]> jfr.im git - yt-dlp.git/blame - yt_dlp/YoutubeDL.py
[cleanup] Misc fixes and cleanup
[yt-dlp.git] / yt_dlp / YoutubeDL.py
CommitLineData
cc52de43 1#!/usr/bin/env python3
26e63931 2import collections
31bd3925 3import contextlib
9d2ecdbc 4import datetime
c1c9a79c 5import errno
31bd3925 6import fileinput
b5ae35ee 7import functools
8222d8de 8import io
b82f815f 9import itertools
8694c600 10import json
62fec3b2 11import locale
083c9df9 12import operator
8222d8de 13import os
dca08720 14import platform
f8271158 15import random
8222d8de
JMF
16import re
17import shutil
dca08720 18import subprocess
8222d8de 19import sys
21cd8fae 20import tempfile
8222d8de 21import time
67134eab 22import tokenize
8222d8de 23import traceback
524e2e4f 24import unicodedata
f9934b96 25import urllib.request
961ea474
S
26from string import ascii_letters
27
f8271158 28from .cache import Cache
8c25f81b 29from .compat import (
8a82af35 30 HAS_LEGACY as compat_has_legacy,
003c69a8 31 compat_get_terminal_size,
e9c0cdd3 32 compat_os_name,
7d1eb38a 33 compat_shlex_quote,
ce02ed60
PH
34 compat_str,
35 compat_urllib_error,
36 compat_urllib_request,
8c25f81b 37)
982ee69a 38from .cookies import load_cookies
f8271158 39from .downloader import FFmpegFD, get_suitable_downloader, shorten_protocol_name
40from .downloader.rtmp import rtmpdump_version
41from .extractor import _LAZY_LOADER
42from .extractor import _PLUGIN_CLASSES as plugin_extractors
43from .extractor import gen_extractor_classes, get_info_extractor
44from .extractor.openload import PhantomJSwrapper
45from .minicurses import format_text
46from .postprocessor import _PLUGIN_CLASSES as plugin_postprocessors
47from .postprocessor import (
48 EmbedThumbnailPP,
49 FFmpegFixupDuplicateMoovPP,
50 FFmpegFixupDurationPP,
51 FFmpegFixupM3u8PP,
52 FFmpegFixupM4aPP,
53 FFmpegFixupStretchedPP,
54 FFmpegFixupTimestampPP,
55 FFmpegMergerPP,
56 FFmpegPostProcessor,
57 MoveFilesAfterDownloadPP,
58 get_postprocessor,
59)
60from .update import detect_variant
8c25f81b 61from .utils import (
f8271158 62 DEFAULT_OUTTMPL,
63 LINK_TEMPLATES,
64 NO_DEFAULT,
1d485a1a 65 NUMBER_RE,
f8271158 66 OUTTMPL_TYPES,
67 POSTPROCESS_WHEN,
68 STR_FORMAT_RE_TMPL,
69 STR_FORMAT_TYPES,
70 ContentTooShortError,
71 DateRange,
72 DownloadCancelled,
73 DownloadError,
74 EntryNotInPlaylist,
75 ExistingVideoReached,
76 ExtractorError,
77 GeoRestrictedError,
78 HEADRequest,
79 InAdvancePagedList,
80 ISO3166Utils,
81 LazyList,
82 MaxDownloadsReached,
19a03940 83 Namespace,
f8271158 84 PagedList,
85 PerRequestProxyHandler,
86 Popen,
87 PostProcessingError,
88 ReExtractInfo,
89 RejectedVideoReached,
90 SameFileError,
91 UnavailableVideoError,
92 YoutubeDLCookieProcessor,
93 YoutubeDLHandler,
94 YoutubeDLRedirectHandler,
eedb7ba5
S
95 age_restricted,
96 args_to_str,
ce02ed60 97 date_from_str,
ce02ed60 98 determine_ext,
b5559424 99 determine_protocol,
c0384f22 100 encode_compat_str,
ce02ed60 101 encodeFilename,
a06916d9 102 error_to_compat_str,
590bc6f6 103 expand_path,
90137ca4 104 filter_dict,
e29663c6 105 float_or_none,
02dbf93f 106 format_bytes,
e0fd9573 107 format_decimal_suffix,
f8271158 108 format_field,
525ef922 109 formatSeconds,
0bb322b9 110 get_domain,
c9969434 111 int_or_none,
732044af 112 iri_to_uri,
34921b43 113 join_nonempty,
ce02ed60 114 locked_file,
0202b52a 115 make_dir,
dca08720 116 make_HTTPS_handler,
8b7539d2 117 merge_headers,
3158150c 118 network_exceptions,
ec11a9f4 119 number_of_digits,
cd6fc19e 120 orderedSet,
083c9df9 121 parse_filesize,
dca08720 122 platform_name,
ce02ed60 123 preferredencoding,
eedb7ba5 124 prepend_extension,
51fb4995 125 register_socks_protocols,
3efb96a6 126 remove_terminal_sequences,
cfb56d1a 127 render_table,
eedb7ba5 128 replace_extension,
ce02ed60 129 sanitize_filename,
1bb5c511 130 sanitize_path,
dcf77cf1 131 sanitize_url,
67dda517 132 sanitized_Request,
e5660ee6 133 std_headers,
1211bb6d 134 str_or_none,
e29663c6 135 strftime_or_none,
ce02ed60 136 subtitles_filename,
819e0531 137 supports_terminal_sequences,
f2ebc5c7 138 timetuple_from_msec,
732044af 139 to_high_limit_path,
324ad820 140 traverse_obj,
6033d980 141 try_get,
29eb5174 142 url_basename,
7d1eb38a 143 variadic,
58b1f00d 144 version_tuple,
53973b4d 145 windows_enable_vt_mode,
ce02ed60
PH
146 write_json_file,
147 write_string,
4f026faf 148)
f8271158 149from .version import RELEASE_GIT_HEAD, __version__
8222d8de 150
e9c0cdd3
YCH
151if compat_os_name == 'nt':
152 import ctypes
153
2459b6e1 154
86e5f3ed 155class YoutubeDL:
8222d8de
JMF
156 """YoutubeDL class.
157
158 YoutubeDL objects are the ones responsible of downloading the
159 actual video file and writing it to disk if the user has requested
160 it, among some other tasks. In most cases there should be one per
161 program. As, given a video URL, the downloader doesn't know how to
162 extract all the needed information, task that InfoExtractors do, it
163 has to pass the URL to one of them.
164
165 For this, YoutubeDL objects have a method that allows
166 InfoExtractors to be registered in a given order. When it is passed
167 a URL, the YoutubeDL object handles it to the first InfoExtractor it
168 finds that reports being able to handle it. The InfoExtractor extracts
169 all the information about the video or videos the URL refers to, and
170 YoutubeDL process the extracted information, possibly using a File
171 Downloader to download the video.
172
173 YoutubeDL objects accept a lot of parameters. In order not to saturate
174 the object constructor with arguments, it receives a dictionary of
175 options instead. These options are available through the params
176 attribute for the InfoExtractors to use. The YoutubeDL also
177 registers itself as the downloader in charge for the InfoExtractors
178 that are added to it, so this is a "mutual registration".
179
180 Available options:
181
182 username: Username for authentication purposes.
183 password: Password for authentication purposes.
180940e0 184 videopassword: Password for accessing a video.
1da50aa3
S
185 ap_mso: Adobe Pass multiple-system operator identifier.
186 ap_username: Multiple-system operator account username.
187 ap_password: Multiple-system operator account password.
8222d8de
JMF
188 usenetrc: Use netrc for authentication instead.
189 verbose: Print additional info to stdout.
190 quiet: Do not print messages to stdout.
ad8915b7 191 no_warnings: Do not print out anything for warnings.
bb66c247 192 forceprint: A dict with keys WHEN mapped to a list of templates to
193 print to stdout. The allowed keys are video or any of the
194 items in utils.POSTPROCESS_WHEN.
ca30f449 195 For compatibility, a single list is also accepted
bb66c247 196 print_to_file: A dict with keys WHEN (same as forceprint) mapped to
197 a list of tuples with (template, filename)
53c18592 198 forceurl: Force printing final URL. (Deprecated)
199 forcetitle: Force printing title. (Deprecated)
200 forceid: Force printing ID. (Deprecated)
201 forcethumbnail: Force printing thumbnail URL. (Deprecated)
202 forcedescription: Force printing description. (Deprecated)
203 forcefilename: Force printing final filename. (Deprecated)
204 forceduration: Force printing duration. (Deprecated)
8694c600 205 forcejson: Force printing info_dict as JSON.
63e0be34
PH
206 dump_single_json: Force printing the info_dict of the whole playlist
207 (or video) as a single JSON line.
c25228e5 208 force_write_download_archive: Force writing download archive regardless
209 of 'skip_download' or 'simulate'.
b7b04c78 210 simulate: Do not download the video files. If unset (or None),
211 simulate only if listsubtitles, listformats or list_thumbnails is used
eb8a4433 212 format: Video format code. see "FORMAT SELECTION" for more details.
093a1710 213 You can also pass a function. The function takes 'ctx' as
214 argument and returns the formats to download.
215 See "build_format_selector" for an implementation
63ad4d43 216 allow_unplayable_formats: Allow unplayable formats to be extracted and downloaded.
b7da73eb 217 ignore_no_formats_error: Ignore "No video formats" error. Usefull for
218 extracting metadata even if the video is not actually
219 available for download (experimental)
0930b11f 220 format_sort: A list of fields by which to sort the video formats.
221 See "Sorting Formats" for more details.
c25228e5 222 format_sort_force: Force the given format_sort. see "Sorting Formats"
223 for more details.
08d30158 224 prefer_free_formats: Whether to prefer video formats with free containers
225 over non-free ones of same quality.
c25228e5 226 allow_multiple_video_streams: Allow multiple video streams to be merged
227 into a single file
228 allow_multiple_audio_streams: Allow multiple audio streams to be merged
229 into a single file
0ba692ac 230 check_formats Whether to test if the formats are downloadable.
9f1a1c36 231 Can be True (check all), False (check none),
232 'selected' (check selected formats),
0ba692ac 233 or None (check only if requested by extractor)
4524baf0 234 paths: Dictionary of output paths. The allowed keys are 'home'
235 'temp' and the keys of OUTTMPL_TYPES (in utils.py)
de6000d9 236 outtmpl: Dictionary of templates for output names. Allowed keys
4524baf0 237 are 'default' and the keys of OUTTMPL_TYPES (in utils.py).
34488702 238 For compatibility with youtube-dl, a single string can also be used
a820dc72
RA
239 outtmpl_na_placeholder: Placeholder for unavailable meta fields.
240 restrictfilenames: Do not allow "&" and spaces in file names
241 trim_file_name: Limit length of filename (extension excluded)
4524baf0 242 windowsfilenames: Force the filenames to be windows compatible
b1940459 243 ignoreerrors: Do not stop on download/postprocessing errors.
244 Can be 'only_download' to ignore only download errors.
245 Default is 'only_download' for CLI, but False for API
26e2805c 246 skip_playlist_after_errors: Number of allowed failures until the rest of
247 the playlist is skipped
d22dec74 248 force_generic_extractor: Force downloader to use the generic extractor
0c3d0f51 249 overwrites: Overwrite all video and metadata files if True,
250 overwrite only non-video files if None
251 and don't overwrite any file if False
34488702 252 For compatibility with youtube-dl,
253 "nooverwrites" may also be used instead
8222d8de
JMF
254 playliststart: Playlist item to start at.
255 playlistend: Playlist item to end at.
c14e88f0 256 playlist_items: Specific indices of playlist to download.
ff815fe6 257 playlistreverse: Download playlist items in reverse order.
75822ca7 258 playlistrandom: Download playlist items in random order.
8222d8de
JMF
259 matchtitle: Download only matching titles.
260 rejecttitle: Reject downloads for matching titles.
8bf9319e 261 logger: Log messages to a logging.Logger instance.
8222d8de 262 logtostderr: Log messages to stderr instead of stdout.
819e0531 263 consoletitle: Display progress in console window's titlebar.
8222d8de
JMF
264 writedescription: Write the video description to a .description file
265 writeinfojson: Write the video description to a .info.json file
75d43ca0 266 clean_infojson: Remove private fields from the infojson
34488702 267 getcomments: Extract video comments. This will not be written to disk
06167fbb 268 unless writeinfojson is also given
1fb07d10 269 writeannotations: Write the video annotations to a .annotations.xml file
8222d8de 270 writethumbnail: Write the thumbnail image to a file
c25228e5 271 allow_playlist_files: Whether to write playlists' description, infojson etc
272 also to disk when using the 'write*' options
ec82d85a 273 write_all_thumbnails: Write all thumbnail formats to files
732044af 274 writelink: Write an internet shortcut file, depending on the
275 current platform (.url/.webloc/.desktop)
276 writeurllink: Write a Windows internet shortcut file (.url)
277 writewebloclink: Write a macOS internet shortcut file (.webloc)
278 writedesktoplink: Write a Linux internet shortcut file (.desktop)
8222d8de 279 writesubtitles: Write the video subtitles to a file
741dd8ea 280 writeautomaticsub: Write the automatically generated subtitles to a file
245524e6 281 allsubtitles: Deprecated - Use subtitleslangs = ['all']
c32b0aab 282 Downloads all the subtitles of the video
0b7f3118 283 (requires writesubtitles or writeautomaticsub)
8222d8de 284 listsubtitles: Lists all available subtitles for the video
a504ced0 285 subtitlesformat: The format code for subtitles
c32b0aab 286 subtitleslangs: List of languages of the subtitles to download (can be regex).
287 The list may contain "all" to refer to all the available
288 subtitles. The language can be prefixed with a "-" to
289 exclude it from the requested languages. Eg: ['all', '-live_chat']
8222d8de
JMF
290 keepvideo: Keep the video file after post-processing
291 daterange: A DateRange object, download only if the upload_date is in the range.
292 skip_download: Skip the actual download of the video file
c35f9e72 293 cachedir: Location of the cache files in the filesystem.
a0e07d31 294 False to disable filesystem cache.
47192f92 295 noplaylist: Download single video instead of a playlist if in doubt.
8dbe9899
PH
296 age_limit: An integer representing the user's age in years.
297 Unsuitable videos for the given age are skipped.
5fe18bdb
PH
298 min_views: An integer representing the minimum view count the video
299 must have in order to not be skipped.
300 Videos without view count information are always
301 downloaded. None for no limit.
302 max_views: An integer representing the maximum view count.
303 Videos that are more popular than that are not
304 downloaded.
305 Videos without view count information are always
306 downloaded. None for no limit.
307 download_archive: File name of a file where all downloads are recorded.
c1c9a79c
PH
308 Videos already present in the file are not downloaded
309 again.
8a51f564 310 break_on_existing: Stop the download process after attempting to download a
311 file that is in the archive.
312 break_on_reject: Stop the download process when encountering a video that
313 has been filtered out.
b222c271 314 break_per_url: Whether break_on_reject and break_on_existing
315 should act on each input URL as opposed to for the entire queue
d76fa1f3 316 cookiefile: File name or text stream from where cookies should be read and dumped to
f59f5ef8
MB
317 cookiesfrombrowser: A tuple containing the name of the browser, the profile
318 name/pathfrom where cookies are loaded, and the name of the
319 keyring. Eg: ('chrome', ) or ('vivaldi', 'default', 'BASICTEXT')
f81c62a6 320 legacyserverconnect: Explicitly allow HTTPS connection to servers that do not
321 support RFC 5746 secure renegotiation
f59f5ef8 322 nocheckcertificate: Do not verify SSL certificates
bb58c9ed 323 client_certificate: Path to client certificate file in PEM format. May include the private key
324 client_certificate_key: Path to private key file for client certificate
325 client_certificate_password: Password for client certificate private key, if encrypted.
326 If not provided and the key is encrypted, yt-dlp will ask interactively
7e8c0af0
PH
327 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
328 At the moment, this is only supported by YouTube.
8b7539d2 329 http_headers: A dictionary of custom headers to be used for all requests
a1ee09e8 330 proxy: URL of the proxy server to use
38cce791 331 geo_verification_proxy: URL of the proxy to use for IP address verification
504f20dd 332 on geo-restricted sites.
e344693b 333 socket_timeout: Time to wait for unresponsive hosts, in seconds
0783b09b
PH
334 bidi_workaround: Work around buggy terminals without bidirectional text
335 support, using fridibi
a0ddb8a2 336 debug_printtraffic:Print out sent and received HTTP traffic
91f071af 337 include_ads: Download ads as well (deprecated)
04b4d394
PH
338 default_search: Prepend this string if an input url is not valid.
339 'auto' for elaborate guessing
62fec3b2 340 encoding: Use this encoding instead of the system-specified.
e8ee972c 341 extract_flat: Do not resolve URLs, return the immediate result.
057a5206
PH
342 Pass in 'in_playlist' to only show this behavior for
343 playlist items.
f2ebc5c7 344 wait_for_video: If given, wait for scheduled streams to become available.
345 The value should be a tuple containing the range
346 (min_secs, max_secs) to wait between retries
4f026faf 347 postprocessors: A list of dictionaries, each with an entry
71b640cc 348 * key: The name of the postprocessor. See
7a5c1cfe 349 yt_dlp/postprocessor/__init__.py for a list.
bb66c247 350 * when: When to run the postprocessor. Allowed values are
351 the entries of utils.POSTPROCESS_WHEN
56d868db 352 Assumed to be 'post_process' if not given
b5ae35ee 353 post_hooks: Deprecated - Register a custom postprocessor instead
354 A list of functions that get called as the final step
ab8e5e51
AM
355 for each video file, after all postprocessors have been
356 called. The filename will be passed as the only argument.
71b640cc
PH
357 progress_hooks: A list of functions that get called on download
358 progress, with a dictionary with the entries
5cda4eda 359 * status: One of "downloading", "error", or "finished".
ee69b99a 360 Check this first and ignore unknown values.
3ba7740d 361 * info_dict: The extracted info_dict
71b640cc 362
5cda4eda 363 If status is one of "downloading", or "finished", the
ee69b99a
PH
364 following properties may also be present:
365 * filename: The final filename (always present)
5cda4eda 366 * tmpfilename: The filename we're currently writing to
71b640cc
PH
367 * downloaded_bytes: Bytes on disk
368 * total_bytes: Size of the whole file, None if unknown
5cda4eda
PH
369 * total_bytes_estimate: Guess of the eventual file size,
370 None if unavailable.
371 * elapsed: The number of seconds since download started.
71b640cc
PH
372 * eta: The estimated time in seconds, None if unknown
373 * speed: The download speed in bytes/second, None if
374 unknown
5cda4eda
PH
375 * fragment_index: The counter of the currently
376 downloaded video fragment.
377 * fragment_count: The number of fragments (= individual
378 files that will be merged)
71b640cc
PH
379
380 Progress hooks are guaranteed to be called at least once
381 (with status "finished") if the download is successful.
819e0531 382 postprocessor_hooks: A list of functions that get called on postprocessing
383 progress, with a dictionary with the entries
384 * status: One of "started", "processing", or "finished".
385 Check this first and ignore unknown values.
386 * postprocessor: Name of the postprocessor
387 * info_dict: The extracted info_dict
388
389 Progress hooks are guaranteed to be called at least twice
390 (with status "started" and "finished") if the processing is successful.
45598f15 391 merge_output_format: Extension to use when merging formats.
6b591b29 392 final_ext: Expected final extension; used to detect when the file was
59a7a13e 393 already downloaded and converted
6271f1ca
PH
394 fixup: Automatically correct known faults of the file.
395 One of:
396 - "never": do nothing
397 - "warn": only emit a warning
398 - "detect_or_warn": check whether we can do anything
62cd676c 399 about it, warn otherwise (default)
504f20dd 400 source_address: Client-side IP address to bind to.
6ec6cb4e 401 call_home: Boolean, true iff we are allowed to contact the
7a5c1cfe 402 yt-dlp servers for debugging. (BROKEN)
1cf376f5 403 sleep_interval_requests: Number of seconds to sleep between requests
404 during extraction
7aa589a5
S
405 sleep_interval: Number of seconds to sleep before each download when
406 used alone or a lower bound of a range for randomized
407 sleep before each download (minimum possible number
408 of seconds to sleep) when used along with
409 max_sleep_interval.
410 max_sleep_interval:Upper bound of a range for randomized sleep before each
411 download (maximum possible number of seconds to sleep).
412 Must only be used along with sleep_interval.
413 Actual sleep time will be a random float from range
414 [sleep_interval; max_sleep_interval].
1cf376f5 415 sleep_interval_subtitles: Number of seconds to sleep before each subtitle download
cfb56d1a
PH
416 listformats: Print an overview of available video formats and exit.
417 list_thumbnails: Print a table of all thumbnails and exit.
0a41f331 418 match_filter: A function that gets called for every video with the signature
419 (info_dict, *, incomplete: bool) -> Optional[str]
420 For backward compatibility with youtube-dl, the signature
421 (info_dict) -> Optional[str] is also allowed.
422 - If it returns a message, the video is ignored.
423 - If it returns None, the video is downloaded.
424 - If it returns utils.NO_DEFAULT, the user is interactively
425 asked whether to download the video.
347de493 426 match_filter_func in utils.py is one example for this.
7e5db8c9 427 no_color: Do not emit color codes in output.
0a840f58 428 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
504f20dd 429 HTTP header
0a840f58 430 geo_bypass_country:
773f291d
S
431 Two-letter ISO 3166-2 country code that will be used for
432 explicit geographic restriction bypassing via faking
504f20dd 433 X-Forwarded-For HTTP header
5f95927a
S
434 geo_bypass_ip_block:
435 IP range in CIDR notation that will be used similarly to
504f20dd 436 geo_bypass_country
71b640cc 437
85729c51 438 The following options determine which downloader is picked:
52a8a1e1 439 external_downloader: A dictionary of protocol keys and the executable of the
440 external downloader to use for it. The allowed protocols
441 are default|http|ftp|m3u8|dash|rtsp|rtmp|mms.
442 Set the value to 'native' to use the native downloader
443 hls_prefer_native: Deprecated - Use external_downloader = {'m3u8': 'native'}
444 or {'m3u8': 'ffmpeg'} instead.
445 Use the native HLS downloader instead of ffmpeg/avconv
bf09af3a
S
446 if True, otherwise use ffmpeg/avconv if False, otherwise
447 use downloader suggested by extractor if None.
53ed7066 448 compat_opts: Compatibility options. See "Differences in default behavior".
3acf6d38 449 The following options do not work when used through the API:
b5ae35ee 450 filename, abort-on-error, multistreams, no-live-chat, format-sort
dac5df5a 451 no-clean-infojson, no-playlist-metafiles, no-keep-subs, no-attach-info-json.
e4f02757 452 Refer __init__.py for their implementation
819e0531 453 progress_template: Dictionary of templates for progress outputs.
454 Allowed keys are 'download', 'postprocess',
455 'download-title' (console title) and 'postprocess-title'.
456 The template is mapped on a dictionary with keys 'progress' and 'info'
23326151 457 retry_sleep_functions: Dictionary of functions that takes the number of attempts
458 as argument and returns the time to sleep in seconds.
459 Allowed keys are 'http', 'fragment', 'file_access'
fe7e0c98 460
8222d8de 461 The following parameters are not used by YoutubeDL itself, they are used by
7a5c1cfe 462 the downloader (see yt_dlp/downloader/common.py):
51d9739f 463 nopart, updatetime, buffersize, ratelimit, throttledratelimit, min_filesize,
205a0654
EH
464 max_filesize, test, noresizebuffer, retries, file_access_retries, fragment_retries,
465 continuedl, noprogress, xattr_set_filesize, hls_use_mpegts, http_chunk_size,
59a7a13e 466 external_downloader_args, concurrent_fragment_downloads.
76b1bd67
JMF
467
468 The following options are used by the post processors:
d4a24f40 469 prefer_ffmpeg: If False, use avconv instead of ffmpeg if both are available,
e4172ac9 470 otherwise prefer ffmpeg. (avconv support is deprecated)
c0b7d117
S
471 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
472 to the binary or its containing directory.
43820c03 473 postprocessor_args: A dictionary of postprocessor/executable keys (in lower case)
34488702 474 and a list of additional command-line arguments for the
475 postprocessor/executable. The dict can also have "PP+EXE" keys
476 which are used when the given exe is used by the given PP.
