]> jfr.im git - yt-dlp.git/blame - yt_dlp/YoutubeDL.py
[YoutubeDL] Fill more fields for playlists (#2824)
[yt-dlp.git] / yt_dlp / YoutubeDL.py
CommitLineData
cc52de43 1#!/usr/bin/env python3
dcdb292f 2# coding: utf-8
8222d8de 3
6febd1c1 4from __future__ import absolute_import, unicode_literals
8222d8de 5
26e63931 6import collections
31bd3925 7import contextlib
9d2ecdbc 8import datetime
c1c9a79c 9import errno
31bd3925 10import fileinput
b5ae35ee 11import functools
8222d8de 12import io
b82f815f 13import itertools
8694c600 14import json
62fec3b2 15import locale
083c9df9 16import operator
8222d8de 17import os
dca08720 18import platform
8222d8de
JMF
19import re
20import shutil
dca08720 21import subprocess
8222d8de 22import sys
21cd8fae 23import tempfile
8222d8de 24import time
67134eab 25import tokenize
8222d8de 26import traceback
75822ca7 27import random
524e2e4f 28import unicodedata
8222d8de 29
ec11a9f4 30from enum import Enum
961ea474
S
31from string import ascii_letters
32
8c25f81b 33from .compat import (
82d8a8b6 34 compat_basestring,
003c69a8 35 compat_get_terminal_size,
4f026faf 36 compat_kwargs,
d0d9ade4 37 compat_numeric_types,
e9c0cdd3 38 compat_os_name,
edf65256 39 compat_pycrypto_AES,
7d1eb38a 40 compat_shlex_quote,
ce02ed60 41 compat_str,
67134eab 42 compat_tokenize_tokenize,
ce02ed60
PH
43 compat_urllib_error,
44 compat_urllib_request,
8b172c2e 45 compat_urllib_request_DataHandler,
819e0531 46 windows_enable_vt_mode,
8c25f81b 47)
982ee69a 48from .cookies import load_cookies
8c25f81b 49from .utils import (
eedb7ba5
S
50 age_restricted,
51 args_to_str,
ce02ed60
PH
52 ContentTooShortError,
53 date_from_str,
54 DateRange,
acd69589 55 DEFAULT_OUTTMPL,
ce02ed60 56 determine_ext,
b5559424 57 determine_protocol,
48f79687 58 DownloadCancelled,
ce02ed60 59 DownloadError,
c0384f22 60 encode_compat_str,
ce02ed60 61 encodeFilename,
498f5606 62 EntryNotInPlaylist,
a06916d9 63 error_to_compat_str,
8b0d7497 64 ExistingVideoReached,
590bc6f6 65 expand_path,
ce02ed60 66 ExtractorError,
e29663c6 67 float_or_none,
02dbf93f 68 format_bytes,
76d321f6 69 format_field,
e0fd9573 70 format_decimal_suffix,
525ef922 71 formatSeconds,
773f291d 72 GeoRestrictedError,
0bb322b9 73 get_domain,
b0249bca 74 HEADRequest,
d37707bd 75 InAdvancePagedList,
c9969434 76 int_or_none,
732044af 77 iri_to_uri,
773f291d 78 ISO3166Utils,
34921b43 79 join_nonempty,
56a8fb4f 80 LazyList,
08438d2c 81 LINK_TEMPLATES,
ce02ed60 82 locked_file,
0202b52a 83 make_dir,
dca08720 84 make_HTTPS_handler,
ce02ed60 85 MaxDownloadsReached,
3158150c 86 network_exceptions,
ec11a9f4 87 number_of_digits,
cd6fc19e 88 orderedSet,
a06916d9 89 OUTTMPL_TYPES,
b7ab0590 90 PagedList,
083c9df9 91 parse_filesize,
91410c9b 92 PerRequestProxyHandler,
dca08720 93 platform_name,
d3c93ec2 94 Popen,
1e43a6f7 95 POSTPROCESS_WHEN,
eedb7ba5 96 PostProcessingError,
ce02ed60 97 preferredencoding,
eedb7ba5 98 prepend_extension,
f2ebc5c7 99 ReExtractInfo,
51fb4995 100 register_socks_protocols,
a06916d9 101 RejectedVideoReached,
3efb96a6 102 remove_terminal_sequences,
cfb56d1a 103 render_table,
eedb7ba5 104 replace_extension,
ce02ed60
PH
105 SameFileError,
106 sanitize_filename,
1bb5c511 107 sanitize_path,
dcf77cf1 108 sanitize_url,
67dda517 109 sanitized_Request,
e5660ee6 110 std_headers,
819e0531 111 STR_FORMAT_RE_TMPL,
112 STR_FORMAT_TYPES,
1211bb6d 113 str_or_none,
e29663c6 114 strftime_or_none,
ce02ed60 115 subtitles_filename,
819e0531 116 supports_terminal_sequences,
f2ebc5c7 117 timetuple_from_msec,
732044af 118 to_high_limit_path,
324ad820 119 traverse_obj,
6033d980 120 try_get,
ce02ed60 121 UnavailableVideoError,
29eb5174 122 url_basename,
7d1eb38a 123 variadic,
58b1f00d 124 version_tuple,
ce02ed60
PH
125 write_json_file,
126 write_string,
6a3f4c3f 127 YoutubeDLCookieProcessor,
dca08720 128 YoutubeDLHandler,
fca6dba8 129 YoutubeDLRedirectHandler,
ce02ed60 130)
a0e07d31 131from .cache import Cache
ec11a9f4 132from .minicurses import format_text
52a8a1e1 133from .extractor import (
134 gen_extractor_classes,
135 get_info_extractor,
136 _LAZY_LOADER,
3ae5e797 137 _PLUGIN_CLASSES as plugin_extractors
52a8a1e1 138)
4c54b89e 139from .extractor.openload import PhantomJSwrapper
52a8a1e1 140from .downloader import (
dbf5416a 141 FFmpegFD,
52a8a1e1 142 get_suitable_downloader,
143 shorten_protocol_name
144)
4c83c967 145from .downloader.rtmp import rtmpdump_version
4f026faf 146from .postprocessor import (
e36d50c5 147 get_postprocessor,
4e3b637d 148 EmbedThumbnailPP,
adbc4ec4 149 FFmpegFixupDuplicateMoovPP,
e36d50c5 150 FFmpegFixupDurationPP,
f17f8651 151 FFmpegFixupM3u8PP,
62cd676c 152 FFmpegFixupM4aPP,
6271f1ca 153 FFmpegFixupStretchedPP,
e36d50c5 154 FFmpegFixupTimestampPP,
4f026faf
PH
155 FFmpegMergerPP,
156 FFmpegPostProcessor,
0202b52a 157 MoveFilesAfterDownloadPP,
3ae5e797 158 _PLUGIN_CLASSES as plugin_postprocessors
4f026faf 159)
4c88ff87 160from .update import detect_variant
36eaf303 161from .version import __version__, RELEASE_GIT_HEAD
8222d8de 162
e9c0cdd3
YCH
163if compat_os_name == 'nt':
164 import ctypes
165
2459b6e1 166
8222d8de
JMF
167class YoutubeDL(object):
168 """YoutubeDL class.
