, {
1618 'extractor': ie_result['extractor'],
1619 'webpage_url': ie_result['webpage_url'],
1620 'webpage_url_basename': url_basename(ie_result['webpage_url']),
0bb322b9 1621 'webpage_url_domain': get_domain(ie_result['webpage_url']),
b868936c 1622 'extractor_key': ie_result['extractor_key'],
1623 })
8222d8de
JMF
1624 return r
1625 ie_result['entries'] = [
b6c45014 1626 self.process_ie_result(_fixup(r), download, extra_info)
8222d8de
JMF
1627 for r in ie_result['entries']
1628 ]
1629 return ie_result
1630 else:
1631 raise Exception('Invalid result type: %s' % result_type)
1632
e92caff5 1633 def _ensure_dir_exists(self, path):
1634 return make_dir(path, self.report_error)
1635
3b603dbd 1636 @staticmethod
1637 def _playlist_infodict(ie_result, **kwargs):
1638 return {
1639 **ie_result,
1640 'playlist': ie_result.get('title') or ie_result.get('id'),
1641 'playlist_id': ie_result.get('id'),
1642 'playlist_title': ie_result.get('title'),
1643 'playlist_uploader': ie_result.get('uploader'),
1644 'playlist_uploader_id': ie_result.get('uploader_id'),
1645 'playlist_index': 0,
1646 **kwargs,
1647 }
1648
30a074c2 1649 def __process_playlist(self, ie_result, download):
1650 # We process each entry in the playlist
1651 playlist = ie_result.get('title') or ie_result.get('id')
1652 self.to_screen('[download] Downloading playlist: %s' % playlist)
1653
498f5606 1654 if 'entries' not in ie_result:
aa9369a2 1655 raise EntryNotInPlaylist('There are no entries')
7c7f7161 1656
1657 MissingEntry = object()
498f5606 1658 incomplete_entries = bool(ie_result.get('requested_entries'))
1659 if incomplete_entries:
bf5f605e 1660 def fill_missing_entries(entries, indices):
7c7f7161 1661 ret = [MissingEntry] * max(indices)
bf5f605e 1662 for i, entry in zip(indices, entries):
498f5606 1663 ret[i - 1] = entry
1664 return ret
1665 ie_result['entries'] = fill_missing_entries(ie_result['entries'], ie_result['requested_entries'])
02fd60d3 1666
30a074c2 1667 playlist_results = []
1668
56a8fb4f 1669 playliststart = self.params.get('playliststart', 1)
30a074c2 1670 playlistend = self.params.get('playlistend')
1671 # For backwards compatibility, interpret -1 as whole list
1672 if playlistend == -1:
1673 playlistend = None
1674
1675 playlistitems_str = self.params.get('playlist_items')
1676 playlistitems = None
1677 if playlistitems_str is not None:
1678 def iter_playlistitems(format):
1679 for string_segment in format.split(','):
1680 if '-' in string_segment:
1681 start, end = string_segment.split('-')
1682 for item in range(int(start), int(end) + 1):
1683 yield int(item)
1684 else:
1685 yield int(string_segment)
1686 playlistitems = orderedSet(iter_playlistitems(playlistitems_str))
1687
1688 ie_entries = ie_result['entries']
8e5fecc8 1689 if isinstance(ie_entries, list):
ed8d87f9 1690 playlist_count = len(ie_entries)
f0d785d3 1691 msg = f'Collected {playlist_count} videos; downloading %d of them'
1692 ie_result['playlist_count'] = ie_result.get('playlist_count') or playlist_count
1693
8e5fecc8 1694 def get_entry(i):
1695 return ie_entries[i - 1]
1696 else:
f0d785d3 1697 msg = 'Downloading %d videos'
c586f9e8 1698 if not isinstance(ie_entries, (PagedList, LazyList)):
8e5fecc8 1699 ie_entries = LazyList(ie_entries)
d37707bd 1700 elif isinstance(ie_entries, InAdvancePagedList):
1701 if ie_entries._pagesize == 1:
1702 playlist_count = ie_entries._pagecount
8e5fecc8 1703
1704 def get_entry(i):
1705 return YoutubeDL.__handle_extraction_exceptions(
1706 lambda self, i: ie_entries[i - 1]
1707 )(self, i)
50fed816 1708
f0d785d3 1709 entries, broken = [], False
ff1c7fc9 1710 items = playlistitems if playlistitems is not None else itertools.count(playliststart)
1711 for i in items:
1712 if i == 0:
1713 continue
56a8fb4f 1714 if playlistitems is None and playlistend is not None and playlistend < i:
1715 break
1716 entry = None
1717 try:
50fed816 1718 entry = get_entry(i)
7c7f7161 1719 if entry is MissingEntry:
498f5606 1720 raise EntryNotInPlaylist()
56a8fb4f 1721 except (IndexError, EntryNotInPlaylist):
1722 if incomplete_entries:
aa9369a2 1723 raise EntryNotInPlaylist(f'Entry {i} cannot be found')
56a8fb4f 1724 elif not playlistitems:
1725 break
1726 entries.append(entry)
120fe513 1727 try:
1728 if entry is not None:
e5a998f3 1729 # TODO: Add auto-generated fields
120fe513 1730 self._match_entry(entry, incomplete=True, silent=True)
1731 except (ExistingVideoReached, RejectedVideoReached):
f0d785d3 1732 broken = True
120fe513 1733 break
56a8fb4f 1734 ie_result['entries'] = entries
30a074c2 1735
56a8fb4f 1736 # Save playlist_index before re-ordering
1737 entries = [
9e598870 1738 ((playlistitems[i - 1] if playlistitems else i + playliststart - 1), entry)
56a8fb4f 1739 for i, entry in enumerate(entries, 1)
1740 if entry is not None]
1741 n_entries = len(entries)
498f5606 1742
f0d785d3 1743 if not (ie_result.get('playlist_count') or broken or playlistitems or playlistend):
1744 ie_result['playlist_count'] = n_entries
1745
e08a85d8 1746 if not playlistitems and (playliststart != 1 or playlistend):
56a8fb4f 1747 playlistitems = list(range(playliststart, playliststart + n_entries))
498f5606 1748 ie_result['requested_entries'] = playlistitems
1749
e08a85d8 1750 _infojson_written = False
0bfc53d0 1751 write_playlist_files = self.params.get('allow_playlist_files', True)
1752 if write_playlist_files and self.params.get('list_thumbnails'):
1753 self.list_thumbnails(ie_result)
1754 if write_playlist_files and not self.params.get('simulate'):
3b603dbd 1755 ie_copy = self._playlist_infodict(ie_result, n_entries=n_entries)
e08a85d8 1756 _infojson_written = self._write_info_json(
1757 'playlist', ie_result, self.prepare_filename(ie_copy, 'pl_infojson'))
1758 if _infojson_written is None:
80c03fa9 1759 return
1760 if self._write_description('playlist', ie_result,
1761 self.prepare_filename(ie_copy, 'pl_description')) is None:
1762 return
681de68e 1763 # TODO: This should be passed to ThumbnailsConvertor if necessary
80c03fa9 1764 self._write_thumbnails('playlist', ie_copy, self.prepare_filename(ie_copy, 'pl_thumbnail'))
30a074c2 1765
1766 if self.params.get('playlistreverse', False):
1767 entries = entries[::-1]
30a074c2 1768 if self.params.get('playlistrandom', False):
1769 random.shuffle(entries)
1770
1771 x_forwarded_for = ie_result.get('__x_forwarded_for_ip')
1772
86e5f3ed 1773 self.to_screen(f'[{ie_result["extractor"]}] playlist {playlist}: {msg % n_entries}')
26e2805c 1774 failures = 0
1775 max_failures = self.params.get('skip_playlist_after_errors') or float('inf')
71729754 1776 for i, entry_tuple in enumerate(entries, 1):
1777 playlist_index, entry = entry_tuple
81139999 1778 if 'playlist-index' in self.params.get('compat_opts', []):
1779 playlist_index = playlistitems[i - 1] if playlistitems else i + playliststart - 1
19a03940 1780 self.to_screen('[download] Downloading video %s of %s' % (
1781 self._format_screen(i, self.Styles.ID), self._format_screen(n_entries, self.Styles.EMPHASIS)))
30a074c2 1782 # This __x_forwarded_for_ip thing is a bit ugly but requires
1783 # minimal changes
1784 if x_forwarded_for:
1785 entry['__x_forwarded_for_ip'] = x_forwarded_for
1786 extra = {
1787 'n_entries': n_entries,
0a5a191a 1788 '__last_playlist_index': max(playlistitems) if playlistitems else (playlistend or n_entries),
f0d785d3 1789 'playlist_count': ie_result.get('playlist_count'),
71729754 1790 'playlist_index': playlist_index,
1791 'playlist_autonumber': i,
30a074c2 1792 'playlist': playlist,
1793 'playlist_id': ie_result.get('id'),
1794 'playlist_title': ie_result.get('title'),
1795 'playlist_uploader': ie_result.get('uploader'),
1796 'playlist_uploader_id': ie_result.get('uploader_id'),
30a074c2 1797 'extractor': ie_result['extractor'],
1798 'webpage_url': ie_result['webpage_url'],
1799 'webpage_url_basename': url_basename(ie_result['webpage_url']),
0bb322b9 1800 'webpage_url_domain': get_domain(ie_result['webpage_url']),
30a074c2 1801 'extractor_key': ie_result['extractor_key'],
1802 }
1803
1804 if self._match_entry(entry, incomplete=True) is not None:
1805 continue
1806
1807 entry_result = self.__process_iterable_entry(entry, download, extra)
26e2805c 1808 if not entry_result:
1809 failures += 1
1810 if failures >= max_failures:
1811 self.report_error(
1812 'Skipping the remaining entries in playlist "%s" since %d items failed extraction' % (playlist, failures))
1813 break
30a074c2 1814 playlist_results.append(entry_result)
1815 ie_result['entries'] = playlist_results
e08a85d8 1816
1817 # Write the updated info to json
cb96c5be 1818 if _infojson_written is True and self._write_info_json(
e08a85d8 1819 'updated playlist', ie_result,
1820 self.prepare_filename(ie_copy, 'pl_infojson'), overwrite=True) is None:
1821 return
ca30f449 1822
ed5835b4 1823 ie_result = self.run_all_pps('playlist', ie_result)
1824 self.to_screen(f'[download] Finished downloading playlist: {playlist}')
30a074c2 1825 return ie_result
1826
a0566bbf 1827 @__handle_extraction_exceptions
1828 def __process_iterable_entry(self, entry, download, extra_info):
1829 return self.process_ie_result(
1830 entry, download=download, extra_info=extra_info)
1831
67134eab
JMF
1832 def _build_format_filter(self, filter_spec):
1833 " Returns a function to filter the formats according to the filter_spec "
083c9df9
PH
1834
1835 OPERATORS = {
1836 '<': operator.lt,
1837 '<=': operator.le,
1838 '>': operator.gt,
1839 '>=': operator.ge,
1840 '=': operator.eq,
1841 '!=': operator.ne,
1842 }
67134eab 1843 operator_rex = re.compile(r'''(?x)\s*
187986a8 1844 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)\s*
1845 (?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1846 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)\s*
083c9df9 1847 ''' % '|'.join(map(re.escape, OPERATORS.keys())))
187986a8 1848 m = operator_rex.fullmatch(filter_spec)
9ddb6925
S
1849 if m:
1850 try:
1851 comparison_value = int(m.group('value'))
1852 except ValueError:
1853 comparison_value = parse_filesize(m.group('value'))
1854 if comparison_value is None:
1855 comparison_value = parse_filesize(m.group('value') + 'B')
1856 if comparison_value is None:
1857 raise ValueError(
1858 'Invalid value %r in format specification %r' % (
67134eab 1859 m.group('value'), filter_spec))
9ddb6925
S
1860 op = OPERATORS[m.group('op')]
1861
083c9df9 1862 if not m:
9ddb6925
S
1863 STR_OPERATORS = {
1864 '=': operator.eq,
10d33b34
YCH
1865 '^=': lambda attr, value: attr.startswith(value),
1866 '$=': lambda attr, value: attr.endswith(value),
1867 '*=': lambda attr, value: value in attr,
1ce9a3cb 1868 '~=': lambda attr, value: value.search(attr) is not None
9ddb6925 1869 }
187986a8 1870 str_operator_rex = re.compile(r'''(?x)\s*
1871 (?P<key>[a-zA-Z0-9._-]+)\s*
1ce9a3cb
LF
1872 (?P<negation>!\s*)?(?P<op>%s)\s*(?P<none_inclusive>\?\s*)?
1873 (?P<quote>["'])?