477 Use 'default' as the name for arguments to passed to all PP
478 For compatibility with youtube-dl, a single list of args
479 can also be used
e409895f 480
481 The following options are used by the extractors:
62bff2c1 482 extractor_retries: Number of times to retry for known errors
483 dynamic_mpd: Whether to process dynamic DASH manifests (default: True)
e409895f 484 hls_split_discontinuity: Split HLS playlists to different formats at
62bff2c1 485 discontinuities such as ad breaks (default: False)
5d3a0e79 486 extractor_args: A dictionary of arguments to be passed to the extractors.
487 See "EXTRACTOR ARGUMENTS" for details.
488 Eg: {'youtube': {'skip': ['dash', 'hls']}}
88f23a18 489 mark_watched: Mark videos watched (even with --simulate). Only for YouTube
5d3a0e79 490 youtube_include_dash_manifest: Deprecated - Use extractor_args instead.
491 If True (default), DASH manifests and related
62bff2c1 492 data will be downloaded and processed by extractor.
493 You can reduce network I/O by disabling it if you don't
494 care about DASH. (only for youtube)
5d3a0e79 495 youtube_include_hls_manifest: Deprecated - Use extractor_args instead.
496 If True (default), HLS manifests and related
62bff2c1 497 data will be downloaded and processed by extractor.
498 You can reduce network I/O by disabling it if you don't
499 care about HLS. (only for youtube)
8222d8de
JMF
500 """
501
86e5f3ed 502 _NUMERIC_FIELDS = {
c9969434 503 'width', 'height', 'tbr', 'abr', 'asr', 'vbr', 'fps', 'filesize', 'filesize_approx',
e6f21b3d 504 'timestamp', 'release_timestamp',
c9969434
S
505 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
506 'average_rating', 'comment_count', 'age_limit',
507 'start_time', 'end_time',
508 'chapter_number', 'season_number', 'episode_number',
509 'track_number', 'disc_number', 'release_year',
86e5f3ed 510 }
c9969434 511
6db9c4d5 512 _format_fields = {
513 # NB: Keep in sync with the docstring of extractor/common.py
a44ca5a4 514 'url', 'manifest_url', 'manifest_stream_number', 'ext', 'format', 'format_id', 'format_note',
6db9c4d5 515 'width', 'height', 'resolution', 'dynamic_range', 'tbr', 'abr', 'acodec', 'asr',
516 'vbr', 'fps', 'vcodec', 'container', 'filesize', 'filesize_approx',
517 'player_url', 'protocol', 'fragment_base_url', 'fragments', 'is_from_start',
518 'preference', 'language', 'language_preference', 'quality', 'source_preference',
519 'http_headers', 'stretched_ratio', 'no_resume', 'has_drm', 'downloader_options',
520 'page_url', 'app', 'play_path', 'tc_url', 'flash_version', 'rtmp_live', 'rtmp_conn', 'rtmp_protocol', 'rtmp_real_time'
521 }
48ee10ee 522 _format_selection_exts = {
523 'audio': {'m4a', 'mp3', 'ogg', 'aac'},
524 'video': {'mp4', 'flv', 'webm', '3gp'},
525 'storyboards': {'mhtml'},
526 }
527
3511266b 528 def __init__(self, params=None, auto_init=True):
883d4b1e 529 """Create a FileDownloader object with the given options.
530 @param auto_init Whether to load the default extractors and print header (if verbose).
49a57e70 531 Set to 'no_verbose_header' to not print the header
883d4b1e 532 """
e9f9a10f
JMF
533 if params is None:
534 params = {}
592b7485 535 self.params = params
8b7491c8 536 self._ies = {}
56c73665 537 self._ies_instances = {}
1e43a6f7 538 self._pps = {k: [] for k in POSTPROCESS_WHEN}
b35496d8 539 self._printed_messages = set()
1cf376f5 540 self._first_webpage_request = True
ab8e5e51 541 self._post_hooks = []
933605d7 542 self._progress_hooks = []
819e0531 543 self._postprocessor_hooks = []
8222d8de
JMF
544 self._download_retcode = 0
545 self._num_downloads = 0
9c906919 546 self._num_videos = 0
592b7485 547 self._playlist_level = 0
548 self._playlist_urls = set()
a0e07d31 549 self.cache = Cache(self)
34308b30 550
819e0531 551 windows_enable_vt_mode()
591bb9d3 552 stdout = sys.stderr if self.params.get('logtostderr') else sys.stdout
553 self._out_files = Namespace(
554 out=stdout,
555 error=sys.stderr,
556 screen=sys.stderr if self.params.get('quiet') else stdout,
557 console=None if compat_os_name == 'nt' else next(
cf4f42cb 558 filter(supports_terminal_sequences, (sys.stderr, sys.stdout)), None)
591bb9d3 559 )
560 self._allow_colors = Namespace(**{
561 type_: not self.params.get('no_color') and supports_terminal_sequences(stream)
7896214c 562 for type_, stream in self._out_files if type_ != 'console'
591bb9d3 563 })
819e0531 564
a61f4b28 565 if sys.version_info < (3, 6):
566 self.report_warning(
0181adef 567 'Python version %d.%d is not supported! Please update to Python 3.6 or above' % sys.version_info[:2])
a61f4b28 568
88acdbc2 569 if self.params.get('allow_unplayable_formats'):
570 self.report_warning(
ec11a9f4 571 f'You have asked for {self._format_err("UNPLAYABLE", self.Styles.EMPHASIS)} formats to be listed/downloaded. '
819e0531 572 'This is a developer option intended for debugging. \n'
573 ' If you experience any issues while using this option, '
ec11a9f4 574 f'{self._format_err("DO NOT", self.Styles.ERROR)} open a bug report')
88acdbc2 575
be5df5ee
S
576 def check_deprecated(param, option, suggestion):
577 if self.params.get(param) is not None:
86e5f3ed 578 self.report_warning(f'{option} is deprecated. Use {suggestion} instead')
be5df5ee
S
579 return True
580 return False
581
582 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
38cce791
YCH
583 if self.params.get('geo_verification_proxy') is None:
584 self.params['geo_verification_proxy'] = self.params['cn_verification_proxy']
585
0d1bb027 586 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
587 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
53ed7066 588 check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"')
0d1bb027 589
49a57e70 590 for msg in self.params.get('_warnings', []):
0d1bb027 591 self.report_warning(msg)
ee8dd27a 592 for msg in self.params.get('_deprecation_warnings', []):
593 self.deprecation_warning(msg)
0d1bb027 594
8a82af35 595 self.params['compat_opts'] = set(self.params.get('compat_opts', ()))
596 if not compat_has_legacy:
597 self.params['compat_opts'].add('no-compat-legacy')
598 if 'list-formats' in self.params['compat_opts']:
ec11a9f4 599 self.params['listformats_table'] = False
600
b5ae35ee 601 if 'overwrites' not in self.params and self.params.get('nooverwrites') is not None:
b868936c 602 # nooverwrites was unnecessarily changed to overwrites
603 # in 0c3d0f51778b153f65c21906031c2e091fcfb641
604 # This ensures compatibility with both keys
605 self.params['overwrites'] = not self.params['nooverwrites']
b5ae35ee 606 elif self.params.get('overwrites') is None:
607 self.params.pop('overwrites', None)
b868936c 608 else:
609 self.params['nooverwrites'] = not self.params['overwrites']
b9d973be 610
455a15e2 611 self.params.setdefault('forceprint', {})
612 self.params.setdefault('print_to_file', {})
bb66c247 613
614 # Compatibility with older syntax
ca30f449 615 if not isinstance(params['forceprint'], dict):
455a15e2 616 self.params['forceprint'] = {'video': params['forceprint']}
ca30f449 617
455a15e2 618 if self.params.get('bidi_workaround', False):
1c088fa8
PH
619 try:
620 import pty
621 master, slave = pty.openpty()
003c69a8 622 width = compat_get_terminal_size().columns
591bb9d3 623 width_args = [] if width is None else ['-w', str(width)]
624 sp_kwargs = {'stdin': subprocess.PIPE, 'stdout': slave, 'stderr': self._out_files.error}
5d681e96 625 try:
d3c93ec2 626 self._output_process = Popen(['bidiv'] + width_args, **sp_kwargs)
5d681e96 627 except OSError:
d3c93ec2 628 self._output_process = Popen(['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
5d681e96 629 self._output_channel = os.fdopen(master, 'rb')
1c088fa8 630 except OSError as ose:
66e7ace1 631 if ose.errno == errno.ENOENT:
49a57e70 632 self.report_warning(
633 'Could not find fribidi executable, ignoring --bidi-workaround. '
634 'Make sure that fribidi is an executable file in one of the directories in your $PATH.')
1c088fa8
PH
635 else:
636 raise
0783b09b 637
97ec5bc5 638 if auto_init:
639 if auto_init != 'no_verbose_header':
640 self.print_debug_header()
641 self.add_default_info_extractors()
642
3089bc74
S
643 if (sys.platform != 'win32'
644 and sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
455a15e2 645 and not self.params.get('restrictfilenames', False)):
e9137224 646 # Unicode filesystem API will throw errors (#1474, #13027)
34308b30 647 self.report_warning(
6febd1c1 648 'Assuming --restrict-filenames since file system encoding '
1b725173 649 'cannot encode all characters. '
6febd1c1 650 'Set the LC_ALL environment variable to fix this.')
4a98cdbf 651 self.params['restrictfilenames'] = True
34308b30 652
de6000d9 653 self.outtmpl_dict = self.parse_outtmpl()
486dd09e 654
187986a8 655 # Creating format selector here allows us to catch syntax errors before the extraction
656 self.format_selector = (
fa9f30b8 657 self.params.get('format') if self.params.get('format') in (None, '-')
093a1710 658 else self.params['format'] if callable(self.params['format'])
187986a8 659 else self.build_format_selector(self.params['format']))
660
8b7539d2 661 # Set http_headers defaults according to std_headers
662 self.params['http_headers'] = merge_headers(std_headers, self.params.get('http_headers', {}))
663
013b50b7 664 hooks = {
665 'post_hooks': self.add_post_hook,
666 'progress_hooks': self.add_progress_hook,
667 'postprocessor_hooks': self.add_postprocessor_hook,
668 }
669 for opt, fn in hooks.items():
670 for ph in self.params.get(opt, []):
671 fn(ph)
71b640cc 672
5bfc8bee 673 for pp_def_raw in self.params.get('postprocessors', []):
674 pp_def = dict(pp_def_raw)
675 when = pp_def.pop('when', 'post_process')
676 self.add_post_processor(
f9934b96 677 get_postprocessor(pp_def.pop('key'))(self, **pp_def),
5bfc8bee 678 when=when)
679
97ec5bc5 680 self._setup_opener()
51fb4995
YCH
681 register_socks_protocols()
682
ed39cac5 683 def preload_download_archive(fn):
684 """Preload the archive, if any is specified"""
685 if fn is None:
686 return False
49a57e70 687 self.write_debug(f'Loading archive file {fn!r}')
ed39cac5 688 try:
689 with locked_file(fn, 'r', encoding='utf-8') as archive_file:
690 for line in archive_file:
691 self.archive.add(line.strip())
86e5f3ed 692 except OSError as ioe:
ed39cac5 693 if ioe.errno != errno.ENOENT:
694 raise
695 return False
696 return True
697
698 self.archive = set()
699 preload_download_archive(self.params.get('download_archive'))
700
7d4111ed
PH
701 def warn_if_short_id(self, argv):
702 # short YouTube ID starting with dash?
703 idxs = [
704 i for i, a in enumerate(argv)
705 if re.match(r'^-[0-9A-Za-z_-]{10}$', a)]
706 if idxs:
707 correct_argv = (
7a5c1cfe 708 ['yt-dlp']
3089bc74
S
709 + [a for i, a in enumerate(argv) if i not in idxs]
710 + ['--'] + [argv[i] for i in idxs]
7d4111ed
PH
711 )
712 self.report_warning(
713 'Long argument string detected. '
49a57e70 714 'Use -- to separate parameters and URLs, like this:\n%s' %
7d4111ed
PH
715 args_to_str(correct_argv))
716
8222d8de
JMF
717 def add_info_extractor(self, ie):
718 """Add an InfoExtractor object to the end of the list."""
8b7491c8 719 ie_key = ie.ie_key()
720 self._ies[ie_key] = ie
e52d7f85 721 if not isinstance(ie, type):
8b7491c8 722 self._ies_instances[ie_key] = ie
e52d7f85 723 ie.set_downloader(self)
8222d8de 724
8b7491c8 725 def _get_info_extractor_class(self, ie_key):
726 ie = self._ies.get(ie_key)
727 if ie is None:
728 ie = get_info_extractor(ie_key)
729 self.add_info_extractor(ie)
730 return ie
731
56c73665
JMF
732 def get_info_extractor(self, ie_key):
733 """
734 Get an instance of an IE with name ie_key, it will try to get one from
735 the _ies list, if there's no instance it will create a new one and add
736 it to the extractor list.
737 """
738 ie = self._ies_instances.get(ie_key)
739 if ie is None:
740 ie = get_info_extractor(ie_key)()
741 self.add_info_extractor(ie)
742 return ie
743
023fa8c4
JMF
744 def add_default_info_extractors(self):
745 """
746 Add the InfoExtractors returned by gen_extractors to the end of the list
747 """
e52d7f85 748 for ie in gen_extractor_classes():
023fa8c4
JMF
749 self.add_info_extractor(ie)
750
56d868db 751 def add_post_processor(self, pp, when='post_process'):
8222d8de 752 """Add a PostProcessor object to the end of the chain."""
5bfa4862 753 self._pps[when].append(pp)
8222d8de
JMF
754 pp.set_downloader(self)
755
ab8e5e51
AM
756 def add_post_hook(self, ph):
757 """Add the post hook"""
758 self._post_hooks.append(ph)
759
933605d7 760 def add_progress_hook(self, ph):
819e0531 761 """Add the download progress hook"""
933605d7 762 self._progress_hooks.append(ph)
8ab470f1 763
819e0531 764 def add_postprocessor_hook(self, ph):
765 """Add the postprocessing progress hook"""
766 self._postprocessor_hooks.append(ph)
5bfc8bee 767 for pps in self._pps.values():
768 for pp in pps:
769 pp.add_progress_hook(ph)
819e0531 770
1c088fa8 771 def _bidi_workaround(self, message):
5d681e96 772 if not hasattr(self, '_output_channel'):
1c088fa8
PH
773 return message
774
5d681e96 775 assert hasattr(self, '_output_process')
11b85ce6 776 assert isinstance(message, compat_str)
6febd1c1 777 line_count = message.count('\n') + 1
0f06bcd7 778 self._output_process.stdin.write((message + '\n').encode())
5d681e96 779 self._output_process.stdin.flush()
0f06bcd7 780 res = ''.join(self._output_channel.readline().decode()
9e1a5b84 781 for _ in range(line_count))
6febd1c1 782 return res[:-len('\n')]
1c088fa8 783
b35496d8 784 def _write_string(self, message, out=None, only_once=False):
785 if only_once:
786 if message in self._printed_messages:
787 return
788 self._printed_messages.add(message)
789 write_string(message, out=out, encoding=self.params.get('encoding'))
734f90bb 790
cf4f42cb 791 def to_stdout(self, message, skip_eol=False, quiet=None):
0760b0a7 792 """Print message to stdout"""
cf4f42cb 793 if quiet is not None:
ae6a1b95 794 self.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument quiet. Use "YoutubeDL.to_screen" instead')
8a82af35 795 if skip_eol is not False:
796 self.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument skip_eol. Use "YoutubeDL.to_screen" instead')
797 self._write_string(self._bidi_workaround(message), self._out_files.out)
cf4f42cb 798
799 def to_screen(self, message, skip_eol=False, quiet=None):
800 """Print message to screen if not in quiet mode"""
8bf9319e 801 if self.params.get('logger'):
43afe285 802 self.params['logger'].debug(message)
cf4f42cb 803 return
804 if (self.params.get('quiet') if quiet is None else quiet) and not self.params.get('verbose'):
805 return
806 self._write_string(
807 '%s%s' % (self._bidi_workaround(message), ('' if skip_eol else '\n')),
591bb9d3 808 self._out_files.screen)
8222d8de 809
b35496d8 810 def to_stderr(self, message, only_once=False):
0760b0a7 811 """Print message to stderr"""
11b85ce6 812 assert isinstance(message, compat_str)
8bf9319e 813 if self.params.get('logger'):
43afe285
IB
814 self.params['logger'].error(message)
815 else:
5792c950 816 self._write_string(f'{self._bidi_workaround(message)}\n', self._out_files.error, only_once=only_once)
cf4f42cb 817
818 def _send_console_code(self, code):
591bb9d3 819 if compat_os_name == 'nt' or not self._out_files.console:
cf4f42cb 820 return
591bb9d3 821 self._write_string(code, self._out_files.console)
8222d8de 822
1e5b9a95
PH
823 def to_console_title(self, message):
824 if not self.params.get('consoletitle', False):
825 return
3efb96a6 826 message = remove_terminal_sequences(message)
4bede0d8
C
827 if compat_os_name == 'nt':
828 if ctypes.windll.kernel32.GetConsoleWindow():
829 # c_wchar_p() might not be necessary if `message` is
830 # already of type unicode()
831 ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
cf4f42cb 832 else:
833 self._send_console_code(f'\033]0;{message}\007')
1e5b9a95 834
bdde425c 835 def save_console_title(self):
cf4f42cb 836 if not self.params.get('consoletitle') or self.params.get('simulate'):
bdde425c 837 return
592b7485 838 self._send_console_code('\033[22;0t') # Save the title on stack
bdde425c
PH
839
840 def restore_console_title(self):
cf4f42cb 841 if not self.params.get('consoletitle') or self.params.get('simulate'):
bdde425c 842 return
592b7485 843 self._send_console_code('\033[23;0t') # Restore the title from stack
bdde425c
PH
844
845 def __enter__(self):
846 self.save_console_title()
847 return self
848
849 def __exit__(self, *args):
850 self.restore_console_title()
f89197d7 851
dca08720 852 if self.params.get('cookiefile') is not None:
1bab3437 853 self.cookiejar.save(ignore_discard=True, ignore_expires=True)
bdde425c 854
fa9f30b8 855 def trouble(self, message=None, tb=None, is_error=True):
8222d8de
JMF
856 """Determine action to take when a download problem appears.
857
858 Depending on if the downloader has been configured to ignore
859 download errors or not, this method may throw an exception or
860 not when errors are found, after printing the message.
861
fa9f30b8 862 @param tb If given, is additional traceback information
863 @param is_error Whether to raise error according to ignorerrors
8222d8de
JMF
864 """
865 if message is not None:
866 self.to_stderr(message)
867 if self.params.get('verbose'):
868 if tb is None:
869 if sys.exc_info()[0]: # if .trouble has been called from an except block
6febd1c1 870 tb = ''
8222d8de 871 if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
6febd1c1 872 tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
c0384f22 873 tb += encode_compat_str(traceback.format_exc())
8222d8de
JMF
874 else:
875 tb_data = traceback.format_list(traceback.extract_stack())
6febd1c1 876 tb = ''.join(tb_data)
c19bc311 877 if tb:
878 self.to_stderr(tb)
fa9f30b8 879 if not is_error:
880 return
b1940459 881 if not self.params.get('ignoreerrors'):
8222d8de
JMF
882 if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
883 exc_info = sys.exc_info()[1].exc_info
884 else:
885 exc_info = sys.exc_info()
886 raise DownloadError(message, exc_info)
887 self._download_retcode = 1
888
19a03940 889 Styles = Namespace(
890 HEADERS='yellow',
891 EMPHASIS='light blue',
492272fe 892 FILENAME='green',
19a03940 893 ID='green',
894 DELIM='blue',
895 ERROR='red',
896 WARNING='yellow',
897 SUPPRESS='light black',
898 )
ec11a9f4 899
7578d77d 900 def _format_text(self, handle, allow_colors, text, f, fallback=None, *, test_encoding=False):
e5a998f3 901 text = str(text)
ec11a9f4 902 if test_encoding:
903 original_text = text
5c104538 904 # handle.encoding can be None. See https://github.com/yt-dlp/yt-dlp/issues/2711
905 encoding = self.params.get('encoding') or getattr(handle, 'encoding', None) or 'ascii'
ec11a9f4 906 text = text.encode(encoding, 'ignore').decode(encoding)
907 if fallback is not None and text != original_text:
908 text = fallback
7578d77d 909 return format_text(text, f) if allow_colors else text if fallback is None else fallback
ec11a9f4 910
591bb9d3 911 def _format_out(self, *args, **kwargs):
912 return self._format_text(self._out_files.out, self._allow_colors.out, *args, **kwargs)
913
ec11a9f4 914 def _format_screen(self, *args, **kwargs):
591bb9d3 915 return self._format_text(self._out_files.screen, self._allow_colors.screen, *args, **kwargs)
ec11a9f4 916
917 def _format_err(self, *args, **kwargs):
591bb9d3 918 return self._format_text(self._out_files.error, self._allow_colors.error, *args, **kwargs)
819e0531 919
c84aeac6 920 def report_warning(self, message, only_once=False):
8222d8de
JMF
921 '''
922 Print the message to stderr, it will be prefixed with 'WARNING:'
923 If stderr is a tty file the 'WARNING:' will be colored
924 '''
6d07ce01
JMF
925 if self.params.get('logger') is not None:
926 self.params['logger'].warning(message)
8222d8de 927 else:
ad8915b7
PH
928 if self.params.get('no_warnings'):
929 return
ec11a9f4 930 self.to_stderr(f'{self._format_err("WARNING:", self.Styles.WARNING)} {message}', only_once)
8222d8de 931
ee8dd27a 932 def deprecation_warning(self, message):
933 if self.params.get('logger') is not None:
a44ca5a4 934 self.params['logger'].warning(f'DeprecationWarning: {message}')
ee8dd27a 935 else:
936 self.to_stderr(f'{self._format_err("DeprecationWarning:", self.Styles.ERROR)} {message}', True)
937
fa9f30b8 938 def report_error(self, message, *args, **kwargs):
8222d8de
JMF
939 '''
940 Do the same as trouble, but prefixes the message with 'ERROR:', colored
941 in red if stderr is a tty file.
942 '''
fa9f30b8 943 self.trouble(f'{self._format_err("ERROR:", self.Styles.ERROR)} {message}', *args, **kwargs)
8222d8de 944
b35496d8 945 def write_debug(self, message, only_once=False):
0760b0a7 946 '''Log debug message or Print message to stderr'''
947 if not self.params.get('verbose', False):
948 return
8a82af35 949 message = f'[debug] {message}'
0760b0a7 950 if self.params.get('logger'):
951 self.params['logger'].debug(message)
952 else:
b35496d8 953 self.to_stderr(message, only_once)
0760b0a7 954
8222d8de
JMF
955 def report_file_already_downloaded(self, file_name):
956 """Report file has already been fully downloaded."""
957 try:
6febd1c1 958 self.to_screen('[download] %s has already been downloaded' % file_name)
ce02ed60 959 except UnicodeEncodeError:
6febd1c1 960 self.to_screen('[download] The file has already been downloaded')
8222d8de 961
0c3d0f51 962 def report_file_delete(self, file_name):
963 """Report that existing file will be deleted."""
964 try:
c25228e5 965 self.to_screen('Deleting existing file %s' % file_name)
0c3d0f51 966 except UnicodeEncodeError:
c25228e5 967 self.to_screen('Deleting existing file')
0c3d0f51 968
319b6059 969 def raise_no_formats(self, info, forced=False, *, msg=None):
0a5a191a 970 has_drm = info.get('_has_drm')
319b6059 971 ignored, expected = self.params.get('ignore_no_formats_error'), bool(msg)
972 msg = msg or has_drm and 'This video is DRM protected' or 'No video formats found!'