169
170 YoutubeDL objects are the ones responsible of downloading the
171 actual video file and writing it to disk if the user has requested
172 it, among some other tasks. In most cases there should be one per
173 program. As, given a video URL, the downloader doesn't know how to
174 extract all the needed information, task that InfoExtractors do, it
175 has to pass the URL to one of them.
176
177 For this, YoutubeDL objects have a method that allows
178 InfoExtractors to be registered in a given order. When it is passed
179 a URL, the YoutubeDL object handles it to the first InfoExtractor it
180 finds that reports being able to handle it. The InfoExtractor extracts
181 all the information about the video or videos the URL refers to, and
182 YoutubeDL process the extracted information, possibly using a File
183 Downloader to download the video.
184
185 YoutubeDL objects accept a lot of parameters. In order not to saturate
186 the object constructor with arguments, it receives a dictionary of
187 options instead. These options are available through the params
188 attribute for the InfoExtractors to use. The YoutubeDL also
189 registers itself as the downloader in charge for the InfoExtractors
190 that are added to it, so this is a "mutual registration".
191
192 Available options:
193
194 username: Username for authentication purposes.
195 password: Password for authentication purposes.
180940e0 196 videopassword: Password for accessing a video.
1da50aa3
S
197 ap_mso: Adobe Pass multiple-system operator identifier.
198 ap_username: Multiple-system operator account username.
199 ap_password: Multiple-system operator account password.
8222d8de
JMF
200 usenetrc: Use netrc for authentication instead.
201 verbose: Print additional info to stdout.
202 quiet: Do not print messages to stdout.
ad8915b7 203 no_warnings: Do not print out anything for warnings.
bb66c247 204 forceprint: A dict with keys WHEN mapped to a list of templates to
205 print to stdout. The allowed keys are video or any of the
206 items in utils.POSTPROCESS_WHEN.
ca30f449 207 For compatibility, a single list is also accepted
bb66c247 208 print_to_file: A dict with keys WHEN (same as forceprint) mapped to
209 a list of tuples with (template, filename)
53c18592 210 forceurl: Force printing final URL. (Deprecated)
211 forcetitle: Force printing title. (Deprecated)
212 forceid: Force printing ID. (Deprecated)
213 forcethumbnail: Force printing thumbnail URL. (Deprecated)
214 forcedescription: Force printing description. (Deprecated)
215 forcefilename: Force printing final filename. (Deprecated)
216 forceduration: Force printing duration. (Deprecated)
8694c600 217 forcejson: Force printing info_dict as JSON.
63e0be34
PH
218 dump_single_json: Force printing the info_dict of the whole playlist
219 (or video) as a single JSON line.
c25228e5 220 force_write_download_archive: Force writing download archive regardless
221 of 'skip_download' or 'simulate'.
b7b04c78 222 simulate: Do not download the video files. If unset (or None),
223 simulate only if listsubtitles, listformats or list_thumbnails is used
eb8a4433 224 format: Video format code. see "FORMAT SELECTION" for more details.
093a1710 225 You can also pass a function. The function takes 'ctx' as
226 argument and returns the formats to download.
227 See "build_format_selector" for an implementation
63ad4d43 228 allow_unplayable_formats: Allow unplayable formats to be extracted and downloaded.
b7da73eb 229 ignore_no_formats_error: Ignore "No video formats" error. Usefull for
230 extracting metadata even if the video is not actually
231 available for download (experimental)
0930b11f 232 format_sort: A list of fields by which to sort the video formats.
233 See "Sorting Formats" for more details.
c25228e5 234 format_sort_force: Force the given format_sort. see "Sorting Formats"
235 for more details.
236 allow_multiple_video_streams: Allow multiple video streams to be merged
237 into a single file
238 allow_multiple_audio_streams: Allow multiple audio streams to be merged
239 into a single file
0ba692ac 240 check_formats Whether to test if the formats are downloadable.
9f1a1c36 241 Can be True (check all), False (check none),
242 'selected' (check selected formats),
0ba692ac 243 or None (check only if requested by extractor)
4524baf0 244 paths: Dictionary of output paths. The allowed keys are 'home'
245 'temp' and the keys of OUTTMPL_TYPES (in utils.py)
de6000d9 246 outtmpl: Dictionary of templates for output names. Allowed keys
4524baf0 247 are 'default' and the keys of OUTTMPL_TYPES (in utils.py).
34488702 248 For compatibility with youtube-dl, a single string can also be used
a820dc72
RA
249 outtmpl_na_placeholder: Placeholder for unavailable meta fields.
250 restrictfilenames: Do not allow "&" and spaces in file names
251 trim_file_name: Limit length of filename (extension excluded)
4524baf0 252 windowsfilenames: Force the filenames to be windows compatible
b1940459 253 ignoreerrors: Do not stop on download/postprocessing errors.
254 Can be 'only_download' to ignore only download errors.
255 Default is 'only_download' for CLI, but False for API
26e2805c 256 skip_playlist_after_errors: Number of allowed failures until the rest of
257 the playlist is skipped
d22dec74 258 force_generic_extractor: Force downloader to use the generic extractor
0c3d0f51 259 overwrites: Overwrite all video and metadata files if True,
260 overwrite only non-video files if None
261 and don't overwrite any file if False
34488702 262 For compatibility with youtube-dl,
263 "nooverwrites" may also be used instead
8222d8de
JMF
264 playliststart: Playlist item to start at.
265 playlistend: Playlist item to end at.
c14e88f0 266 playlist_items: Specific indices of playlist to download.
ff815fe6 267 playlistreverse: Download playlist items in reverse order.
75822ca7 268 playlistrandom: Download playlist items in random order.
8222d8de
JMF
269 matchtitle: Download only matching titles.
270 rejecttitle: Reject downloads for matching titles.
8bf9319e 271 logger: Log messages to a logging.Logger instance.
8222d8de 272 logtostderr: Log messages to stderr instead of stdout.
819e0531 273 consoletitle: Display progress in console window's titlebar.
8222d8de
JMF
274 writedescription: Write the video description to a .description file
275 writeinfojson: Write the video description to a .info.json file
75d43ca0 276 clean_infojson: Remove private fields from the infojson
34488702 277 getcomments: Extract video comments. This will not be written to disk
06167fbb 278 unless writeinfojson is also given
1fb07d10 279 writeannotations: Write the video annotations to a .annotations.xml file
8222d8de 280 writethumbnail: Write the thumbnail image to a file
c25228e5 281 allow_playlist_files: Whether to write playlists' description, infojson etc
282 also to disk when using the 'write*' options
ec82d85a 283 write_all_thumbnails: Write all thumbnail formats to files
732044af 284 writelink: Write an internet shortcut file, depending on the
285 current platform (.url/.webloc/.desktop)
286 writeurllink: Write a Windows internet shortcut file (.url)
287 writewebloclink: Write a macOS internet shortcut file (.webloc)
288 writedesktoplink: Write a Linux internet shortcut file (.desktop)
8222d8de 289 writesubtitles: Write the video subtitles to a file
741dd8ea 290 writeautomaticsub: Write the automatically generated subtitles to a file
245524e6 291 allsubtitles: Deprecated - Use subtitleslangs = ['all']
c32b0aab 292 Downloads all the subtitles of the video
0b7f3118 293 (requires writesubtitles or writeautomaticsub)
8222d8de 294 listsubtitles: Lists all available subtitles for the video
a504ced0 295 subtitlesformat: The format code for subtitles
c32b0aab 296 subtitleslangs: List of languages of the subtitles to download (can be regex).