1874 (?P<value>(?(quote)(?:(?!(?P=quote))[^\\]|\\.)+|[\w.-]+))
1875 (?(quote)(?P=quote))\s*
9ddb6925 1876 ''' % '|'.join(map(re.escape, STR_OPERATORS.keys())))
187986a8 1877 m = str_operator_rex.fullmatch(filter_spec)
9ddb6925 1878 if m:
1ce9a3cb
LF
1879 if m.group('op') == '~=':
1880 comparison_value = re.compile(m.group('value'))
1881 else:
1882 comparison_value = re.sub(r'''\\([\\"'])''', r'\1', m.group('value'))
2cc779f4
S
1883 str_op = STR_OPERATORS[m.group('op')]
1884 if m.group('negation'):
e118a879 1885 op = lambda attr, value: not str_op(attr, value)
2cc779f4
S
1886 else:
1887 op = str_op
083c9df9 1888
9ddb6925 1889 if not m:
187986a8 1890 raise SyntaxError('Invalid filter specification %r' % filter_spec)
083c9df9
PH
1891
1892 def _filter(f):
1893 actual_value = f.get(m.group('key'))
1894 if actual_value is None:
1895 return m.group('none_inclusive')
1896 return op(actual_value, comparison_value)
67134eab
JMF
1897 return _filter
1898
9f1a1c36 1899 def _check_formats(self, formats):
1900 for f in formats:
1901 self.to_screen('[info] Testing format %s' % f['format_id'])
75689fe5 1902 path = self.get_output_path('temp')
1903 if not self._ensure_dir_exists(f'{path}/'):
1904 continue
1905 temp_file = tempfile.NamedTemporaryFile(suffix='.tmp', delete=False, dir=path or None)
9f1a1c36 1906 temp_file.close()
1907 try:
1908 success, _ = self.dl(temp_file.name, f, test=True)
1909 except (DownloadError, IOError, OSError, ValueError) + network_exceptions:
1910 success = False
1911 finally:
1912 if os.path.exists(temp_file.name):
1913 try:
1914 os.remove(temp_file.name)
1915 except OSError:
1916 self.report_warning('Unable to delete temporary file "%s"' % temp_file.name)
1917 if success:
1918 yield f
1919 else:
1920 self.to_screen('[info] Unable to download format %s. Skipping...' % f['format_id'])
1921
0017d9ad 1922 def _default_format_spec(self, info_dict, download=True):
0017d9ad 1923
af0f7428
S
1924 def can_merge():
1925 merger = FFmpegMergerPP(self)
1926 return merger.available and merger.can_merge()
1927
91ebc640 1928 prefer_best = (
b7b04c78 1929 not self.params.get('simulate')
91ebc640 1930 and download
1931 and (
1932 not can_merge()
21633673 1933 or info_dict.get('is_live') and not self.params.get('live_from_start')
de6000d9 1934 or self.outtmpl_dict['default'] == '-'))
53ed7066 1935 compat = (
1936 prefer_best
1937 or self.params.get('allow_multiple_audio_streams', False)
1938 or 'format-spec' in self.params.get('compat_opts', []))
91ebc640 1939
1940 return (
53ed7066 1941 'best/bestvideo+bestaudio' if prefer_best
1942 else 'bestvideo*+bestaudio/best' if not compat
91ebc640 1943 else 'bestvideo+bestaudio/best')
0017d9ad 1944
67134eab
JMF
1945 def build_format_selector(self, format_spec):
1946 def syntax_error(note, start):
1947 message = (
1948 'Invalid format specification: '
86e5f3ed 1949 '{}\n\t{}\n\t{}^'.format(note, format_spec, ' ' * start[1]))
67134eab
JMF
1950 return SyntaxError(message)
1951
1952 PICKFIRST = 'PICKFIRST'
1953 MERGE = 'MERGE'
1954 SINGLE = 'SINGLE'
0130afb7 1955 GROUP = 'GROUP'
67134eab
JMF
1956 FormatSelector = collections.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
1957
91ebc640 1958 allow_multiple_streams = {'audio': self.params.get('allow_multiple_audio_streams', False),
1959 'video': self.params.get('allow_multiple_video_streams', False)}
909d24dd 1960
9f1a1c36 1961 check_formats = self.params.get('check_formats') == 'selected'
e8e73840 1962
67134eab
JMF
1963 def _parse_filter(tokens):
1964 filter_parts = []
1965 for type, string, start, _, _ in tokens:
1966 if type == tokenize.OP and string == ']':
1967 return ''.join(filter_parts)
1968 else:
1969 filter_parts.append(string)
1970
232541df 1971 def _remove_unused_ops(tokens):
17cc1534 1972 # Remove operators that we don't use and join them with the surrounding strings
232541df
JMF
1973 # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
1974 ALLOWED_OPS = ('/', '+', ',', '(', ')')
1975 last_string, last_start, last_end, last_line = None, None, None, None
1976 for type, string, start, end, line in tokens:
1977 if type == tokenize.OP and string == '[':
1978 if last_string:
1979 yield tokenize.NAME, last_string, last_start, last_end, last_line
1980 last_string = None
1981 yield type, string, start, end, line
1982 # everything inside brackets will be handled by _parse_filter
1983 for type, string, start, end, line in tokens:
1984 yield type, string, start, end, line
1985 if type == tokenize.OP and string == ']':
1986 break
1987 elif type == tokenize.OP and string in ALLOWED_OPS:
1988 if last_string:
1989 yield tokenize.NAME, last_string, last_start, last_end, last_line
1990 last_string = None
1991 yield type, string, start, end, line
1992 elif type in [tokenize.NAME, tokenize.NUMBER, tokenize.OP]:
1993 if not last_string:
1994 last_string = string
1995 last_start = start
1996 last_end = end
1997 else:
1998 last_string += string
1999 if last_string:
2000 yield tokenize.NAME, last_string, last_start, last_end, last_line
2001
cf2ac6df 2002 def _parse_format_selection(tokens, inside_merge=False, inside_choice=False, inside_group=False):
67134eab
JMF
2003 selectors = []
2004 current_selector = None
2005 for type, string, start, _, _ in tokens:
2006 # ENCODING is only defined in python 3.x
2007 if type == getattr(tokenize, 'ENCODING', None):
2008 continue
2009 elif type in [tokenize.NAME, tokenize.NUMBER]:
2010 current_selector = FormatSelector(SINGLE, string, [])
2011 elif type == tokenize.OP:
cf2ac6df
JMF
2012 if string == ')':
2013 if not inside_group:
2014 # ')' will be handled by the parentheses group
2015 tokens.restore_last_token()
67134eab 2016 break
cf2ac6df 2017 elif inside_merge and string in ['/', ',']:
0130afb7
JMF
2018 tokens.restore_last_token()
2019 break
cf2ac6df
JMF
2020 elif inside_choice and string == ',':
2021 tokens.restore_last_token()
2022 break
2023 elif string == ',':
0a31a350
JMF
2024 if not current_selector:
2025 raise syntax_error('"," must follow a format selector', start)
67134eab
JMF
2026 selectors.append(current_selector)
2027 current_selector = None
2028 elif string == '/':
d96d604e
JMF
2029 if not current_selector:
2030 raise syntax_error('"/" must follow a format selector', start)
67134eab 2031 first_choice = current_selector
cf2ac6df 2032 second_choice = _parse_format_selection(tokens, inside_choice=True)
f5f4a27a 2033 current_selector = FormatSelector(PICKFIRST, (first_choice, second_choice), [])
67134eab
JMF
2034 elif string == '[':
2035 if not current_selector:
2036 current_selector = FormatSelector(SINGLE, 'best', [])
2037 format_filter = _parse_filter(tokens)
2038 current_selector.filters.append(format_filter)
0130afb7
JMF
2039 elif string == '(':
2040 if current_selector:
2041 raise syntax_error('Unexpected "("', start)
cf2ac6df
JMF
2042 group = _parse_format_selection(tokens, inside_group=True)
2043 current_selector = FormatSelector(GROUP, group, [])
67134eab 2044 elif string == '+':
d03cfdce 2045 if not current_selector:
2046 raise syntax_error('Unexpected "+"', start)
2047 selector_1 = current_selector
2048 selector_2 = _parse_format_selection(tokens, inside_merge=True)
2049 if not selector_2:
2050 raise syntax_error('Expected a selector', start)
2051 current_selector = FormatSelector(MERGE, (selector_1, selector_2), [])
67134eab 2052 else:
86e5f3ed 2053 raise syntax_error(f'Operator not recognized: "{string}"', start)
67134eab
JMF
2054 elif type == tokenize.ENDMARKER:
2055 break
2056 if current_selector:
2057 selectors.append(current_selector)
2058 return selectors
2059
f8d4ad9a 2060 def _merge(formats_pair):
2061 format_1, format_2 = formats_pair
2062
2063 formats_info = []
2064 formats_info.extend(format_1.get('requested_formats', (format_1,)))
2065 formats_info.extend(format_2.get('requested_formats', (format_2,)))
2066
2067 if not allow_multiple_streams['video'] or not allow_multiple_streams['audio']:
551f9388 2068 get_no_more = {'video': False, 'audio': False}
f8d4ad9a 2069 for (i, fmt_info) in enumerate(formats_info):
551f9388 2070 if fmt_info.get('acodec') == fmt_info.get('vcodec') == 'none':
2071 formats_info.pop(i)
2072 continue
2073 for aud_vid in ['audio', 'video']:
f8d4ad9a 2074 if not allow_multiple_streams[aud_vid] and fmt_info.get(aud_vid[0] + 'codec') != 'none':
2075 if get_no_more[aud_vid]:
2076 formats_info.pop(i)
f5510afe 2077 break
f8d4ad9a 2078 get_no_more[aud_vid] = True
2079
2080 if len(formats_info) == 1:
2081 return formats_info[0]
2082
2083 video_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('vcodec') != 'none']
2084 audio_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('acodec') != 'none']
2085
2086 the_only_video = video_fmts[0] if len(video_fmts) == 1 else None
2087 the_only_audio = audio_fmts[0] if len(audio_fmts) == 1 else None
2088
2089 output_ext = self.params.get('merge_output_format')
2090 if not output_ext:
2091 if the_only_video:
2092 output_ext = the_only_video['ext']
2093 elif the_only_audio and not video_fmts:
2094 output_ext = the_only_audio['ext']
2095 else:
2096 output_ext = 'mkv'
2097
975a0d0d 2098 filtered = lambda *keys: filter(None, (traverse_obj(fmt, *keys) for fmt in formats_info))
2099
f8d4ad9a 2100 new_dict = {
2101 'requested_formats': formats_info,
975a0d0d 2102 'format': '+'.join(filtered('format')),
2103 'format_id': '+'.join(filtered('format_id')),
f8d4ad9a 2104 'ext': output_ext,
975a0d0d 2105 'protocol': '+'.join(map(determine_protocol, formats_info)),
093a1710 2106 'language': '+'.join(orderedSet(filtered('language'))) or None,
2107 'format_note': '+'.join(orderedSet(filtered('format_note'))) or None,
2108 'filesize_approx': sum(filtered('filesize', 'filesize_approx')) or None,
975a0d0d 2109 'tbr': sum(filtered('tbr', 'vbr', 'abr')),
f8d4ad9a 2110 }
2111
2112 if the_only_video:
2113 new_dict.update({
2114 'width': the_only_video.get('width'),
2115 'height': the_only_video.get('height'),
2116 'resolution': the_only_video.get('resolution') or self.format_resolution(the_only_video),
2117 'fps': the_only_video.get('fps'),
49a57e70 2118 'dynamic_range': the_only_video.get('dynamic_range'),
f8d4ad9a 2119 'vcodec': the_only_video.get('vcodec'),
2120 'vbr': the_only_video.get('vbr'),
2121 'stretched_ratio': the_only_video.get('stretched_ratio'),
2122 })
2123
2124 if the_only_audio:
2125 new_dict.update({
2126 'acodec': the_only_audio.get('acodec'),
2127 'abr': the_only_audio.get('abr'),
975a0d0d 2128 'asr': the_only_audio.get('asr'),
f8d4ad9a 2129 })
2130
2131 return new_dict
2132
e8e73840 2133 def _check_formats(formats):
981052c9 2134 if not check_formats:
2135 yield from formats
b5ac45b1 2136 return
9f1a1c36 2137 yield from self._check_formats(formats)
e8e73840 2138
67134eab 2139 def _build_selector_function(selector):
909d24dd 2140 if isinstance(selector, list): # ,
67134eab
JMF
2141 fs = [_build_selector_function(s) for s in selector]
2142
317f7ab6 2143 def selector_function(ctx):
67134eab 2144 for f in fs:
981052c9 2145 yield from f(ctx)
67134eab 2146 return selector_function
909d24dd 2147
2148 elif selector.type == GROUP: # ()
0130afb7 2149 selector_function = _build_selector_function(selector.selector)
909d24dd 2150
2151 elif selector.type == PICKFIRST: # /
67134eab
JMF
2152 fs = [_build_selector_function(s) for s in selector.selector]
2153
317f7ab6 2154 def selector_function(ctx):
67134eab 2155 for f in fs:
317f7ab6 2156 picked_formats = list(f(ctx))
67134eab
JMF
2157 if picked_formats:
2158 return picked_formats
2159 return []
67134eab 2160
981052c9 2161 elif selector.type == MERGE: # +
2162 selector_1, selector_2 = map(_build_selector_function, selector.selector)
2163
2164 def selector_function(ctx):
adbc4ec4 2165 for pair in itertools.product(selector_1(ctx), selector_2(ctx)):
981052c9 2166 yield _merge(pair)
2167
909d24dd 2168 elif selector.type == SINGLE: # atom
598d185d 2169 format_spec = selector.selector or 'best'
909d24dd 2170
f8d4ad9a 2171 # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
909d24dd 2172 if format_spec == 'all':
2173 def selector_function(ctx):
9222c381 2174 yield from _check_formats(ctx['formats'][::-1])
f8d4ad9a 2175 elif format_spec == 'mergeall':
2176 def selector_function(ctx):
316f2650 2177 formats = list(_check_formats(
2178 f for f in ctx['formats'] if f.get('vcodec') != 'none' or f.get('acodec') != 'none'))
e01d6aa4 2179 if not formats:
2180 return
921b76ca 2181 merged_format = formats[-1]
2182 for f in formats[-2::-1]:
f8d4ad9a 2183 merged_format = _merge((merged_format, f))
2184 yield merged_format
909d24dd 2185
2186 else:
85e801a9 2187 format_fallback, seperate_fallback, format_reverse, format_idx = False, None, True, 1
eff63539 2188 mobj = re.match(
2189 r'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
2190 format_spec)
2191 if mobj is not None:
2192 format_idx = int_or_none(mobj.group('n'), default=1)
e8e73840 2193 format_reverse = mobj.group('bw')[0] == 'b'
eff63539 2194 format_type = (mobj.group('type') or [None])[0]
2195 not_format_type = {'v': 'a', 'a': 'v'}.get(format_type)
2196 format_modified = mobj.group('mod') is not None
909d24dd 2197
2198 format_fallback = not format_type and not format_modified # for b, w
8326b00a 2199 _filter_f = (
eff63539 2200 (lambda f: f.get('%scodec' % format_type) != 'none')
2201 if format_type and format_modified # bv*, ba*, wv*, wa*
2202 else (lambda f: f.get('%scodec' % not_format_type) == 'none')
2203 if format_type # bv, ba, wv, wa
2204 else (lambda f: f.get('vcodec') != 'none' and f.get('acodec') != 'none')
2205 if not format_modified # b, w
8326b00a 2206 else lambda f: True) # b*, w*
2207 filter_f = lambda f: _filter_f(f) and (
2208 f.get('vcodec') != 'none' or f.get('acodec') != 'none')
67134eab 2209 else:
48ee10ee 2210 if format_spec in self._format_selection_exts['audio']:
b11c04a8 2211 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none'
48ee10ee 2212 elif format_spec in self._format_selection_exts['video']:
b11c04a8 2213 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none' and f.get('vcodec') != 'none'
85e801a9 2214 seperate_fallback = lambda f: f.get('ext') == format_spec and f.get('vcodec') != 'none'
48ee10ee 2215 elif format_spec in self._format_selection_exts['storyboards']:
b11c04a8 2216 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') == 'none' and f.get('vcodec') == 'none'
2217 else:
b5ae35ee 2218 filter_f = lambda f: f.get('format_id') == format_spec # id
909d24dd 2219
2220 def selector_function(ctx):
2221 formats = list(ctx['formats'])
909d24dd 2222 matches = list(filter(filter_f, formats)) if filter_f is not None else formats
85e801a9 2223 if not matches:
2224 if format_fallback and ctx['incomplete_formats']:
2225 # for extractors with incomplete formats (audio only (soundcloud)
2226 # or video only (imgur)) best/worst will fallback to
2227 # best/worst {video,audio}-only format
2228 matches = formats
2229 elif seperate_fallback and not ctx['has_merged_format']:
2230 # for compatibility with youtube-dl when there is no pre-merged format
2231 matches = list(filter(seperate_fallback, formats))
981052c9 2232 matches = LazyList(_check_formats(matches[::-1 if format_reverse else 1]))
2233 try:
e8e73840 2234 yield matches[format_idx - 1]
4abea8ca 2235 except LazyList.IndexError:
981052c9 2236 return
083c9df9 2237
67134eab 2238 filters = [self._build_format_filter(f) for f in selector.filters]
083c9df9 2239
317f7ab6 2240 def final_selector(ctx):
adbc4ec4 2241 ctx_copy = dict(ctx)
67134eab 2242 for _filter in filters:
317f7ab6
S
2243 ctx_copy['formats'] = list(filter(_filter, ctx_copy['formats']))
2244 return selector_function(ctx_copy)
67134eab 2245 return final_selector
083c9df9 2246
0f06bcd7 2247 stream = io.BytesIO(format_spec.encode())
0130afb7 2248 try:
f9934b96 2249 tokens = list(_remove_unused_ops(tokenize.tokenize(stream.readline)))
0130afb7
JMF
2250 except tokenize.TokenError:
2251 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec)))
2252
86e5f3ed 2253 class TokenIterator:
0130afb7
JMF
2254 def __init__(self, tokens):
2255 self.tokens = tokens
2256 self.counter = 0
2257
2258 def __iter__(self):
2259 return self
2260
2261 def __next__(self):
2262 if self.counter >= len(self.tokens):
2263 raise StopIteration()
2264 value = self.tokens[self.counter]
2265 self.counter += 1
2266 return value
2267
2268 next = __next__
2269
2270 def restore_last_token(self):
2271 self.counter -= 1
2272
2273 parsed_selector = _parse_format_selection(iter(TokenIterator(tokens)))
67134eab 2274 return _build_selector_function(parsed_selector)
a9c58ad9 2275
e5660ee6 2276 def _calc_headers(self, info_dict):
8b7539d2 2277 res = merge_headers(self.params['http_headers'], info_dict.get('http_headers') or {})
e5660ee6 2278
c487cf00 2279 cookies = self._calc_cookies(info_dict['url'])
e5660ee6
JMF
2280 if cookies:
2281 res['Cookie'] = cookies
2282
0016b84e
S
2283 if 'X-Forwarded-For' not in res:
2284 x_forwarded_for_ip = info_dict.get('__x_forwarded_for_ip')
2285 if x_forwarded_for_ip:
2286 res['X-Forwarded-For'] = x_forwarded_for_ip
2287
e5660ee6
JMF
2288 return res
2289
c487cf00 2290 def _calc_cookies(self, url):
2291 pr = sanitized_Request(url)
e5660ee6 2292 self.cookiejar.add_cookie_header(pr)
662435f7 2293 return pr.get_header('Cookie')
e5660ee6 2294
9f1a1c36 2295 def _sort_thumbnails(self, thumbnails):
2296 thumbnails.sort(key=lambda t: (
2297 t.get('preference') if t.get('preference') is not None else -1,
2298 t.get('width') if t.get('width') is not None else -1,
2299 t.get('height') if t.get('height') is not None else -1,
2300 t.get('id') if t.get('id') is not None else '',
2301 t.get('url')))
2302
b0249bca 2303 def _sanitize_thumbnails(self, info_dict):
bc516a3f 2304 thumbnails = info_dict.get('thumbnails')
2305 if thumbnails is None:
2306 thumbnail = info_dict.get('thumbnail')
2307 if thumbnail:
2308 info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}]
9f1a1c36 2309 if not thumbnails:
2310 return
2311
2312 def check_thumbnails(thumbnails):
2313 for t in thumbnails:
2314 self.to_screen(f'[info] Testing thumbnail {t["id"]}')
2315 try:
2316 self.urlopen(HEADRequest(t['url']))
2317 except network_exceptions as err:
2318 self.to_screen(f'[info] Unable to connect to thumbnail {t["id"]} URL {t["url"]!r} - {err}. Skipping...')
2319 continue
2320 yield t
2321
2322 self._sort_thumbnails(thumbnails)
2323 for i, t in enumerate(thumbnails):
2324 if t.get('id') is None:
2325 t['id'] = '%d' % i
2326 if t.get('width') and t.get('height'):
2327 t['resolution'] = '%dx%d' % (t['width'], t['height'])
2328 t['url'] = sanitize_url(t['url'])
2329
2330 if self.params.get('check_formats') is True:
282f5709 2331 info_dict['thumbnails'] = LazyList(check_thumbnails(thumbnails[::-1]), reverse=True)
9f1a1c36 2332 else:
2333 info_dict['thumbnails'] = thumbnails
bc516a3f 2334
03f83004
LNO
2335 def _fill_common_fields(self, info_dict, is_video=True):
2336 # TODO: move sanitization here
2337 if is_video:
2338 # playlists are allowed to lack "title"
d4736fdb 2339 title = info_dict.get('title', NO_DEFAULT)
2340 if title is NO_DEFAULT:
03f83004
LNO
2341 raise ExtractorError('Missing "title" field in extractor result',
2342 video_id=info_dict['id'], ie=info_dict['extractor'])
d4736fdb 2343 info_dict['fulltitle'] = title
2344 if not title:
2345 if title == '':
2346 self.write_debug('Extractor gave empty title. Creating a generic title')
2347 else:
2348 self.report_warning('Extractor failed to obtain "title". Creating a generic title instead')
1d485a1a 2349 info_dict['title'] = f'{info_dict["extractor"].replace(":", "-")} video #{info_dict["id"]}'
03f83004
LNO
2350
2351 if info_dict.get('duration') is not None:
2352 info_dict['duration_string'] = formatSeconds(info_dict['duration'])
2353
2354 for ts_key, date_key in (
2355 ('timestamp', 'upload_date'),
2356 ('release_timestamp', 'release_date'),
2357 ('modified_timestamp', 'modified_date'),
2358 ):
2359 if info_dict.get(date_key) is None and info_dict.get(ts_key) is not None:
2360 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
2361 # see http://bugs.python.org/issue1646728)
19a03940 2362 with contextlib.suppress(ValueError, OverflowError, OSError):
03f83004
LNO
2363 upload_date = datetime.datetime.utcfromtimestamp(info_dict[ts_key])
2364 info_dict[date_key] = upload_date.strftime('%Y%m%d')
03f83004
LNO
2365
2366 live_keys = ('is_live', 'was_live')
2367 live_status = info_dict.get('live_status')
2368 if live_status is None:
2369 for key in live_keys:
2370 if info_dict.get(key) is False:
2371 continue
2372 if info_dict.get(key):
2373 live_status = key
2374 break
2375 if all(info_dict.get(key) is False for key in live_keys):
2376 live_status = 'not_live'
2377 if live_status:
2378 info_dict['live_status'] = live_status
2379 for key in live_keys:
2380 if info_dict.get(key) is None:
2381 info_dict[key] = (live_status == key)
2382
2383 # Auto generate title fields corresponding to the *_number fields when missing
2384 # in order to always have clean titles. This is very common for TV series.
2385 for field in ('chapter', 'season', 'episode'):
2386 if info_dict.get('%s_number' % field) is not None and not info_dict.get(field):
2387 info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field])
2388
415f8d51 2389 def _raise_pending_errors(self, info):
2390 err = info.pop('__pending_error', None)
2391 if err:
2392 self.report_error(err, tb=False)
2393
dd82ffea
JMF
2394 def process_video_result(self, info_dict, download=True):
2395 assert info_dict.get('_type', 'video') == 'video'
9c906919 2396 self._num_videos += 1
dd82ffea 2397
bec1fad2 2398 if 'id' not in info_dict:
fc08bdd6 2399 raise ExtractorError('Missing "id" field in extractor result', ie=info_dict['extractor'])
2400 elif not info_dict.get('id'):
2401 raise ExtractorError('Extractor failed to obtain "id"', ie=info_dict['extractor'])
455a15e2 2402
c9969434
S
2403 def report_force_conversion(field, field_not, conversion):
2404 self.report_warning(
2405 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
2406 % (field, field_not, conversion))
2407
2408 def sanitize_string_field(info, string_field):
2409 field = info.get(string_field)
2410 if field is None or isinstance(field, compat_str):
2411 return
2412 report_force_conversion(string_field, 'a string', 'string')
2413 info[string_field] = compat_str(field)
2414
2415 def sanitize_numeric_fields(info):
2416 for numeric_field in self._NUMERIC_FIELDS:
2417 field = info.get(numeric_field)
f9934b96 2418 if field is None or isinstance(field, (int, float)):
c9969434
S
2419 continue
2420 report_force_conversion(numeric_field, 'numeric', 'int')
2421 info[numeric_field] = int_or_none(field)
2422
2423 sanitize_string_field(info_dict, 'id')
2424 sanitize_numeric_fields(info_dict)
4c3f8c3f 2425 if (info_dict.get('duration') or 0) <= 0 and info_dict.pop('duration', None):
50e93e03 2426 self.report_warning('"duration" field is negative, there is an error in extractor')
be6217b2 2427
dd82ffea
JMF
2428 if 'playlist' not in info_dict:
2429 # It isn't part of a playlist
2430 info_dict['playlist'] = None
2431 info_dict['playlist_index'] = None
2432
bc516a3f 2433 self._sanitize_thumbnails(info_dict)
d5519808 2434
536a55da 2435 thumbnail = info_dict.get('thumbnail')
bc516a3f 2436 thumbnails = info_dict.get('thumbnails')
536a55da
S
2437 if thumbnail:
2438 info_dict['thumbnail'] = sanitize_url(thumbnail)
2439 elif thumbnails:
d5519808
PH
2440 info_dict['thumbnail'] = thumbnails[-1]['url']
2441
ae30b840 2442 if info_dict.get('display_id') is None and 'id' in info_dict:
0afef30b
PH
2443 info_dict['display_id'] = info_dict['id']
2444
03f83004 2445 self._fill_common_fields(info_dict)
33d2fc2f 2446
05108a49
S
2447 for cc_kind in ('subtitles', 'automatic_captions'):
2448 cc = info_dict.get(cc_kind)
2449 if cc:
2450 for _, subtitle in cc.items():
2451 for subtitle_format in subtitle:
2452 if subtitle_format.get('url'):
2453 subtitle_format['url'] = sanitize_url(subtitle_format['url'])
2454 if subtitle_format.get('ext') is None:
2455 subtitle_format['ext'] = determine_ext(subtitle_format['url']).lower()
2456
2457 automatic_captions = info_dict.get('automatic_captions')
4bba3716 2458 subtitles = info_dict.get('subtitles')
4bba3716 2459
360e1ca5 2460 info_dict['requested_subtitles'] = self.process_subtitles(
05108a49 2461 info_dict['id'], subtitles, automatic_captions)
a504ced0 2462
dd82ffea
JMF
2463 if info_dict.get('formats') is None:
2464 # There's only one format available
2465 formats = [info_dict]
2466 else:
2467 formats = info_dict['formats']
2468
0a5a191a 2469 # or None ensures --clean-infojson removes it
2470 info_dict['_has_drm'] = any(f.get('has_drm') for f in formats) or None
88acdbc2 2471 if not self.params.get('allow_unplayable_formats'):
2472 formats = [f for f in formats if not f.get('has_drm')]
0a5a191a 2473 if info_dict['_has_drm'] and all(
c0b6e5c7 2474 f.get('acodec') == f.get('vcodec') == 'none' for f in formats):
2475 self.report_warning(
2476 'This video is DRM protected and only images are available for download. '
2477 'Use --list-formats to see them')
88acdbc2 2478
319b6059 2479 get_from_start = not info_dict.get('is_live') or bool(self.params.get('live_from_start'))
2480 if not get_from_start:
2481 info_dict['title'] += ' ' + datetime.datetime.now().strftime('%Y-%m-%d %H:%M')
2482 if info_dict.get('is_live') and formats:
adbc4ec4 2483 formats = [f for f in formats if bool(f.get('is_from_start')) == get_from_start]
319b6059 2484 if get_from_start and not formats:
a44ca5a4 2485 self.raise_no_formats(info_dict, msg=(
2486 '--live-from-start is passed, but there are no formats that can be downloaded from the start. '
2487 'If you want to download from the current time, use --no-live-from-start'))
adbc4ec4 2488
db95dc13 2489 if not formats:
1151c407 2490 self.raise_no_formats(info_dict)
db95dc13 2491
73af5cc8
S
2492 def is_wellformed(f):
2493 url = f.get('url')
a5ac0c47 2494 if not url:
73af5cc8
S
2495 self.report_warning(
2496 '"url" field is missing or empty - skipping format, '
2497 'there is an error in extractor')
a5ac0c47
S
2498 return False
2499 if isinstance(url, bytes):
2500 sanitize_string_field(f, 'url')
2501 return True
73af5cc8
S
2502
2503 # Filter out malformed formats for better extraction robustness
2504 formats = list(filter(is_wellformed, formats))
2505
181c7053
S
2506 formats_dict = {}
2507
dd82ffea 2508 # We check that all the formats have the format and format_id fields
db95dc13 2509 for i, format in enumerate(formats):
c9969434
S
2510 sanitize_string_field(format, 'format_id')
2511 sanitize_numeric_fields(format)
dcf77cf1 2512 format['url'] = sanitize_url(format['url'])
e74e3b63 2513 if not format.get('format_id'):
8016c922 2514 format['format_id'] = compat_str(i)
e2effb08
S
2515 else:
2516 # Sanitize format_id from characters used in format selector expression
ec85ded8 2517 format['format_id'] = re.sub(r'[\s,/+\[\]()]', '_', format['format_id'])
181c7053
S
2518 format_id = format['format_id']
2519 if format_id not in formats_dict:
2520 formats_dict[format_id] = []
2521 formats_dict[format_id].append(format)
2522
2523 # Make sure all formats have unique format_id
03b4de72 2524 common_exts = set(itertools.chain(*self._format_selection_exts.values()))
181c7053 2525 for format_id, ambiguous_formats in formats_dict.items():
48ee10ee 2526 ambigious_id = len(ambiguous_formats) > 1
2527 for i, format in enumerate(ambiguous_formats):
2528 if ambigious_id:
181c7053 2529 format['format_id'] = '%s-%d' % (format_id, i)
48ee10ee 2530 if format.get('ext') is None:
2531 format['ext'] = determine_ext(format['url']).lower()
2532 # Ensure there is no conflict between id and ext in format selection
2533 # See https://github.com/yt-dlp/yt-dlp/issues/1282
2534 if format['format_id'] != format['ext'] and format['format_id'] in common_exts:
2535 format['format_id'] = 'f%s' % format['format_id']
181c7053
S
2536
2537 for i, format in enumerate(formats):