973 if forced or not ignored:
1151c407 974 raise ExtractorError(msg, video_id=info['id'], ie=info['extractor'],
319b6059 975 expected=has_drm or ignored or expected)
88acdbc2 976 else:
977 self.report_warning(msg)
978
de6000d9 979 def parse_outtmpl(self):
980 outtmpl_dict = self.params.get('outtmpl', {})
981 if not isinstance(outtmpl_dict, dict):
982 outtmpl_dict = {'default': outtmpl_dict}
71ce444a 983 # Remove spaces in the default template
984 if self.params.get('restrictfilenames'):
985 sanitize = lambda x: x.replace(' - ', ' ').replace(' ', '-')
986 else:
987 sanitize = lambda x: x
de6000d9 988 outtmpl_dict.update({
71ce444a 989 k: sanitize(v) for k, v in DEFAULT_OUTTMPL.items()
80c03fa9 990 if outtmpl_dict.get(k) is None})
86e5f3ed 991 for _, val in outtmpl_dict.items():
de6000d9 992 if isinstance(val, bytes):
86e5f3ed 993 self.report_warning('Parameter outtmpl is bytes, but should be a unicode string')
de6000d9 994 return outtmpl_dict
995
21cd8fae 996 def get_output_path(self, dir_type='', filename=None):
997 paths = self.params.get('paths', {})
998 assert isinstance(paths, dict)
999 path = os.path.join(
1000 expand_path(paths.get('home', '').strip()),
1001 expand_path(paths.get(dir_type, '').strip()) if dir_type else '',
1002 filename or '')
21cd8fae 1003 return sanitize_path(path, force=self.params.get('windowsfilenames'))
1004
76a264ac 1005 @staticmethod
901130bb 1006 def _outtmpl_expandpath(outtmpl):
1007 # expand_path translates '%%' into '%' and '$$' into '$'
1008 # correspondingly that is not what we want since we need to keep
1009 # '%%' intact for template dict substitution step. Working around
1010 # with boundary-alike separator hack.
1011 sep = ''.join([random.choice(ascii_letters) for _ in range(32)])
86e5f3ed 1012 outtmpl = outtmpl.replace('%%', f'%{sep}%').replace('$$', f'${sep}$')
901130bb 1013
1014 # outtmpl should be expand_path'ed before template dict substitution
1015 # because meta fields may contain env variables we don't want to
1016 # be expanded. For example, for outtmpl "%(title)s.%(ext)s" and
1017 # title "Hello $PATH", we don't want `$PATH` to be expanded.
1018 return expand_path(outtmpl).replace(sep, '')
1019
1020 @staticmethod
1021 def escape_outtmpl(outtmpl):
1022 ''' Escape any remaining strings like %s, %abc% etc. '''
1023 return re.sub(
1024 STR_FORMAT_RE_TMPL.format('', '(?![%(\0])'),
1025 lambda mobj: ('' if mobj.group('has_key') else '%') + mobj.group(0),
1026 outtmpl)
1027
1028 @classmethod
1029 def validate_outtmpl(cls, outtmpl):
76a264ac 1030 ''' @return None or Exception object '''
7d1eb38a 1031 outtmpl = re.sub(
37893bb0 1032 STR_FORMAT_RE_TMPL.format('[^)]*', '[ljqBUDS]'),
7d1eb38a 1033 lambda mobj: f'{mobj.group(0)[:-1]}s',
1034 cls._outtmpl_expandpath(outtmpl))
76a264ac 1035 try:
7d1eb38a 1036 cls.escape_outtmpl(outtmpl) % collections.defaultdict(int)
76a264ac 1037 return None
1038 except ValueError as err:
1039 return err
1040
03b4de72 1041 @staticmethod
1042 def _copy_infodict(info_dict):
1043 info_dict = dict(info_dict)
09b49e1f 1044 info_dict.pop('__postprocessors', None)
415f8d51 1045 info_dict.pop('__pending_error', None)
03b4de72 1046 return info_dict
1047
e0fd9573 1048 def prepare_outtmpl(self, outtmpl, info_dict, sanitize=False):
1049 """ Make the outtmpl and info_dict suitable for substitution: ydl.escape_outtmpl(outtmpl) % info_dict
1050 @param sanitize Whether to sanitize the output as a filename.
1051 For backward compatibility, a function can also be passed
1052 """
1053
6e84b215 1054 info_dict.setdefault('epoch', int(time.time())) # keep epoch consistent once set
143db31d 1055
03b4de72 1056 info_dict = self._copy_infodict(info_dict)
752cda38 1057 info_dict['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
53c18592 1058 formatSeconds(info_dict['duration'], '-' if sanitize else ':')
143db31d 1059 if info_dict.get('duration', None) is not None
1060 else None)
1d485a1a 1061 info_dict['autonumber'] = int(self.params.get('autonumber_start', 1) - 1 + self._num_downloads)
9c906919 1062 info_dict['video_autonumber'] = self._num_videos
752cda38 1063 if info_dict.get('resolution') is None:
1064 info_dict['resolution'] = self.format_resolution(info_dict, default=None)
143db31d 1065
e6f21b3d 1066 # For fields playlist_index, playlist_autonumber and autonumber convert all occurrences
143db31d 1067 # of %(field)s to %(field)0Nd for backward compatibility
1068 field_size_compat_map = {
0a5a191a 1069 'playlist_index': number_of_digits(info_dict.get('__last_playlist_index') or 0),
ec11a9f4 1070 'playlist_autonumber': number_of_digits(info_dict.get('n_entries') or 0),
752cda38 1071 'autonumber': self.params.get('autonumber_size') or 5,
143db31d 1072 }
752cda38 1073
385a27fa 1074 TMPL_DICT = {}
37893bb0 1075 EXTERNAL_FORMAT_RE = re.compile(STR_FORMAT_RE_TMPL.format('[^)]*', f'[{STR_FORMAT_TYPES}ljqBUDS]'))
385a27fa 1076 MATH_FUNCTIONS = {
1077 '+': float.__add__,
1078 '-': float.__sub__,
1079 }
e625be0d 1080 # Field is of the form key1.key2...
1081 # where keys (except first) can be string, int or slice
2b8a2973 1082 FIELD_RE = r'\w*(?:\.(?:\w+|{num}|{num}?(?::{num}?){{1,2}}))*'.format(num=r'(?:-?\d+)')
1d485a1a 1083 MATH_FIELD_RE = rf'(?:{FIELD_RE}|-?{NUMBER_RE})'
385a27fa 1084 MATH_OPERATORS_RE = r'(?:%s)' % '|'.join(map(re.escape, MATH_FUNCTIONS.keys()))
1d485a1a 1085 INTERNAL_FORMAT_RE = re.compile(rf'''(?x)
e625be0d 1086 (?P<negate>-)?
1d485a1a 1087 (?P<fields>{FIELD_RE})
1088 (?P<maths>(?:{MATH_OPERATORS_RE}{MATH_FIELD_RE})*)
e625be0d 1089 (?:>(?P<strf_format>.+?))?
34baa9fd 1090 (?P<remaining>
1091 (?P<alternate>(?<!\\),[^|&)]+)?
1092 (?:&(?P<replacement>.*?))?
1093 (?:\|(?P<default>.*?))?
1d485a1a 1094 )$''')
752cda38 1095
2b8a2973 1096 def _traverse_infodict(k):
1097 k = k.split('.')
1098 if k[0] == '':
1099 k.pop(0)
1100 return traverse_obj(info_dict, k, is_user_input=True, traverse_string=True)
76a264ac 1101
752cda38 1102 def get_value(mdict):
1103 # Object traversal
2b8a2973 1104 value = _traverse_infodict(mdict['fields'])
752cda38 1105 # Negative
1106 if mdict['negate']:
1107 value = float_or_none(value)
1108 if value is not None:
1109 value *= -1
1110 # Do maths
385a27fa 1111 offset_key = mdict['maths']
1112 if offset_key:
752cda38 1113 value = float_or_none(value)
1114 operator = None
385a27fa 1115 while offset_key:
1116 item = re.match(
1117 MATH_FIELD_RE if operator else MATH_OPERATORS_RE,
1118 offset_key).group(0)
1119 offset_key = offset_key[len(item):]
1120 if operator is None:
752cda38 1121 operator = MATH_FUNCTIONS[item]
385a27fa 1122 continue
1123 item, multiplier = (item[1:], -1) if item[0] == '-' else (item, 1)
1124 offset = float_or_none(item)
1125 if offset is None:
2b8a2973 1126 offset = float_or_none(_traverse_infodict(item))
385a27fa 1127 try:
1128 value = operator(value, multiplier * offset)
1129 except (TypeError, ZeroDivisionError):
1130 return None
1131 operator = None
752cda38 1132 # Datetime formatting
1133 if mdict['strf_format']:
7c37ff97 1134 value = strftime_or_none(value, mdict['strf_format'].replace('\\,', ','))
752cda38 1135
1136 return value
1137
b868936c 1138 na = self.params.get('outtmpl_na_placeholder', 'NA')
1139
e0fd9573 1140 def filename_sanitizer(key, value, restricted=self.params.get('restrictfilenames')):
5c3895ff 1141 return sanitize_filename(str(value), restricted=restricted, is_id=(
1142 bool(re.search(r'(^|[_.])id(\.|$)', key))
8a82af35 1143 if 'filename-sanitization' in self.params['compat_opts']
5c3895ff 1144 else NO_DEFAULT))
e0fd9573 1145
1146 sanitizer = sanitize if callable(sanitize) else filename_sanitizer
1147 sanitize = bool(sanitize)
1148
6e84b215 1149 def _dumpjson_default(obj):
1150 if isinstance(obj, (set, LazyList)):
1151 return list(obj)
adbc4ec4 1152 return repr(obj)
6e84b215 1153
752cda38 1154 def create_key(outer_mobj):
1155 if not outer_mobj.group('has_key'):
b836dc94 1156 return outer_mobj.group(0)
752cda38 1157 key = outer_mobj.group('key')
752cda38 1158 mobj = re.match(INTERNAL_FORMAT_RE, key)
e0fd9573 1159 initial_field = mobj.group('fields') if mobj else ''
e978789f 1160 value, replacement, default = None, None, na
7c37ff97 1161 while mobj:
e625be0d 1162 mobj = mobj.groupdict()
7c37ff97 1163 default = mobj['default'] if mobj['default'] is not None else default
752cda38 1164 value = get_value(mobj)
e978789f 1165 replacement = mobj['replacement']
7c37ff97 1166 if value is None and mobj['alternate']:
34baa9fd 1167 mobj = re.match(INTERNAL_FORMAT_RE, mobj['remaining'][1:])
7c37ff97 1168 else:
1169 break
752cda38 1170
b868936c 1171 fmt = outer_mobj.group('format')
752cda38 1172 if fmt == 's' and value is not None and key in field_size_compat_map.keys():
86e5f3ed 1173 fmt = f'0{field_size_compat_map[key]:d}d'
752cda38 1174
e978789f 1175 value = default if value is None else value if replacement is None else replacement
752cda38 1176
4476d2c7 1177 flags = outer_mobj.group('conversion') or ''
7d1eb38a 1178 str_fmt = f'{fmt[:-1]}s'
524e2e4f 1179 if fmt[-1] == 'l': # list
4476d2c7 1180 delim = '\n' if '#' in flags else ', '
9e907ebd 1181 value, fmt = delim.join(map(str, variadic(value, allowed_types=(str, bytes)))), str_fmt
524e2e4f 1182 elif fmt[-1] == 'j': # json
4476d2c7 1183 value, fmt = json.dumps(value, default=_dumpjson_default, indent=4 if '#' in flags else None), str_fmt
524e2e4f 1184 elif fmt[-1] == 'q': # quoted
4476d2c7 1185 value = map(str, variadic(value) if '#' in flags else [value])
1186 value, fmt = ' '.join(map(compat_shlex_quote, value)), str_fmt
524e2e4f 1187 elif fmt[-1] == 'B': # bytes
0f06bcd7 1188 value = f'%{str_fmt}'.encode() % str(value).encode()
f5aa5cfb 1189 value, fmt = value.decode('utf-8', 'ignore'), 's'
524e2e4f 1190 elif fmt[-1] == 'U': # unicode normalized
524e2e4f 1191 value, fmt = unicodedata.normalize(
1192 # "+" = compatibility equivalence, "#" = NFD
4476d2c7 1193 'NF%s%s' % ('K' if '+' in flags else '', 'D' if '#' in flags else 'C'),
524e2e4f 1194 value), str_fmt
e0fd9573 1195 elif fmt[-1] == 'D': # decimal suffix
abbeeebc 1196 num_fmt, fmt = fmt[:-1].replace('#', ''), 's'
1197 value = format_decimal_suffix(value, f'%{num_fmt}f%s' if num_fmt else '%d%s',
1198 factor=1024 if '#' in flags else 1000)
37893bb0 1199 elif fmt[-1] == 'S': # filename sanitization
e0fd9573 1200 value, fmt = filename_sanitizer(initial_field, value, restricted='#' in flags), str_fmt
7d1eb38a 1201 elif fmt[-1] == 'c':
524e2e4f 1202 if value:
1203 value = str(value)[0]
76a264ac 1204 else:
524e2e4f 1205 fmt = str_fmt
76a264ac 1206 elif fmt[-1] not in 'rs': # numeric
a439a3a4 1207 value = float_or_none(value)
752cda38 1208 if value is None:
1209 value, fmt = default, 's'
901130bb 1210
752cda38 1211 if sanitize:
1212 if fmt[-1] == 'r':
1213 # If value is an object, sanitize might convert it to a string
1214 # So we convert it to repr first
7d1eb38a 1215 value, fmt = repr(value), str_fmt
639f1cea 1216 if fmt[-1] in 'csr':
e0fd9573 1217 value = sanitizer(initial_field, value)
901130bb 1218
b868936c 1219 key = '%s\0%s' % (key.replace('%', '%\0'), outer_mobj.group('format'))
385a27fa 1220 TMPL_DICT[key] = value
b868936c 1221 return '{prefix}%({key}){fmt}'.format(key=key, fmt=fmt, prefix=outer_mobj.group('prefix'))
752cda38 1222
385a27fa 1223 return EXTERNAL_FORMAT_RE.sub(create_key, outtmpl), TMPL_DICT
143db31d 1224
819e0531 1225 def evaluate_outtmpl(self, outtmpl, info_dict, *args, **kwargs):
1226 outtmpl, info_dict = self.prepare_outtmpl(outtmpl, info_dict, *args, **kwargs)
1227 return self.escape_outtmpl(outtmpl) % info_dict
1228
5127e92a 1229 def _prepare_filename(self, info_dict, *, outtmpl=None, tmpl_type=None):
1230 assert None in (outtmpl, tmpl_type), 'outtmpl and tmpl_type are mutually exclusive'
1231 if outtmpl is None:
1232 outtmpl = self.outtmpl_dict.get(tmpl_type or 'default', self.outtmpl_dict['default'])
8222d8de 1233 try:
5127e92a 1234 outtmpl = self._outtmpl_expandpath(outtmpl)
e0fd9573 1235 filename = self.evaluate_outtmpl(outtmpl, info_dict, True)
6a0546e3 1236 if not filename:
1237 return None
15da37c7 1238
5127e92a 1239 if tmpl_type in ('', 'temp'):
6a0546e3 1240 final_ext, ext = self.params.get('final_ext'), info_dict.get('ext')
1241 if final_ext and ext and final_ext != ext and filename.endswith(f'.{final_ext}'):
1242 filename = replace_extension(filename, ext, final_ext)
5127e92a 1243 elif tmpl_type:
6a0546e3 1244 force_ext = OUTTMPL_TYPES[tmpl_type]
1245 if force_ext:
1246 filename = replace_extension(filename, force_ext, info_dict.get('ext'))
de6000d9 1247
bdc3fd2f
U
1248 # https://github.com/blackjack4494/youtube-dlc/issues/85
1249 trim_file_name = self.params.get('trim_file_name', False)
1250 if trim_file_name:
5c22c63d 1251 no_ext, *ext = filename.rsplit('.', 2)
1252 filename = join_nonempty(no_ext[:trim_file_name], *ext, delim='.')
bdc3fd2f 1253
0202b52a 1254 return filename
8222d8de 1255 except ValueError as err:
6febd1c1 1256 self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')')
8222d8de
JMF
1257 return None
1258
5127e92a 1259 def prepare_filename(self, info_dict, dir_type='', *, outtmpl=None, warn=False):
1260 """Generate the output filename"""
1261 if outtmpl:
1262 assert not dir_type, 'outtmpl and dir_type are mutually exclusive'
1263 dir_type = None
1264 filename = self._prepare_filename(info_dict, tmpl_type=dir_type, outtmpl=outtmpl)
80c03fa9 1265 if not filename and dir_type not in ('', 'temp'):
1266 return ''
de6000d9 1267
c84aeac6 1268 if warn:
21cd8fae 1269 if not self.params.get('paths'):
de6000d9 1270 pass
1271 elif filename == '-':
c84aeac6 1272 self.report_warning('--paths is ignored when an outputting to stdout', only_once=True)
de6000d9 1273 elif os.path.isabs(filename):
c84aeac6 1274 self.report_warning('--paths is ignored since an absolute path is given in output template', only_once=True)
de6000d9 1275 if filename == '-' or not filename:
1276 return filename
1277
21cd8fae 1278 return self.get_output_path(dir_type, filename)
0202b52a 1279
120fe513 1280 def _match_entry(self, info_dict, incomplete=False, silent=False):
ecdec191 1281 """ Returns None if the file should be downloaded """
8222d8de 1282
c77495e3 1283 video_title = info_dict.get('title', info_dict.get('id', 'video'))
1284
8b0d7497 1285 def check_filter():
8b0d7497 1286 if 'title' in info_dict:
1287 # This can happen when we're just evaluating the playlist
1288 title = info_dict['title']
1289 matchtitle = self.params.get('matchtitle', False)
1290 if matchtitle:
1291 if not re.search(matchtitle, title, re.IGNORECASE):
1292 return '"' + title + '" title did not match pattern "' + matchtitle + '"'
1293 rejecttitle = self.params.get('rejecttitle', False)
1294 if rejecttitle:
1295 if re.search(rejecttitle, title, re.IGNORECASE):
1296 return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
1297 date = info_dict.get('upload_date')
1298 if date is not None:
1299 dateRange = self.params.get('daterange', DateRange())
1300 if date not in dateRange:
86e5f3ed 1301 return f'{date_from_str(date).isoformat()} upload date is not in range {dateRange}'
8b0d7497 1302 view_count = info_dict.get('view_count')
1303 if view_count is not None:
1304 min_views = self.params.get('min_views')
1305 if min_views is not None and view_count < min_views:
1306 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
1307 max_views = self.params.get('max_views')
1308 if max_views is not None and view_count > max_views:
1309 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
1310 if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')):
1311 return 'Skipping "%s" because it is age restricted' % video_title
8b0d7497 1312
8f18aca8 1313 match_filter = self.params.get('match_filter')
1314 if match_filter is not None:
1315 try:
1316 ret = match_filter(info_dict, incomplete=incomplete)
1317 except TypeError:
1318 # For backward compatibility
1319 ret = None if incomplete else match_filter(info_dict)
492272fe 1320 if ret is NO_DEFAULT:
1321 while True:
1322 filename = self._format_screen(self.prepare_filename(info_dict), self.Styles.FILENAME)
1323 reply = input(self._format_screen(
1324 f'Download "{filename}"? (Y/n): ', self.Styles.EMPHASIS)).lower().strip()
1325 if reply in {'y', ''}:
1326 return None
1327 elif reply == 'n':
1328 return f'Skipping {video_title}'
492272fe 1329 elif ret is not None:
8f18aca8 1330 return ret
8b0d7497 1331 return None
1332
c77495e3 1333 if self.in_download_archive(info_dict):
1334 reason = '%s has already been recorded in the archive' % video_title
1335 break_opt, break_err = 'break_on_existing', ExistingVideoReached
1336 else:
1337 reason = check_filter()
1338 break_opt, break_err = 'break_on_reject', RejectedVideoReached
8b0d7497 1339 if reason is not None:
120fe513 1340 if not silent:
1341 self.to_screen('[download] ' + reason)
c77495e3 1342 if self.params.get(break_opt, False):
1343 raise break_err()
8b0d7497 1344 return reason
fe7e0c98 1345
b6c45014
JMF
1346 @staticmethod
1347 def add_extra_info(info_dict, extra_info):
1348 '''Set the keys from extra_info in info dict if they are missing'''
1349 for key, value in extra_info.items():
1350 info_dict.setdefault(key, value)
1351
409e1828 1352 def extract_info(self, url, download=True, ie_key=None, extra_info=None,
61aa5ba3 1353 process=True, force_generic_extractor=False):
41d1cca3 1354 """
1355 Return a list with a dictionary for each video extracted.
1356
1357 Arguments:
1358 url -- URL to extract
1359
1360 Keyword arguments:
1361 download -- whether to download videos during extraction
1362 ie_key -- extractor key hint
1363 extra_info -- dictionary containing the extra values to add to each result
1364 process -- whether to resolve all unresolved references (URLs, playlist items),
1365 must be True for download to work.
1366 force_generic_extractor -- force using the generic extractor
1367 """
fe7e0c98 1368
409e1828 1369 if extra_info is None:
1370 extra_info = {}
1371
61aa5ba3 1372 if not ie_key and force_generic_extractor:
d22dec74
S
1373 ie_key = 'Generic'
1374
8222d8de 1375 if ie_key:
8b7491c8 1376 ies = {ie_key: self._get_info_extractor_class(ie_key)}
8222d8de
JMF
1377 else:
1378 ies = self._ies
1379
8b7491c8 1380 for ie_key, ie in ies.items():
8222d8de
JMF
1381 if not ie.suitable(url):
1382 continue
1383
1384 if not ie.working():
6febd1c1
PH
1385 self.report_warning('The program functionality for this site has been marked as broken, '
1386 'and will probably not work.')
8222d8de 1387
1151c407 1388 temp_id = ie.get_temp_id(url)
a0566bbf 1389 if temp_id is not None and self.in_download_archive({'id': temp_id, 'ie_key': ie_key}):
5e5be0c0 1390 self.to_screen(f'[{ie_key}] {temp_id}: has already been recorded in the archive')
1391 if self.params.get('break_on_existing', False):
1392 raise ExistingVideoReached()
a0566bbf 1393 break
8b7491c8 1394 return self.__extract_info(url, self.get_info_extractor(ie_key), download, extra_info, process)
a0566bbf 1395 else:
1396 self.report_error('no suitable InfoExtractor for URL %s' % url)
1397
8e5fecc8 1398 def __handle_extraction_exceptions(func):
b5ae35ee 1399 @functools.wraps(func)
a0566bbf 1400 def wrapper(self, *args, **kwargs):
6da22e7d 1401 while True:
1402 try:
1403 return func(self, *args, **kwargs)
1404 except (DownloadCancelled, LazyList.IndexError, PagedList.IndexError):
8222d8de 1405 raise
6da22e7d 1406 except ReExtractInfo as e:
1407 if e.expected:
1408 self.to_screen(f'{e}; Re-extracting data')
1409 else:
1410 self.to_stderr('\r')
1411 self.report_warning(f'{e}; Re-extracting data')
1412 continue
1413 except GeoRestrictedError as e:
1414 msg = e.msg
1415 if e.countries:
1416 msg += '\nThis video is available in %s.' % ', '.join(
1417 map(ISO3166Utils.short2full, e.countries))
1418 msg += '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1419 self.report_error(msg)
1420 except ExtractorError as e: # An error we somewhat expected
1421 self.report_error(str(e), e.format_traceback())
1422 except Exception as e:
1423 if self.params.get('ignoreerrors'):
1424 self.report_error(str(e), tb=encode_compat_str(traceback.format_exc()))
1425 else:
1426 raise
1427 break
a0566bbf 1428 return wrapper
1429
f2ebc5c7 1430 def _wait_for_video(self, ie_result):
1431 if (not self.params.get('wait_for_video')
1432 or ie_result.get('_type', 'video') != 'video'
1433 or ie_result.get('formats') or ie_result.get('url')):
1434 return
1435
1436 format_dur = lambda dur: '%02d:%02d:%02d' % timetuple_from_msec(dur * 1000)[:-1]
1437 last_msg = ''
1438
1439 def progress(msg):
1440 nonlocal last_msg
1441 self.to_screen(msg + ' ' * (len(last_msg) - len(msg)) + '\r', skip_eol=True)
1442 last_msg = msg
1443
1444 min_wait, max_wait = self.params.get('wait_for_video')
1445 diff = try_get(ie_result, lambda x: x['release_timestamp'] - time.time())
1446 if diff is None and ie_result.get('live_status') == 'is_upcoming':
16c620bc 1447 diff = round(random.uniform(min_wait, max_wait) if (max_wait and min_wait) else (max_wait or min_wait), 0)
f2ebc5c7 1448 self.report_warning('Release time of video is not known')
1449 elif (diff or 0) <= 0:
1450 self.report_warning('Video should already be available according to extracted info')
38d79fd1 1451 diff = min(max(diff or 0, min_wait or 0), max_wait or float('inf'))
f2ebc5c7 1452 self.to_screen(f'[wait] Waiting for {format_dur(diff)} - Press Ctrl+C to try now')
1453
1454 wait_till = time.time() + diff
1455 try:
1456 while True:
1457 diff = wait_till - time.time()
1458 if diff <= 0:
1459 progress('')
1460 raise ReExtractInfo('[wait] Wait period ended', expected=True)
1461 progress(f'[wait] Remaining time until next attempt: {self._format_screen(format_dur(diff), self.Styles.EMPHASIS)}')
1462 time.sleep(1)
1463 except KeyboardInterrupt:
1464 progress('')
1465 raise ReExtractInfo('[wait] Interrupted by user', expected=True)
1466 except BaseException as e:
1467 if not isinstance(e, ReExtractInfo):
1468 self.to_screen('')
1469 raise
1470
a0566bbf 1471 @__handle_extraction_exceptions
58f197b7 1472 def __extract_info(self, url, ie, download, extra_info, process):
a0566bbf 1473 ie_result = ie.extract(url)
1474 if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
1475 return
1476 if isinstance(ie_result, list):
1477 # Backwards compatibility: old IE result format
1478 ie_result = {
1479 '_type': 'compat_list',
1480 'entries': ie_result,
1481 }
e37d0efb 1482 if extra_info.get('original_url'):
1483 ie_result.setdefault('original_url', extra_info['original_url'])
a0566bbf 1484 self.add_default_extra_info(ie_result, ie, url)
1485 if process:
f2ebc5c7 1486 self._wait_for_video(ie_result)
a0566bbf 1487 return self.process_ie_result(ie_result, download, extra_info)
8222d8de 1488 else:
a0566bbf 1489 return ie_result
fe7e0c98 1490
ea38e55f 1491 def add_default_extra_info(self, ie_result, ie, url):
6033d980 1492 if url is not None:
1493 self.add_extra_info(ie_result, {
1494 'webpage_url': url,
1495 'original_url': url,
57ebfca3 1496 })
1497 webpage_url = ie_result.get('webpage_url')
1498 if webpage_url:
1499 self.add_extra_info(ie_result, {
1500 'webpage_url_basename': url_basename(webpage_url),
1501 'webpage_url_domain': get_domain(webpage_url),
6033d980 1502 })
1503 if ie is not None:
1504 self.add_extra_info(ie_result, {
1505 'extractor': ie.IE_NAME,
1506 'extractor_key': ie.ie_key(),
1507 })
ea38e55f 1508
58adec46 1509 def process_ie_result(self, ie_result, download=True, extra_info=None):
8222d8de
JMF
1510 """
1511 Take the result of the ie(may be modified) and resolve all unresolved
1512 references (URLs, playlist items).