297 The list may contain "all" to refer to all the available
298 subtitles. The language can be prefixed with a "-" to
299 exclude it from the requested languages. Eg: ['all', '-live_chat']
8222d8de
JMF
300 keepvideo: Keep the video file after post-processing
301 daterange: A DateRange object, download only if the upload_date is in the range.
302 skip_download: Skip the actual download of the video file
c35f9e72 303 cachedir: Location of the cache files in the filesystem.
a0e07d31 304 False to disable filesystem cache.
47192f92 305 noplaylist: Download single video instead of a playlist if in doubt.
8dbe9899
PH
306 age_limit: An integer representing the user's age in years.
307 Unsuitable videos for the given age are skipped.
5fe18bdb
PH
308 min_views: An integer representing the minimum view count the video
309 must have in order to not be skipped.
310 Videos without view count information are always
311 downloaded. None for no limit.
312 max_views: An integer representing the maximum view count.
313 Videos that are more popular than that are not
314 downloaded.
315 Videos without view count information are always
316 downloaded. None for no limit.
317 download_archive: File name of a file where all downloads are recorded.
c1c9a79c
PH
318 Videos already present in the file are not downloaded
319 again.
8a51f564 320 break_on_existing: Stop the download process after attempting to download a
321 file that is in the archive.
322 break_on_reject: Stop the download process when encountering a video that
323 has been filtered out.
b222c271 324 break_per_url: Whether break_on_reject and break_on_existing
325 should act on each input URL as opposed to for the entire queue
8a51f564 326 cookiefile: File name where cookies should be read from and dumped to
f59f5ef8
MB
327 cookiesfrombrowser: A tuple containing the name of the browser, the profile
328 name/pathfrom where cookies are loaded, and the name of the
329 keyring. Eg: ('chrome', ) or ('vivaldi', 'default', 'BASICTEXT')
f81c62a6 330 legacyserverconnect: Explicitly allow HTTPS connection to servers that do not
331 support RFC 5746 secure renegotiation
f59f5ef8 332 nocheckcertificate: Do not verify SSL certificates
7e8c0af0
PH
333 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
334 At the moment, this is only supported by YouTube.
a1ee09e8 335 proxy: URL of the proxy server to use
38cce791 336 geo_verification_proxy: URL of the proxy to use for IP address verification
504f20dd 337 on geo-restricted sites.
e344693b 338 socket_timeout: Time to wait for unresponsive hosts, in seconds
0783b09b
PH
339 bidi_workaround: Work around buggy terminals without bidirectional text
340 support, using fridibi
a0ddb8a2 341 debug_printtraffic:Print out sent and received HTTP traffic
91f071af 342 include_ads: Download ads as well (deprecated)
04b4d394
PH
343 default_search: Prepend this string if an input url is not valid.
344 'auto' for elaborate guessing
62fec3b2 345 encoding: Use this encoding instead of the system-specified.
e8ee972c 346 extract_flat: Do not resolve URLs, return the immediate result.
057a5206
PH
347 Pass in 'in_playlist' to only show this behavior for
348 playlist items.
f2ebc5c7 349 wait_for_video: If given, wait for scheduled streams to become available.
350 The value should be a tuple containing the range
351 (min_secs, max_secs) to wait between retries
4f026faf 352 postprocessors: A list of dictionaries, each with an entry
71b640cc 353 * key: The name of the postprocessor. See
7a5c1cfe 354 yt_dlp/postprocessor/__init__.py for a list.
bb66c247 355 * when: When to run the postprocessor. Allowed values are
356 the entries of utils.POSTPROCESS_WHEN
56d868db 357 Assumed to be 'post_process' if not given
b5ae35ee 358 post_hooks: Deprecated - Register a custom postprocessor instead
359 A list of functions that get called as the final step
ab8e5e51
AM
360 for each video file, after all postprocessors have been
361 called. The filename will be passed as the only argument.
71b640cc
PH
362 progress_hooks: A list of functions that get called on download
363 progress, with a dictionary with the entries
5cda4eda 364 * status: One of "downloading", "error", or "finished".
ee69b99a 365 Check this first and ignore unknown values.
3ba7740d 366 * info_dict: The extracted info_dict
71b640cc 367
5cda4eda 368 If status is one of "downloading", or "finished", the
ee69b99a
PH
369 following properties may also be present:
370 * filename: The final filename (always present)
5cda4eda 371 * tmpfilename: The filename we're currently writing to
71b640cc
PH
372 * downloaded_bytes: Bytes on disk
373 * total_bytes: Size of the whole file, None if unknown
5cda4eda
PH
374 * total_bytes_estimate: Guess of the eventual file size,
375 None if unavailable.
376 * elapsed: The number of seconds since download started.
71b640cc
PH
377 * eta: The estimated time in seconds, None if unknown
378 * speed: The download speed in bytes/second, None if
379 unknown
5cda4eda
PH
380 * fragment_index: The counter of the currently
381 downloaded video fragment.
382 * fragment_count: The number of fragments (= individual
383 files that will be merged)
71b640cc
PH
384
385 Progress hooks are guaranteed to be called at least once
386 (with status "finished") if the download is successful.
819e0531 387 postprocessor_hooks: A list of functions that get called on postprocessing
388 progress, with a dictionary with the entries
389 * status: One of "started", "processing", or "finished".
390 Check this first and ignore unknown values.
391 * postprocessor: Name of the postprocessor
392 * info_dict: The extracted info_dict
393
394 Progress hooks are guaranteed to be called at least twice
395 (with status "started" and "finished") if the processing is successful.
45598f15 396 merge_output_format: Extension to use when merging formats.
6b591b29 397 final_ext: Expected final extension; used to detect when the file was
59a7a13e 398 already downloaded and converted
6271f1ca
PH
399 fixup: Automatically correct known faults of the file.
400 One of:
401 - "never": do nothing
402 - "warn": only emit a warning
403 - "detect_or_warn": check whether we can do anything
62cd676c 404 about it, warn otherwise (default)
504f20dd 405 source_address: Client-side IP address to bind to.
6ec6cb4e 406 call_home: Boolean, true iff we are allowed to contact the
7a5c1cfe 407 yt-dlp servers for debugging. (BROKEN)
1cf376f5 408 sleep_interval_requests: Number of seconds to sleep between requests
409 during extraction
7aa589a5
S
410 sleep_interval: Number of seconds to sleep before each download when
411 used alone or a lower bound of a range for randomized
412 sleep before each download (minimum possible number
413 of seconds to sleep) when used along with
414 max_sleep_interval.
415 max_sleep_interval:Upper bound of a range for randomized sleep before each
416 download (maximum possible number of seconds to sleep).
417 Must only be used along with sleep_interval.
418 Actual sleep time will be a random float from range
419 [sleep_interval; max_sleep_interval].
1cf376f5 420 sleep_interval_subtitles: Number of seconds to sleep before each subtitle download
cfb56d1a
PH
421 listformats: Print an overview of available video formats and exit.
422 list_thumbnails: Print a table of all thumbnails and exit.