8c51aa65 2538 if format.get('format') is None:
6febd1c1 2539 format['format'] = '{id} - {res}{note}'.format(
8c51aa65
JMF
2540 id=format['format_id'],
2541 res=self.format_resolution(format),
b868936c 2542 note=format_field(format, 'format_note', ' (%s)'),
8c51aa65 2543 )
6f0be937 2544 if format.get('protocol') is None:
b5559424 2545 format['protocol'] = determine_protocol(format)
239df021 2546 if format.get('resolution') is None:
2547 format['resolution'] = self.format_resolution(format, default=None)
176f1866 2548 if format.get('dynamic_range') is None and format.get('vcodec') != 'none':
2549 format['dynamic_range'] = 'SDR'
f2fe69c7 2550 if (info_dict.get('duration') and format.get('tbr')
2551 and not format.get('filesize') and not format.get('filesize_approx')):
2552 format['filesize_approx'] = info_dict['duration'] * format['tbr'] * (1024 / 8)
2553
e5660ee6
JMF
2554 # Add HTTP headers, so that external programs can use them from the
2555 # json output
2556 full_format_info = info_dict.copy()
2557 full_format_info.update(format)
2558 format['http_headers'] = self._calc_headers(full_format_info)
0016b84e
S
2559 # Remove private housekeeping stuff
2560 if '__x_forwarded_for_ip' in info_dict:
2561 del info_dict['__x_forwarded_for_ip']
dd82ffea 2562
9f1a1c36 2563 if self.params.get('check_formats') is True:
282f5709 2564 formats = LazyList(self._check_formats(formats[::-1]), reverse=True)
9f1a1c36 2565
88acdbc2 2566 if not formats or formats[0] is not info_dict:
b3d9ef88
JMF
2567 # only set the 'formats' fields if the original info_dict list them
2568 # otherwise we end up with a circular reference, the first (and unique)
f89197d7 2569 # element in the 'formats' field in info_dict is info_dict itself,
dfb1b146 2570 # which can't be exported to json
b3d9ef88 2571 info_dict['formats'] = formats
4ec82a72 2572
2573 info_dict, _ = self.pre_process(info_dict)
2574
6db9c4d5 2575 if self._match_entry(info_dict, incomplete=self._format_fields) is not None:
09b49e1f 2576 return info_dict
2577
2578 self.post_extract(info_dict)
2579 info_dict, _ = self.pre_process(info_dict, 'after_filter')
2580
093a1710 2581 # The pre-processors may have modified the formats
2582 formats = info_dict.get('formats', [info_dict])
2583
fa9f30b8 2584 list_only = self.params.get('simulate') is None and (
2585 self.params.get('list_thumbnails') or self.params.get('listformats') or self.params.get('listsubtitles'))
2586 interactive_format_selection = not list_only and self.format_selector == '-'
b7b04c78 2587 if self.params.get('list_thumbnails'):
2588 self.list_thumbnails(info_dict)
b7b04c78 2589 if self.params.get('listsubtitles'):
2590 if 'automatic_captions' in info_dict:
2591 self.list_subtitles(
2592 info_dict['id'], automatic_captions, 'automatic captions')
2593 self.list_subtitles(info_dict['id'], subtitles, 'subtitles')
fa9f30b8 2594 if self.params.get('listformats') or interactive_format_selection:
b69fd25c 2595 self.list_formats(info_dict)
169dbde9 2596 if list_only:
b7b04c78 2597 # Without this printing, -F --print-json will not work
169dbde9 2598 self.__forced_printings(info_dict, self.prepare_filename(info_dict), incomplete=True)
c487cf00 2599 return info_dict
bfaae0a7 2600
187986a8 2601 format_selector = self.format_selector
2602 if format_selector is None:
0017d9ad 2603 req_format = self._default_format_spec(info_dict, download=download)
0760b0a7 2604 self.write_debug('Default format spec: %s' % req_format)
187986a8 2605 format_selector = self.build_format_selector(req_format)
317f7ab6 2606
fa9f30b8 2607 while True:
2608 if interactive_format_selection:
2609 req_format = input(
2610 self._format_screen('\nEnter format selector: ', self.Styles.EMPHASIS))
2611 try:
2612 format_selector = self.build_format_selector(req_format)
2613 except SyntaxError as err:
2614 self.report_error(err, tb=False, is_error=False)
2615 continue
2616
85e801a9 2617 formats_to_download = list(format_selector({
fa9f30b8 2618 'formats': formats,
85e801a9 2619 'has_merged_format': any('none' not in (f.get('acodec'), f.get('vcodec')) for f in formats),
2620 'incomplete_formats': (
2621 # All formats are video-only or
2622 all(f.get('vcodec') != 'none' and f.get('acodec') == 'none' for f in formats)
2623 # all formats are audio-only
2624 or all(f.get('vcodec') == 'none' and f.get('acodec') != 'none' for f in formats)),
2625 }))
fa9f30b8 2626 if interactive_format_selection and not formats_to_download:
2627 self.report_error('Requested format is not available', tb=False, is_error=False)
2628 continue
2629 break
317f7ab6 2630
dd82ffea 2631 if not formats_to_download:
b7da73eb 2632 if not self.params.get('ignore_no_formats_error'):
c0b6e5c7 2633 raise ExtractorError(
2634 'Requested format is not available. Use --list-formats for a list of available formats',
2635 expected=True, video_id=info_dict['id'], ie=info_dict['extractor'])
b62fa6d7 2636 self.report_warning('Requested format is not available')
2637 # Process what we can, even without any available formats.
2638 formats_to_download = [{}]
a13e6848 2639
b62fa6d7 2640 best_format = formats_to_download[-1]
2641 if download:
2642 if best_format:
2643 self.to_screen(
2644 f'[info] {info_dict["id"]}: Downloading {len(formats_to_download)} format(s): '
2645 + ', '.join([f['format_id'] for f in formats_to_download]))
a13e6848 2646 max_downloads_reached = False
f46e2f9d 2647 for i, fmt in enumerate(formats_to_download):
09b49e1f 2648 formats_to_download[i] = new_info = self._copy_infodict(info_dict)
b7da73eb 2649 new_info.update(fmt)
a13e6848 2650 try:
2651 self.process_info(new_info)
2652 except MaxDownloadsReached:
2653 max_downloads_reached = True
415f8d51 2654 self._raise_pending_errors(new_info)
f46e2f9d 2655 # Remove copied info
2656 for key, val in tuple(new_info.items()):
2657 if info_dict.get(key) == val:
2658 new_info.pop(key)
a13e6848 2659 if max_downloads_reached:
2660 break
ebed8b37 2661
86e5f3ed 2662 write_archive = {f.get('__write_download_archive', False) for f in formats_to_download}
a13e6848 2663 assert write_archive.issubset({True, False, 'ignore'})
2664 if True in write_archive and False not in write_archive:
2665 self.record_download_archive(info_dict)
be72c624 2666
2667 info_dict['requested_downloads'] = formats_to_download
ed5835b4 2668 info_dict = self.run_all_pps('after_video', info_dict)
a13e6848 2669 if max_downloads_reached:
2670 raise MaxDownloadsReached()
ebed8b37 2671
49a57e70 2672 # We update the info dict with the selected best quality format (backwards compatibility)
be72c624 2673 info_dict.update(best_format)
dd82ffea
JMF
2674 return info_dict
2675
98c70d6f 2676 def process_subtitles(self, video_id, normal_subtitles, automatic_captions):
a504ced0 2677 """Select the requested subtitles and their format"""
d8a58ddc 2678 available_subs, normal_sub_langs = {}, []
98c70d6f
JMF
2679 if normal_subtitles and self.params.get('writesubtitles'):
2680 available_subs.update(normal_subtitles)
d8a58ddc 2681 normal_sub_langs = tuple(normal_subtitles.keys())
98c70d6f
JMF
2682 if automatic_captions and self.params.get('writeautomaticsub'):
2683 for lang, cap_info in automatic_captions.items():
360e1ca5
JMF
2684 if lang not in available_subs:
2685 available_subs[lang] = cap_info
2686
4d171848
JMF
2687 if (not self.params.get('writesubtitles') and not
2688 self.params.get('writeautomaticsub') or not
2689 available_subs):
2690 return None
a504ced0 2691
d8a58ddc 2692 all_sub_langs = tuple(available_subs.keys())
a504ced0 2693 if self.params.get('allsubtitles', False):
c32b0aab 2694 requested_langs = all_sub_langs
2695 elif self.params.get('subtitleslangs', False):
77c4a9ef 2696 # A list is used so that the order of languages will be the same as
2697 # given in subtitleslangs. See https://github.com/yt-dlp/yt-dlp/issues/1041
2698 requested_langs = []
2699 for lang_re in self.params.get('subtitleslangs'):
77c4a9ef 2700 discard = lang_re[0] == '-'
c32b0aab 2701 if discard:
77c4a9ef 2702 lang_re = lang_re[1:]
3aa91540 2703 if lang_re == 'all':
2704 if discard:
2705 requested_langs = []
2706 else:
2707 requested_langs.extend(all_sub_langs)
2708 continue
77c4a9ef 2709 current_langs = filter(re.compile(lang_re + '$').match, all_sub_langs)
c32b0aab 2710 if discard:
2711 for lang in current_langs:
77c4a9ef 2712 while lang in requested_langs:
2713 requested_langs.remove(lang)
c32b0aab 2714 else:
77c4a9ef 2715 requested_langs.extend(current_langs)
2716 requested_langs = orderedSet(requested_langs)
d8a58ddc 2717 elif normal_sub_langs:
2718 requested_langs = ['en'] if 'en' in normal_sub_langs else normal_sub_langs[:1]
a504ced0 2719 else:
d8a58ddc 2720 requested_langs = ['en'] if 'en' in all_sub_langs else all_sub_langs[:1]
ad3dc496 2721 if requested_langs:
2722 self.write_debug('Downloading subtitles: %s' % ', '.join(requested_langs))
a504ced0
JMF
2723
2724 formats_query = self.params.get('subtitlesformat', 'best')
2725 formats_preference = formats_query.split('/') if formats_query else []
2726 subs = {}
2727 for lang in requested_langs:
2728 formats = available_subs.get(lang)
2729 if formats is None:
86e5f3ed 2730 self.report_warning(f'{lang} subtitles not available for {video_id}')
a504ced0 2731 continue
a504ced0
JMF
2732 for ext in formats_preference:
2733 if ext == 'best':
2734 f = formats[-1]
2735 break
2736 matches = list(filter(lambda f: f['ext'] == ext, formats))
2737 if matches:
2738 f = matches[-1]
2739 break
2740 else:
2741 f = formats[-1]
2742 self.report_warning(
2743 'No subtitle format found matching "%s" for language %s, '
2744 'using %s' % (formats_query, lang, f['ext']))
2745 subs[lang] = f
2746 return subs
2747
bb66c247 2748 def _forceprint(self, key, info_dict):
2749 if info_dict is None:
2750 return
2751 info_copy = info_dict.copy()
2752 info_copy['formats_table'] = self.render_formats_table(info_dict)
2753 info_copy['thumbnails_table'] = self.render_thumbnails_table(info_dict)
2754 info_copy['subtitles_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('subtitles'))
2755 info_copy['automatic_captions_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('automatic_captions'))
2756
2757 def format_tmpl(tmpl):
2758 mobj = re.match(r'\w+(=?)$', tmpl)
2759 if mobj and mobj.group(1):
2760 return f'{tmpl[:-1]} = %({tmpl[:-1]})r'
2761 elif mobj:
2762 return f'%({tmpl})s'
2763 return tmpl
8130779d 2764
bb66c247 2765 for tmpl in self.params['forceprint'].get(key, []):
2766 self.to_stdout(self.evaluate_outtmpl(format_tmpl(tmpl), info_copy))
2767
2768 for tmpl, file_tmpl in self.params['print_to_file'].get(key, []):
5127e92a 2769 filename = self.prepare_filename(info_dict, outtmpl=file_tmpl)
bb66c247 2770 tmpl = format_tmpl(tmpl)
2771 self.to_screen(f'[info] Writing {tmpl!r} to: {filename}')
8d93e69d 2772 if self._ensure_dir_exists(filename):
86e5f3ed 2773 with open(filename, 'a', encoding='utf-8') as f:
8d93e69d 2774 f.write(self.evaluate_outtmpl(tmpl, info_copy) + '\n')
ca30f449 2775
d06daf23 2776 def __forced_printings(self, info_dict, filename, incomplete):
53c18592 2777 def print_mandatory(field, actual_field=None):
2778 if actual_field is None:
2779 actual_field = field
d06daf23 2780 if (self.params.get('force%s' % field, False)
53c18592 2781 and (not incomplete or info_dict.get(actual_field) is not None)):
2782 self.to_stdout(info_dict[actual_field])
d06daf23
S
2783
2784 def print_optional(field):
2785 if (self.params.get('force%s' % field, False)
2786 and info_dict.get(field) is not None):
2787 self.to_stdout(info_dict[field])
2788
53c18592 2789 info_dict = info_dict.copy()
2790 if filename is not None:
2791 info_dict['filename'] = filename
2792 if info_dict.get('requested_formats') is not None:
2793 # For RTMP URLs, also include the playpath
2794 info_dict['urls'] = '\n'.join(f['url'] + f.get('play_path', '') for f in info_dict['requested_formats'])
10331a26 2795 elif info_dict.get('url'):
53c18592 2796 info_dict['urls'] = info_dict['url'] + info_dict.get('play_path', '')
2797
bb66c247 2798 if (self.params.get('forcejson')
2799 or self.params['forceprint'].get('video')
2800 or self.params['print_to_file'].get('video')):
2b8a2973 2801 self.post_extract(info_dict)
bb66c247 2802 self._forceprint('video', info_dict)
53c18592 2803
d06daf23
S
2804 print_mandatory('title')
2805 print_mandatory('id')
53c18592 2806 print_mandatory('url', 'urls')
d06daf23
S
2807 print_optional('thumbnail')
2808 print_optional('description')
53c18592 2809 print_optional('filename')
b868936c 2810 if self.params.get('forceduration') and info_dict.get('duration') is not None:
d06daf23
S
2811 self.to_stdout(formatSeconds(info_dict['duration']))
2812 print_mandatory('format')
53c18592 2813
2b8a2973 2814 if self.params.get('forcejson'):
6e84b215 2815 self.to_stdout(json.dumps(self.sanitize_info(info_dict)))
d06daf23 2816
e8e73840 2817 def dl(self, name, info, subtitle=False, test=False):
88acdbc2 2818 if not info.get('url'):
1151c407 2819 self.raise_no_formats(info, True)
e8e73840 2820
2821 if test:
2822 verbose = self.params.get('verbose')
2823 params = {
2824 'test': True,
a169858f 2825 'quiet': self.params.get('quiet') or not verbose,
e8e73840 2826 'verbose': verbose,
2827 'noprogress': not verbose,
2828 'nopart': True,
2829 'skip_unavailable_fragments': False,
2830 'keep_fragments': False,
2831 'overwrites': True,
2832 '_no_ytdl_file': True,
2833 }
2834 else:
2835 params = self.params
96fccc10 2836 fd = get_suitable_downloader(info, params, to_stdout=(name == '-'))(self, params)
e8e73840 2837 if not test:
2838 for ph in self._progress_hooks:
2839 fd.add_progress_hook(ph)
42676437
M
2840 urls = '", "'.join(
2841 (f['url'].split(',')[0] + ',<data>' if f['url'].startswith('data:') else f['url'])
2842 for f in info.get('requested_formats', []) or [info])
3a408f9d 2843 self.write_debug(f'Invoking {fd.FD_NAME} downloader on "{urls}"')
03b4de72 2844
adbc4ec4
THD
2845 # Note: Ideally info should be a deep-copied so that hooks cannot modify it.