1513
1514 It will also download the videos if 'download'.
1515 Returns the resolved ie_result.
1516 """
58adec46 1517 if extra_info is None:
1518 extra_info = {}
e8ee972c
PH
1519 result_type = ie_result.get('_type', 'video')
1520
057a5206 1521 if result_type in ('url', 'url_transparent'):
134c6ea8 1522 ie_result['url'] = sanitize_url(ie_result['url'])
e37d0efb 1523 if ie_result.get('original_url'):
1524 extra_info.setdefault('original_url', ie_result['original_url'])
1525
057a5206 1526 extract_flat = self.params.get('extract_flat', False)
3089bc74
S
1527 if ((extract_flat == 'in_playlist' and 'playlist' in extra_info)
1528 or extract_flat is True):
ecb54191 1529 info_copy = ie_result.copy()
6033d980 1530 ie = try_get(ie_result.get('ie_key'), self.get_info_extractor)
360167b9 1531 if ie and not ie_result.get('id'):
4614bc22 1532 info_copy['id'] = ie.get_temp_id(ie_result['url'])
6033d980 1533 self.add_default_extra_info(info_copy, ie, ie_result['url'])
4614bc22 1534 self.add_extra_info(info_copy, extra_info)
b5475f11 1535 info_copy, _ = self.pre_process(info_copy)
ecb54191 1536 self.__forced_printings(info_copy, self.prepare_filename(info_copy), incomplete=True)
415f8d51 1537 self._raise_pending_errors(info_copy)
4614bc22 1538 if self.params.get('force_write_download_archive', False):
1539 self.record_download_archive(info_copy)
e8ee972c
PH
1540 return ie_result
1541
8222d8de 1542 if result_type == 'video':
b6c45014 1543 self.add_extra_info(ie_result, extra_info)
9c2b75b5 1544 ie_result = self.process_video_result(ie_result, download=download)
415f8d51 1545 self._raise_pending_errors(ie_result)
28b0eb0f 1546 additional_urls = (ie_result or {}).get('additional_urls')
9c2b75b5 1547 if additional_urls:
e9f4ccd1 1548 # TODO: Improve MetadataParserPP to allow setting a list
9c2b75b5 1549 if isinstance(additional_urls, compat_str):
1550 additional_urls = [additional_urls]
1551 self.to_screen(
1552 '[info] %s: %d additional URL(s) requested' % (ie_result['id'], len(additional_urls)))
1553 self.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls))
1554 ie_result['additional_entries'] = [
1555 self.extract_info(
b69fd25c 1556 url, download, extra_info=extra_info,
9c2b75b5 1557 force_generic_extractor=self.params.get('force_generic_extractor'))
1558 for url in additional_urls
1559 ]
1560 return ie_result
8222d8de
JMF
1561 elif result_type == 'url':
1562 # We have to add extra_info to the results because it may be
1563 # contained in a playlist
07cce701 1564 return self.extract_info(
1565 ie_result['url'], download,
1566 ie_key=ie_result.get('ie_key'),
1567 extra_info=extra_info)
7fc3fa05
PH
1568 elif result_type == 'url_transparent':
1569 # Use the information from the embedding page
1570 info = self.extract_info(
1571 ie_result['url'], ie_key=ie_result.get('ie_key'),
1572 extra_info=extra_info, download=False, process=False)
1573
1640eb09
S
1574 # extract_info may return None when ignoreerrors is enabled and
1575 # extraction failed with an error, don't crash and return early
1576 # in this case
1577 if not info:
1578 return info
1579
412c617d 1580 new_result = info.copy()
90137ca4 1581 new_result.update(filter_dict(ie_result, lambda k, v: (
1582 v is not None and k not in {'_type', 'url', 'id', 'extractor', 'extractor_key', 'ie_key'})))
7fc3fa05 1583
0563f7ac
S
1584 # Extracted info may not be a video result (i.e.
1585 # info.get('_type', 'video') != video) but rather an url or
1586 # url_transparent. In such cases outer metadata (from ie_result)
1587 # should be propagated to inner one (info). For this to happen
1588 # _type of info should be overridden with url_transparent. This
067aa17e 1589 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
0563f7ac
S
1590 if new_result.get('_type') == 'url':
1591 new_result['_type'] = 'url_transparent'
7fc3fa05
PH
1592
1593 return self.process_ie_result(
1594 new_result, download=download, extra_info=extra_info)
40fcba5e 1595 elif result_type in ('playlist', 'multi_video'):
30a074c2 1596 # Protect from infinite recursion due to recursively nested playlists
1597 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
1598 webpage_url = ie_result['webpage_url']
1599 if webpage_url in self._playlist_urls:
7e85e872 1600 self.to_screen(
30a074c2 1601 '[download] Skipping already downloaded playlist: %s'
1602 % ie_result.get('title') or ie_result.get('id'))
1603 return
7e85e872 1604
30a074c2 1605 self._playlist_level += 1
1606 self._playlist_urls.add(webpage_url)
03f83004 1607 self._fill_common_fields(ie_result, False)
bc516a3f 1608 self._sanitize_thumbnails(ie_result)
30a074c2 1609 try:
1610 return self.__process_playlist(ie_result, download)
1611 finally:
1612 self._playlist_level -= 1
1613 if not self._playlist_level:
1614 self._playlist_urls.clear()
8222d8de 1615 elif result_type == 'compat_list':
c9bf4114
PH
1616 self.report_warning(
1617 'Extractor %s returned a compat_list result. '
1618 'It needs to be updated.' % ie_result.get('extractor'))
5f6a1245 1619
8222d8de 1620 def _fixup(r):
b868936c 1621 self.add_extra_info(r, {
1622 'extractor': ie_result['extractor'],
1623 'webpage_url': ie_result['webpage_url'],
1624 'webpage_url_basename': url_basename(ie_result['webpage_url']),
0bb322b9 1625 'webpage_url_domain': get_domain(ie_result['webpage_url']),
b868936c 1626 'extractor_key': ie_result['extractor_key'],
1627 })
8222d8de
JMF
1628 return r
1629 ie_result['entries'] = [
b6c45014 1630 self.process_ie_result(_fixup(r), download, extra_info)
8222d8de
JMF
1631 for r in ie_result['entries']
1632 ]
1633 return ie_result
1634 else:
1635 raise Exception('Invalid result type: %s' % result_type)
1636
e92caff5 1637 def _ensure_dir_exists(self, path):
1638 return make_dir(path, self.report_error)
1639
3b603dbd 1640 @staticmethod
1641 def _playlist_infodict(ie_result, **kwargs):
1642 return {
1643 **ie_result,
1644 'playlist': ie_result.get('title') or ie_result.get('id'),
1645 'playlist_id': ie_result.get('id'),
1646 'playlist_title': ie_result.get('title'),
1647 'playlist_uploader': ie_result.get('uploader'),
1648 'playlist_uploader_id': ie_result.get('uploader_id'),
1649 'playlist_index': 0,
1650 **kwargs,
1651 }
1652
30a074c2 1653 def __process_playlist(self, ie_result, download):
1654 # We process each entry in the playlist
1655 playlist = ie_result.get('title') or ie_result.get('id')
1656 self.to_screen('[download] Downloading playlist: %s' % playlist)
1657
498f5606 1658 if 'entries' not in ie_result:
aa9369a2 1659 raise EntryNotInPlaylist('There are no entries')
7c7f7161 1660
1661 MissingEntry = object()
498f5606 1662 incomplete_entries = bool(ie_result.get('requested_entries'))
1663 if incomplete_entries:
bf5f605e 1664 def fill_missing_entries(entries, indices):
7c7f7161 1665 ret = [MissingEntry] * max(indices)
bf5f605e 1666 for i, entry in zip(indices, entries):
498f5606 1667 ret[i - 1] = entry
1668 return ret
1669 ie_result['entries'] = fill_missing_entries(ie_result['entries'], ie_result['requested_entries'])
02fd60d3 1670
30a074c2 1671 playlist_results = []
1672
56a8fb4f 1673 playliststart = self.params.get('playliststart', 1)
30a074c2 1674 playlistend = self.params.get('playlistend')
1675 # For backwards compatibility, interpret -1 as whole list
1676 if playlistend == -1:
1677 playlistend = None
1678
1679 playlistitems_str = self.params.get('playlist_items')
1680 playlistitems = None
1681 if playlistitems_str is not None:
1682 def iter_playlistitems(format):
1683 for string_segment in format.split(','):
1684 if '-' in string_segment:
1685 start, end = string_segment.split('-')
1686 for item in range(int(start), int(end) + 1):
1687 yield int(item)
1688 else:
1689 yield int(string_segment)
1690 playlistitems = orderedSet(iter_playlistitems(playlistitems_str))
1691
1692 ie_entries = ie_result['entries']
8e5fecc8 1693 if isinstance(ie_entries, list):
ed8d87f9 1694 playlist_count = len(ie_entries)
f0d785d3 1695 msg = f'Collected {playlist_count} videos; downloading %d of them'
1696 ie_result['playlist_count'] = ie_result.get('playlist_count') or playlist_count
1697
8e5fecc8 1698 def get_entry(i):
1699 return ie_entries[i - 1]
1700 else:
f0d785d3 1701 msg = 'Downloading %d videos'
c586f9e8 1702 if not isinstance(ie_entries, (PagedList, LazyList)):
8e5fecc8 1703 ie_entries = LazyList(ie_entries)
d37707bd 1704 elif isinstance(ie_entries, InAdvancePagedList):
1705 if ie_entries._pagesize == 1:
1706 playlist_count = ie_entries._pagecount
8e5fecc8 1707
1708 def get_entry(i):
1709 return YoutubeDL.__handle_extraction_exceptions(
1710 lambda self, i: ie_entries[i - 1]
1711 )(self, i)
50fed816 1712
f0d785d3 1713 entries, broken = [], False
ff1c7fc9 1714 items = playlistitems if playlistitems is not None else itertools.count(playliststart)
1715 for i in items:
1716 if i == 0:
1717 continue
56a8fb4f 1718 if playlistitems is None and playlistend is not None and playlistend < i:
1719 break
1720 entry = None
1721 try:
50fed816 1722 entry = get_entry(i)
7c7f7161 1723 if entry is MissingEntry:
498f5606 1724 raise EntryNotInPlaylist()
56a8fb4f 1725 except (IndexError, EntryNotInPlaylist):
1726 if incomplete_entries:
aa9369a2 1727 raise EntryNotInPlaylist(f'Entry {i} cannot be found')
56a8fb4f 1728 elif not playlistitems:
1729 break
1730 entries.append(entry)
120fe513 1731 try:
1732 if entry is not None:
e5a998f3 1733 # TODO: Add auto-generated fields
120fe513 1734 self._match_entry(entry, incomplete=True, silent=True)
1735 except (ExistingVideoReached, RejectedVideoReached):
f0d785d3 1736 broken = True
120fe513 1737 break
56a8fb4f 1738 ie_result['entries'] = entries
30a074c2 1739
56a8fb4f 1740 # Save playlist_index before re-ordering
1741 entries = [
9e598870 1742 ((playlistitems[i - 1] if playlistitems else i + playliststart - 1), entry)
56a8fb4f 1743 for i, entry in enumerate(entries, 1)
1744 if entry is not None]
1745 n_entries = len(entries)
498f5606 1746
f0d785d3 1747 if not (ie_result.get('playlist_count') or broken or playlistitems or playlistend):
1748 ie_result['playlist_count'] = n_entries
1749
e08a85d8 1750 if not playlistitems and (playliststart != 1 or playlistend):
56a8fb4f 1751 playlistitems = list(range(playliststart, playliststart + n_entries))
498f5606 1752 ie_result['requested_entries'] = playlistitems
1753
e08a85d8 1754 _infojson_written = False
0bfc53d0 1755 write_playlist_files = self.params.get('allow_playlist_files', True)
1756 if write_playlist_files and self.params.get('list_thumbnails'):
1757 self.list_thumbnails(ie_result)
1758 if write_playlist_files and not self.params.get('simulate'):
3b603dbd 1759 ie_copy = self._playlist_infodict(ie_result, n_entries=n_entries)
e08a85d8 1760 _infojson_written = self._write_info_json(
1761 'playlist', ie_result, self.prepare_filename(ie_copy, 'pl_infojson'))
1762 if _infojson_written is None:
80c03fa9 1763 return
1764 if self._write_description('playlist', ie_result,
1765 self.prepare_filename(ie_copy, 'pl_description')) is None:
1766 return
681de68e 1767 # TODO: This should be passed to ThumbnailsConvertor if necessary
80c03fa9 1768 self._write_thumbnails('playlist', ie_copy, self.prepare_filename(ie_copy, 'pl_thumbnail'))
30a074c2 1769
1770 if self.params.get('playlistreverse', False):
1771 entries = entries[::-1]
30a074c2 1772 if self.params.get('playlistrandom', False):
1773 random.shuffle(entries)
1774
1775 x_forwarded_for = ie_result.get('__x_forwarded_for_ip')
1776
86e5f3ed 1777 self.to_screen(f'[{ie_result["extractor"]}] playlist {playlist}: {msg % n_entries}')
26e2805c 1778 failures = 0
1779 max_failures = self.params.get('skip_playlist_after_errors') or float('inf')
71729754 1780 for i, entry_tuple in enumerate(entries, 1):
1781 playlist_index, entry = entry_tuple
8a82af35 1782 if 'playlist-index' in self.params['compat_opts']:
81139999 1783 playlist_index = playlistitems[i - 1] if playlistitems else i + playliststart - 1
19a03940 1784 self.to_screen('[download] Downloading video %s of %s' % (
1785 self._format_screen(i, self.Styles.ID), self._format_screen(n_entries, self.Styles.EMPHASIS)))
30a074c2 1786 # This __x_forwarded_for_ip thing is a bit ugly but requires
1787 # minimal changes
1788 if x_forwarded_for:
1789 entry['__x_forwarded_for_ip'] = x_forwarded_for
1790 extra = {
1791 'n_entries': n_entries,
0a5a191a 1792 '__last_playlist_index': max(playlistitems) if playlistitems else (playlistend or n_entries),
f0d785d3 1793 'playlist_count': ie_result.get('playlist_count'),
71729754 1794 'playlist_index': playlist_index,
1795 'playlist_autonumber': i,
30a074c2 1796 'playlist': playlist,
1797 'playlist_id': ie_result.get('id'),
1798 'playlist_title': ie_result.get('title'),
1799 'playlist_uploader': ie_result.get('uploader'),
1800 'playlist_uploader_id': ie_result.get('uploader_id'),
30a074c2 1801 'extractor': ie_result['extractor'],
1802 'webpage_url': ie_result['webpage_url'],
1803 'webpage_url_basename': url_basename(ie_result['webpage_url']),
0bb322b9 1804 'webpage_url_domain': get_domain(ie_result['webpage_url']),
30a074c2 1805 'extractor_key': ie_result['extractor_key'],
1806 }
1807
1808 if self._match_entry(entry, incomplete=True) is not None:
1809 continue
1810
1811 entry_result = self.__process_iterable_entry(entry, download, extra)
26e2805c 1812 if not entry_result:
1813 failures += 1
1814 if failures >= max_failures:
1815 self.report_error(
1816 'Skipping the remaining entries in playlist "%s" since %d items failed extraction' % (playlist, failures))
1817 break
30a074c2 1818 playlist_results.append(entry_result)
1819 ie_result['entries'] = playlist_results
e08a85d8 1820
1821 # Write the updated info to json
cb96c5be 1822 if _infojson_written is True and self._write_info_json(
e08a85d8 1823 'updated playlist', ie_result,
1824 self.prepare_filename(ie_copy, 'pl_infojson'), overwrite=True) is None:
1825 return
ca30f449 1826
ed5835b4 1827 ie_result = self.run_all_pps('playlist', ie_result)
1828 self.to_screen(f'[download] Finished downloading playlist: {playlist}')
30a074c2 1829 return ie_result
1830
a0566bbf 1831 @__handle_extraction_exceptions
1832 def __process_iterable_entry(self, entry, download, extra_info):
1833 return self.process_ie_result(
1834 entry, download=download, extra_info=extra_info)
1835
67134eab
JMF
1836 def _build_format_filter(self, filter_spec):
1837 " Returns a function to filter the formats according to the filter_spec "
083c9df9
PH
1838
1839 OPERATORS = {
1840 '<': operator.lt,
1841 '<=': operator.le,
1842 '>': operator.gt,
1843 '>=': operator.ge,
1844 '=': operator.eq,
1845 '!=': operator.ne,
1846 }
67134eab 1847 operator_rex = re.compile(r'''(?x)\s*
187986a8 1848 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)\s*
1849 (?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1850 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)\s*
083c9df9 1851 ''' % '|'.join(map(re.escape, OPERATORS.keys())))
187986a8 1852 m = operator_rex.fullmatch(filter_spec)
9ddb6925
S
1853 if m:
1854 try:
1855 comparison_value = int(m.group('value'))
1856 except ValueError:
1857 comparison_value = parse_filesize(m.group('value'))
1858 if comparison_value is None:
1859 comparison_value = parse_filesize(m.group('value') + 'B')
1860 if comparison_value is None:
1861 raise ValueError(
1862 'Invalid value %r in format specification %r' % (
67134eab 1863 m.group('value'), filter_spec))
9ddb6925
S
1864 op = OPERATORS[m.group('op')]
1865
083c9df9 1866 if not m:
9ddb6925
S
1867 STR_OPERATORS = {
1868 '=': operator.eq,
10d33b34
YCH
1869 '^=': lambda attr, value: attr.startswith(value),
1870 '$=': lambda attr, value: attr.endswith(value),
1871 '*=': lambda attr, value: value in attr,
1ce9a3cb 1872 '~=': lambda attr, value: value.search(attr) is not None
9ddb6925 1873 }
187986a8 1874 str_operator_rex = re.compile(r'''(?x)\s*
1875 (?P<key>[a-zA-Z0-9._-]+)\s*
1ce9a3cb
LF
1876 (?P<negation>!\s*)?(?P<op>%s)\s*(?P<none_inclusive>\?\s*)?
1877 (?P<quote>["'])?