347de493
PH
423 match_filter: A function that gets called with the info_dict of
424 every video.
425 If it returns a message, the video is ignored.
426 If it returns None, the video is downloaded.
427 match_filter_func in utils.py is one example for this.
7e5db8c9 428 no_color: Do not emit color codes in output.
0a840f58 429 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
504f20dd 430 HTTP header
0a840f58 431 geo_bypass_country:
773f291d
S
432 Two-letter ISO 3166-2 country code that will be used for
433 explicit geographic restriction bypassing via faking
504f20dd 434 X-Forwarded-For HTTP header
5f95927a
S
435 geo_bypass_ip_block:
436 IP range in CIDR notation that will be used similarly to
504f20dd 437 geo_bypass_country
71b640cc 438
85729c51 439 The following options determine which downloader is picked:
52a8a1e1 440 external_downloader: A dictionary of protocol keys and the executable of the
441 external downloader to use for it. The allowed protocols
442 are default|http|ftp|m3u8|dash|rtsp|rtmp|mms.
443 Set the value to 'native' to use the native downloader
444 hls_prefer_native: Deprecated - Use external_downloader = {'m3u8': 'native'}
445 or {'m3u8': 'ffmpeg'} instead.
446 Use the native HLS downloader instead of ffmpeg/avconv
bf09af3a
S
447 if True, otherwise use ffmpeg/avconv if False, otherwise
448 use downloader suggested by extractor if None.
53ed7066 449 compat_opts: Compatibility options. See "Differences in default behavior".
3acf6d38 450 The following options do not work when used through the API:
b5ae35ee 451 filename, abort-on-error, multistreams, no-live-chat, format-sort
dac5df5a 452 no-clean-infojson, no-playlist-metafiles, no-keep-subs, no-attach-info-json.
e4f02757 453 Refer __init__.py for their implementation
819e0531 454 progress_template: Dictionary of templates for progress outputs.
455 Allowed keys are 'download', 'postprocess',
456 'download-title' (console title) and 'postprocess-title'.
457 The template is mapped on a dictionary with keys 'progress' and 'info'
fe7e0c98 458
8222d8de 459 The following parameters are not used by YoutubeDL itself, they are used by
7a5c1cfe 460 the downloader (see yt_dlp/downloader/common.py):
51d9739f 461 nopart, updatetime, buffersize, ratelimit, throttledratelimit, min_filesize,
205a0654
EH
462 max_filesize, test, noresizebuffer, retries, file_access_retries, fragment_retries,
463 continuedl, noprogress, xattr_set_filesize, hls_use_mpegts, http_chunk_size,
59a7a13e 464 external_downloader_args, concurrent_fragment_downloads.
76b1bd67
JMF
465
466 The following options are used by the post processors:
d4a24f40 467 prefer_ffmpeg: If False, use avconv instead of ffmpeg if both are available,
e4172ac9 468 otherwise prefer ffmpeg. (avconv support is deprecated)
c0b7d117
S
469 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
470 to the binary or its containing directory.
43820c03 471 postprocessor_args: A dictionary of postprocessor/executable keys (in lower case)
34488702 472 and a list of additional command-line arguments for the
473 postprocessor/executable. The dict can also have "PP+EXE" keys
474 which are used when the given exe is used by the given PP.
475 Use 'default' as the name for arguments to passed to all PP
476 For compatibility with youtube-dl, a single list of args
477 can also be used
e409895f 478
479 The following options are used by the extractors:
62bff2c1 480 extractor_retries: Number of times to retry for known errors
481 dynamic_mpd: Whether to process dynamic DASH manifests (default: True)
e409895f 482 hls_split_discontinuity: Split HLS playlists to different formats at
62bff2c1 483 discontinuities such as ad breaks (default: False)
5d3a0e79 484 extractor_args: A dictionary of arguments to be passed to the extractors.
485 See "EXTRACTOR ARGUMENTS" for details.
486 Eg: {'youtube': {'skip': ['dash', 'hls']}}
88f23a18 487 mark_watched: Mark videos watched (even with --simulate). Only for YouTube
5d3a0e79 488 youtube_include_dash_manifest: Deprecated - Use extractor_args instead.
489 If True (default), DASH manifests and related
62bff2c1 490 data will be downloaded and processed by extractor.
491 You can reduce network I/O by disabling it if you don't
492 care about DASH. (only for youtube)
5d3a0e79 493 youtube_include_hls_manifest: Deprecated - Use extractor_args instead.
494 If True (default), HLS manifests and related
62bff2c1 495 data will be downloaded and processed by extractor.
496 You can reduce network I/O by disabling it if you don't
497 care about HLS. (only for youtube)
8222d8de
JMF
498 """
499
c9969434
S
500 _NUMERIC_FIELDS = set((
501 'width', 'height', 'tbr', 'abr', 'asr', 'vbr', 'fps', 'filesize', 'filesize_approx',
e6f21b3d 502 'timestamp', 'release_timestamp',
c9969434
S
503 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
504 'average_rating', 'comment_count', 'age_limit',
505 'start_time', 'end_time',
506 'chapter_number', 'season_number', 'episode_number',
507 'track_number', 'disc_number', 'release_year',
c9969434
S
508 ))
509
48ee10ee 510 _format_selection_exts = {
511 'audio': {'m4a', 'mp3', 'ogg', 'aac'},
512 'video': {'mp4', 'flv', 'webm', '3gp'},
513 'storyboards': {'mhtml'},
514 }
515
8222d8de 516 params = None
8b7491c8 517 _ies = {}
1e43a6f7 518 _pps = {k: [] for k in POSTPROCESS_WHEN}
b35496d8 519 _printed_messages = set()
1cf376f5 520 _first_webpage_request = True
8222d8de
JMF
521 _download_retcode = None
522 _num_downloads = None
30a074c2 523 _playlist_level = 0
524 _playlist_urls = set()
8222d8de
JMF
525 _screen_file = None
526
3511266b 527 def __init__(self, params=None, auto_init=True):
883d4b1e 528 """Create a FileDownloader object with the given options.
529 @param auto_init Whether to load the default extractors and print header (if verbose).