2846 # But it may contain objects that are not deep-copyable
2847 new_info = self._copy_infodict(info)
e8e73840 2848 if new_info.get('http_headers') is None:
2849 new_info['http_headers'] = self._calc_headers(new_info)
2850 return fd.download(name, new_info, subtitle)
2851
e04938ab 2852 def existing_file(self, filepaths, *, default_overwrite=True):
2853 existing_files = list(filter(os.path.exists, orderedSet(filepaths)))
2854 if existing_files and not self.params.get('overwrites', default_overwrite):
2855 return existing_files[0]
2856
2857 for file in existing_files:
2858 self.report_file_delete(file)
2859 os.remove(file)
2860 return None
2861
8222d8de 2862 def process_info(self, info_dict):
09b49e1f 2863 """Process a single resolved IE result. (Modifies it in-place)"""
8222d8de
JMF
2864
2865 assert info_dict.get('_type', 'video') == 'video'
f46e2f9d 2866 original_infodict = info_dict
fd288278 2867
4513a41a 2868 if 'format' not in info_dict and 'ext' in info_dict:
8222d8de
JMF
2869 info_dict['format'] = info_dict['ext']
2870
09b49e1f 2871 # This is mostly just for backward compatibility of process_info
2872 # As a side-effect, this allows for format-specific filters
c77495e3 2873 if self._match_entry(info_dict) is not None:
9e907ebd 2874 info_dict['__write_download_archive'] = 'ignore'
8222d8de
JMF
2875 return
2876
09b49e1f 2877 # Does nothing under normal operation - for backward compatibility of process_info
277d6ff5 2878 self.post_extract(info_dict)
0c14d66a 2879 self._num_downloads += 1
8222d8de 2880
dcf64d43 2881 # info_dict['_filename'] needs to be set for backward compatibility
de6000d9 2882 info_dict['_filename'] = full_filename = self.prepare_filename(info_dict, warn=True)
2883 temp_filename = self.prepare_filename(info_dict, 'temp')
0202b52a 2884 files_to_move = {}
8222d8de
JMF
2885
2886 # Forced printings
4513a41a 2887 self.__forced_printings(info_dict, full_filename, incomplete=('format' not in info_dict))
8222d8de 2888
ca6d59d2 2889 def check_max_downloads():
2890 if self._num_downloads >= float(self.params.get('max_downloads') or 'inf'):
2891 raise MaxDownloadsReached()
2892
b7b04c78 2893 if self.params.get('simulate'):
9e907ebd 2894 info_dict['__write_download_archive'] = self.params.get('force_write_download_archive')
ca6d59d2 2895 check_max_downloads()
8222d8de
JMF
2896 return
2897
de6000d9 2898 if full_filename is None:
8222d8de 2899 return
e92caff5 2900 if not self._ensure_dir_exists(encodeFilename(full_filename)):
0202b52a 2901 return
e92caff5 2902 if not self._ensure_dir_exists(encodeFilename(temp_filename)):
8222d8de
JMF
2903 return
2904
80c03fa9 2905 if self._write_description('video', info_dict,
2906 self.prepare_filename(info_dict, 'description')) is None:
2907 return
2908
2909 sub_files = self._write_subtitles(info_dict, temp_filename)
2910 if sub_files is None:
2911 return
2912 files_to_move.update(dict(sub_files))
2913
2914 thumb_files = self._write_thumbnails(
2915 'video', info_dict, temp_filename, self.prepare_filename(info_dict, 'thumbnail'))
2916 if thumb_files is None:
2917 return
2918 files_to_move.update(dict(thumb_files))
8222d8de 2919
80c03fa9 2920 infofn = self.prepare_filename(info_dict, 'infojson')
2921 _infojson_written = self._write_info_json('video', info_dict, infofn)
2922 if _infojson_written:
dac5df5a 2923 info_dict['infojson_filename'] = infofn
e75bb0d6 2924 # For backward compatibility, even though it was a private field
80c03fa9 2925 info_dict['__infojson_filename'] = infofn
2926 elif _infojson_written is None:
2927 return
2928
2929 # Note: Annotations are deprecated
2930 annofn = None
1fb07d10 2931 if self.params.get('writeannotations', False):
de6000d9 2932 annofn = self.prepare_filename(info_dict, 'annotation')
80c03fa9 2933 if annofn:
e92caff5 2934 if not self._ensure_dir_exists(encodeFilename(annofn)):
0202b52a 2935 return
0c3d0f51 2936 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(annofn)):
6febd1c1 2937 self.to_screen('[info] Video annotations are already present')
ffddb112
RA
2938 elif not info_dict.get('annotations'):
2939 self.report_warning('There are no annotations to write.')
7b6fefc9
PH
2940 else:
2941 try:
6febd1c1 2942 self.to_screen('[info] Writing video annotations to: ' + annofn)
86e5f3ed 2943 with open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
7b6fefc9
PH
2944 annofile.write(info_dict['annotations'])
2945 except (KeyError, TypeError):
6febd1c1 2946 self.report_warning('There are no annotations to write.')
86e5f3ed 2947 except OSError:
6febd1c1 2948 self.report_error('Cannot write annotations file: ' + annofn)
7b6fefc9 2949 return
1fb07d10 2950
732044af 2951 # Write internet shortcut files
08438d2c 2952 def _write_link_file(link_type):
60f3e995 2953 url = try_get(info_dict['webpage_url'], iri_to_uri)
2954 if not url:
2955 self.report_warning(
2956 f'Cannot write internet shortcut file because the actual URL of "{info_dict["webpage_url"]}" is unknown')
2957 return True
08438d2c 2958 linkfn = replace_extension(self.prepare_filename(info_dict, 'link'), link_type, info_dict.get('ext'))
0e6b018a
Z
2959 if not self._ensure_dir_exists(encodeFilename(linkfn)):
2960 return False
10e3742e 2961 if self.params.get('overwrites', True) and os.path.exists(encodeFilename(linkfn)):
08438d2c 2962 self.to_screen(f'[info] Internet shortcut (.{link_type}) is already present')
2963 return True
2964 try:
2965 self.to_screen(f'[info] Writing internet shortcut (.{link_type}) to: {linkfn}')
86e5f3ed 2966 with open(encodeFilename(to_high_limit_path(linkfn)), 'w', encoding='utf-8',
2967 newline='\r\n' if link_type == 'url' else '\n') as linkfile:
60f3e995 2968 template_vars = {'url': url}
08438d2c 2969 if link_type == 'desktop':
2970 template_vars['filename'] = linkfn[:-(len(link_type) + 1)]
2971 linkfile.write(LINK_TEMPLATES[link_type] % template_vars)
86e5f3ed 2972 except OSError:
08438d2c 2973 self.report_error(f'Cannot write internet shortcut {linkfn}')
2974 return False
732044af 2975 return True
2976
08438d2c 2977 write_links = {
2978 'url': self.params.get('writeurllink'),
2979 'webloc': self.params.get('writewebloclink'),
2980 'desktop': self.params.get('writedesktoplink'),
2981 }
2982 if self.params.get('writelink'):
2983 link_type = ('webloc' if sys.platform == 'darwin'
2984 else 'desktop' if sys.platform.startswith('linux')
2985 else 'url')
2986 write_links[link_type] = True
2987
2988 if any(should_write and not _write_link_file(link_type)
2989 for link_type, should_write in write_links.items()):
2990 return
732044af 2991
f46e2f9d 2992 def replace_info_dict(new_info):
2993 nonlocal info_dict
2994 if new_info == info_dict:
2995 return
2996 info_dict.clear()
2997 info_dict.update(new_info)
2998
415f8d51 2999 new_info, files_to_move = self.pre_process(info_dict, 'before_dl', files_to_move)
3000 replace_info_dict(new_info)
56d868db 3001
a13e6848 3002 if self.params.get('skip_download'):
56d868db 3003 info_dict['filepath'] = temp_filename
3004 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
3005 info_dict['__files_to_move'] = files_to_move
f46e2f9d 3006 replace_info_dict(self.run_pp(MoveFilesAfterDownloadPP(self, False), info_dict))
9e907ebd 3007 info_dict['__write_download_archive'] = self.params.get('force_write_download_archive')
56d868db 3008 else:
3009 # Download
b868936c 3010 info_dict.setdefault('__postprocessors', [])
4340deca 3011 try:
0202b52a 3012
e04938ab 3013 def existing_video_file(*filepaths):
6b591b29 3014 ext = info_dict.get('ext')
e04938ab 3015 converted = lambda file: replace_extension(file, self.params.get('final_ext') or ext, ext)
3016 file = self.existing_file(itertools.chain(*zip(map(converted, filepaths), filepaths)),
3017 default_overwrite=False)
3018 if file:
3019 info_dict['ext'] = os.path.splitext(file)[1][1:]
3020 return file
0202b52a 3021
3022 success = True
4340deca 3023 if info_dict.get('requested_formats') is not None:
81cd954a
S
3024
3025 def compatible_formats(formats):
d03cfdce 3026 # TODO: some formats actually allow this (mkv, webm, ogg, mp4), but not all of them.
3027 video_formats = [format for format in formats if format.get('vcodec') != 'none']
3028 audio_formats = [format for format in formats if format.get('acodec') != 'none']
3029 if len(video_formats) > 2 or len(audio_formats) > 2:
3030 return False
3031
81cd954a 3032 # Check extension
86e5f3ed 3033 exts = {format.get('ext') for format in formats}
d03cfdce 3034 COMPATIBLE_EXTS = (
86e5f3ed 3035 {'mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma'},
3036 {'webm'},
d03cfdce 3037 )
3038 for ext_sets in COMPATIBLE_EXTS:
3039 if ext_sets.issuperset(exts):
3040 return True
81cd954a
S
3041 # TODO: Check acodec/vcodec
3042 return False
3043
3044 requested_formats = info_dict['requested_formats']
0202b52a 3045 old_ext = info_dict['ext']
4e3b637d 3046 if self.params.get('merge_output_format') is None:
3047 if not compatible_formats(requested_formats):
3048 info_dict['ext'] = 'mkv'
3049 self.report_warning(
3050 'Requested formats are incompatible for merge and will be merged into mkv')
3051 if (info_dict['ext'] == 'webm'
3052 and info_dict.get('thumbnails')
3053 # check with type instead of pp_key, __name__, or isinstance
3054 # since we dont want any custom PPs to trigger this
c487cf00 3055 and any(type(pp) == EmbedThumbnailPP for pp in self._pps['post_process'])): # noqa: E721
4e3b637d 3056 info_dict['ext'] = 'mkv'
3057 self.report_warning(
3058 'webm doesn\'t support embedding a thumbnail, mkv will be used')
124bc071 3059 new_ext = info_dict['ext']
0202b52a 3060
124bc071 3061 def correct_ext(filename, ext=new_ext):
96fccc10 3062 if filename == '-':
3063 return filename
0202b52a 3064 filename_real_ext = os.path.splitext(filename)[1][1:]
3065 filename_wo_ext = (
3066 os.path.splitext(filename)[0]
124bc071 3067 if filename_real_ext in (old_ext, new_ext)
0202b52a 3068 else filename)
86e5f3ed 3069 return f'{filename_wo_ext}.{ext}'
0202b52a 3070
38c6902b 3071 # Ensure filename always has a correct extension for successful merge
0202b52a 3072 full_filename = correct_ext(full_filename)
3073 temp_filename = correct_ext(temp_filename)
e04938ab 3074 dl_filename = existing_video_file(full_filename, temp_filename)
1ea24129 3075 info_dict['__real_download'] = False
18e674b4 3076
adbc4ec4
THD
3077 downloaded = []
3078 merger = FFmpegMergerPP(self)
3079
3080 fd = get_suitable_downloader(info_dict, self.params, to_stdout=temp_filename == '-')
dbf5416a 3081 if dl_filename is not None:
6c7274ec 3082 self.report_file_already_downloaded(dl_filename)
adbc4ec4
THD
3083 elif fd:
3084 for f in requested_formats if fd != FFmpegFD else []:
3085 f['filepath'] = fname = prepend_extension(
3086 correct_ext(temp_filename, info_dict['ext']),
3087 'f%s' % f['format_id'], info_dict['ext'])
3088 downloaded.append(fname)
dbf5416a 3089 info_dict['url'] = '\n'.join(f['url'] for f in requested_formats)
3090 success, real_download = self.dl(temp_filename, info_dict)
3091 info_dict['__real_download'] = real_download
18e674b4 3092 else:
18e674b4 3093 if self.params.get('allow_unplayable_formats'):
3094 self.report_warning(
3095 'You have requested merging of multiple formats '
3096 'while also allowing unplayable formats to be downloaded. '
3097 'The formats won\'t be merged to prevent data corruption.')