1878 (?P<value>(?(quote)(?:(?!(?P=quote))[^\\]|\\.)+|[\w.-]+))
1879 (?(quote)(?P=quote))\s*
9ddb6925 1880 ''' % '|'.join(map(re.escape, STR_OPERATORS.keys())))
187986a8 1881 m = str_operator_rex.fullmatch(filter_spec)
9ddb6925 1882 if m:
1ce9a3cb
LF
1883 if m.group('op') == '~=':
1884 comparison_value = re.compile(m.group('value'))
1885 else:
1886 comparison_value = re.sub(r'''\\([\\"'])''', r'\1', m.group('value'))
2cc779f4
S
1887 str_op = STR_OPERATORS[m.group('op')]
1888 if m.group('negation'):
e118a879 1889 op = lambda attr, value: not str_op(attr, value)
2cc779f4
S
1890 else:
1891 op = str_op
083c9df9 1892
9ddb6925 1893 if not m:
187986a8 1894 raise SyntaxError('Invalid filter specification %r' % filter_spec)
083c9df9
PH
1895
1896 def _filter(f):
1897 actual_value = f.get(m.group('key'))
1898 if actual_value is None:
1899 return m.group('none_inclusive')
1900 return op(actual_value, comparison_value)
67134eab
JMF
1901 return _filter
1902
9f1a1c36 1903 def _check_formats(self, formats):
1904 for f in formats:
1905 self.to_screen('[info] Testing format %s' % f['format_id'])
75689fe5 1906 path = self.get_output_path('temp')
1907 if not self._ensure_dir_exists(f'{path}/'):
1908 continue
1909 temp_file = tempfile.NamedTemporaryFile(suffix='.tmp', delete=False, dir=path or None)
9f1a1c36 1910 temp_file.close()
1911 try:
1912 success, _ = self.dl(temp_file.name, f, test=True)
8a82af35 1913 except (DownloadError, OSError, ValueError) + network_exceptions:
9f1a1c36 1914 success = False
1915 finally:
1916 if os.path.exists(temp_file.name):
1917 try:
1918 os.remove(temp_file.name)
1919 except OSError:
1920 self.report_warning('Unable to delete temporary file "%s"' % temp_file.name)
1921 if success:
1922 yield f
1923 else:
1924 self.to_screen('[info] Unable to download format %s. Skipping...' % f['format_id'])
1925
0017d9ad 1926 def _default_format_spec(self, info_dict, download=True):
0017d9ad 1927
af0f7428
S
1928 def can_merge():
1929 merger = FFmpegMergerPP(self)
1930 return merger.available and merger.can_merge()
1931
91ebc640 1932 prefer_best = (
b7b04c78 1933 not self.params.get('simulate')
91ebc640 1934 and download
1935 and (
1936 not can_merge()
21633673 1937 or info_dict.get('is_live') and not self.params.get('live_from_start')
de6000d9 1938 or self.outtmpl_dict['default'] == '-'))
53ed7066 1939 compat = (
1940 prefer_best
1941 or self.params.get('allow_multiple_audio_streams', False)
8a82af35 1942 or 'format-spec' in self.params['compat_opts'])
91ebc640 1943
1944 return (
53ed7066 1945 'best/bestvideo+bestaudio' if prefer_best
1946 else 'bestvideo*+bestaudio/best' if not compat
91ebc640 1947 else 'bestvideo+bestaudio/best')
0017d9ad 1948
67134eab
JMF
1949 def build_format_selector(self, format_spec):
1950 def syntax_error(note, start):
1951 message = (
1952 'Invalid format specification: '
86e5f3ed 1953 '{}\n\t{}\n\t{}^'.format(note, format_spec, ' ' * start[1]))
67134eab
JMF
1954 return SyntaxError(message)
1955
1956 PICKFIRST = 'PICKFIRST'
1957 MERGE = 'MERGE'
1958 SINGLE = 'SINGLE'
0130afb7 1959 GROUP = 'GROUP'
67134eab
JMF
1960 FormatSelector = collections.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
1961
91ebc640 1962 allow_multiple_streams = {'audio': self.params.get('allow_multiple_audio_streams', False),
1963 'video': self.params.get('allow_multiple_video_streams', False)}
909d24dd 1964
9f1a1c36 1965 check_formats = self.params.get('check_formats') == 'selected'
e8e73840 1966
67134eab
JMF
1967 def _parse_filter(tokens):
1968 filter_parts = []
1969 for type, string, start, _, _ in tokens:
1970 if type == tokenize.OP and string == ']':
1971 return ''.join(filter_parts)
1972 else:
1973 filter_parts.append(string)
1974
232541df 1975 def _remove_unused_ops(tokens):
17cc1534 1976 # Remove operators that we don't use and join them with the surrounding strings
232541df
JMF
1977 # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
1978 ALLOWED_OPS = ('/', '+', ',', '(', ')')
1979 last_string, last_start, last_end, last_line = None, None, None, None
1980 for type, string, start, end, line in tokens:
1981 if type == tokenize.OP and string == '[':
1982 if last_string:
1983 yield tokenize.NAME, last_string, last_start, last_end, last_line
1984 last_string = None
1985 yield type, string, start, end, line
1986 # everything inside brackets will be handled by _parse_filter
1987 for type, string, start, end, line in tokens:
1988 yield type, string, start, end, line
1989 if type == tokenize.OP and string == ']':
1990 break
1991 elif type == tokenize.OP and string in ALLOWED_OPS:
1992 if last_string:
1993 yield tokenize.NAME, last_string, last_start, last_end, last_line
1994 last_string = None
1995 yield type, string, start, end, line
1996 elif type in [tokenize.NAME, tokenize.NUMBER, tokenize.OP]:
1997 if not last_string:
1998 last_string = string
1999 last_start = start
2000 last_end = end
2001 else:
2002 last_string += string
2003 if last_string:
2004 yield tokenize.NAME, last_string, last_start, last_end, last_line
2005
cf2ac6df 2006 def _parse_format_selection(tokens, inside_merge=False, inside_choice=False, inside_group=False):
67134eab
JMF
2007 selectors = []
2008 current_selector = None
2009 for type, string, start, _, _ in tokens:
2010 # ENCODING is only defined in python 3.x
2011 if type == getattr(tokenize, 'ENCODING', None):
2012 continue
2013 elif type in [tokenize.NAME, tokenize.NUMBER]:
2014 current_selector = FormatSelector(SINGLE, string, [])
2015 elif type == tokenize.OP:
cf2ac6df
JMF
2016 if string == ')':
2017 if not inside_group:
2018 # ')' will be handled by the parentheses group
2019 tokens.restore_last_token()
67134eab 2020 break
cf2ac6df 2021 elif inside_merge and string in ['/', ',']:
0130afb7
JMF
2022 tokens.restore_last_token()
2023 break
cf2ac6df
JMF
2024 elif inside_choice and string == ',':
2025 tokens.restore_last_token()
2026 break
2027 elif string == ',':
0a31a350
JMF
2028 if not current_selector:
2029 raise syntax_error('"," must follow a format selector', start)
67134eab
JMF
2030 selectors.append(current_selector)
2031 current_selector = None
2032 elif string == '/':
d96d604e
JMF
2033 if not current_selector:
2034 raise syntax_error('"/" must follow a format selector', start)
67134eab 2035 first_choice = current_selector
cf2ac6df 2036 second_choice = _parse_format_selection(tokens, inside_choice=True)
f5f4a27a 2037 current_selector = FormatSelector(PICKFIRST, (first_choice, second_choice), [])
67134eab
JMF
2038 elif string == '[':
2039 if not current_selector:
2040 current_selector = FormatSelector(SINGLE, 'best', [])
2041 format_filter = _parse_filter(tokens)
2042 current_selector.filters.append(format_filter)
0130afb7
JMF
2043 elif string == '(':
2044 if current_selector:
2045 raise syntax_error('Unexpected "("', start)
cf2ac6df
JMF
2046 group = _parse_format_selection(tokens, inside_group=True)
2047 current_selector = FormatSelector(GROUP, group, [])
67134eab 2048 elif string == '+':
d03cfdce 2049 if not current_selector:
2050 raise syntax_error('Unexpected "+"', start)
2051 selector_1 = current_selector
2052 selector_2 = _parse_format_selection(tokens, inside_merge=True)
2053 if not selector_2:
2054 raise syntax_error('Expected a selector', start)
2055 current_selector = FormatSelector(MERGE, (selector_1, selector_2), [])
67134eab 2056 else:
86e5f3ed 2057 raise syntax_error(f'Operator not recognized: "{string}"', start)
67134eab
JMF
2058 elif type == tokenize.ENDMARKER:
2059 break
2060 if current_selector:
2061 selectors.append(current_selector)
2062 return selectors
2063
f8d4ad9a 2064 def _merge(formats_pair):
2065 format_1, format_2 = formats_pair
2066
2067 formats_info = []
2068 formats_info.extend(format_1.get('requested_formats', (format_1,)))
2069 formats_info.extend(format_2.get('requested_formats', (format_2,)))
2070
2071 if not allow_multiple_streams['video'] or not allow_multiple_streams['audio']:
551f9388 2072 get_no_more = {'video': False, 'audio': False}
f8d4ad9a 2073 for (i, fmt_info) in enumerate(formats_info):
551f9388 2074 if fmt_info.get('acodec') == fmt_info.get('vcodec') == 'none':
2075 formats_info.pop(i)
2076 continue
2077 for aud_vid in ['audio', 'video']:
f8d4ad9a 2078 if not allow_multiple_streams[aud_vid] and fmt_info.get(aud_vid[0] + 'codec') != 'none':
2079 if get_no_more[aud_vid]:
2080 formats_info.pop(i)
f5510afe 2081 break
f8d4ad9a 2082 get_no_more[aud_vid] = True
2083
2084 if len(formats_info) == 1:
2085 return formats_info[0]
2086
2087 video_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('vcodec') != 'none']
2088 audio_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('acodec') != 'none']
2089
2090 the_only_video = video_fmts[0] if len(video_fmts) == 1 else None
2091 the_only_audio = audio_fmts[0] if len(audio_fmts) == 1 else None
2092
2093 output_ext = self.params.get('merge_output_format')
2094 if not output_ext:
2095 if the_only_video:
2096 output_ext = the_only_video['ext']
2097 elif the_only_audio and not video_fmts:
2098 output_ext = the_only_audio['ext']
2099 else:
2100 output_ext = 'mkv'
2101
975a0d0d 2102 filtered = lambda *keys: filter(None, (traverse_obj(fmt, *keys) for fmt in formats_info))
2103
f8d4ad9a 2104 new_dict = {
2105 'requested_formats': formats_info,
975a0d0d 2106 'format': '+'.join(filtered('format')),
2107 'format_id': '+'.join(filtered('format_id')),
f8d4ad9a 2108 'ext': output_ext,
975a0d0d 2109 'protocol': '+'.join(map(determine_protocol, formats_info)),
093a1710 2110 'language': '+'.join(orderedSet(filtered('language'))) or None,
2111 'format_note': '+'.join(orderedSet(filtered('format_note'))) or None,
2112 'filesize_approx': sum(filtered('filesize', 'filesize_approx')) or None,
975a0d0d 2113 'tbr': sum(filtered('tbr', 'vbr', 'abr')),
f8d4ad9a 2114 }
2115
2116 if the_only_video:
2117 new_dict.update({
2118 'width': the_only_video.get('width'),
2119 'height': the_only_video.get('height'),
2120 'resolution': the_only_video.get('resolution') or self.format_resolution(the_only_video),
2121 'fps': the_only_video.get('fps'),
49a57e70 2122 'dynamic_range': the_only_video.get('dynamic_range'),
f8d4ad9a 2123 'vcodec': the_only_video.get('vcodec'),
2124 'vbr': the_only_video.get('vbr'),
2125 'stretched_ratio': the_only_video.get('stretched_ratio'),
2126 })
2127
2128 if the_only_audio:
2129 new_dict.update({
2130 'acodec': the_only_audio.get('acodec'),
2131 'abr': the_only_audio.get('abr'),
975a0d0d 2132 'asr': the_only_audio.get('asr'),
f8d4ad9a 2133 })
2134
2135 return new_dict
2136
e8e73840 2137 def _check_formats(formats):
981052c9 2138 if not check_formats:
2139 yield from formats
b5ac45b1 2140 return
9f1a1c36 2141 yield from self._check_formats(formats)
e8e73840 2142
67134eab 2143 def _build_selector_function(selector):
909d24dd 2144 if isinstance(selector, list): # ,
67134eab
JMF
2145 fs = [_build_selector_function(s) for s in selector]
2146
317f7ab6 2147 def selector_function(ctx):
67134eab 2148 for f in fs:
981052c9 2149 yield from f(ctx)
67134eab 2150 return selector_function
909d24dd 2151
2152 elif selector.type == GROUP: # ()
0130afb7 2153 selector_function = _build_selector_function(selector.selector)
909d24dd 2154
2155 elif selector.type == PICKFIRST: # /
67134eab
JMF
2156 fs = [_build_selector_function(s) for s in selector.selector]
2157
317f7ab6 2158 def selector_function(ctx):
67134eab 2159 for f in fs:
317f7ab6 2160 picked_formats = list(f(ctx))
67134eab
JMF
2161 if picked_formats:
2162 return picked_formats
2163 return []
67134eab 2164
981052c9 2165 elif selector.type == MERGE: # +
2166 selector_1, selector_2 = map(_build_selector_function, selector.selector)
2167
2168 def selector_function(ctx):
adbc4ec4 2169 for pair in itertools.product(selector_1(ctx), selector_2(ctx)):
981052c9 2170 yield _merge(pair)
2171
909d24dd 2172 elif selector.type == SINGLE: # atom
598d185d 2173 format_spec = selector.selector or 'best'
909d24dd 2174
f8d4ad9a 2175 # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
909d24dd 2176 if format_spec == 'all':
2177 def selector_function(ctx):
9222c381 2178 yield from _check_formats(ctx['formats'][::-1])
f8d4ad9a 2179 elif format_spec == 'mergeall':
2180 def selector_function(ctx):
316f2650 2181 formats = list(_check_formats(
2182 f for f in ctx['formats'] if f.get('vcodec') != 'none' or f.get('acodec') != 'none'))
e01d6aa4 2183 if not formats:
2184 return
921b76ca 2185 merged_format = formats[-1]
2186 for f in formats[-2::-1]:
f8d4ad9a 2187 merged_format = _merge((merged_format, f))
2188 yield merged_format
909d24dd 2189
2190 else:
85e801a9 2191 format_fallback, seperate_fallback, format_reverse, format_idx = False, None, True, 1
eff63539 2192 mobj = re.match(
2193 r'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
2194 format_spec)
2195 if mobj is not None:
2196 format_idx = int_or_none(mobj.group('n'), default=1)
e8e73840 2197 format_reverse = mobj.group('bw')[0] == 'b'
eff63539 2198 format_type = (mobj.group('type') or [None])[0]
2199 not_format_type = {'v': 'a', 'a': 'v'}.get(format_type)
2200 format_modified = mobj.group('mod') is not None
909d24dd 2201
2202 format_fallback = not format_type and not format_modified # for b, w
8326b00a 2203 _filter_f = (
eff63539 2204 (lambda f: f.get('%scodec' % format_type) != 'none')
2205 if format_type and format_modified # bv*, ba*, wv*, wa*
2206 else (lambda f: f.get('%scodec' % not_format_type) == 'none')
2207 if format_type # bv, ba, wv, wa
2208 else (lambda f: f.get('vcodec') != 'none' and f.get('acodec') != 'none')
2209 if not format_modified # b, w
8326b00a 2210 else lambda f: True) # b*, w*
2211 filter_f = lambda f: _filter_f(f) and (
2212 f.get('vcodec') != 'none' or f.get('acodec') != 'none')
67134eab 2213 else:
48ee10ee 2214 if format_spec in self._format_selection_exts['audio']:
b11c04a8 2215 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none'
48ee10ee 2216 elif format_spec in self._format_selection_exts['video']:
b11c04a8 2217 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none' and f.get('vcodec') != 'none'
85e801a9 2218 seperate_fallback = lambda f: f.get('ext') == format_spec and f.get('vcodec') != 'none'
48ee10ee 2219 elif format_spec in self._format_selection_exts['storyboards']:
b11c04a8 2220 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') == 'none' and f.get('vcodec') == 'none'
2221 else:
b5ae35ee 2222 filter_f = lambda f: f.get('format_id') == format_spec # id
909d24dd 2223
2224 def selector_function(ctx):
2225 formats = list(ctx['formats'])
909d24dd 2226 matches = list(filter(filter_f, formats)) if filter_f is not None else formats
85e801a9 2227 if not matches:
2228 if format_fallback and ctx['incomplete_formats']:
2229 # for extractors with incomplete formats (audio only (soundcloud)
2230 # or video only (imgur)) best/worst will fallback to
2231 # best/worst {video,audio}-only format
2232 matches = formats
2233 elif seperate_fallback and not ctx['has_merged_format']:
2234 # for compatibility with youtube-dl when there is no pre-merged format
2235 matches = list(filter(seperate_fallback, formats))
981052c9 2236 matches = LazyList(_check_formats(matches[::-1 if format_reverse else 1]))
2237 try:
e8e73840 2238 yield matches[format_idx - 1]
4abea8ca 2239 except LazyList.IndexError:
981052c9 2240 return
083c9df9 2241
67134eab 2242 filters = [self._build_format_filter(f) for f in selector.filters]
083c9df9 2243
317f7ab6 2244 def final_selector(ctx):
adbc4ec4 2245 ctx_copy = dict(ctx)
67134eab 2246 for _filter in filters:
317f7ab6
S
2247 ctx_copy['formats'] = list(filter(_filter, ctx_copy['formats']))
2248 return selector_function(ctx_copy)
67134eab 2249 return final_selector
083c9df9 2250
0f06bcd7 2251 stream = io.BytesIO(format_spec.encode())
0130afb7 2252 try:
f9934b96 2253 tokens = list(_remove_unused_ops(tokenize.tokenize(stream.readline)))
0130afb7
JMF
2254 except tokenize.TokenError:
2255 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec)))
2256
86e5f3ed 2257 class TokenIterator:
0130afb7
JMF
2258 def __init__(self, tokens):
2259 self.tokens = tokens
2260 self.counter = 0
2261
2262 def __iter__(self):
2263 return self
2264
2265 def __next__(self):
2266 if self.counter >= len(self.tokens):
2267 raise StopIteration()
2268 value = self.tokens[self.counter]
2269 self.counter += 1
2270 return value
2271
2272 next = __next__
2273
2274 def restore_last_token(self):
2275 self.counter -= 1
2276
2277 parsed_selector = _parse_format_selection(iter(TokenIterator(tokens)))
67134eab 2278 return _build_selector_function(parsed_selector)
a9c58ad9 2279
e5660ee6 2280 def _calc_headers(self, info_dict):
8b7539d2 2281 res = merge_headers(self.params['http_headers'], info_dict.get('http_headers') or {})
e5660ee6 2282
c487cf00 2283 cookies = self._calc_cookies(info_dict['url'])
e5660ee6
JMF
2284 if cookies:
2285 res['Cookie'] = cookies
2286
0016b84e
S
2287 if 'X-Forwarded-For' not in res:
2288 x_forwarded_for_ip = info_dict.get('__x_forwarded_for_ip')
2289 if x_forwarded_for_ip:
2290 res['X-Forwarded-For'] = x_forwarded_for_ip
2291
e5660ee6
JMF
2292 return res
2293
c487cf00 2294 def _calc_cookies(self, url):
2295 pr = sanitized_Request(url)
e5660ee6 2296 self.cookiejar.add_cookie_header(pr)
662435f7 2297 return pr.get_header('Cookie')
e5660ee6 2298
9f1a1c36 2299 def _sort_thumbnails(self, thumbnails):
2300 thumbnails.sort(key=lambda t: (
2301 t.get('preference') if t.get('preference') is not None else -1,
2302 t.get('width') if t.get('width') is not None else -1,
2303 t.get('height') if t.get('height') is not None else -1,
2304 t.get('id') if t.get('id') is not None else '',
2305 t.get('url')))
2306
b0249bca 2307 def _sanitize_thumbnails(self, info_dict):
bc516a3f 2308 thumbnails = info_dict.get('thumbnails')
2309 if thumbnails is None:
2310 thumbnail = info_dict.get('thumbnail')
2311 if thumbnail:
2312 info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}]
9f1a1c36 2313 if not thumbnails:
2314 return
2315
2316 def check_thumbnails(thumbnails):
2317 for t in thumbnails:
2318 self.to_screen(f'[info] Testing thumbnail {t["id"]}')
2319 try:
2320 self.urlopen(HEADRequest(t['url']))
2321 except network_exceptions as err:
2322 self.to_screen(f'[info] Unable to connect to thumbnail {t["id"]} URL {t["url"]!r} - {err}. Skipping...')
2323 continue
2324 yield t
2325
2326 self._sort_thumbnails(thumbnails)
2327 for i, t in enumerate(thumbnails):
2328 if t.get('id') is None:
2329 t['id'] = '%d' % i
2330 if t.get('width') and t.get('height'):
2331 t['resolution'] = '%dx%d' % (t['width'], t['height'])
2332 t['url'] = sanitize_url(t['url'])
2333
2334 if self.params.get('check_formats') is True:
282f5709 2335 info_dict['thumbnails'] = LazyList(check_thumbnails(thumbnails[::-1]), reverse=True)
9f1a1c36 2336 else:
2337 info_dict['thumbnails'] = thumbnails
bc516a3f 2338
03f83004
LNO
2339 def _fill_common_fields(self, info_dict, is_video=True):
2340 # TODO: move sanitization here
2341 if is_video:
2342 # playlists are allowed to lack "title"
d4736fdb 2343 title = info_dict.get('title', NO_DEFAULT)
2344 if title is NO_DEFAULT:
03f83004
LNO
2345 raise ExtractorError('Missing "title" field in extractor result',
2346 video_id=info_dict['id'], ie=info_dict['extractor'])
d4736fdb 2347 info_dict['fulltitle'] = title
2348 if not title:
2349 if title == '':
2350 self.write_debug('Extractor gave empty title. Creating a generic title')
2351 else:
2352 self.report_warning('Extractor failed to obtain "title". Creating a generic title instead')
1d485a1a 2353 info_dict['title'] = f'{info_dict["extractor"].replace(":", "-")} video #{info_dict["id"]}'
03f83004
LNO
2354
2355 if info_dict.get('duration') is not None:
2356 info_dict['duration_string'] = formatSeconds(info_dict['duration'])
2357
2358 for ts_key, date_key in (
2359 ('timestamp', 'upload_date'),
2360 ('release_timestamp', 'release_date'),
2361 ('modified_timestamp', 'modified_date'),
2362 ):
2363 if info_dict.get(date_key) is None and info_dict.get(ts_key) is not None:
2364 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
2365 # see http://bugs.python.org/issue1646728)
19a03940 2366 with contextlib.suppress(ValueError, OverflowError, OSError):
03f83004
LNO
2367 upload_date = datetime.datetime.utcfromtimestamp(info_dict[ts_key])
2368 info_dict[date_key] = upload_date.strftime('%Y%m%d')
03f83004
LNO
2369
2370 live_keys = ('is_live', 'was_live')
2371 live_status = info_dict.get('live_status')
2372 if live_status is None:
2373 for key in live_keys:
2374 if info_dict.get(key) is False:
2375 continue
2376 if info_dict.get(key):
2377 live_status = key
2378 break
2379 if all(info_dict.get(key) is False for key in live_keys):
2380 live_status = 'not_live'
2381 if live_status:
2382 info_dict['live_status'] = live_status
2383 for key in live_keys:
2384 if info_dict.get(key) is None:
2385 info_dict[key] = (live_status == key)
2386
2387 # Auto generate title fields corresponding to the *_number fields when missing
2388 # in order to always have clean titles. This is very common for TV series.