49a57e70 530 Set to 'no_verbose_header' to not print the header
883d4b1e 531 """
e9f9a10f
JMF
532 if params is None:
533 params = {}
8b7491c8 534 self._ies = {}
56c73665 535 self._ies_instances = {}
1e43a6f7 536 self._pps = {k: [] for k in POSTPROCESS_WHEN}
b35496d8 537 self._printed_messages = set()
1cf376f5 538 self._first_webpage_request = True
ab8e5e51 539 self._post_hooks = []
933605d7 540 self._progress_hooks = []
819e0531 541 self._postprocessor_hooks = []
8222d8de
JMF
542 self._download_retcode = 0
543 self._num_downloads = 0
9c906919 544 self._num_videos = 0
8222d8de 545 self._screen_file = [sys.stdout, sys.stderr][params.get('logtostderr', False)]
0783b09b 546 self._err_file = sys.stderr
819e0531 547 self.params = params
a0e07d31 548 self.cache = Cache(self)
34308b30 549
819e0531 550 windows_enable_vt_mode()
ec11a9f4 551 self._allow_colors = {
552 'screen': not self.params.get('no_color') and supports_terminal_sequences(self._screen_file),
553 'err': not self.params.get('no_color') and supports_terminal_sequences(self._err_file),
554 }
819e0531 555
a61f4b28 556 if sys.version_info < (3, 6):
557 self.report_warning(
0181adef 558 'Python version %d.%d is not supported! Please update to Python 3.6 or above' % sys.version_info[:2])
a61f4b28 559
88acdbc2 560 if self.params.get('allow_unplayable_formats'):
561 self.report_warning(
ec11a9f4 562 f'You have asked for {self._format_err("UNPLAYABLE", self.Styles.EMPHASIS)} formats to be listed/downloaded. '
819e0531 563 'This is a developer option intended for debugging. \n'
564 ' If you experience any issues while using this option, '
ec11a9f4 565 f'{self._format_err("DO NOT", self.Styles.ERROR)} open a bug report')
88acdbc2 566
be5df5ee
S
567 def check_deprecated(param, option, suggestion):
568 if self.params.get(param) is not None:
53ed7066 569 self.report_warning('%s is deprecated. Use %s instead' % (option, suggestion))
be5df5ee
S
570 return True
571 return False
572
573 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
38cce791
YCH
574 if self.params.get('geo_verification_proxy') is None:
575 self.params['geo_verification_proxy'] = self.params['cn_verification_proxy']
576
0d1bb027 577 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
578 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
53ed7066 579 check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"')
0d1bb027 580
49a57e70 581 for msg in self.params.get('_warnings', []):
0d1bb027 582 self.report_warning(msg)
ee8dd27a 583 for msg in self.params.get('_deprecation_warnings', []):
584 self.deprecation_warning(msg)
0d1bb027 585
ec11a9f4 586 if 'list-formats' in self.params.get('compat_opts', []):
587 self.params['listformats_table'] = False
588
b5ae35ee 589 if 'overwrites' not in self.params and self.params.get('nooverwrites') is not None:
b868936c 590 # nooverwrites was unnecessarily changed to overwrites
591 # in 0c3d0f51778b153f65c21906031c2e091fcfb641
592 # This ensures compatibility with both keys
593 self.params['overwrites'] = not self.params['nooverwrites']
b5ae35ee 594 elif self.params.get('overwrites') is None:
595 self.params.pop('overwrites', None)
b868936c 596 else:
597 self.params['nooverwrites'] = not self.params['overwrites']
b9d973be 598
455a15e2 599 self.params.setdefault('forceprint', {})
600 self.params.setdefault('print_to_file', {})
bb66c247 601
602 # Compatibility with older syntax
ca30f449 603 if not isinstance(params['forceprint'], dict):
455a15e2 604 self.params['forceprint'] = {'video': params['forceprint']}
ca30f449 605
455a15e2 606 if self.params.get('bidi_workaround', False):
1c088fa8
PH
607 try:
608 import pty
609 master, slave = pty.openpty()
003c69a8 610 width = compat_get_terminal_size().columns
1c088fa8
PH
611 if width is None:
612 width_args = []
613 else:
614 width_args = ['-w', str(width)]
5d681e96 615 sp_kwargs = dict(
1c088fa8
PH
616 stdin=subprocess.PIPE,
617 stdout=slave,
618 stderr=self._err_file)
5d681e96 619 try:
d3c93ec2 620 self._output_process = Popen(['bidiv'] + width_args, **sp_kwargs)
5d681e96 621 except OSError:
d3c93ec2 622 self._output_process = Popen(['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
5d681e96 623 self._output_channel = os.fdopen(master, 'rb')
1c088fa8 624 except OSError as ose:
66e7ace1 625 if ose.errno == errno.ENOENT:
49a57e70 626 self.report_warning(
627 'Could not find fribidi executable, ignoring --bidi-workaround. '
628 'Make sure that fribidi is an executable file in one of the directories in your $PATH.')
1c088fa8
PH
629 else:
630 raise
0783b09b 631
3089bc74
S
632 if (sys.platform != 'win32'
633 and sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
455a15e2 634 and not self.params.get('restrictfilenames', False)):
e9137224 635 # Unicode filesystem API will throw errors (#1474, #13027)
34308b30 636 self.report_warning(
6febd1c1 637 'Assuming --restrict-filenames since file system encoding '
1b725173 638 'cannot encode all characters. '
6febd1c1 639 'Set the LC_ALL environment variable to fix this.')
4a98cdbf 640 self.params['restrictfilenames'] = True
34308b30 641
de6000d9 642 self.outtmpl_dict = self.parse_outtmpl()
486dd09e 643
187986a8 644 # Creating format selector here allows us to catch syntax errors before the extraction
645 self.format_selector = (
fa9f30b8 646 self.params.get('format') if self.params.get('format') in (None, '-')
093a1710 647 else self.params['format'] if callable(self.params['format'])
187986a8 648 else self.build_format_selector(self.params['format']))
649
dca08720
PH
650 self._setup_opener()
651
3511266b 652 if auto_init:
883d4b1e 653 if auto_init != 'no_verbose_header':
654 self.print_debug_header()
3511266b
PH
655 self.add_default_info_extractors()
656
013b50b7 657 hooks = {
658 'post_hooks': self.add_post_hook,
659 'progress_hooks': self.add_progress_hook,
660 'postprocessor_hooks': self.add_postprocessor_hook,
661 }
662 for opt, fn in hooks.items():
663 for ph in self.params.get(opt, []):
664 fn(ph)
71b640cc 665
5bfc8bee 666 for pp_def_raw in self.params.get('postprocessors', []):
667 pp_def = dict(pp_def_raw)
668 when = pp_def.pop('when', 'post_process')
669 self.add_post_processor(
670 get_postprocessor(pp_def.pop('key'))(self, **compat_kwargs(pp_def)),
671 when=when)
672
51fb4995
YCH
673 register_socks_protocols()
674
ed39cac5 675 def preload_download_archive(fn):
676 """Preload the archive, if any is specified"""
677 if fn is None:
678 return False
49a57e70 679 self.write_debug(f'Loading archive file {fn!r}')
ed39cac5 680 try:
681 with locked_file(fn, 'r', encoding='utf-8') as archive_file:
682 for line in archive_file:
683 self.archive.add(line.strip())
684 except IOError as ioe:
685 if ioe.errno != errno.ENOENT:
686 raise
687 return False
688 return True
689
690 self.archive = set()
691 preload_download_archive(self.params.get('download_archive'))
692
7d4111ed
PH
693 def warn_if_short_id(self, argv):
694 # short YouTube ID starting with dash?
695 idxs = [
696 i for i, a in enumerate(argv)
697 if re.match(r'^-[0-9A-Za-z_-]{10}$', a)]
698 if idxs:
699 correct_argv = (
7a5c1cfe 700 ['yt-dlp']
3089bc74
S
701 + [a for i, a in enumerate(argv) if i not in idxs]
702 + ['--'] + [argv[i] for i in idxs]
7d4111ed
PH
703 )
704 self.report_warning(
705 'Long argument string detected. '
49a57e70 706 'Use -- to separate parameters and URLs, like this:\n%s' %
7d4111ed
PH
707 args_to_str(correct_argv))
708
8222d8de
JMF
709 def add_info_extractor(self, ie):
710 """Add an InfoExtractor object to the end of the list."""