3098 elif not merger.available:
e8969bda 3099 msg = 'You have requested merging of multiple formats but ffmpeg is not installed'
3100 if not self.params.get('ignoreerrors'):
3101 self.report_error(f'{msg}. Aborting due to --abort-on-error')
3102 return
3103 self.report_warning(f'{msg}. The formats won\'t be merged')
18e674b4 3104
96fccc10 3105 if temp_filename == '-':
adbc4ec4 3106 reason = ('using a downloader other than ffmpeg' if FFmpegFD.can_merge_formats(info_dict, self.params)
96fccc10 3107 else 'but the formats are incompatible for simultaneous download' if merger.available
3108 else 'but ffmpeg is not installed')
3109 self.report_warning(
3110 f'You have requested downloading multiple formats to stdout {reason}. '
3111 'The formats will be streamed one after the other')
3112 fname = temp_filename
dbf5416a 3113 for f in requested_formats:
3114 new_info = dict(info_dict)
3115 del new_info['requested_formats']
3116 new_info.update(f)
96fccc10 3117 if temp_filename != '-':
124bc071 3118 fname = prepend_extension(
3119 correct_ext(temp_filename, new_info['ext']),
3120 'f%s' % f['format_id'], new_info['ext'])
96fccc10 3121 if not self._ensure_dir_exists(fname):
3122 return
a21e0ab1 3123 f['filepath'] = fname
96fccc10 3124 downloaded.append(fname)
dbf5416a 3125 partial_success, real_download = self.dl(fname, new_info)
3126 info_dict['__real_download'] = info_dict['__real_download'] or real_download
3127 success = success and partial_success
adbc4ec4
THD
3128
3129 if downloaded and merger.available and not self.params.get('allow_unplayable_formats'):
3130 info_dict['__postprocessors'].append(merger)
3131 info_dict['__files_to_merge'] = downloaded
3132 # Even if there were no downloads, it is being merged only now
3133 info_dict['__real_download'] = True
3134 else:
3135 for file in downloaded:
3136 files_to_move[file] = None
4340deca
P
3137 else:
3138 # Just a single file
e04938ab 3139 dl_filename = existing_video_file(full_filename, temp_filename)
6c7274ec 3140 if dl_filename is None or dl_filename == temp_filename:
3141 # dl_filename == temp_filename could mean that the file was partially downloaded with --no-part.
3142 # So we should try to resume the download
e8e73840 3143 success, real_download = self.dl(temp_filename, info_dict)
0202b52a 3144 info_dict['__real_download'] = real_download
6c7274ec 3145 else:
3146 self.report_file_already_downloaded(dl_filename)
0202b52a 3147
0202b52a 3148 dl_filename = dl_filename or temp_filename
c571435f 3149 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
0202b52a 3150
3158150c 3151 except network_exceptions as err:
7960b056 3152 self.report_error('unable to download video data: %s' % error_to_compat_str(err))
4340deca 3153 return
86e5f3ed 3154 except OSError as err:
4340deca
P
3155 raise UnavailableVideoError(err)
3156 except (ContentTooShortError, ) as err:
86e5f3ed 3157 self.report_error(f'content too short (expected {err.expected} bytes and served {err.downloaded})')
4340deca 3158 return
8222d8de 3159
415f8d51 3160 self._raise_pending_errors(info_dict)
de6000d9 3161 if success and full_filename != '-':
f17f8651 3162
fd7cfb64 3163 def fixup():
3164 do_fixup = True
3165 fixup_policy = self.params.get('fixup')
3166 vid = info_dict['id']
3167
3168 if fixup_policy in ('ignore', 'never'):
3169 return
3170 elif fixup_policy == 'warn':
3fe75fdc 3171 do_fixup = 'warn'
f89b3e2d 3172 elif fixup_policy != 'force':
3173 assert fixup_policy in ('detect_or_warn', None)
3174 if not info_dict.get('__real_download'):
3175 do_fixup = False
fd7cfb64 3176
3177 def ffmpeg_fixup(cndn, msg, cls):
3fe75fdc 3178 if not (do_fixup and cndn):
fd7cfb64 3179 return
3fe75fdc 3180 elif do_fixup == 'warn':
fd7cfb64 3181 self.report_warning(f'{vid}: {msg}')
3182 return
3183 pp = cls(self)
3184 if pp.available:
3185 info_dict['__postprocessors'].append(pp)
3186 else:
3187 self.report_warning(f'{vid}: {msg}. Install ffmpeg to fix this automatically')
3188
3189 stretched_ratio = info_dict.get('stretched_ratio')
3190 ffmpeg_fixup(
3191 stretched_ratio not in (1, None),
3192 f'Non-uniform pixel ratio {stretched_ratio}',
3193 FFmpegFixupStretchedPP)
3194
3195 ffmpeg_fixup(
3196 (info_dict.get('requested_formats') is None
3197 and info_dict.get('container') == 'm4a_dash'
3198 and info_dict.get('ext') == 'm4a'),
3199 'writing DASH m4a. Only some players support this container',
3200 FFmpegFixupM4aPP)
3201
993191c0 3202 downloader = get_suitable_downloader(info_dict, self.params) if 'protocol' in info_dict else None
24146491 3203 downloader = downloader.FD_NAME if downloader else None
adbc4ec4
THD
3204
3205 if info_dict.get('requested_formats') is None: # Not necessary if doing merger
24146491 3206 ffmpeg_fixup(downloader == 'hlsnative' and not self.params.get('hls_use_mpegts')
494f5230 3207 or info_dict.get('is_live') and self.params.get('hls_use_mpegts') is None,
adbc4ec4
THD
3208 'Possible MPEG-TS in MP4 container or malformed AAC timestamps',
3209 FFmpegFixupM3u8PP)
3210 ffmpeg_fixup(info_dict.get('is_live') and downloader == 'DashSegmentsFD',
3211 'Possible duplicate MOOV atoms', FFmpegFixupDuplicateMoovPP)
3212
24146491 3213 ffmpeg_fixup(downloader == 'web_socket_fragment', 'Malformed timestamps detected', FFmpegFixupTimestampPP)
3214 ffmpeg_fixup(downloader == 'web_socket_fragment', 'Malformed duration detected', FFmpegFixupDurationPP)
fd7cfb64 3215
3216 fixup()
8222d8de 3217 try:
f46e2f9d 3218 replace_info_dict(self.post_process(dl_filename, info_dict, files_to_move))
af819c21 3219 except PostProcessingError as err:
3220 self.report_error('Postprocessing: %s' % str(err))
8222d8de 3221 return
ab8e5e51
AM
3222 try:
3223 for ph in self._post_hooks:
23c1a667 3224 ph(info_dict['filepath'])
ab8e5e51
AM
3225 except Exception as err:
3226 self.report_error('post hooks: %s' % str(err))
3227 return
9e907ebd 3228 info_dict['__write_download_archive'] = True
2d30509f 3229
c487cf00 3230 assert info_dict is original_infodict # Make sure the info_dict was modified in-place
a13e6848 3231 if self.params.get('force_write_download_archive'):
9e907ebd 3232 info_dict['__write_download_archive'] = True
ca6d59d2 3233 check_max_downloads()
8222d8de 3234
aa9369a2 3235 def __download_wrapper(self, func):
3236 @functools.wraps(func)
3237 def wrapper(*args, **kwargs):
3238 try:
3239 res = func(*args, **kwargs)
3240 except UnavailableVideoError as e:
3241 self.report_error(e)
b222c271 3242 except DownloadCancelled as e:
3243 self.to_screen(f'[info] {e}')
3244 if not self.params.get('break_per_url'):
3245 raise
aa9369a2 3246 else:
3247 if self.params.get('dump_single_json', False):
3248 self.post_extract(res)
3249 self.to_stdout(json.dumps(self.sanitize_info(res)))
3250 return wrapper
3251
8222d8de
JMF
3252 def download(self, url_list):
3253 """Download a given list of URLs."""
aa9369a2 3254 url_list = variadic(url_list) # Passing a single URL is a common mistake
de6000d9 3255 outtmpl = self.outtmpl_dict['default']
3089bc74
S
3256 if (len(url_list) > 1
3257 and outtmpl != '-'
3258 and '%' not in outtmpl
3259 and self.params.get('max_downloads') != 1):
acd69589 3260 raise SameFileError(outtmpl)
8222d8de
JMF
3261
3262 for url in url_list:
aa9369a2 3263 self.__download_wrapper(self.extract_info)(
3264 url, force_generic_extractor=self.params.get('force_generic_extractor', False))
8222d8de
JMF
3265
3266 return self._download_retcode
3267
1dcc4c0c 3268 def download_with_info_file(self, info_filename):
31bd3925
JMF
3269 with contextlib.closing(fileinput.FileInput(
3270 [info_filename], mode='r',
3271 openhook=fileinput.hook_encoded('utf-8'))) as f:
3272 # FileInput doesn't have a read method, we can't call json.load
8012d892 3273 info = self.sanitize_info(json.loads('\n'.join(f)), self.params.get('clean_infojson', True))
d4943898 3274 try:
aa9369a2 3275 self.__download_wrapper(self.process_ie_result)(info, download=True)
f2ebc5c7 3276 except (DownloadError, EntryNotInPlaylist, ReExtractInfo) as e:
bf5f605e 3277 if not isinstance(e, EntryNotInPlaylist):
3278 self.to_stderr('\r')
d4943898
JMF
3279 webpage_url = info.get('webpage_url')
3280 if webpage_url is not None:
aa9369a2 3281 self.report_warning(f'The info failed to download: {e}; trying with URL {webpage_url}')
d4943898
JMF
3282 return self.download([webpage_url])
3283 else:
3284 raise
3285 return self._download_retcode
1dcc4c0c 3286
cb202fd2 3287 @staticmethod
8012d892 3288 def sanitize_info(info_dict, remove_private_keys=False):
3289 ''' Sanitize the infodict for converting to json '''
3ad56b42 3290 if info_dict is None:
3291 return info_dict
6e84b215 3292 info_dict.setdefault('epoch', int(time.time()))
6a5a30f9 3293 info_dict.setdefault('_type', 'video')
09b49e1f 3294
8012d892 3295 if remove_private_keys:
0a5a191a 3296 reject = lambda k, v: v is None or k.startswith('__') or k in {
f46e2f9d 3297 'requested_downloads', 'requested_formats', 'requested_subtitles', 'requested_entries',
0a5a191a 3298 'entries', 'filepath', '_filename', 'infojson_filename', 'original_url', 'playlist_autonumber',
6e84b215 3299 }
ae8f99e6 3300 else:
09b49e1f 3301 reject = lambda k, v: False
adbc4ec4
THD
3302
3303 def filter_fn(obj):
3304 if isinstance(obj, dict):
3305 return {k: filter_fn(v) for k, v in obj.items() if not reject(k, v)}
3306 elif isinstance(obj, (list, tuple, set, LazyList)):
3307 return list(map(filter_fn, obj))
3308 elif obj is None or isinstance(obj, (str, int, float, bool)):
3309 return obj
3310 else:
3311 return repr(obj)
3312
5226731e 3313 return filter_fn(info_dict)
cb202fd2 3314
8012d892 3315 @staticmethod
3316 def filter_requested_info(info_dict, actually_filter=True):
3317 ''' Alias of sanitize_info for backward compatibility '''
3318 return YoutubeDL.sanitize_info(info_dict, actually_filter)
3319
43d7f5a5 3320 def _delete_downloaded_files(self, *files_to_delete, info={}, msg=None):
3321 for filename in set(filter(None, files_to_delete)):
3322 if msg:
3323 self.to_screen(msg % filename)
3324 try:
3325 os.remove(filename)
3326 except OSError:
3327 self.report_warning(f'Unable to delete file {filename}')
3328 if filename in info.get('__files_to_move', []): # NB: Delete even if None
3329 del info['__files_to_move'][filename]
3330
ed5835b4 3331 @staticmethod
3332 def post_extract(info_dict):
3333 def actual_post_extract(info_dict):
3334 if info_dict.get('_type') in ('playlist', 'multi_video'):
3335 for video_dict in info_dict.get('entries', {}):
3336 actual_post_extract(video_dict or {})
3337 return
3338
09b49e1f 3339 post_extractor = info_dict.pop('__post_extractor', None) or (lambda: {})
3340 info_dict.update(post_extractor())
ed5835b4 3341
3342 actual_post_extract(info_dict or {})
3343
dcf64d43 3344 def run_pp(self, pp, infodict):
5bfa4862 3345 files_to_delete = []
dcf64d43 3346 if '__files_to_move' not in infodict:
3347 infodict['__files_to_move'] = {}
b1940459 3348 try:
3349 files_to_delete, infodict = pp.run(infodict)
3350 except PostProcessingError as e:
3351 # Must be True and not 'only_download'
3352 if self.params.get('ignoreerrors') is True:
3353 self.report_error(e)
3354 return infodict
3355 raise
3356
5bfa4862 3357 if not files_to_delete:
dcf64d43 3358 return infodict
5bfa4862 3359 if self.params.get('keepvideo', False):
3360 for f in files_to_delete:
dcf64d43 3361 infodict['__files_to_move'].setdefault(f, '')
5bfa4862 3362 else:
43d7f5a5 3363 self._delete_downloaded_files(
3364 *files_to_delete, info=infodict, msg='Deleting original file %s (pass -k to keep)')
dcf64d43 3365 return infodict
5bfa4862 3366
ed5835b4 3367 def run_all_pps(self, key, info, *, additional_pps=None):
bb66c247 3368 self._forceprint(key, info)
ed5835b4 3369 for pp in (additional_pps or []) + self._pps[key]:
dc5f409c 3370 info = self.run_pp(pp, info)
ed5835b4 3371 return info
277d6ff5 3372
56d868db 3373 def pre_process(self, ie_info, key='pre_process', files_to_move=None):
5bfa4862 3374 info = dict(ie_info)
56d868db 3375 info['__files_to_move'] = files_to_move or {}
415f8d51 3376 try:
3377 info = self.run_all_pps(key, info)
3378 except PostProcessingError as err:
3379 msg = f'Preprocessing: {err}'
3380 info.setdefault('__pending_error', msg)
3381 self.report_error(msg, is_error=False)
56d868db 3382 return info, info.pop('__files_to_move', None)
5bfa4862 3383
f46e2f9d 3384 def post_process(self, filename, info, files_to_move=None):
8222d8de 3385 """Run all the postprocessors on the given file."""