2389 for field in ('chapter', 'season', 'episode'):
2390 if info_dict.get('%s_number' % field) is not None and not info_dict.get(field):
2391 info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field])
2392
415f8d51 2393 def _raise_pending_errors(self, info):
2394 err = info.pop('__pending_error', None)
2395 if err:
2396 self.report_error(err, tb=False)
2397
dd82ffea
JMF
2398 def process_video_result(self, info_dict, download=True):
2399 assert info_dict.get('_type', 'video') == 'video'
9c906919 2400 self._num_videos += 1
dd82ffea 2401
bec1fad2 2402 if 'id' not in info_dict:
fc08bdd6 2403 raise ExtractorError('Missing "id" field in extractor result', ie=info_dict['extractor'])
2404 elif not info_dict.get('id'):
2405 raise ExtractorError('Extractor failed to obtain "id"', ie=info_dict['extractor'])
455a15e2 2406
c9969434
S
2407 def report_force_conversion(field, field_not, conversion):
2408 self.report_warning(
2409 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
2410 % (field, field_not, conversion))
2411
2412 def sanitize_string_field(info, string_field):
2413 field = info.get(string_field)
2414 if field is None or isinstance(field, compat_str):
2415 return
2416 report_force_conversion(string_field, 'a string', 'string')
2417 info[string_field] = compat_str(field)
2418
2419 def sanitize_numeric_fields(info):
2420 for numeric_field in self._NUMERIC_FIELDS:
2421 field = info.get(numeric_field)
f9934b96 2422 if field is None or isinstance(field, (int, float)):
c9969434
S
2423 continue
2424 report_force_conversion(numeric_field, 'numeric', 'int')
2425 info[numeric_field] = int_or_none(field)
2426
2427 sanitize_string_field(info_dict, 'id')
2428 sanitize_numeric_fields(info_dict)
4c3f8c3f 2429 if (info_dict.get('duration') or 0) <= 0 and info_dict.pop('duration', None):
50e93e03 2430 self.report_warning('"duration" field is negative, there is an error in extractor')
be6217b2 2431
dd82ffea
JMF
2432 if 'playlist' not in info_dict:
2433 # It isn't part of a playlist
2434 info_dict['playlist'] = None
2435 info_dict['playlist_index'] = None
2436
bc516a3f 2437 self._sanitize_thumbnails(info_dict)
d5519808 2438
536a55da 2439 thumbnail = info_dict.get('thumbnail')
bc516a3f 2440 thumbnails = info_dict.get('thumbnails')
536a55da
S
2441 if thumbnail:
2442 info_dict['thumbnail'] = sanitize_url(thumbnail)
2443 elif thumbnails:
d5519808
PH
2444 info_dict['thumbnail'] = thumbnails[-1]['url']
2445
ae30b840 2446 if info_dict.get('display_id') is None and 'id' in info_dict:
0afef30b
PH
2447 info_dict['display_id'] = info_dict['id']
2448
03f83004 2449 self._fill_common_fields(info_dict)
33d2fc2f 2450
05108a49
S
2451 for cc_kind in ('subtitles', 'automatic_captions'):
2452 cc = info_dict.get(cc_kind)
2453 if cc:
2454 for _, subtitle in cc.items():
2455 for subtitle_format in subtitle:
2456 if subtitle_format.get('url'):
2457 subtitle_format['url'] = sanitize_url(subtitle_format['url'])
2458 if subtitle_format.get('ext') is None:
2459 subtitle_format['ext'] = determine_ext(subtitle_format['url']).lower()
2460
2461 automatic_captions = info_dict.get('automatic_captions')
4bba3716 2462 subtitles = info_dict.get('subtitles')
4bba3716 2463
360e1ca5 2464 info_dict['requested_subtitles'] = self.process_subtitles(
05108a49 2465 info_dict['id'], subtitles, automatic_captions)
a504ced0 2466
dd82ffea
JMF
2467 if info_dict.get('formats') is None:
2468 # There's only one format available
2469 formats = [info_dict]
2470 else:
2471 formats = info_dict['formats']
2472
0a5a191a 2473 # or None ensures --clean-infojson removes it
2474 info_dict['_has_drm'] = any(f.get('has_drm') for f in formats) or None
88acdbc2 2475 if not self.params.get('allow_unplayable_formats'):
2476 formats = [f for f in formats if not f.get('has_drm')]
0a5a191a 2477 if info_dict['_has_drm'] and all(
c0b6e5c7 2478 f.get('acodec') == f.get('vcodec') == 'none' for f in formats):
2479 self.report_warning(
2480 'This video is DRM protected and only images are available for download. '
2481 'Use --list-formats to see them')
88acdbc2 2482
319b6059 2483 get_from_start = not info_dict.get('is_live') or bool(self.params.get('live_from_start'))
2484 if not get_from_start:
2485 info_dict['title'] += ' ' + datetime.datetime.now().strftime('%Y-%m-%d %H:%M')
2486 if info_dict.get('is_live') and formats:
adbc4ec4 2487 formats = [f for f in formats if bool(f.get('is_from_start')) == get_from_start]
319b6059 2488 if get_from_start and not formats:
a44ca5a4 2489 self.raise_no_formats(info_dict, msg=(
2490 '--live-from-start is passed, but there are no formats that can be downloaded from the start. '
2491 'If you want to download from the current time, use --no-live-from-start'))
adbc4ec4 2492
db95dc13 2493 if not formats:
1151c407 2494 self.raise_no_formats(info_dict)
db95dc13 2495
73af5cc8
S
2496 def is_wellformed(f):
2497 url = f.get('url')
a5ac0c47 2498 if not url:
73af5cc8
S
2499 self.report_warning(
2500 '"url" field is missing or empty - skipping format, '
2501 'there is an error in extractor')
a5ac0c47
S
2502 return False
2503 if isinstance(url, bytes):
2504 sanitize_string_field(f, 'url')
2505 return True
73af5cc8
S
2506
2507 # Filter out malformed formats for better extraction robustness
2508 formats = list(filter(is_wellformed, formats))
2509
181c7053
S
2510 formats_dict = {}
2511
dd82ffea 2512 # We check that all the formats have the format and format_id fields
db95dc13 2513 for i, format in enumerate(formats):
c9969434
S
2514 sanitize_string_field(format, 'format_id')
2515 sanitize_numeric_fields(format)
dcf77cf1 2516 format['url'] = sanitize_url(format['url'])
e74e3b63 2517 if not format.get('format_id'):
8016c922 2518 format['format_id'] = compat_str(i)
e2effb08
S
2519 else:
2520 # Sanitize format_id from characters used in format selector expression
ec85ded8 2521 format['format_id'] = re.sub(r'[\s,/+\[\]()]', '_', format['format_id'])
181c7053
S
2522 format_id = format['format_id']
2523 if format_id not in formats_dict:
2524 formats_dict[format_id] = []
2525 formats_dict[format_id].append(format)
2526
2527 # Make sure all formats have unique format_id
03b4de72 2528 common_exts = set(itertools.chain(*self._format_selection_exts.values()))
181c7053 2529 for format_id, ambiguous_formats in formats_dict.items():
48ee10ee 2530 ambigious_id = len(ambiguous_formats) > 1
2531 for i, format in enumerate(ambiguous_formats):
2532 if ambigious_id:
181c7053 2533 format['format_id'] = '%s-%d' % (format_id, i)
48ee10ee 2534 if format.get('ext') is None:
2535 format['ext'] = determine_ext(format['url']).lower()
2536 # Ensure there is no conflict between id and ext in format selection
2537 # See https://github.com/yt-dlp/yt-dlp/issues/1282
2538 if format['format_id'] != format['ext'] and format['format_id'] in common_exts:
2539 format['format_id'] = 'f%s' % format['format_id']
181c7053
S
2540
2541 for i, format in enumerate(formats):
8c51aa65 2542 if format.get('format') is None:
6febd1c1 2543 format['format'] = '{id} - {res}{note}'.format(
8c51aa65
JMF
2544 id=format['format_id'],
2545 res=self.format_resolution(format),
b868936c 2546 note=format_field(format, 'format_note', ' (%s)'),
8c51aa65 2547 )
6f0be937 2548 if format.get('protocol') is None:
b5559424 2549 format['protocol'] = determine_protocol(format)
239df021 2550 if format.get('resolution') is None:
2551 format['resolution'] = self.format_resolution(format, default=None)
176f1866 2552 if format.get('dynamic_range') is None and format.get('vcodec') != 'none':
2553 format['dynamic_range'] = 'SDR'
f2fe69c7 2554 if (info_dict.get('duration') and format.get('tbr')
2555 and not format.get('filesize') and not format.get('filesize_approx')):
2556 format['filesize_approx'] = info_dict['duration'] * format['tbr'] * (1024 / 8)
2557
e5660ee6
JMF
2558 # Add HTTP headers, so that external programs can use them from the
2559 # json output
2560 full_format_info = info_dict.copy()
2561 full_format_info.update(format)
2562 format['http_headers'] = self._calc_headers(full_format_info)
0016b84e
S
2563 # Remove private housekeeping stuff
2564 if '__x_forwarded_for_ip' in info_dict:
2565 del info_dict['__x_forwarded_for_ip']
dd82ffea 2566
9f1a1c36 2567 if self.params.get('check_formats') is True:
282f5709 2568 formats = LazyList(self._check_formats(formats[::-1]), reverse=True)
9f1a1c36 2569
88acdbc2 2570 if not formats or formats[0] is not info_dict:
b3d9ef88
JMF
2571 # only set the 'formats' fields if the original info_dict list them
2572 # otherwise we end up with a circular reference, the first (and unique)
f89197d7 2573 # element in the 'formats' field in info_dict is info_dict itself,
dfb1b146 2574 # which can't be exported to json
b3d9ef88 2575 info_dict['formats'] = formats
4ec82a72 2576
2577 info_dict, _ = self.pre_process(info_dict)
2578
6db9c4d5 2579 if self._match_entry(info_dict, incomplete=self._format_fields) is not None:
09b49e1f 2580 return info_dict
2581
2582 self.post_extract(info_dict)
2583 info_dict, _ = self.pre_process(info_dict, 'after_filter')
2584
093a1710 2585 # The pre-processors may have modified the formats
2586 formats = info_dict.get('formats', [info_dict])
2587
fa9f30b8 2588 list_only = self.params.get('simulate') is None and (
2589 self.params.get('list_thumbnails') or self.params.get('listformats') or self.params.get('listsubtitles'))
2590 interactive_format_selection = not list_only and self.format_selector == '-'
b7b04c78 2591 if self.params.get('list_thumbnails'):
2592 self.list_thumbnails(info_dict)
b7b04c78 2593 if self.params.get('listsubtitles'):
2594 if 'automatic_captions' in info_dict:
2595 self.list_subtitles(
2596 info_dict['id'], automatic_captions, 'automatic captions')
2597 self.list_subtitles(info_dict['id'], subtitles, 'subtitles')
fa9f30b8 2598 if self.params.get('listformats') or interactive_format_selection:
b69fd25c 2599 self.list_formats(info_dict)
169dbde9 2600 if list_only:
b7b04c78 2601 # Without this printing, -F --print-json will not work
169dbde9 2602 self.__forced_printings(info_dict, self.prepare_filename(info_dict), incomplete=True)
c487cf00 2603 return info_dict
bfaae0a7 2604
187986a8 2605 format_selector = self.format_selector
2606 if format_selector is None:
0017d9ad 2607 req_format = self._default_format_spec(info_dict, download=download)
0760b0a7 2608 self.write_debug('Default format spec: %s' % req_format)
187986a8 2609 format_selector = self.build_format_selector(req_format)
317f7ab6 2610
fa9f30b8 2611 while True:
2612 if interactive_format_selection:
2613 req_format = input(
2614 self._format_screen('\nEnter format selector: ', self.Styles.EMPHASIS))
2615 try:
2616 format_selector = self.build_format_selector(req_format)
2617 except SyntaxError as err:
2618 self.report_error(err, tb=False, is_error=False)
2619 continue
2620
85e801a9 2621 formats_to_download = list(format_selector({
fa9f30b8 2622 'formats': formats,
85e801a9 2623 'has_merged_format': any('none' not in (f.get('acodec'), f.get('vcodec')) for f in formats),
2624 'incomplete_formats': (
2625 # All formats are video-only or
2626 all(f.get('vcodec') != 'none' and f.get('acodec') == 'none' for f in formats)
2627 # all formats are audio-only
2628 or all(f.get('vcodec') == 'none' and f.get('acodec') != 'none' for f in formats)),
2629 }))
fa9f30b8 2630 if interactive_format_selection and not formats_to_download:
2631 self.report_error('Requested format is not available', tb=False, is_error=False)
2632 continue
2633 break
317f7ab6 2634
dd82ffea 2635 if not formats_to_download:
b7da73eb 2636 if not self.params.get('ignore_no_formats_error'):
c0b6e5c7 2637 raise ExtractorError(
2638 'Requested format is not available. Use --list-formats for a list of available formats',
2639 expected=True, video_id=info_dict['id'], ie=info_dict['extractor'])
b62fa6d7 2640 self.report_warning('Requested format is not available')
2641 # Process what we can, even without any available formats.
2642 formats_to_download = [{}]
a13e6848 2643
b62fa6d7 2644 best_format = formats_to_download[-1]
2645 if download:
2646 if best_format:
2647 self.to_screen(
2648 f'[info] {info_dict["id"]}: Downloading {len(formats_to_download)} format(s): '
2649 + ', '.join([f['format_id'] for f in formats_to_download]))
a13e6848 2650 max_downloads_reached = False
f46e2f9d 2651 for i, fmt in enumerate(formats_to_download):
09b49e1f 2652 formats_to_download[i] = new_info = self._copy_infodict(info_dict)
b7da73eb 2653 new_info.update(fmt)
a13e6848 2654 try:
2655 self.process_info(new_info)
2656 except MaxDownloadsReached:
2657 max_downloads_reached = True
415f8d51 2658 self._raise_pending_errors(new_info)
f46e2f9d 2659 # Remove copied info
2660 for key, val in tuple(new_info.items()):
2661 if info_dict.get(key) == val:
2662 new_info.pop(key)
a13e6848 2663 if max_downloads_reached:
2664 break
ebed8b37 2665
86e5f3ed 2666 write_archive = {f.get('__write_download_archive', False) for f in formats_to_download}
a13e6848 2667 assert write_archive.issubset({True, False, 'ignore'})
2668 if True in write_archive and False not in write_archive:
2669 self.record_download_archive(info_dict)
be72c624 2670
2671 info_dict['requested_downloads'] = formats_to_download
ed5835b4 2672 info_dict = self.run_all_pps('after_video', info_dict)
a13e6848 2673 if max_downloads_reached:
2674 raise MaxDownloadsReached()
ebed8b37 2675
49a57e70 2676 # We update the info dict with the selected best quality format (backwards compatibility)
be72c624 2677 info_dict.update(best_format)
dd82ffea
JMF
2678 return info_dict
2679
98c70d6f 2680 def process_subtitles(self, video_id, normal_subtitles, automatic_captions):
a504ced0 2681 """Select the requested subtitles and their format"""
d8a58ddc 2682 available_subs, normal_sub_langs = {}, []
98c70d6f
JMF
2683 if normal_subtitles and self.params.get('writesubtitles'):
2684 available_subs.update(normal_subtitles)
d8a58ddc 2685 normal_sub_langs = tuple(normal_subtitles.keys())
98c70d6f
JMF
2686 if automatic_captions and self.params.get('writeautomaticsub'):
2687 for lang, cap_info in automatic_captions.items():
360e1ca5
JMF
2688 if lang not in available_subs:
2689 available_subs[lang] = cap_info
2690
4d171848
JMF
2691 if (not self.params.get('writesubtitles') and not
2692 self.params.get('writeautomaticsub') or not
2693 available_subs):
2694 return None
a504ced0 2695
d8a58ddc 2696 all_sub_langs = tuple(available_subs.keys())
a504ced0 2697 if self.params.get('allsubtitles', False):
c32b0aab 2698 requested_langs = all_sub_langs
2699 elif self.params.get('subtitleslangs', False):
77c4a9ef 2700 # A list is used so that the order of languages will be the same as
2701 # given in subtitleslangs. See https://github.com/yt-dlp/yt-dlp/issues/1041
2702 requested_langs = []
2703 for lang_re in self.params.get('subtitleslangs'):
77c4a9ef 2704 discard = lang_re[0] == '-'
c32b0aab 2705 if discard:
77c4a9ef 2706 lang_re = lang_re[1:]
3aa91540 2707 if lang_re == 'all':
2708 if discard:
2709 requested_langs = []
2710 else:
2711 requested_langs.extend(all_sub_langs)
2712 continue
77c4a9ef 2713 current_langs = filter(re.compile(lang_re + '$').match, all_sub_langs)
c32b0aab 2714 if discard:
2715 for lang in current_langs:
77c4a9ef 2716 while lang in requested_langs:
2717 requested_langs.remove(lang)
c32b0aab 2718 else:
77c4a9ef 2719 requested_langs.extend(current_langs)
2720 requested_langs = orderedSet(requested_langs)
d8a58ddc 2721 elif normal_sub_langs:
2722 requested_langs = ['en'] if 'en' in normal_sub_langs else normal_sub_langs[:1]
a504ced0 2723 else:
d8a58ddc 2724 requested_langs = ['en'] if 'en' in all_sub_langs else all_sub_langs[:1]
ad3dc496 2725 if requested_langs:
2726 self.write_debug('Downloading subtitles: %s' % ', '.join(requested_langs))
a504ced0
JMF
2727
2728 formats_query = self.params.get('subtitlesformat', 'best')
2729 formats_preference = formats_query.split('/') if formats_query else []
2730 subs = {}
2731 for lang in requested_langs:
2732 formats = available_subs.get(lang)
2733 if formats is None:
86e5f3ed 2734 self.report_warning(f'{lang} subtitles not available for {video_id}')
a504ced0 2735 continue
a504ced0
JMF
2736 for ext in formats_preference:
2737 if ext == 'best':
2738 f = formats[-1]
2739 break
2740 matches = list(filter(lambda f: f['ext'] == ext, formats))
2741 if matches:
2742 f = matches[-1]
2743 break
2744 else:
2745 f = formats[-1]
2746 self.report_warning(
2747 'No subtitle format found matching "%s" for language %s, '
2748 'using %s' % (formats_query, lang, f['ext']))
2749 subs[lang] = f
2750 return subs
2751
bb66c247 2752 def _forceprint(self, key, info_dict):
2753 if info_dict is None:
2754 return
2755 info_copy = info_dict.copy()
2756 info_copy['formats_table'] = self.render_formats_table(info_dict)
2757 info_copy['thumbnails_table'] = self.render_thumbnails_table(info_dict)
2758 info_copy['subtitles_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('subtitles'))
2759 info_copy['automatic_captions_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('automatic_captions'))
2760
2761 def format_tmpl(tmpl):
2762 mobj = re.match(r'\w+(=?)$', tmpl)
2763 if mobj and mobj.group(1):
2764 return f'{tmpl[:-1]} = %({tmpl[:-1]})r'
2765 elif mobj:
2766 return f'%({tmpl})s'
2767 return tmpl
8130779d 2768
bb66c247 2769 for tmpl in self.params['forceprint'].get(key, []):
2770 self.to_stdout(self.evaluate_outtmpl(format_tmpl(tmpl), info_copy))
2771
2772 for tmpl, file_tmpl in self.params['print_to_file'].get(key, []):
5127e92a 2773 filename = self.prepare_filename(info_dict, outtmpl=file_tmpl)
bb66c247 2774 tmpl = format_tmpl(tmpl)
2775 self.to_screen(f'[info] Writing {tmpl!r} to: {filename}')
8d93e69d 2776 if self._ensure_dir_exists(filename):
86e5f3ed 2777 with open(filename, 'a', encoding='utf-8') as f:
8d93e69d 2778 f.write(self.evaluate_outtmpl(tmpl, info_copy) + '\n')
ca30f449 2779
d06daf23 2780 def __forced_printings(self, info_dict, filename, incomplete):
53c18592 2781 def print_mandatory(field, actual_field=None):
2782 if actual_field is None:
2783 actual_field = field
d06daf23 2784 if (self.params.get('force%s' % field, False)
53c18592 2785 and (not incomplete or info_dict.get(actual_field) is not None)):
2786 self.to_stdout(info_dict[actual_field])
d06daf23
S
2787
2788 def print_optional(field):
2789 if (self.params.get('force%s' % field, False)
2790 and info_dict.get(field) is not None):
2791 self.to_stdout(info_dict[field])
2792
53c18592 2793 info_dict = info_dict.copy()
2794 if filename is not None:
2795 info_dict['filename'] = filename
2796 if info_dict.get('requested_formats') is not None:
2797 # For RTMP URLs, also include the playpath
2798 info_dict['urls'] = '\n'.join(f['url'] + f.get('play_path', '') for f in info_dict['requested_formats'])
10331a26 2799 elif info_dict.get('url'):
53c18592 2800 info_dict['urls'] = info_dict['url'] + info_dict.get('play_path', '')
2801
bb66c247 2802 if (self.params.get('forcejson')
2803 or self.params['forceprint'].get('video')
2804 or self.params['print_to_file'].get('video')):
2b8a2973 2805 self.post_extract(info_dict)
bb66c247 2806 self._forceprint('video', info_dict)
53c18592 2807
d06daf23
S
2808 print_mandatory('title')
2809 print_mandatory('id')
53c18592 2810 print_mandatory('url', 'urls')
d06daf23
S
2811 print_optional('thumbnail')
2812 print_optional('description')
53c18592 2813 print_optional('filename')
b868936c 2814 if self.params.get('forceduration') and info_dict.get('duration') is not None:
d06daf23
S
2815 self.to_stdout(formatSeconds(info_dict['duration']))
2816 print_mandatory('format')
53c18592 2817
2b8a2973 2818 if self.params.get('forcejson'):
6e84b215 2819 self.to_stdout(json.dumps(self.sanitize_info(info_dict)))
d06daf23 2820
e8e73840 2821 def dl(self, name, info, subtitle=False, test=False):
88acdbc2 2822 if not info.get('url'):
1151c407 2823 self.raise_no_formats(info, True)
e8e73840 2824
2825 if test:
2826 verbose = self.params.get('verbose')
2827 params = {
2828 'test': True,
a169858f 2829 'quiet': self.params.get('quiet') or not verbose,
e8e73840 2830 'verbose': verbose,
2831 'noprogress': not verbose,
2832 'nopart': True,
2833 'skip_unavailable_fragments': False,
2834 'keep_fragments': False,
2835 'overwrites': True,
2836 '_no_ytdl_file': True,
2837 }
2838 else:
2839 params = self.params
96fccc10 2840 fd = get_suitable_downloader(info, params, to_stdout=(name == '-'))(self, params)
e8e73840 2841 if not test:
2842 for ph in self._progress_hooks:
2843 fd.add_progress_hook(ph)
42676437
M
2844 urls = '", "'.join(
2845 (f['url'].split(',')[0] + ',<data>' if f['url'].startswith('data:') else f['url'])
2846 for f in info.get('requested_formats', []) or [info])
3a408f9d 2847 self.write_debug(f'Invoking {fd.FD_NAME} downloader on "{urls}"')
03b4de72 2848
adbc4ec4
THD
2849 # Note: Ideally info should be a deep-copied so that hooks cannot modify it.
2850 # But it may contain objects that are not deep-copyable
2851 new_info = self._copy_infodict(info)
e8e73840 2852 if new_info.get('http_headers') is None:
2853 new_info['http_headers'] = self._calc_headers(new_info)
2854 return fd.download(name, new_info, subtitle)
2855
e04938ab 2856 def existing_file(self, filepaths, *, default_overwrite=True):
2857 existing_files = list(filter(os.path.exists, orderedSet(filepaths)))
2858 if existing_files and not self.params.get('overwrites', default_overwrite):
2859 return existing_files[0]
2860
2861 for file in existing_files:
2862 self.report_file_delete(file)
2863 os.remove(file)
2864 return None
2865
8222d8de 2866 def process_info(self, info_dict):
09b49e1f 2867 """Process a single resolved IE result. (Modifies it in-place)"""
8222d8de
JMF
2868
2869 assert info_dict.get('_type', 'video') == 'video'
f46e2f9d 2870 original_infodict = info_dict
fd288278 2871
4513a41a 2872 if 'format' not in info_dict and 'ext' in info_dict:
8222d8de
JMF
2873 info_dict['format'] = info_dict['ext']
2874
09b49e1f 2875 # This is mostly just for backward compatibility of process_info
2876 # As a side-effect, this allows for format-specific filters
c77495e3 2877 if self._match_entry(info_dict) is not None:
9e907ebd 2878 info_dict['__write_download_archive'] = 'ignore'
8222d8de
JMF
2879 return
2880
09b49e1f 2881 # Does nothing under normal operation - for backward compatibility of process_info
277d6ff5 2882 self.post_extract(info_dict)
0c14d66a 2883 self._num_downloads += 1
8222d8de 2884
dcf64d43 2885 # info_dict['_filename'] needs to be set for backward compatibility
de6000d9 2886 info_dict['_filename'] = full_filename = self.prepare_filename(info_dict, warn=True)
2887 temp_filename = self.prepare_filename(info_dict, 'temp')
0202b52a 2888 files_to_move = {}
8222d8de
JMF
2889
2890 # Forced printings
4513a41a 2891 self.__forced_printings(info_dict, full_filename, incomplete=('format' not in info_dict))
8222d8de 2892
ca6d59d2 2893 def check_max_downloads():
2894 if self._num_downloads >= float(self.params.get('max_downloads') or 'inf'):
2895 raise MaxDownloadsReached()
2896
b7b04c78 2897 if self.params.get('simulate'):
9e907ebd 2898 info_dict['__write_download_archive'] = self.params.get('force_write_download_archive')
ca6d59d2 2899 check_max_downloads()
8222d8de
JMF
2900 return
2901
de6000d9 2902 if full_filename is None:
8222d8de 2903 return
e92caff5 2904 if not self._ensure_dir_exists(encodeFilename(full_filename)):
0202b52a 2905 return
e92caff5 2906 if not self._ensure_dir_exists(encodeFilename(temp_filename)):
8222d8de
JMF
2907 return
2908
80c03fa9 2909 if self._write_description('video', info_dict,
2910 self.prepare_filename(info_dict, 'description')) is None:
2911 return
2912
2913 sub_files = self._write_subtitles(info_dict, temp_filename)
2914 if sub_files is None:
2915 return
2916 files_to_move.update(dict(sub_files))
2917
2918 thumb_files = self._write_thumbnails(
2919 'video', info_dict, temp_filename, self.prepare_filename(info_dict, 'thumbnail'))
2920 if thumb_files is None:
2921 return
2922 files_to_move.update(dict(thumb_files))
8222d8de 2923
80c03fa9 2924 infofn = self.prepare_filename(info_dict, 'infojson')
2925 _infojson_written = self._write_info_json('video', info_dict, infofn)
2926 if _infojson_written:
dac5df5a 2927 info_dict['infojson_filename'] = infofn
e75bb0d6 2928 # For backward compatibility, even though it was a private field
80c03fa9 2929 info_dict['__infojson_filename'] = infofn
2930 elif _infojson_written is None:
2931 return
2932
2933 # Note: Annotations are deprecated
2934 annofn = None
1fb07d10 2935 if self.params.get('writeannotations', False):
de6000d9 2936 annofn = self.prepare_filename(info_dict, 'annotation')
80c03fa9 2937 if annofn:
e92caff5 2938 if not self._ensure_dir_exists(encodeFilename(annofn)):
0202b52a 2939 return
0c3d0f51 2940 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(annofn)):
6febd1c1 2941 self.to_screen('[info] Video annotations are already present')
ffddb112
RA
2942 elif not info_dict.get('annotations'):
2943 self.report_warning('There are no annotations to write.')
7b6fefc9
PH
2944 else:
2945 try:
6febd1c1 2946 self.to_screen('[info] Writing video annotations to: ' + annofn)
86e5f3ed 2947 with open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
7b6fefc9
PH
2948 annofile.write(info_dict['annotations'])
2949 except (KeyError, TypeError):
6febd1c1 2950 self.report_warning('There are no annotations to write.')