8b7491c8 711 ie_key = ie.ie_key()
712 self._ies[ie_key] = ie
e52d7f85 713 if not isinstance(ie, type):
8b7491c8 714 self._ies_instances[ie_key] = ie
e52d7f85 715 ie.set_downloader(self)
8222d8de 716
8b7491c8 717 def _get_info_extractor_class(self, ie_key):
718 ie = self._ies.get(ie_key)
719 if ie is None:
720 ie = get_info_extractor(ie_key)
721 self.add_info_extractor(ie)
722 return ie
723
56c73665
JMF
724 def get_info_extractor(self, ie_key):
725 """
726 Get an instance of an IE with name ie_key, it will try to get one from
727 the _ies list, if there's no instance it will create a new one and add
728 it to the extractor list.
729 """
730 ie = self._ies_instances.get(ie_key)
731 if ie is None:
732 ie = get_info_extractor(ie_key)()
733 self.add_info_extractor(ie)
734 return ie
735
023fa8c4
JMF
736 def add_default_info_extractors(self):
737 """
738 Add the InfoExtractors returned by gen_extractors to the end of the list
739 """
e52d7f85 740 for ie in gen_extractor_classes():
023fa8c4
JMF
741 self.add_info_extractor(ie)
742
56d868db 743 def add_post_processor(self, pp, when='post_process'):
8222d8de 744 """Add a PostProcessor object to the end of the chain."""
5bfa4862 745 self._pps[when].append(pp)
8222d8de
JMF
746 pp.set_downloader(self)
747
ab8e5e51
AM
748 def add_post_hook(self, ph):
749 """Add the post hook"""
750 self._post_hooks.append(ph)
751
933605d7 752 def add_progress_hook(self, ph):
819e0531 753 """Add the download progress hook"""
933605d7 754 self._progress_hooks.append(ph)
8ab470f1 755
819e0531 756 def add_postprocessor_hook(self, ph):
757 """Add the postprocessing progress hook"""
758 self._postprocessor_hooks.append(ph)
5bfc8bee 759 for pps in self._pps.values():
760 for pp in pps:
761 pp.add_progress_hook(ph)
819e0531 762
1c088fa8 763 def _bidi_workaround(self, message):
5d681e96 764 if not hasattr(self, '_output_channel'):
1c088fa8
PH
765 return message
766
5d681e96 767 assert hasattr(self, '_output_process')
11b85ce6 768 assert isinstance(message, compat_str)
6febd1c1
PH
769 line_count = message.count('\n') + 1
770 self._output_process.stdin.write((message + '\n').encode('utf-8'))
5d681e96 771 self._output_process.stdin.flush()
6febd1c1 772 res = ''.join(self._output_channel.readline().decode('utf-8')
9e1a5b84 773 for _ in range(line_count))
6febd1c1 774 return res[:-len('\n')]
1c088fa8 775
b35496d8 776 def _write_string(self, message, out=None, only_once=False):
777 if only_once:
778 if message in self._printed_messages:
779 return
780 self._printed_messages.add(message)
781 write_string(message, out=out, encoding=self.params.get('encoding'))
734f90bb 782
848887eb 783 def to_stdout(self, message, skip_eol=False, quiet=False):
0760b0a7 784 """Print message to stdout"""
8bf9319e 785 if self.params.get('logger'):
43afe285 786 self.params['logger'].debug(message)
835a1478 787 elif not quiet or self.params.get('verbose'):
788 self._write_string(
789 '%s%s' % (self._bidi_workaround(message), ('' if skip_eol else '\n')),
790 self._err_file if quiet else self._screen_file)
8222d8de 791
b35496d8 792 def to_stderr(self, message, only_once=False):
0760b0a7 793 """Print message to stderr"""
11b85ce6 794 assert isinstance(message, compat_str)
8bf9319e 795 if self.params.get('logger'):
43afe285
IB
796 self.params['logger'].error(message)
797 else:
b35496d8 798 self._write_string('%s\n' % self._bidi_workaround(message), self._err_file, only_once=only_once)
8222d8de 799
1e5b9a95
PH
800 def to_console_title(self, message):
801 if not self.params.get('consoletitle', False):
802 return
3efb96a6 803 message = remove_terminal_sequences(message)
4bede0d8
C
804 if compat_os_name == 'nt':
805 if ctypes.windll.kernel32.GetConsoleWindow():
806 # c_wchar_p() might not be necessary if `message` is
807 # already of type unicode()
808 ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
1e5b9a95 809 elif 'TERM' in os.environ:
b46696bd 810 self._write_string('\033]0;%s\007' % message, self._screen_file)
1e5b9a95 811
bdde425c
PH
812 def save_console_title(self):
813 if not self.params.get('consoletitle', False):
814 return
b7b04c78 815 if self.params.get('simulate'):
94c3442e 816 return
4bede0d8 817 if compat_os_name != 'nt' and 'TERM' in os.environ:
efd6c574 818 # Save the title on stack
734f90bb 819 self._write_string('\033[22;0t', self._screen_file)
bdde425c
PH
820
821 def restore_console_title(self):
822 if not self.params.get('consoletitle', False):
823 return
b7b04c78 824 if self.params.get('simulate'):
94c3442e 825 return
4bede0d8 826 if compat_os_name != 'nt' and 'TERM' in os.environ:
efd6c574 827 # Restore the title from stack
734f90bb 828 self._write_string('\033[23;0t', self._screen_file)
bdde425c
PH
829
830 def __enter__(self):
831 self.save_console_title()
832 return self
833
834 def __exit__(self, *args):
835 self.restore_console_title()
f89197d7 836
dca08720 837 if self.params.get('cookiefile') is not None:
1bab3437 838 self.cookiejar.save(ignore_discard=True, ignore_expires=True)
bdde425c 839
fa9f30b8 840 def trouble(self, message=None, tb=None, is_error=True):
8222d8de
JMF
841 """Determine action to take when a download problem appears.
842
843 Depending on if the downloader has been configured to ignore
844 download errors or not, this method may throw an exception or
845 not when errors are found, after printing the message.