8222d8de 3386 info['filepath'] = filename
dcf64d43 3387 info['__files_to_move'] = files_to_move or {}
ed5835b4 3388 info = self.run_all_pps('post_process', info, additional_pps=info.get('__postprocessors'))
dcf64d43 3389 info = self.run_pp(MoveFilesAfterDownloadPP(self), info)
3390 del info['__files_to_move']
ed5835b4 3391 return self.run_all_pps('after_move', info)
c1c9a79c 3392
5db07df6 3393 def _make_archive_id(self, info_dict):
e9fef7ee
S
3394 video_id = info_dict.get('id')
3395 if not video_id:
3396 return
5db07df6
PH
3397 # Future-proof against any change in case
3398 # and backwards compatibility with prior versions
e9fef7ee 3399 extractor = info_dict.get('extractor_key') or info_dict.get('ie_key') # key in a playlist
7012b23c 3400 if extractor is None:
1211bb6d
S
3401 url = str_or_none(info_dict.get('url'))
3402 if not url:
3403 return
e9fef7ee 3404 # Try to find matching extractor for the URL and take its ie_key
8b7491c8 3405 for ie_key, ie in self._ies.items():
1211bb6d 3406 if ie.suitable(url):
8b7491c8 3407 extractor = ie_key
e9fef7ee
S
3408 break
3409 else:
3410 return
86e5f3ed 3411 return f'{extractor.lower()} {video_id}'
5db07df6
PH
3412
3413 def in_download_archive(self, info_dict):
3414 fn = self.params.get('download_archive')
3415 if fn is None:
3416 return False
3417
3418 vid_id = self._make_archive_id(info_dict)
e9fef7ee 3419 if not vid_id:
7012b23c 3420 return False # Incomplete video information
5db07df6 3421
a45e8619 3422 return vid_id in self.archive
c1c9a79c
PH
3423
3424 def record_download_archive(self, info_dict):
3425 fn = self.params.get('download_archive')
3426 if fn is None:
3427 return
5db07df6
PH
3428 vid_id = self._make_archive_id(info_dict)
3429 assert vid_id
a13e6848 3430 self.write_debug(f'Adding to archive: {vid_id}')
c1c9a79c 3431 with locked_file(fn, 'a', encoding='utf-8') as archive_file:
6febd1c1 3432 archive_file.write(vid_id + '\n')
a45e8619 3433 self.archive.add(vid_id)
dd82ffea 3434
8c51aa65 3435 @staticmethod
8abeeb94 3436 def format_resolution(format, default='unknown'):
9359f3d4 3437 if format.get('vcodec') == 'none' and format.get('acodec') != 'none':
fb04e403 3438 return 'audio only'
f49d89ee
PH
3439 if format.get('resolution') is not None:
3440 return format['resolution']
35615307 3441 if format.get('width') and format.get('height'):
ff51ed58 3442 return '%dx%d' % (format['width'], format['height'])
35615307 3443 elif format.get('height'):
ff51ed58 3444 return '%sp' % format['height']
35615307 3445 elif format.get('width'):
ff51ed58 3446 return '%dx?' % format['width']
3447 return default
8c51aa65 3448
8130779d 3449 def _list_format_headers(self, *headers):
3450 if self.params.get('listformats_table', True) is not False:
591bb9d3 3451 return [self._format_out(header, self.Styles.HEADERS) for header in headers]
8130779d 3452 return headers
3453
c57f7757
PH
3454 def _format_note(self, fdict):
3455 res = ''
3456 if fdict.get('ext') in ['f4f', 'f4m']:
f304da8a 3457 res += '(unsupported)'
32f90364
PH
3458 if fdict.get('language'):
3459 if res:
3460 res += ' '
f304da8a 3461 res += '[%s]' % fdict['language']
c57f7757 3462 if fdict.get('format_note') is not None:
f304da8a 3463 if res:
3464 res += ' '
3465 res += fdict['format_note']
c57f7757 3466 if fdict.get('tbr') is not None:
f304da8a 3467 if res:
3468 res += ', '
3469 res += '%4dk' % fdict['tbr']
c57f7757
PH
3470 if fdict.get('container') is not None:
3471 if res:
3472 res += ', '
3473 res += '%s container' % fdict['container']
3089bc74
S
3474 if (fdict.get('vcodec') is not None
3475 and fdict.get('vcodec') != 'none'):
c57f7757
PH
3476 if res:
3477 res += ', '
3478 res += fdict['vcodec']
91c7271a 3479 if fdict.get('vbr') is not None:
c57f7757
PH
3480 res += '@'
3481 elif fdict.get('vbr') is not None and fdict.get('abr') is not None:
3482 res += 'video@'
3483 if fdict.get('vbr') is not None:
3484 res += '%4dk' % fdict['vbr']
fbb21cf5 3485 if fdict.get('fps') is not None:
5d583bdf
S
3486 if res:
3487 res += ', '
3488 res += '%sfps' % fdict['fps']
c57f7757
PH
3489 if fdict.get('acodec') is not None:
3490 if res:
3491 res += ', '
3492 if fdict['acodec'] == 'none':
3493 res += 'video only'
3494 else:
3495 res += '%-5s' % fdict['acodec']
3496 elif fdict.get('abr') is not None:
3497 if res:
3498 res += ', '
3499 res += 'audio'
3500 if fdict.get('abr') is not None:
3501 res += '@%3dk' % fdict['abr']
3502 if fdict.get('asr') is not None:
3503 res += ' (%5dHz)' % fdict['asr']
3504 if fdict.get('filesize') is not None:
3505 if res:
3506 res += ', '
3507 res += format_bytes(fdict['filesize'])
9732d77e
PH
3508 elif fdict.get('filesize_approx') is not None:
3509 if res:
3510 res += ', '
3511 res += '~' + format_bytes(fdict['filesize_approx'])
c57f7757 3512 return res
91c7271a 3513
8130779d 3514 def render_formats_table(self, info_dict):
b69fd25c 3515 if not info_dict.get('formats') and not info_dict.get('url'):
8130779d 3516 return None
b69fd25c 3517
94badb25 3518 formats = info_dict.get('formats', [info_dict])
8130779d 3519 if not self.params.get('listformats_table', True) is not False:
76d321f6 3520 table = [
3521 [
3522 format_field(f, 'format_id'),
3523 format_field(f, 'ext'),
3524 self.format_resolution(f),
8130779d 3525 self._format_note(f)
3526 ] for f in formats if f.get('preference') is None or f['preference'] >= -1000]
3527 return render_table(['format code', 'extension', 'resolution', 'note'], table, extra_gap=1)
3528
591bb9d3 3529 delim = self._format_out('\u2502', self.Styles.DELIM, '|', test_encoding=True)
8130779d 3530 table = [
3531 [
591bb9d3 3532 self._format_out(format_field(f, 'format_id'), self.Styles.ID),
8130779d 3533 format_field(f, 'ext'),
3534 format_field(f, func=self.format_resolution, ignore=('audio only', 'images')),
3535 format_field(f, 'fps', '\t%d'),
3536 format_field(f, 'dynamic_range', '%s', ignore=(None, 'SDR')).replace('HDR', ''),
3537 delim,
3538 format_field(f, 'filesize', ' \t%s', func=format_bytes) + format_field(f, 'filesize_approx', '~\t%s', func=format_bytes),
3539 format_field(f, 'tbr', '\t%dk'),
3540 shorten_protocol_name(f.get('protocol', '')),
3541 delim,
3542 format_field(f, 'vcodec', default='unknown').replace(
3543 'none', 'images' if f.get('acodec') == 'none'
591bb9d3 3544 else self._format_out('audio only', self.Styles.SUPPRESS)),
8130779d 3545 format_field(f, 'vbr', '\t%dk'),
3546 format_field(f, 'acodec', default='unknown').replace(
3547 'none', '' if f.get('vcodec') == 'none'
591bb9d3 3548 else self._format_out('video only', self.Styles.SUPPRESS)),
8130779d 3549 format_field(f, 'abr', '\t%dk'),
3550 format_field(f, 'asr', '\t%dHz'),
3551 join_nonempty(
591bb9d3 3552 self._format_out('UNSUPPORTED', 'light red') if f.get('ext') in ('f4f', 'f4m') else None,
8130779d 3553 format_field(f, 'language', '[%s]'),
3554 join_nonempty(format_field(f, 'format_note'),
3555 format_field(f, 'container', ignore=(None, f.get('ext'))),
3556 delim=', '),
3557 delim=' '),
3558 ] for f in formats if f.get('preference') is None or f['preference'] >= -1000]
3559 header_line = self._list_format_headers(
3560 'ID', 'EXT', 'RESOLUTION', '\tFPS', 'HDR', delim, '\tFILESIZE', '\tTBR', 'PROTO',
3561 delim, 'VCODEC', '\tVBR', 'ACODEC', '\tABR', '\tASR', 'MORE INFO')
3562
3563 return render_table(
3564 header_line, table, hide_empty=True,
591bb9d3 3565 delim=self._format_out('\u2500', self.Styles.DELIM, '-', test_encoding=True))
8130779d 3566
3567 def render_thumbnails_table(self, info_dict):
88f23a18 3568 thumbnails = list(info_dict.get('thumbnails') or [])
cfb56d1a 3569 if not thumbnails:
8130779d 3570 return None
3571 return render_table(
ec11a9f4 3572 self._list_format_headers('ID', 'Width', 'Height', 'URL'),
6970b600 3573 [[t.get('id'), t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails])
2412044c 3574
8130779d 3575 def render_subtitles_table(self, video_id, subtitles):
2412044c 3576 def _row(lang, formats):
49c258e1 3577 exts, names = zip(*((f['ext'], f.get('name') or 'unknown') for f in reversed(formats)))
2412044c 3578 if len(set(names)) == 1:
7aee40c1 3579 names = [] if names[0] == 'unknown' else names[:1]
2412044c 3580 return [lang, ', '.join(names), ', '.join(exts)]
3581
8130779d 3582 if not subtitles:
3583 return None
3584 return render_table(
ec11a9f4 3585 self._list_format_headers('Language', 'Name', 'Formats'),
2412044c 3586 [_row(lang, formats) for lang, formats in subtitles.items()],
8130779d 3587 hide_empty=True)
3588
3589 def __list_table(self, video_id, name, func, *args):
3590 table = func(*args)
3591 if not table:
3592 self.to_screen(f'{video_id} has no {name}')
3593 return
3594 self.to_screen(f'[info] Available {name} for {video_id}:')
3595 self.to_stdout(table)
3596
3597 def list_formats(self, info_dict):
3598 self.__list_table(info_dict['id'], 'formats', self.render_formats_table, info_dict)
3599
3600 def list_thumbnails(self, info_dict):
3601 self.__list_table(info_dict['id'], 'thumbnails', self.render_thumbnails_table, info_dict)
3602
3603 def list_subtitles(self, video_id, subtitles, name='subtitles'):
3604 self.__list_table(video_id, name, self.render_subtitles_table, video_id, subtitles)
a504ced0 3605
dca08720
PH
3606 def urlopen(self, req):
3607 """ Start an HTTP download """
f9934b96 3608 if isinstance(req, str):
67dda517 3609 req = sanitized_Request(req)
19a41fc6 3610 return self._opener.open(req, timeout=self._socket_timeout)
dca08720
PH
3611
3612 def print_debug_header(self):
3613 if not self.params.get('verbose'):
3614 return
49a57e70 3615
3616 def get_encoding(stream):
2a938746 3617 ret = str(getattr(stream, 'encoding', 'missing (%s)' % type(stream).__name__))
49a57e70 3618 if not supports_terminal_sequences(stream):
53973b4d 3619 from .utils import WINDOWS_VT_MODE # Must be imported locally
e3c7d495 3620 ret += ' (No VT)' if WINDOWS_VT_MODE is False else ' (No ANSI)'
49a57e70 3621 return ret
3622
591bb9d3 3623 encoding_str = 'Encodings: locale %s, fs %s, pref %s, %s' % (
49a57e70 3624 locale.getpreferredencoding(),
3625 sys.getfilesystemencoding(),
591bb9d3 3626 self.get_encoding(),
3627 ', '.join(
7896214c 3628 f'{key} {get_encoding(stream)}' for key, stream in self._out_files
591bb9d3 3629 if stream is not None and key != 'console')
3630 )
883d4b1e 3631
3632 logger = self.params.get('logger')
3633 if logger:
3634 write_debug = lambda msg: logger.debug(f'[debug] {msg}')
3635 write_debug(encoding_str)
3636 else:
96565c7e 3637 write_string(f'[debug] {encoding_str}\n', encoding=None)
49a57e70 3638 write_debug = lambda msg: self._write_string(f'[debug] {msg}\n')
734f90bb 3639
4c88ff87 3640 source = detect_variant()
36eaf303 3641 write_debug(join_nonempty(
3642 'yt-dlp version', __version__,
3643 f'[{RELEASE_GIT_HEAD}]' if RELEASE_GIT_HEAD else '',
3644 '' if source == 'unknown' else f'({source})',
3645 delim=' '))
6e21fdd2 3646 if not _LAZY_LOADER:
3647 if os.environ.get('YTDLP_NO_LAZY_EXTRACTORS'):
49a57e70 3648 write_debug('Lazy loading extractors is forcibly disabled')
6e21fdd2 3649 else:
49a57e70 3650 write_debug('Lazy loading extractors is disabled')
3ae5e797 3651 if plugin_extractors or plugin_postprocessors:
49a57e70 3652 write_debug('Plugins: %s' % [
3ae5e797 3653 '%s%s' % (klass.__name__, '' if klass.__name__ == name else f' as {name}')
3654 for name, klass in itertools.chain(plugin_extractors.items(), plugin_postprocessors.items())])
53ed7066 3655 if self.params.get('compat_opts'):
49a57e70 3656 write_debug('Compatibility options: %s' % ', '.join(self.params.get('compat_opts')))
36eaf303 3657
3658 if source == 'source':
dca08720 3659 try:
36eaf303 3660 sp = Popen(
3661 ['git', 'rev-parse', '--short', 'HEAD'],
3662 stdout=subprocess.PIPE, stderr=subprocess.PIPE,
3663 cwd=os.path.dirname(os.path.abspath(__file__)))
3664 out, err = sp.communicate_or_kill()
3665 out = out.decode().strip()
3666 if re.match('[0-9a-f]+', out):
3667 write_debug('Git HEAD: %s' % out)
70a1165b 3668 except Exception:
19a03940 3669 with contextlib.suppress(Exception):
36eaf303 3670 sys.exc_clear()
b300cda4
S
3671
3672 def python_implementation():
3673 impl_name = platform.python_implementation()
3674 if impl_name == 'PyPy' and hasattr(sys, 'pypy_version_info'):