86e5f3ed 2951 except OSError:
6febd1c1 2952 self.report_error('Cannot write annotations file: ' + annofn)
7b6fefc9 2953 return
1fb07d10 2954
732044af 2955 # Write internet shortcut files
08438d2c 2956 def _write_link_file(link_type):
60f3e995 2957 url = try_get(info_dict['webpage_url'], iri_to_uri)
2958 if not url:
2959 self.report_warning(
2960 f'Cannot write internet shortcut file because the actual URL of "{info_dict["webpage_url"]}" is unknown')
2961 return True
08438d2c 2962 linkfn = replace_extension(self.prepare_filename(info_dict, 'link'), link_type, info_dict.get('ext'))
0e6b018a
Z
2963 if not self._ensure_dir_exists(encodeFilename(linkfn)):
2964 return False
10e3742e 2965 if self.params.get('overwrites', True) and os.path.exists(encodeFilename(linkfn)):
08438d2c 2966 self.to_screen(f'[info] Internet shortcut (.{link_type}) is already present')
2967 return True
2968 try:
2969 self.to_screen(f'[info] Writing internet shortcut (.{link_type}) to: {linkfn}')
86e5f3ed 2970 with open(encodeFilename(to_high_limit_path(linkfn)), 'w', encoding='utf-8',
2971 newline='\r\n' if link_type == 'url' else '\n') as linkfile:
60f3e995 2972 template_vars = {'url': url}
08438d2c 2973 if link_type == 'desktop':
2974 template_vars['filename'] = linkfn[:-(len(link_type) + 1)]
2975 linkfile.write(LINK_TEMPLATES[link_type] % template_vars)
86e5f3ed 2976 except OSError:
08438d2c 2977 self.report_error(f'Cannot write internet shortcut {linkfn}')
2978 return False
732044af 2979 return True
2980
08438d2c 2981 write_links = {
2982 'url': self.params.get('writeurllink'),
2983 'webloc': self.params.get('writewebloclink'),
2984 'desktop': self.params.get('writedesktoplink'),
2985 }
2986 if self.params.get('writelink'):
2987 link_type = ('webloc' if sys.platform == 'darwin'
2988 else 'desktop' if sys.platform.startswith('linux')
2989 else 'url')
2990 write_links[link_type] = True
2991
2992 if any(should_write and not _write_link_file(link_type)
2993 for link_type, should_write in write_links.items()):
2994 return
732044af 2995
f46e2f9d 2996 def replace_info_dict(new_info):
2997 nonlocal info_dict
2998 if new_info == info_dict:
2999 return
3000 info_dict.clear()
3001 info_dict.update(new_info)
3002
415f8d51 3003 new_info, files_to_move = self.pre_process(info_dict, 'before_dl', files_to_move)
3004 replace_info_dict(new_info)
56d868db 3005
a13e6848 3006 if self.params.get('skip_download'):
56d868db 3007 info_dict['filepath'] = temp_filename
3008 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
3009 info_dict['__files_to_move'] = files_to_move
f46e2f9d 3010 replace_info_dict(self.run_pp(MoveFilesAfterDownloadPP(self, False), info_dict))
9e907ebd 3011 info_dict['__write_download_archive'] = self.params.get('force_write_download_archive')
56d868db 3012 else:
3013 # Download
b868936c 3014 info_dict.setdefault('__postprocessors', [])
4340deca 3015 try:
0202b52a 3016
e04938ab 3017 def existing_video_file(*filepaths):
6b591b29 3018 ext = info_dict.get('ext')
e04938ab 3019 converted = lambda file: replace_extension(file, self.params.get('final_ext') or ext, ext)
3020 file = self.existing_file(itertools.chain(*zip(map(converted, filepaths), filepaths)),
3021 default_overwrite=False)
3022 if file:
3023 info_dict['ext'] = os.path.splitext(file)[1][1:]
3024 return file
0202b52a 3025
3026 success = True
4340deca 3027 if info_dict.get('requested_formats') is not None:
81cd954a
S
3028
3029 def compatible_formats(formats):
d03cfdce 3030 # TODO: some formats actually allow this (mkv, webm, ogg, mp4), but not all of them.
3031 video_formats = [format for format in formats if format.get('vcodec') != 'none']
3032 audio_formats = [format for format in formats if format.get('acodec') != 'none']
3033 if len(video_formats) > 2 or len(audio_formats) > 2:
3034 return False
3035
81cd954a 3036 # Check extension
86e5f3ed 3037 exts = {format.get('ext') for format in formats}
d03cfdce 3038 COMPATIBLE_EXTS = (
86e5f3ed 3039 {'mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma'},
3040 {'webm'},
d03cfdce 3041 )
3042 for ext_sets in COMPATIBLE_EXTS:
3043 if ext_sets.issuperset(exts):
3044 return True
81cd954a
S
3045 # TODO: Check acodec/vcodec
3046 return False
3047
3048 requested_formats = info_dict['requested_formats']
0202b52a 3049 old_ext = info_dict['ext']
4e3b637d 3050 if self.params.get('merge_output_format') is None:
3051 if not compatible_formats(requested_formats):
3052 info_dict['ext'] = 'mkv'
3053 self.report_warning(
3054 'Requested formats are incompatible for merge and will be merged into mkv')
3055 if (info_dict['ext'] == 'webm'
3056 and info_dict.get('thumbnails')
3057 # check with type instead of pp_key, __name__, or isinstance
3058 # since we dont want any custom PPs to trigger this
c487cf00 3059 and any(type(pp) == EmbedThumbnailPP for pp in self._pps['post_process'])): # noqa: E721
4e3b637d 3060 info_dict['ext'] = 'mkv'
3061 self.report_warning(
3062 'webm doesn\'t support embedding a thumbnail, mkv will be used')
124bc071 3063 new_ext = info_dict['ext']
0202b52a 3064
124bc071 3065 def correct_ext(filename, ext=new_ext):
96fccc10 3066 if filename == '-':
3067 return filename
0202b52a 3068 filename_real_ext = os.path.splitext(filename)[1][1:]
3069 filename_wo_ext = (
3070 os.path.splitext(filename)[0]
124bc071 3071 if filename_real_ext in (old_ext, new_ext)
0202b52a 3072 else filename)
86e5f3ed 3073 return f'{filename_wo_ext}.{ext}'
0202b52a 3074
38c6902b 3075 # Ensure filename always has a correct extension for successful merge
0202b52a 3076 full_filename = correct_ext(full_filename)
3077 temp_filename = correct_ext(temp_filename)
e04938ab 3078 dl_filename = existing_video_file(full_filename, temp_filename)
1ea24129 3079 info_dict['__real_download'] = False
18e674b4 3080
adbc4ec4
THD
3081 downloaded = []
3082 merger = FFmpegMergerPP(self)
3083
3084 fd = get_suitable_downloader(info_dict, self.params, to_stdout=temp_filename == '-')
dbf5416a 3085 if dl_filename is not None:
6c7274ec 3086 self.report_file_already_downloaded(dl_filename)
adbc4ec4
THD
3087 elif fd:
3088 for f in requested_formats if fd != FFmpegFD else []:
3089 f['filepath'] = fname = prepend_extension(
3090 correct_ext(temp_filename, info_dict['ext']),
3091 'f%s' % f['format_id'], info_dict['ext'])
3092 downloaded.append(fname)
dbf5416a 3093 info_dict['url'] = '\n'.join(f['url'] for f in requested_formats)
3094 success, real_download = self.dl(temp_filename, info_dict)
3095 info_dict['__real_download'] = real_download
18e674b4 3096 else:
18e674b4 3097 if self.params.get('allow_unplayable_formats'):
3098 self.report_warning(
3099 'You have requested merging of multiple formats '
3100 'while also allowing unplayable formats to be downloaded. '
3101 'The formats won\'t be merged to prevent data corruption.')
3102 elif not merger.available:
e8969bda 3103 msg = 'You have requested merging of multiple formats but ffmpeg is not installed'
3104 if not self.params.get('ignoreerrors'):
3105 self.report_error(f'{msg}. Aborting due to --abort-on-error')
3106 return
3107 self.report_warning(f'{msg}. The formats won\'t be merged')
18e674b4 3108
96fccc10 3109 if temp_filename == '-':
adbc4ec4 3110 reason = ('using a downloader other than ffmpeg' if FFmpegFD.can_merge_formats(info_dict, self.params)
96fccc10 3111 else 'but the formats are incompatible for simultaneous download' if merger.available
3112 else 'but ffmpeg is not installed')
3113 self.report_warning(
3114 f'You have requested downloading multiple formats to stdout {reason}. '
3115 'The formats will be streamed one after the other')
3116 fname = temp_filename
dbf5416a 3117 for f in requested_formats:
3118 new_info = dict(info_dict)
3119 del new_info['requested_formats']
3120 new_info.update(f)
96fccc10 3121 if temp_filename != '-':
124bc071 3122 fname = prepend_extension(
3123 correct_ext(temp_filename, new_info['ext']),
3124 'f%s' % f['format_id'], new_info['ext'])
96fccc10 3125 if not self._ensure_dir_exists(fname):
3126 return
a21e0ab1 3127 f['filepath'] = fname
96fccc10 3128 downloaded.append(fname)
dbf5416a 3129 partial_success, real_download = self.dl(fname, new_info)
3130 info_dict['__real_download'] = info_dict['__real_download'] or real_download
3131 success = success and partial_success
adbc4ec4
THD
3132
3133 if downloaded and merger.available and not self.params.get('allow_unplayable_formats'):
3134 info_dict['__postprocessors'].append(merger)
3135 info_dict['__files_to_merge'] = downloaded
3136 # Even if there were no downloads, it is being merged only now
3137 info_dict['__real_download'] = True
3138 else:
3139 for file in downloaded:
3140 files_to_move[file] = None
4340deca
P
3141 else:
3142 # Just a single file
e04938ab 3143 dl_filename = existing_video_file(full_filename, temp_filename)
6c7274ec 3144 if dl_filename is None or dl_filename == temp_filename:
3145 # dl_filename == temp_filename could mean that the file was partially downloaded with --no-part.
3146 # So we should try to resume the download
e8e73840 3147 success, real_download = self.dl(temp_filename, info_dict)
0202b52a 3148 info_dict['__real_download'] = real_download
6c7274ec 3149 else:
3150 self.report_file_already_downloaded(dl_filename)
0202b52a 3151
0202b52a 3152 dl_filename = dl_filename or temp_filename
c571435f 3153 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
0202b52a 3154
3158150c 3155 except network_exceptions as err:
7960b056 3156 self.report_error('unable to download video data: %s' % error_to_compat_str(err))
4340deca 3157 return
86e5f3ed 3158 except OSError as err:
4340deca
P
3159 raise UnavailableVideoError(err)
3160 except (ContentTooShortError, ) as err:
86e5f3ed 3161 self.report_error(f'content too short (expected {err.expected} bytes and served {err.downloaded})')
4340deca 3162 return
8222d8de 3163
415f8d51 3164 self._raise_pending_errors(info_dict)
de6000d9 3165 if success and full_filename != '-':
f17f8651 3166
fd7cfb64 3167 def fixup():
3168 do_fixup = True
3169 fixup_policy = self.params.get('fixup')
3170 vid = info_dict['id']
3171
3172 if fixup_policy in ('ignore', 'never'):
3173 return
3174 elif fixup_policy == 'warn':
3fe75fdc 3175 do_fixup = 'warn'
f89b3e2d 3176 elif fixup_policy != 'force':
3177 assert fixup_policy in ('detect_or_warn', None)
3178 if not info_dict.get('__real_download'):
3179 do_fixup = False
fd7cfb64 3180
3181 def ffmpeg_fixup(cndn, msg, cls):
3fe75fdc 3182 if not (do_fixup and cndn):
fd7cfb64 3183 return
3fe75fdc 3184 elif do_fixup == 'warn':
fd7cfb64 3185 self.report_warning(f'{vid}: {msg}')
3186 return
3187 pp = cls(self)
3188 if pp.available:
3189 info_dict['__postprocessors'].append(pp)
3190 else:
3191 self.report_warning(f'{vid}: {msg}. Install ffmpeg to fix this automatically')
3192
3193 stretched_ratio = info_dict.get('stretched_ratio')
3194 ffmpeg_fixup(
3195 stretched_ratio not in (1, None),
3196 f'Non-uniform pixel ratio {stretched_ratio}',
3197 FFmpegFixupStretchedPP)
3198
3199 ffmpeg_fixup(
3200 (info_dict.get('requested_formats') is None
3201 and info_dict.get('container') == 'm4a_dash'
3202 and info_dict.get('ext') == 'm4a'),
3203 'writing DASH m4a. Only some players support this container',
3204 FFmpegFixupM4aPP)
3205
993191c0 3206 downloader = get_suitable_downloader(info_dict, self.params) if 'protocol' in info_dict else None
24146491 3207 downloader = downloader.FD_NAME if downloader else None
adbc4ec4
THD
3208
3209 if info_dict.get('requested_formats') is None: # Not necessary if doing merger
24146491 3210 ffmpeg_fixup(downloader == 'hlsnative' and not self.params.get('hls_use_mpegts')
494f5230 3211 or info_dict.get('is_live') and self.params.get('hls_use_mpegts') is None,
adbc4ec4
THD
3212 'Possible MPEG-TS in MP4 container or malformed AAC timestamps',
3213 FFmpegFixupM3u8PP)
3214 ffmpeg_fixup(info_dict.get('is_live') and downloader == 'DashSegmentsFD',
3215 'Possible duplicate MOOV atoms', FFmpegFixupDuplicateMoovPP)
3216
24146491 3217 ffmpeg_fixup(downloader == 'web_socket_fragment', 'Malformed timestamps detected', FFmpegFixupTimestampPP)
3218 ffmpeg_fixup(downloader == 'web_socket_fragment', 'Malformed duration detected', FFmpegFixupDurationPP)
fd7cfb64 3219
3220 fixup()
8222d8de 3221 try:
f46e2f9d 3222 replace_info_dict(self.post_process(dl_filename, info_dict, files_to_move))
af819c21 3223 except PostProcessingError as err:
3224 self.report_error('Postprocessing: %s' % str(err))
8222d8de 3225 return
ab8e5e51
AM
3226 try:
3227 for ph in self._post_hooks:
23c1a667 3228 ph(info_dict['filepath'])
ab8e5e51
AM
3229 except Exception as err:
3230 self.report_error('post hooks: %s' % str(err))
3231 return
9e907ebd 3232 info_dict['__write_download_archive'] = True
2d30509f 3233
c487cf00 3234 assert info_dict is original_infodict # Make sure the info_dict was modified in-place
a13e6848 3235 if self.params.get('force_write_download_archive'):
9e907ebd 3236 info_dict['__write_download_archive'] = True
ca6d59d2 3237 check_max_downloads()
8222d8de 3238
aa9369a2 3239 def __download_wrapper(self, func):
3240 @functools.wraps(func)
3241 def wrapper(*args, **kwargs):
3242 try:
3243 res = func(*args, **kwargs)
3244 except UnavailableVideoError as e:
3245 self.report_error(e)
b222c271 3246 except DownloadCancelled as e:
3247 self.to_screen(f'[info] {e}')
3248 if not self.params.get('break_per_url'):
3249 raise
aa9369a2 3250 else:
3251 if self.params.get('dump_single_json', False):
3252 self.post_extract(res)
3253 self.to_stdout(json.dumps(self.sanitize_info(res)))
3254 return wrapper
3255
8222d8de
JMF
3256 def download(self, url_list):
3257 """Download a given list of URLs."""
aa9369a2 3258 url_list = variadic(url_list) # Passing a single URL is a common mistake
de6000d9 3259 outtmpl = self.outtmpl_dict['default']
3089bc74
S
3260 if (len(url_list) > 1
3261 and outtmpl != '-'
3262 and '%' not in outtmpl
3263 and self.params.get('max_downloads') != 1):
acd69589 3264 raise SameFileError(outtmpl)
8222d8de
JMF
3265
3266 for url in url_list:
aa9369a2 3267 self.__download_wrapper(self.extract_info)(
3268 url, force_generic_extractor=self.params.get('force_generic_extractor', False))
8222d8de
JMF
3269
3270 return self._download_retcode
3271
1dcc4c0c 3272 def download_with_info_file(self, info_filename):
31bd3925
JMF
3273 with contextlib.closing(fileinput.FileInput(
3274 [info_filename], mode='r',
3275 openhook=fileinput.hook_encoded('utf-8'))) as f:
3276 # FileInput doesn't have a read method, we can't call json.load
8012d892 3277 info = self.sanitize_info(json.loads('\n'.join(f)), self.params.get('clean_infojson', True))
d4943898 3278 try:
aa9369a2 3279 self.__download_wrapper(self.process_ie_result)(info, download=True)
f2ebc5c7 3280 except (DownloadError, EntryNotInPlaylist, ReExtractInfo) as e:
bf5f605e 3281 if not isinstance(e, EntryNotInPlaylist):
3282 self.to_stderr('\r')
d4943898
JMF
3283 webpage_url = info.get('webpage_url')
3284 if webpage_url is not None:
aa9369a2 3285 self.report_warning(f'The info failed to download: {e}; trying with URL {webpage_url}')
d4943898
JMF
3286 return self.download([webpage_url])
3287 else:
3288 raise
3289 return self._download_retcode
1dcc4c0c 3290
cb202fd2 3291 @staticmethod
8012d892 3292 def sanitize_info(info_dict, remove_private_keys=False):
3293 ''' Sanitize the infodict for converting to json '''
3ad56b42 3294 if info_dict is None:
3295 return info_dict
6e84b215 3296 info_dict.setdefault('epoch', int(time.time()))
6a5a30f9 3297 info_dict.setdefault('_type', 'video')
09b49e1f 3298
8012d892 3299 if remove_private_keys:
0a5a191a 3300 reject = lambda k, v: v is None or k.startswith('__') or k in {
f46e2f9d 3301 'requested_downloads', 'requested_formats', 'requested_subtitles', 'requested_entries',
0a5a191a 3302 'entries', 'filepath', '_filename', 'infojson_filename', 'original_url', 'playlist_autonumber',
6e84b215 3303 }
ae8f99e6 3304 else:
09b49e1f 3305 reject = lambda k, v: False
adbc4ec4
THD
3306
3307 def filter_fn(obj):
3308 if isinstance(obj, dict):
3309 return {k: filter_fn(v) for k, v in obj.items() if not reject(k, v)}
3310 elif isinstance(obj, (list, tuple, set, LazyList)):
3311 return list(map(filter_fn, obj))
3312 elif obj is None or isinstance(obj, (str, int, float, bool)):
3313 return obj
3314 else:
3315 return repr(obj)
3316
5226731e 3317 return filter_fn(info_dict)
cb202fd2 3318
8012d892 3319 @staticmethod
3320 def filter_requested_info(info_dict, actually_filter=True):
3321 ''' Alias of sanitize_info for backward compatibility '''
3322 return YoutubeDL.sanitize_info(info_dict, actually_filter)
3323
43d7f5a5 3324 def _delete_downloaded_files(self, *files_to_delete, info={}, msg=None):
3325 for filename in set(filter(None, files_to_delete)):
3326 if msg:
3327 self.to_screen(msg % filename)
3328 try:
3329 os.remove(filename)
3330 except OSError:
3331 self.report_warning(f'Unable to delete file {filename}')
3332 if filename in info.get('__files_to_move', []): # NB: Delete even if None
3333 del info['__files_to_move'][filename]
3334
ed5835b4 3335 @staticmethod
3336 def post_extract(info_dict):
3337 def actual_post_extract(info_dict):
3338 if info_dict.get('_type') in ('playlist', 'multi_video'):
3339 for video_dict in info_dict.get('entries', {}):
3340 actual_post_extract(video_dict or {})
3341 return
3342
09b49e1f 3343 post_extractor = info_dict.pop('__post_extractor', None) or (lambda: {})
3344 info_dict.update(post_extractor())
ed5835b4 3345
3346 actual_post_extract(info_dict or {})
3347
dcf64d43 3348 def run_pp(self, pp, infodict):
5bfa4862 3349 files_to_delete = []
dcf64d43 3350 if '__files_to_move' not in infodict:
3351 infodict['__files_to_move'] = {}
b1940459 3352 try:
3353 files_to_delete, infodict = pp.run(infodict)
3354 except PostProcessingError as e:
3355 # Must be True and not 'only_download'
3356 if self.params.get('ignoreerrors') is True:
3357 self.report_error(e)
3358 return infodict
3359 raise
3360
5bfa4862 3361 if not files_to_delete:
dcf64d43 3362 return infodict
5bfa4862 3363 if self.params.get('keepvideo', False):
3364 for f in files_to_delete:
dcf64d43 3365 infodict['__files_to_move'].setdefault(f, '')
5bfa4862 3366 else:
43d7f5a5 3367 self._delete_downloaded_files(
3368 *files_to_delete, info=infodict, msg='Deleting original file %s (pass -k to keep)')
dcf64d43 3369 return infodict
5bfa4862 3370
ed5835b4 3371 def run_all_pps(self, key, info, *, additional_pps=None):
bb66c247 3372 self._forceprint(key, info)
ed5835b4 3373 for pp in (additional_pps or []) + self._pps[key]:
dc5f409c 3374 info = self.run_pp(pp, info)
ed5835b4 3375 return info
277d6ff5 3376
56d868db 3377 def pre_process(self, ie_info, key='pre_process', files_to_move=None):
5bfa4862 3378 info = dict(ie_info)
56d868db 3379 info['__files_to_move'] = files_to_move or {}
415f8d51 3380 try:
3381 info = self.run_all_pps(key, info)
3382 except PostProcessingError as err:
3383 msg = f'Preprocessing: {err}'
3384 info.setdefault('__pending_error', msg)
3385 self.report_error(msg, is_error=False)
56d868db 3386 return info, info.pop('__files_to_move', None)
5bfa4862 3387
f46e2f9d 3388 def post_process(self, filename, info, files_to_move=None):
8222d8de 3389 """Run all the postprocessors on the given file."""