846
fa9f30b8 847 @param tb If given, is additional traceback information
848 @param is_error Whether to raise error according to ignorerrors
8222d8de
JMF
849 """
850 if message is not None:
851 self.to_stderr(message)
852 if self.params.get('verbose'):
853 if tb is None:
854 if sys.exc_info()[0]: # if .trouble has been called from an except block
6febd1c1 855 tb = ''
8222d8de 856 if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
6febd1c1 857 tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
c0384f22 858 tb += encode_compat_str(traceback.format_exc())
8222d8de
JMF
859 else:
860 tb_data = traceback.format_list(traceback.extract_stack())
6febd1c1 861 tb = ''.join(tb_data)
c19bc311 862 if tb:
863 self.to_stderr(tb)
fa9f30b8 864 if not is_error:
865 return
b1940459 866 if not self.params.get('ignoreerrors'):
8222d8de
JMF
867 if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
868 exc_info = sys.exc_info()[1].exc_info
869 else:
870 exc_info = sys.exc_info()
871 raise DownloadError(message, exc_info)
872 self._download_retcode = 1
873
0760b0a7 874 def to_screen(self, message, skip_eol=False):
875 """Print message to stdout if not in quiet mode"""
876 self.to_stdout(
877 message, skip_eol, quiet=self.params.get('quiet', False))
878
ec11a9f4 879 class Styles(Enum):
880 HEADERS = 'yellow'
f304da8a 881 EMPHASIS = 'light blue'
ec11a9f4 882 ID = 'green'
883 DELIM = 'blue'
884 ERROR = 'red'
885 WARNING = 'yellow'
ff51ed58 886 SUPPRESS = 'light black'
ec11a9f4 887
7578d77d 888 def _format_text(self, handle, allow_colors, text, f, fallback=None, *, test_encoding=False):
ec11a9f4 889 if test_encoding:
890 original_text = text
5c104538 891 # handle.encoding can be None. See https://github.com/yt-dlp/yt-dlp/issues/2711
892 encoding = self.params.get('encoding') or getattr(handle, 'encoding', None) or 'ascii'
ec11a9f4 893 text = text.encode(encoding, 'ignore').decode(encoding)
894 if fallback is not None and text != original_text:
895 text = fallback
896 if isinstance(f, self.Styles):
f304da8a 897 f = f.value
7578d77d 898 return format_text(text, f) if allow_colors else text if fallback is None else fallback
ec11a9f4 899
900 def _format_screen(self, *args, **kwargs):
7578d77d 901 return self._format_text(
902 self._screen_file, self._allow_colors['screen'], *args, **kwargs)
ec11a9f4 903
904 def _format_err(self, *args, **kwargs):
7578d77d 905 return self._format_text(
906 self._err_file, self._allow_colors['err'], *args, **kwargs)
819e0531 907
c84aeac6 908 def report_warning(self, message, only_once=False):
8222d8de
JMF
909 '''
910 Print the message to stderr, it will be prefixed with 'WARNING:'
911 If stderr is a tty file the 'WARNING:' will be colored
912 '''
6d07ce01
JMF
913 if self.params.get('logger') is not None:
914 self.params['logger'].warning(message)
8222d8de 915 else:
ad8915b7
PH
916 if self.params.get('no_warnings'):
917 return
ec11a9f4 918 self.to_stderr(f'{self._format_err("WARNING:", self.Styles.WARNING)} {message}', only_once)
8222d8de 919
ee8dd27a 920 def deprecation_warning(self, message):
921 if self.params.get('logger') is not None:
922 self.params['logger'].warning('DeprecationWarning: {message}')
923 else:
924 self.to_stderr(f'{self._format_err("DeprecationWarning:", self.Styles.ERROR)} {message}', True)
925
fa9f30b8 926 def report_error(self, message, *args, **kwargs):
8222d8de
JMF
927 '''
928 Do the same as trouble, but prefixes the message with 'ERROR:', colored
929 in red if stderr is a tty file.
930 '''
fa9f30b8 931 self.trouble(f'{self._format_err("ERROR:", self.Styles.ERROR)} {message}', *args, **kwargs)
8222d8de 932
b35496d8 933 def write_debug(self, message, only_once=False):
0760b0a7 934 '''Log debug message or Print message to stderr'''
935 if not self.params.get('verbose', False):
936 return
937 message = '[debug] %s' % message
938 if self.params.get('logger'):
939 self.params['logger'].debug(message)
940 else:
b35496d8 941 self.to_stderr(message, only_once)
0760b0a7 942
8222d8de
JMF
943 def report_file_already_downloaded(self, file_name):
944 """Report file has already been fully downloaded."""
945 try:
6febd1c1 946 self.to_screen('[download] %s has already been downloaded' % file_name)
ce02ed60 947 except UnicodeEncodeError:
6febd1c1 948 self.to_screen('[download] The file has already been downloaded')
8222d8de 949
0c3d0f51 950 def report_file_delete(self, file_name):
951 """Report that existing file will be deleted."""
952 try:
c25228e5 953 self.to_screen('Deleting existing file %s' % file_name)
0c3d0f51 954 except UnicodeEncodeError:
c25228e5 955 self.to_screen('Deleting existing file')
0c3d0f51 956
1151c407 957 def raise_no_formats(self, info, forced=False):
958 has_drm = info.get('__has_drm')
88acdbc2 959 msg = 'This video is DRM protected' if has_drm else 'No video formats found!'
960 expected = self.params.get('ignore_no_formats_error')
961 if forced or not expected:
1151c407 962 raise ExtractorError(msg, video_id=info['id'], ie=info['extractor'],
963 expected=has_drm or expected)
88acdbc2 964 else:
965 self.report_warning(msg)
966
de6000d9 967 def parse_outtmpl(self):
968 outtmpl_dict = self.params.get('outtmpl', {})
969 if not isinstance(outtmpl_dict, dict):
970 outtmpl_dict = {'default': outtmpl_dict}
71ce444a 971 # Remove spaces in the default template
972 if self.params.get('restrictfilenames'):
973 sanitize = lambda x: x.replace(' - ', ' ').replace(' ', '-')
974 else:
975 sanitize = lambda x: x
de6000d9 976 outtmpl_dict.update({
71ce444a 977 k: sanitize(v) for k, v in DEFAULT_OUTTMPL.items()
80c03fa9 978 if outtmpl_dict.get(k) is None})
de6000d9 979 for key, val in outtmpl_dict.items():
980 if isinstance(val, bytes):
981 self.report_warning(
982 'Parameter outtmpl is bytes, but should be a unicode string. '
983 'Put from __future__ import unicode_literals at the top of your code file or consider switching to Python 3.x.')
984 return outtmpl_dict
985
21cd8fae 986 def get_output_path(self, dir_type='', filename=None):
987 paths = self.params.get('paths', {})
988 assert isinstance(paths, dict)
989 path = os.path.join(
990 expand_path(paths.get('home', '').strip()),
991 expand_path(paths.get(dir_type, '').strip()) if dir_type else '',
992 filename or '')
993
994 # Temporary fix for #4787
995 # 'Treat' all problem characters by passing filename through preferredencoding
996 # to workaround encoding issues with subprocess on python2 @ Windows
997 if sys.version_info < (3, 0) and sys.platform == 'win32':
998 path = encodeFilename(path, True).decode(preferredencoding())
999 return sanitize_path(path, force=self.params.get('windowsfilenames'))
1000
76a264ac 1001 @staticmethod
901130bb 1002 def _outtmpl_expandpath(outtmpl):
1003 # expand_path translates '%%' into '%' and '$$' into '$'
1004 # correspondingly that is not what we want since we need to keep
1005 # '%%' intact for template dict substitution step. Working around
1006 # with boundary-alike separator hack.
1007 sep = ''.join([random.choice(ascii_letters) for _ in range(32)])
1008 outtmpl = outtmpl.replace('%%', '%{0}%'.format(sep)).replace('$$', '${0}$'.format(sep))
1009
1010 # outtmpl should be expand_path'ed before template dict substitution
1011 # because meta fields may contain env variables we don't want to
1012 # be expanded. For example, for outtmpl "%(title)s.%(ext)s" and
1013 # title "Hello $PATH", we don't want `$PATH` to be expanded.