3675 return impl_name + ' version %d.%d.%d' % sys.pypy_version_info[:3]
3676 return impl_name
3677
49a57e70 3678 write_debug('Python version %s (%s %s) - %s' % (
e5813e53 3679 platform.python_version(),
3680 python_implementation(),
3681 platform.architecture()[0],
b300cda4 3682 platform_name()))
d28b5171 3683
8913ef74 3684 exe_versions, ffmpeg_features = FFmpegPostProcessor.get_versions_and_features(self)
3685 ffmpeg_features = {key for key, val in ffmpeg_features.items() if val}
3686 if ffmpeg_features:
19a03940 3687 exe_versions['ffmpeg'] += ' (%s)' % ','.join(sorted(ffmpeg_features))
8913ef74 3688
4c83c967 3689 exe_versions['rtmpdump'] = rtmpdump_version()
feee8d32 3690 exe_versions['phantomjs'] = PhantomJSwrapper._version()
d28b5171 3691 exe_str = ', '.join(
2831b468 3692 f'{exe} {v}' for exe, v in sorted(exe_versions.items()) if v
3693 ) or 'none'
49a57e70 3694 write_debug('exe versions: %s' % exe_str)
dca08720 3695
1d485a1a 3696 from .compat.compat_utils import get_package_info
9b8ee23b 3697 from .dependencies import available_dependencies
3698
3699 write_debug('Optional libraries: %s' % (', '.join(sorted({
1d485a1a 3700 join_nonempty(*get_package_info(m)) for m in available_dependencies.values()
9b8ee23b 3701 })) or 'none'))
2831b468 3702
97ec5bc5 3703 self._setup_opener()
dca08720
PH
3704 proxy_map = {}
3705 for handler in self._opener.handlers:
3706 if hasattr(handler, 'proxies'):
3707 proxy_map.update(handler.proxies)
49a57e70 3708 write_debug(f'Proxy map: {proxy_map}')
dca08720 3709
49a57e70 3710 # Not implemented
3711 if False and self.params.get('call_home'):
0f06bcd7 3712 ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode()
49a57e70 3713 write_debug('Public IP address: %s' % ipaddr)
58b1f00d 3714 latest_version = self.urlopen(
0f06bcd7 3715 'https://yt-dl.org/latest/version').read().decode()
58b1f00d
PH
3716 if version_tuple(latest_version) > version_tuple(__version__):
3717 self.report_warning(
3718 'You are using an outdated version (newest version: %s)! '
3719 'See https://yt-dl.org/update if you need help updating.' %
3720 latest_version)
3721
e344693b 3722 def _setup_opener(self):
97ec5bc5 3723 if hasattr(self, '_opener'):
3724 return
6ad14cab 3725 timeout_val = self.params.get('socket_timeout')
17bddf3e 3726 self._socket_timeout = 20 if timeout_val is None else float(timeout_val)
6ad14cab 3727
982ee69a 3728 opts_cookiesfrombrowser = self.params.get('cookiesfrombrowser')
dca08720
PH
3729 opts_cookiefile = self.params.get('cookiefile')
3730 opts_proxy = self.params.get('proxy')
3731
982ee69a 3732 self.cookiejar = load_cookies(opts_cookiefile, opts_cookiesfrombrowser, self)
dca08720 3733
6a3f4c3f 3734 cookie_processor = YoutubeDLCookieProcessor(self.cookiejar)
dca08720
PH
3735 if opts_proxy is not None:
3736 if opts_proxy == '':
3737 proxies = {}
3738 else:
3739 proxies = {'http': opts_proxy, 'https': opts_proxy}
3740 else:
3741 proxies = compat_urllib_request.getproxies()
067aa17e 3742 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
dca08720
PH
3743 if 'http' in proxies and 'https' not in proxies:
3744 proxies['https'] = proxies['http']
91410c9b 3745 proxy_handler = PerRequestProxyHandler(proxies)
a0ddb8a2
PH
3746
3747 debuglevel = 1 if self.params.get('debug_printtraffic') else 0
be4a824d
PH
3748 https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel)
3749 ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel)
fca6dba8 3750 redirect_handler = YoutubeDLRedirectHandler()
f9934b96 3751 data_handler = urllib.request.DataHandler()
6240b0a2
JMF
3752
3753 # When passing our own FileHandler instance, build_opener won't add the
3754 # default FileHandler and allows us to disable the file protocol, which
3755 # can be used for malicious purposes (see
067aa17e 3756 # https://github.com/ytdl-org/youtube-dl/issues/8227)
6240b0a2
JMF
3757 file_handler = compat_urllib_request.FileHandler()
3758
3759 def file_open(*args, **kwargs):
7a5c1cfe 3760 raise compat_urllib_error.URLError('file:// scheme is explicitly disabled in yt-dlp for security reasons')
6240b0a2
JMF
3761 file_handler.file_open = file_open
3762
3763 opener = compat_urllib_request.build_opener(
fca6dba8 3764 proxy_handler, https_handler, cookie_processor, ydlh, redirect_handler, data_handler, file_handler)
2461f79d 3765
dca08720
PH
3766 # Delete the default user-agent header, which would otherwise apply in
3767 # cases where our custom HTTP handler doesn't come into play
067aa17e 3768 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
dca08720
PH
3769 opener.addheaders = []
3770 self._opener = opener
62fec3b2
PH
3771
3772 def encode(self, s):
3773 if isinstance(s, bytes):
3774 return s # Already encoded
3775
3776 try:
3777 return s.encode(self.get_encoding())
3778 except UnicodeEncodeError as err:
3779 err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.'
3780 raise
3781
3782 def get_encoding(self):
3783 encoding = self.params.get('encoding')
3784 if encoding is None:
3785 encoding = preferredencoding()
3786 return encoding
ec82d85a 3787
e08a85d8 3788 def _write_info_json(self, label, ie_result, infofn, overwrite=None):
cb96c5be 3789 ''' Write infojson and returns True = written, 'exists' = Already exists, False = skip, None = error '''
e08a85d8 3790 if overwrite is None:
3791 overwrite = self.params.get('overwrites', True)
80c03fa9 3792 if not self.params.get('writeinfojson'):
3793 return False
3794 elif not infofn:
3795 self.write_debug(f'Skipping writing {label} infojson')
3796 return False
3797 elif not self._ensure_dir_exists(infofn):
3798 return None
e08a85d8 3799 elif not overwrite and os.path.exists(infofn):
80c03fa9 3800 self.to_screen(f'[info] {label.title()} metadata is already present')
cb96c5be 3801 return 'exists'
3802
3803 self.to_screen(f'[info] Writing {label} metadata as JSON to: {infofn}')
3804 try:
3805 write_json_file(self.sanitize_info(ie_result, self.params.get('clean_infojson', True)), infofn)
3806 return True
86e5f3ed 3807 except OSError:
cb96c5be 3808 self.report_error(f'Cannot write {label} metadata to JSON file {infofn}')
3809 return None
80c03fa9 3810
3811 def _write_description(self, label, ie_result, descfn):
3812 ''' Write description and returns True = written, False = skip, None = error '''
3813 if not self.params.get('writedescription'):
3814 return False
3815 elif not descfn:
3816 self.write_debug(f'Skipping writing {label} description')
3817 return False
3818 elif not self._ensure_dir_exists(descfn):
3819 return None
3820 elif not self.params.get('overwrites', True) and os.path.exists(descfn):
3821 self.to_screen(f'[info] {label.title()} description is already present')
3822 elif ie_result.get('description') is None:
3823 self.report_warning(f'There\'s no {label} description to write')
3824 return False
3825 else:
3826 try:
3827 self.to_screen(f'[info] Writing {label} description to: {descfn}')
86e5f3ed 3828 with open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
80c03fa9 3829 descfile.write(ie_result['description'])
86e5f3ed 3830 except OSError:
80c03fa9 3831 self.report_error(f'Cannot write {label} description file {descfn}')
3832 return None
3833 return True
3834
3835 def _write_subtitles(self, info_dict, filename):
3836 ''' Write subtitles to file and return list of (sub_filename, final_sub_filename); or None if error'''
3837 ret = []
3838 subtitles = info_dict.get('requested_subtitles')
3839 if not subtitles or not (self.params.get('writesubtitles') or self.params.get('writeautomaticsub')):
3840 # subtitles download errors are already managed as troubles in relevant IE
3841 # that way it will silently go on when used with unsupporting IE
3842 return ret
3843
3844 sub_filename_base = self.prepare_filename(info_dict, 'subtitle')
3845 if not sub_filename_base:
3846 self.to_screen('[info] Skipping writing video subtitles')
3847 return ret
3848 for sub_lang, sub_info in subtitles.items():
3849 sub_format = sub_info['ext']
3850 sub_filename = subtitles_filename(filename, sub_lang, sub_format, info_dict.get('ext'))
3851 sub_filename_final = subtitles_filename(sub_filename_base, sub_lang, sub_format, info_dict.get('ext'))
e04938ab 3852 existing_sub = self.existing_file((sub_filename_final, sub_filename))
3853 if existing_sub:
80c03fa9 3854 self.to_screen(f'[info] Video subtitle {sub_lang}.{sub_format} is already present')
e04938ab 3855 sub_info['filepath'] = existing_sub
3856 ret.append((existing_sub, sub_filename_final))
80c03fa9 3857 continue
3858
3859 self.to_screen(f'[info] Writing video subtitles to: {sub_filename}')
3860 if sub_info.get('data') is not None:
3861 try:
3862 # Use newline='' to prevent conversion of newline characters
3863 # See https://github.com/ytdl-org/youtube-dl/issues/10268
86e5f3ed 3864 with open(sub_filename, 'w', encoding='utf-8', newline='') as subfile:
80c03fa9 3865 subfile.write(sub_info['data'])
3866 sub_info['filepath'] = sub_filename
3867 ret.append((sub_filename, sub_filename_final))
3868 continue
86e5f3ed 3869 except OSError:
80c03fa9 3870 self.report_error(f'Cannot write video subtitles file {sub_filename}')
3871 return None
3872
3873 try:
3874 sub_copy = sub_info.copy()
3875 sub_copy.setdefault('http_headers', info_dict.get('http_headers'))
3876 self.dl(sub_filename, sub_copy, subtitle=True)
3877 sub_info['filepath'] = sub_filename
3878 ret.append((sub_filename, sub_filename_final))
6020e05d 3879 except (DownloadError, ExtractorError, IOError, OSError, ValueError) + network_exceptions as err:
c70c418d 3880 msg = f'Unable to download video subtitles for {sub_lang!r}: {err}'
6020e05d 3881 if self.params.get('ignoreerrors') is not True: # False or 'only_download'
c70c418d 3882 if not self.params.get('ignoreerrors'):
3883 self.report_error(msg)
3884 raise DownloadError(msg)
3885 self.report_warning(msg)
519804a9 3886 return ret
80c03fa9 3887
3888 def _write_thumbnails(self, label, info_dict, filename, thumb_filename_base=None):
3889 ''' Write thumbnails to file and return list of (thumb_filename, final_thumb_filename) '''
6c4fd172 3890 write_all = self.params.get('write_all_thumbnails', False)
80c03fa9 3891 thumbnails, ret = [], []
6c4fd172 3892 if write_all or self.params.get('writethumbnail', False):
0202b52a 3893 thumbnails = info_dict.get('thumbnails') or []
6c4fd172 3894 multiple = write_all and len(thumbnails) > 1
ec82d85a 3895
80c03fa9 3896 if thumb_filename_base is None:
3897 thumb_filename_base = filename
3898 if thumbnails and not thumb_filename_base:
3899 self.write_debug(f'Skipping writing {label} thumbnail')
3900 return ret
3901
dd0228ce 3902 for idx, t in list(enumerate(thumbnails))[::-1]:
80c03fa9 3903 thumb_ext = (f'{t["id"]}.' if multiple else '') + determine_ext(t['url'], 'jpg')
aa9369a2 3904 thumb_display_id = f'{label} thumbnail {t["id"]}'
80c03fa9 3905 thumb_filename = replace_extension(filename, thumb_ext, info_dict.get('ext'))
3906 thumb_filename_final = replace_extension(thumb_filename_base, thumb_ext, info_dict.get('ext'))
ec82d85a 3907
e04938ab 3908 existing_thumb = self.existing_file((thumb_filename_final, thumb_filename))
3909 if existing_thumb:
aa9369a2 3910 self.to_screen('[info] %s is already present' % (
3911 thumb_display_id if multiple else f'{label} thumbnail').capitalize())
e04938ab 3912 t['filepath'] = existing_thumb
3913 ret.append((existing_thumb, thumb_filename_final))
ec82d85a 3914 else:
80c03fa9 3915 self.to_screen(f'[info] Downloading {thumb_display_id} ...')
ec82d85a 3916 try:
297e9952 3917 uf = self.urlopen(sanitized_Request(t['url'], headers=t.get('http_headers', {})))
80c03fa9 3918 self.to_screen(f'[info] Writing {thumb_display_id} to: {thumb_filename}')
d3d89c32 3919 with open(encodeFilename(thumb_filename), 'wb') as thumbf:
ec82d85a 3920 shutil.copyfileobj(uf, thumbf)
80c03fa9 3921 ret.append((thumb_filename, thumb_filename_final))
885cc0b7 3922 t['filepath'] = thumb_filename
3158150c 3923 except network_exceptions as err:
dd0228ce 3924 thumbnails.pop(idx)
80c03fa9 3925 self.report_warning(f'Unable to download {thumb_display_id}: {err}')
6c4fd172 3926 if ret and not write_all:
3927 break
0202b52a 3928 return ret