8222d8de 3390 info['filepath'] = filename
dcf64d43 3391 info['__files_to_move'] = files_to_move or {}
ed5835b4 3392 info = self.run_all_pps('post_process', info, additional_pps=info.get('__postprocessors'))
dcf64d43 3393 info = self.run_pp(MoveFilesAfterDownloadPP(self), info)
3394 del info['__files_to_move']
ed5835b4 3395 return self.run_all_pps('after_move', info)
c1c9a79c 3396
5db07df6 3397 def _make_archive_id(self, info_dict):
e9fef7ee
S
3398 video_id = info_dict.get('id')
3399 if not video_id:
3400 return
5db07df6
PH
3401 # Future-proof against any change in case
3402 # and backwards compatibility with prior versions
e9fef7ee 3403 extractor = info_dict.get('extractor_key') or info_dict.get('ie_key') # key in a playlist
7012b23c 3404 if extractor is None:
1211bb6d
S
3405 url = str_or_none(info_dict.get('url'))
3406 if not url:
3407 return
e9fef7ee 3408 # Try to find matching extractor for the URL and take its ie_key
8b7491c8 3409 for ie_key, ie in self._ies.items():
1211bb6d 3410 if ie.suitable(url):
8b7491c8 3411 extractor = ie_key
e9fef7ee
S
3412 break
3413 else:
3414 return
86e5f3ed 3415 return f'{extractor.lower()} {video_id}'
5db07df6
PH
3416
3417 def in_download_archive(self, info_dict):
3418 fn = self.params.get('download_archive')
3419 if fn is None:
3420 return False
3421
3422 vid_id = self._make_archive_id(info_dict)
e9fef7ee 3423 if not vid_id:
7012b23c 3424 return False # Incomplete video information
5db07df6 3425
a45e8619 3426 return vid_id in self.archive
c1c9a79c
PH
3427
3428 def record_download_archive(self, info_dict):
3429 fn = self.params.get('download_archive')
3430 if fn is None:
3431 return
5db07df6
PH
3432 vid_id = self._make_archive_id(info_dict)
3433 assert vid_id
a13e6848 3434 self.write_debug(f'Adding to archive: {vid_id}')
c1c9a79c 3435 with locked_file(fn, 'a', encoding='utf-8') as archive_file:
6febd1c1 3436 archive_file.write(vid_id + '\n')
a45e8619 3437 self.archive.add(vid_id)
dd82ffea 3438
8c51aa65 3439 @staticmethod
8abeeb94 3440 def format_resolution(format, default='unknown'):
9359f3d4 3441 if format.get('vcodec') == 'none' and format.get('acodec') != 'none':
fb04e403 3442 return 'audio only'
f49d89ee
PH
3443 if format.get('resolution') is not None:
3444 return format['resolution']
35615307 3445 if format.get('width') and format.get('height'):
ff51ed58 3446 return '%dx%d' % (format['width'], format['height'])
35615307 3447 elif format.get('height'):
ff51ed58 3448 return '%sp' % format['height']
35615307 3449 elif format.get('width'):
ff51ed58 3450 return '%dx?' % format['width']
3451 return default
8c51aa65 3452
8130779d 3453 def _list_format_headers(self, *headers):
3454 if self.params.get('listformats_table', True) is not False:
591bb9d3 3455 return [self._format_out(header, self.Styles.HEADERS) for header in headers]
8130779d 3456 return headers
3457
c57f7757
PH
3458 def _format_note(self, fdict):
3459 res = ''
3460 if fdict.get('ext') in ['f4f', 'f4m']:
f304da8a 3461 res += '(unsupported)'
32f90364
PH
3462 if fdict.get('language'):
3463 if res:
3464 res += ' '
f304da8a 3465 res += '[%s]' % fdict['language']
c57f7757 3466 if fdict.get('format_note') is not None:
f304da8a 3467 if res:
3468 res += ' '
3469 res += fdict['format_note']
c57f7757 3470 if fdict.get('tbr') is not None:
f304da8a 3471 if res:
3472 res += ', '
3473 res += '%4dk' % fdict['tbr']
c57f7757
PH
3474 if fdict.get('container') is not None:
3475 if res:
3476 res += ', '
3477 res += '%s container' % fdict['container']
3089bc74
S
3478 if (fdict.get('vcodec') is not None
3479 and fdict.get('vcodec') != 'none'):
c57f7757
PH
3480 if res:
3481 res += ', '
3482 res += fdict['vcodec']
91c7271a 3483 if fdict.get('vbr') is not None:
c57f7757
PH
3484 res += '@'
3485 elif fdict.get('vbr') is not None and fdict.get('abr') is not None:
3486 res += 'video@'
3487 if fdict.get('vbr') is not None:
3488 res += '%4dk' % fdict['vbr']
fbb21cf5 3489 if fdict.get('fps') is not None:
5d583bdf
S
3490 if res:
3491 res += ', '
3492 res += '%sfps' % fdict['fps']
c57f7757
PH
3493 if fdict.get('acodec') is not None:
3494 if res:
3495 res += ', '
3496 if fdict['acodec'] == 'none':
3497 res += 'video only'
3498 else:
3499 res += '%-5s' % fdict['acodec']
3500 elif fdict.get('abr') is not None:
3501 if res:
3502 res += ', '
3503 res += 'audio'
3504 if fdict.get('abr') is not None:
3505 res += '@%3dk' % fdict['abr']
3506 if fdict.get('asr') is not None:
3507 res += ' (%5dHz)' % fdict['asr']
3508 if fdict.get('filesize') is not None:
3509 if res:
3510 res += ', '
3511 res += format_bytes(fdict['filesize'])
9732d77e
PH
3512 elif fdict.get('filesize_approx') is not None:
3513 if res:
3514 res += ', '
3515 res += '~' + format_bytes(fdict['filesize_approx'])
c57f7757 3516 return res
91c7271a 3517
8130779d 3518 def render_formats_table(self, info_dict):
b69fd25c 3519 if not info_dict.get('formats') and not info_dict.get('url'):
8130779d 3520 return None
b69fd25c 3521
94badb25 3522 formats = info_dict.get('formats', [info_dict])
8130779d 3523 if not self.params.get('listformats_table', True) is not False:
76d321f6 3524 table = [
3525 [
3526 format_field(f, 'format_id'),
3527 format_field(f, 'ext'),
3528 self.format_resolution(f),
8130779d 3529 self._format_note(f)
3530 ] for f in formats if f.get('preference') is None or f['preference'] >= -1000]
3531 return render_table(['format code', 'extension', 'resolution', 'note'], table, extra_gap=1)
3532
591bb9d3 3533 delim = self._format_out('\u2502', self.Styles.DELIM, '|', test_encoding=True)
8130779d 3534 table = [
3535 [
591bb9d3 3536 self._format_out(format_field(f, 'format_id'), self.Styles.ID),
8130779d 3537 format_field(f, 'ext'),
3538 format_field(f, func=self.format_resolution, ignore=('audio only', 'images')),
3539 format_field(f, 'fps', '\t%d'),
3540 format_field(f, 'dynamic_range', '%s', ignore=(None, 'SDR')).replace('HDR', ''),
3541 delim,
3542 format_field(f, 'filesize', ' \t%s', func=format_bytes) + format_field(f, 'filesize_approx', '~\t%s', func=format_bytes),
3543 format_field(f, 'tbr', '\t%dk'),
3544 shorten_protocol_name(f.get('protocol', '')),
3545 delim,
3546 format_field(f, 'vcodec', default='unknown').replace(
3547 'none', 'images' if f.get('acodec') == 'none'
591bb9d3 3548 else self._format_out('audio only', self.Styles.SUPPRESS)),
8130779d 3549 format_field(f, 'vbr', '\t%dk'),
3550 format_field(f, 'acodec', default='unknown').replace(
3551 'none', '' if f.get('vcodec') == 'none'
591bb9d3 3552 else self._format_out('video only', self.Styles.SUPPRESS)),
8130779d 3553 format_field(f, 'abr', '\t%dk'),
3554 format_field(f, 'asr', '\t%dHz'),
3555 join_nonempty(
591bb9d3 3556 self._format_out('UNSUPPORTED', 'light red') if f.get('ext') in ('f4f', 'f4m') else None,
8130779d 3557 format_field(f, 'language', '[%s]'),
3558 join_nonempty(format_field(f, 'format_note'),
3559 format_field(f, 'container', ignore=(None, f.get('ext'))),
3560 delim=', '),
3561 delim=' '),
3562 ] for f in formats if f.get('preference') is None or f['preference'] >= -1000]
3563 header_line = self._list_format_headers(
3564 'ID', 'EXT', 'RESOLUTION', '\tFPS', 'HDR', delim, '\tFILESIZE', '\tTBR', 'PROTO',
3565 delim, 'VCODEC', '\tVBR', 'ACODEC', '\tABR', '\tASR', 'MORE INFO')
3566
3567 return render_table(
3568 header_line, table, hide_empty=True,
591bb9d3 3569 delim=self._format_out('\u2500', self.Styles.DELIM, '-', test_encoding=True))
8130779d 3570
3571 def render_thumbnails_table(self, info_dict):
88f23a18 3572 thumbnails = list(info_dict.get('thumbnails') or [])
cfb56d1a 3573 if not thumbnails:
8130779d 3574 return None
3575 return render_table(
ec11a9f4 3576 self._list_format_headers('ID', 'Width', 'Height', 'URL'),
6970b600 3577 [[t.get('id'), t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails])
2412044c 3578
8130779d 3579 def render_subtitles_table(self, video_id, subtitles):
2412044c 3580 def _row(lang, formats):
49c258e1 3581 exts, names = zip(*((f['ext'], f.get('name') or 'unknown') for f in reversed(formats)))
2412044c 3582 if len(set(names)) == 1:
7aee40c1 3583 names = [] if names[0] == 'unknown' else names[:1]
2412044c 3584 return [lang, ', '.join(names), ', '.join(exts)]
3585
8130779d 3586 if not subtitles:
3587 return None
3588 return render_table(
ec11a9f4 3589 self._list_format_headers('Language', 'Name', 'Formats'),
2412044c 3590 [_row(lang, formats) for lang, formats in subtitles.items()],
8130779d 3591 hide_empty=True)
3592
3593 def __list_table(self, video_id, name, func, *args):
3594 table = func(*args)
3595 if not table:
3596 self.to_screen(f'{video_id} has no {name}')
3597 return
3598 self.to_screen(f'[info] Available {name} for {video_id}:')
3599 self.to_stdout(table)
3600
3601 def list_formats(self, info_dict):
3602 self.__list_table(info_dict['id'], 'formats', self.render_formats_table, info_dict)
3603
3604 def list_thumbnails(self, info_dict):
3605 self.__list_table(info_dict['id'], 'thumbnails', self.render_thumbnails_table, info_dict)
3606
3607 def list_subtitles(self, video_id, subtitles, name='subtitles'):
3608 self.__list_table(video_id, name, self.render_subtitles_table, video_id, subtitles)
a504ced0 3609
dca08720
PH
3610 def urlopen(self, req):
3611 """ Start an HTTP download """
f9934b96 3612 if isinstance(req, str):
67dda517 3613 req = sanitized_Request(req)
19a41fc6 3614 return self._opener.open(req, timeout=self._socket_timeout)
dca08720
PH
3615
3616 def print_debug_header(self):
3617 if not self.params.get('verbose'):
3618 return
49a57e70 3619
3620 def get_encoding(stream):
2a938746 3621 ret = str(getattr(stream, 'encoding', 'missing (%s)' % type(stream).__name__))
49a57e70 3622 if not supports_terminal_sequences(stream):
53973b4d 3623 from .utils import WINDOWS_VT_MODE # Must be imported locally
e3c7d495 3624 ret += ' (No VT)' if WINDOWS_VT_MODE is False else ' (No ANSI)'
49a57e70 3625 return ret
3626
591bb9d3 3627 encoding_str = 'Encodings: locale %s, fs %s, pref %s, %s' % (
49a57e70 3628 locale.getpreferredencoding(),
3629 sys.getfilesystemencoding(),
591bb9d3 3630 self.get_encoding(),
3631 ', '.join(
7896214c 3632 f'{key} {get_encoding(stream)}' for key, stream in self._out_files
591bb9d3 3633 if stream is not None and key != 'console')
3634 )
883d4b1e 3635
3636 logger = self.params.get('logger')
3637 if logger:
3638 write_debug = lambda msg: logger.debug(f'[debug] {msg}')
3639 write_debug(encoding_str)
3640 else:
96565c7e 3641 write_string(f'[debug] {encoding_str}\n', encoding=None)
49a57e70 3642 write_debug = lambda msg: self._write_string(f'[debug] {msg}\n')
734f90bb 3643
4c88ff87 3644 source = detect_variant()
36eaf303 3645 write_debug(join_nonempty(
3646 'yt-dlp version', __version__,
3647 f'[{RELEASE_GIT_HEAD}]' if RELEASE_GIT_HEAD else '',
3648 '' if source == 'unknown' else f'({source})',
3649 delim=' '))
6e21fdd2 3650 if not _LAZY_LOADER:
3651 if os.environ.get('YTDLP_NO_LAZY_EXTRACTORS'):
49a57e70 3652 write_debug('Lazy loading extractors is forcibly disabled')
6e21fdd2 3653 else:
49a57e70 3654 write_debug('Lazy loading extractors is disabled')
3ae5e797 3655 if plugin_extractors or plugin_postprocessors:
49a57e70 3656 write_debug('Plugins: %s' % [
3ae5e797 3657 '%s%s' % (klass.__name__, '' if klass.__name__ == name else f' as {name}')
3658 for name, klass in itertools.chain(plugin_extractors.items(), plugin_postprocessors.items())])
8a82af35 3659 if self.params['compat_opts']:
3660 write_debug('Compatibility options: %s' % ', '.join(self.params['compat_opts']))
36eaf303 3661
3662 if source == 'source':
dca08720 3663 try:
36eaf303 3664 sp = Popen(
3665 ['git', 'rev-parse', '--short', 'HEAD'],
3666 stdout=subprocess.PIPE, stderr=subprocess.PIPE,
3667 cwd=os.path.dirname(os.path.abspath(__file__)))
3668 out, err = sp.communicate_or_kill()
3669 out = out.decode().strip()
3670 if re.match('[0-9a-f]+', out):
3671 write_debug('Git HEAD: %s' % out)
70a1165b 3672 except Exception:
19a03940 3673 with contextlib.suppress(Exception):
36eaf303 3674 sys.exc_clear()
b300cda4
S
3675
3676 def python_implementation():
3677 impl_name = platform.python_implementation()
3678 if impl_name == 'PyPy' and hasattr(sys, 'pypy_version_info'):
3679 return impl_name + ' version %d.%d.%d' % sys.pypy_version_info[:3]
3680 return impl_name
3681
49a57e70 3682 write_debug('Python version %s (%s %s) - %s' % (
e5813e53 3683 platform.python_version(),
3684 python_implementation(),
3685 platform.architecture()[0],
b300cda4 3686 platform_name()))
d28b5171 3687
8913ef74 3688 exe_versions, ffmpeg_features = FFmpegPostProcessor.get_versions_and_features(self)
3689 ffmpeg_features = {key for key, val in ffmpeg_features.items() if val}
3690 if ffmpeg_features:
19a03940 3691 exe_versions['ffmpeg'] += ' (%s)' % ','.join(sorted(ffmpeg_features))
8913ef74 3692
4c83c967 3693 exe_versions['rtmpdump'] = rtmpdump_version()
feee8d32 3694 exe_versions['phantomjs'] = PhantomJSwrapper._version()
d28b5171 3695 exe_str = ', '.join(
2831b468 3696 f'{exe} {v}' for exe, v in sorted(exe_versions.items()) if v
3697 ) or 'none'
49a57e70 3698 write_debug('exe versions: %s' % exe_str)
dca08720 3699
1d485a1a 3700 from .compat.compat_utils import get_package_info
9b8ee23b 3701 from .dependencies import available_dependencies
3702
3703 write_debug('Optional libraries: %s' % (', '.join(sorted({
1d485a1a 3704 join_nonempty(*get_package_info(m)) for m in available_dependencies.values()
9b8ee23b 3705 })) or 'none'))
2831b468 3706
97ec5bc5 3707 self._setup_opener()
dca08720
PH
3708 proxy_map = {}
3709 for handler in self._opener.handlers:
3710 if hasattr(handler, 'proxies'):
3711 proxy_map.update(handler.proxies)
49a57e70 3712 write_debug(f'Proxy map: {proxy_map}')
dca08720 3713
49a57e70 3714 # Not implemented
3715 if False and self.params.get('call_home'):
0f06bcd7 3716 ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode()
49a57e70 3717 write_debug('Public IP address: %s' % ipaddr)
58b1f00d 3718 latest_version = self.urlopen(
0f06bcd7 3719 'https://yt-dl.org/latest/version').read().decode()
58b1f00d
PH
3720 if version_tuple(latest_version) > version_tuple(__version__):
3721 self.report_warning(
3722 'You are using an outdated version (newest version: %s)! '
3723 'See https://yt-dl.org/update if you need help updating.' %
3724 latest_version)
3725
e344693b 3726 def _setup_opener(self):
97ec5bc5 3727 if hasattr(self, '_opener'):
3728 return
6ad14cab 3729 timeout_val = self.params.get('socket_timeout')
17bddf3e 3730 self._socket_timeout = 20 if timeout_val is None else float(timeout_val)
6ad14cab 3731
982ee69a 3732 opts_cookiesfrombrowser = self.params.get('cookiesfrombrowser')
dca08720
PH
3733 opts_cookiefile = self.params.get('cookiefile')
3734 opts_proxy = self.params.get('proxy')
3735
982ee69a 3736 self.cookiejar = load_cookies(opts_cookiefile, opts_cookiesfrombrowser, self)
dca08720 3737
6a3f4c3f 3738 cookie_processor = YoutubeDLCookieProcessor(self.cookiejar)
dca08720
PH
3739 if opts_proxy is not None:
3740 if opts_proxy == '':
3741 proxies = {}
3742 else:
3743 proxies = {'http': opts_proxy, 'https': opts_proxy}
3744 else:
3745 proxies = compat_urllib_request.getproxies()
067aa17e 3746 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
dca08720
PH
3747 if 'http' in proxies and 'https' not in proxies:
3748 proxies['https'] = proxies['http']
91410c9b 3749 proxy_handler = PerRequestProxyHandler(proxies)
a0ddb8a2
PH
3750
3751 debuglevel = 1 if self.params.get('debug_printtraffic') else 0
be4a824d
PH
3752 https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel)
3753 ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel)
fca6dba8 3754 redirect_handler = YoutubeDLRedirectHandler()
f9934b96 3755 data_handler = urllib.request.DataHandler()
6240b0a2
JMF
3756
3757 # When passing our own FileHandler instance, build_opener won't add the
3758 # default FileHandler and allows us to disable the file protocol, which
3759 # can be used for malicious purposes (see
067aa17e 3760 # https://github.com/ytdl-org/youtube-dl/issues/8227)
6240b0a2
JMF
3761 file_handler = compat_urllib_request.FileHandler()
3762
3763 def file_open(*args, **kwargs):
7a5c1cfe 3764 raise compat_urllib_error.URLError('file:// scheme is explicitly disabled in yt-dlp for security reasons')
6240b0a2
JMF
3765 file_handler.file_open = file_open
3766
3767 opener = compat_urllib_request.build_opener(
fca6dba8 3768 proxy_handler, https_handler, cookie_processor, ydlh, redirect_handler, data_handler, file_handler)
2461f79d 3769
dca08720
PH
3770 # Delete the default user-agent header, which would otherwise apply in
3771 # cases where our custom HTTP handler doesn't come into play
067aa17e 3772 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
dca08720
PH
3773 opener.addheaders = []
3774 self._opener = opener
62fec3b2
PH
3775
3776 def encode(self, s):
3777 if isinstance(s, bytes):
3778 return s # Already encoded
3779
3780 try:
3781 return s.encode(self.get_encoding())
3782 except UnicodeEncodeError as err:
3783 err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.'
3784 raise
3785
3786 def get_encoding(self):
3787 encoding = self.params.get('encoding')
3788 if encoding is None:
3789 encoding = preferredencoding()
3790 return encoding
ec82d85a 3791
e08a85d8 3792 def _write_info_json(self, label, ie_result, infofn, overwrite=None):
cb96c5be 3793 ''' Write infojson and returns True = written, 'exists' = Already exists, False = skip, None = error '''
e08a85d8 3794 if overwrite is None:
3795 overwrite = self.params.get('overwrites', True)
80c03fa9 3796 if not self.params.get('writeinfojson'):
3797 return False
3798 elif not infofn:
3799 self.write_debug(f'Skipping writing {label} infojson')
3800 return False
3801 elif not self._ensure_dir_exists(infofn):
3802 return None
e08a85d8 3803 elif not overwrite and os.path.exists(infofn):
80c03fa9 3804 self.to_screen(f'[info] {label.title()} metadata is already present')
cb96c5be 3805 return 'exists'
3806
3807 self.to_screen(f'[info] Writing {label} metadata as JSON to: {infofn}')
3808 try:
3809 write_json_file(self.sanitize_info(ie_result, self.params.get('clean_infojson', True)), infofn)
3810 return True
86e5f3ed 3811 except OSError:
cb96c5be 3812 self.report_error(f'Cannot write {label} metadata to JSON file {infofn}')
3813 return None
80c03fa9 3814
3815 def _write_description(self, label, ie_result, descfn):
3816 ''' Write description and returns True = written, False = skip, None = error '''
3817 if not self.params.get('writedescription'):
3818 return False
3819 elif not descfn:
3820 self.write_debug(f'Skipping writing {label} description')
3821 return False
3822 elif not self._ensure_dir_exists(descfn):
3823 return None
3824 elif not self.params.get('overwrites', True) and os.path.exists(descfn):
3825 self.to_screen(f'[info] {label.title()} description is already present')
3826 elif ie_result.get('description') is None:
3827 self.report_warning(f'There\'s no {label} description to write')
3828 return False
3829 else:
3830 try:
3831 self.to_screen(f'[info] Writing {label} description to: {descfn}')
86e5f3ed 3832 with open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
80c03fa9 3833 descfile.write(ie_result['description'])
86e5f3ed 3834 except OSError:
80c03fa9 3835 self.report_error(f'Cannot write {label} description file {descfn}')
3836 return None
3837 return True
3838
3839 def _write_subtitles(self, info_dict, filename):
3840 ''' Write subtitles to file and return list of (sub_filename, final_sub_filename); or None if error'''
3841 ret = []
3842 subtitles = info_dict.get('requested_subtitles')
3843 if not subtitles or not (self.params.get('writesubtitles') or self.params.get('writeautomaticsub')):
3844 # subtitles download errors are already managed as troubles in relevant IE
3845 # that way it will silently go on when used with unsupporting IE
3846 return ret
3847
3848 sub_filename_base = self.prepare_filename(info_dict, 'subtitle')
3849 if not sub_filename_base:
3850 self.to_screen('[info] Skipping writing video subtitles')
3851 return ret
3852 for sub_lang, sub_info in subtitles.items():
3853 sub_format = sub_info['ext']
3854 sub_filename = subtitles_filename(filename, sub_lang, sub_format, info_dict.get('ext'))
3855 sub_filename_final = subtitles_filename(sub_filename_base, sub_lang, sub_format, info_dict.get('ext'))
e04938ab 3856 existing_sub = self.existing_file((sub_filename_final, sub_filename))
3857 if existing_sub:
80c03fa9 3858 self.to_screen(f'[info] Video subtitle {sub_lang}.{sub_format} is already present')
e04938ab 3859 sub_info['filepath'] = existing_sub
3860 ret.append((existing_sub, sub_filename_final))
80c03fa9 3861 continue
3862
3863 self.to_screen(f'[info] Writing video subtitles to: {sub_filename}')
3864 if sub_info.get('data') is not None:
3865 try:
3866 # Use newline='' to prevent conversion of newline characters
3867 # See https://github.com/ytdl-org/youtube-dl/issues/10268
86e5f3ed 3868 with open(sub_filename, 'w', encoding='utf-8', newline='') as subfile:
80c03fa9 3869 subfile.write(sub_info['data'])
3870 sub_info['filepath'] = sub_filename
3871 ret.append((sub_filename, sub_filename_final))
3872 continue
86e5f3ed 3873 except OSError:
80c03fa9 3874 self.report_error(f'Cannot write video subtitles file {sub_filename}')
3875 return None
3876
3877 try:
3878 sub_copy = sub_info.copy()
3879 sub_copy.setdefault('http_headers', info_dict.get('http_headers'))
3880 self.dl(sub_filename, sub_copy, subtitle=True)
3881 sub_info['filepath'] = sub_filename
3882 ret.append((sub_filename, sub_filename_final))
6020e05d 3883 except (DownloadError, ExtractorError, IOError, OSError, ValueError) + network_exceptions as err:
c70c418d 3884 msg = f'Unable to download video subtitles for {sub_lang!r}: {err}'
6020e05d 3885 if self.params.get('ignoreerrors') is not True: # False or 'only_download'
c70c418d 3886 if not self.params.get('ignoreerrors'):
3887 self.report_error(msg)
3888 raise DownloadError(msg)
3889 self.report_warning(msg)
519804a9 3890 return ret
80c03fa9 3891
3892 def _write_thumbnails(self, label, info_dict, filename, thumb_filename_base=None):
3893 ''' Write thumbnails to file and return list of (thumb_filename, final_thumb_filename) '''
6c4fd172 3894 write_all = self.params.get('write_all_thumbnails', False)
80c03fa9 3895 thumbnails, ret = [], []
6c4fd172 3896 if write_all or self.params.get('writethumbnail', False):
0202b52a 3897 thumbnails = info_dict.get('thumbnails') or []
6c4fd172 3898 multiple = write_all and len(thumbnails) > 1
ec82d85a 3899
80c03fa9 3900 if thumb_filename_base is None:
3901 thumb_filename_base = filename
3902 if thumbnails and not thumb_filename_base:
3903 self.write_debug(f'Skipping writing {label} thumbnail')
3904 return ret
3905
dd0228ce 3906 for idx, t in list(enumerate(thumbnails))[::-1]:
80c03fa9 3907 thumb_ext = (f'{t["id"]}.' if multiple else '') + determine_ext(t['url'], 'jpg')
aa9369a2 3908 thumb_display_id = f'{label} thumbnail {t["id"]}'
80c03fa9 3909 thumb_filename = replace_extension(filename, thumb_ext, info_dict.get('ext'))
3910 thumb_filename_final = replace_extension(thumb_filename_base, thumb_ext, info_dict.get('ext'))
ec82d85a 3911
e04938ab 3912 existing_thumb = self.existing_file((thumb_filename_final, thumb_filename))
3913 if existing_thumb:
aa9369a2 3914 self.to_screen('[info] %s is already present' % (
3915 thumb_display_id if multiple else f'{label} thumbnail').capitalize())
e04938ab 3916 t['filepath'] = existing_thumb
3917 ret.append((existing_thumb, thumb_filename_final))
ec82d85a 3918 else:
80c03fa9 3919 self.to_screen(f'[info] Downloading {thumb_display_id} ...')
ec82d85a 3920 try:
297e9952 3921 uf = self.urlopen(sanitized_Request(t['url'], headers=t.get('http_headers', {})))
80c03fa9 3922 self.to_screen(f'[info] Writing {thumb_display_id} to: {thumb_filename}')
d3d89c32 3923 with open(encodeFilename(thumb_filename), 'wb') as thumbf:
ec82d85a 3924 shutil.copyfileobj(uf, thumbf)
80c03fa9 3925 ret.append((thumb_filename, thumb_filename_final))
885cc0b7 3926 t['filepath'] = thumb_filename
3158150c 3927 except network_exceptions as err:
dd0228ce 3928 thumbnails.pop(idx)
80c03fa9 3929 self.report_warning(f'Unable to download {thumb_display_id}: {err}')
6c4fd172 3930 if ret and not write_all:
3931 break
0202b52a 3932 return ret