1014 return expand_path(outtmpl).replace(sep, '')
1015
1016 @staticmethod
1017 def escape_outtmpl(outtmpl):
1018 ''' Escape any remaining strings like %s, %abc% etc. '''
1019 return re.sub(
1020 STR_FORMAT_RE_TMPL.format('', '(?![%(\0])'),
1021 lambda mobj: ('' if mobj.group('has_key') else '%') + mobj.group(0),
1022 outtmpl)
1023
1024 @classmethod
1025 def validate_outtmpl(cls, outtmpl):
76a264ac 1026 ''' @return None or Exception object '''
7d1eb38a 1027 outtmpl = re.sub(
37893bb0 1028 STR_FORMAT_RE_TMPL.format('[^)]*', '[ljqBUDS]'),
7d1eb38a 1029 lambda mobj: f'{mobj.group(0)[:-1]}s',
1030 cls._outtmpl_expandpath(outtmpl))
76a264ac 1031 try:
7d1eb38a 1032 cls.escape_outtmpl(outtmpl) % collections.defaultdict(int)
76a264ac 1033 return None
1034 except ValueError as err:
1035 return err
1036
03b4de72 1037 @staticmethod
1038 def _copy_infodict(info_dict):
1039 info_dict = dict(info_dict)
09b49e1f 1040 info_dict.pop('__postprocessors', None)
03b4de72 1041 return info_dict
1042
e0fd9573 1043 def prepare_outtmpl(self, outtmpl, info_dict, sanitize=False):
1044 """ Make the outtmpl and info_dict suitable for substitution: ydl.escape_outtmpl(outtmpl) % info_dict
1045 @param sanitize Whether to sanitize the output as a filename.
1046 For backward compatibility, a function can also be passed
1047 """
1048
6e84b215 1049 info_dict.setdefault('epoch', int(time.time())) # keep epoch consistent once set
143db31d 1050
03b4de72 1051 info_dict = self._copy_infodict(info_dict)
752cda38 1052 info_dict['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
53c18592 1053 formatSeconds(info_dict['duration'], '-' if sanitize else ':')
143db31d 1054 if info_dict.get('duration', None) is not None
1055 else None)
752cda38 1056 info_dict['autonumber'] = self.params.get('autonumber_start', 1) - 1 + self._num_downloads
9c906919 1057 info_dict['video_autonumber'] = self._num_videos
752cda38 1058 if info_dict.get('resolution') is None:
1059 info_dict['resolution'] = self.format_resolution(info_dict, default=None)
143db31d 1060
e6f21b3d 1061 # For fields playlist_index, playlist_autonumber and autonumber convert all occurrences
143db31d 1062 # of %(field)s to %(field)0Nd for backward compatibility
1063 field_size_compat_map = {
ec11a9f4 1064 'playlist_index': number_of_digits(info_dict.get('_last_playlist_index') or 0),
1065 'playlist_autonumber': number_of_digits(info_dict.get('n_entries') or 0),
752cda38 1066 'autonumber': self.params.get('autonumber_size') or 5,
143db31d 1067 }
752cda38 1068
385a27fa 1069 TMPL_DICT = {}
37893bb0 1070 EXTERNAL_FORMAT_RE = re.compile(STR_FORMAT_RE_TMPL.format('[^)]*', f'[{STR_FORMAT_TYPES}ljqBUDS]'))
385a27fa 1071 MATH_FUNCTIONS = {
1072 '+': float.__add__,
1073 '-': float.__sub__,
1074 }
e625be0d 1075 # Field is of the form key1.key2...
1076 # where keys (except first) can be string, int or slice
2b8a2973 1077 FIELD_RE = r'\w*(?:\.(?:\w+|{num}|{num}?(?::{num}?){{1,2}}))*'.format(num=r'(?:-?\d+)')
e0fd9573 1078 MATH_FIELD_RE = r'''(?:{field}|{num})'''.format(field=FIELD_RE, num=r'-?\d+(?:.\d+)?')
385a27fa 1079 MATH_OPERATORS_RE = r'(?:%s)' % '|'.join(map(re.escape, MATH_FUNCTIONS.keys()))
e625be0d 1080 INTERNAL_FORMAT_RE = re.compile(r'''(?x)
1081 (?P<negate>-)?
385a27fa 1082 (?P<fields>{field})
1083 (?P<maths>(?:{math_op}{math_field})*)
e625be0d 1084 (?:>(?P<strf_format>.+?))?
e978789f
P
1085 (?P<alternate>(?<!\\),[^|&)]+)?
1086 (?:&(?P<replacement>.*?))?
e625be0d 1087 (?:\|(?P<default>.*?))?
385a27fa 1088 $'''.format(field=FIELD_RE, math_op=MATH_OPERATORS_RE, math_field=MATH_FIELD_RE))
752cda38 1089
2b8a2973 1090 def _traverse_infodict(k):
1091 k = k.split('.')
1092 if k[0] == '':
1093 k.pop(0)
1094 return traverse_obj(info_dict, k, is_user_input=True, traverse_string=True)
76a264ac 1095
752cda38 1096 def get_value(mdict):
1097 # Object traversal
2b8a2973 1098 value = _traverse_infodict(mdict['fields'])
752cda38 1099 # Negative
1100 if mdict['negate']:
1101 value = float_or_none(value)
1102 if value is not None:
1103 value *= -1
1104 # Do maths
385a27fa 1105 offset_key = mdict['maths']
1106 if offset_key:
752cda38 1107 value = float_or_none(value)
1108 operator = None
385a27fa 1109 while offset_key:
1110 item = re.match(
1111 MATH_FIELD_RE if operator else MATH_OPERATORS_RE,
1112 offset_key).group(0)
1113 offset_key = offset_key[len(item):]
1114 if operator is None:
752cda38 1115 operator = MATH_FUNCTIONS[item]
385a27fa 1116 continue
1117 item, multiplier = (item[1:], -1) if item[0] == '-' else (item, 1)
1118 offset = float_or_none(item)
1119 if offset is None:
2b8a2973 1120 offset = float_or_none(_traverse_infodict(item))
385a27fa 1121 try:
1122 value = operator(value, multiplier * offset)
1123 except (TypeError, ZeroDivisionError):
1124 return None
1125 operator = None
752cda38 1126 # Datetime formatting
1127 if mdict['strf_format']:
7c37ff97 1128 value = strftime_or_none(value, mdict['strf_format'].replace('\\,', ','))
752cda38 1129
1130 return value
1131
b868936c 1132 na = self.params.get('outtmpl_na_placeholder', 'NA')
1133
e0fd9573 1134 def filename_sanitizer(key, value, restricted=self.params.get('restrictfilenames')):
1135 return sanitize_filename(str(value), restricted=restricted,
1136 is_id=re.search(r'(^|[_.])id(\.|$)', key))
1137
1138 sanitizer = sanitize if callable(sanitize) else filename_sanitizer
1139 sanitize = bool(sanitize)
1140
6e84b215 1141 def _dumpjson_default(obj):
1142 if isinstance(obj, (set, LazyList)):
1143 return list(obj)
adbc4ec4 1144 return repr(obj)
6e84b215 1145
752cda38 1146 def create_key(outer_mobj):
1147 if not outer_mobj.group('has_key'):
b836dc94 1148 return outer_mobj.group(0)
752cda38 1149 key = outer_mobj.group('key')
752cda38 1150 mobj = re.match(INTERNAL_FORMAT_RE, key)
e0fd9573 1151 initial_field = mobj.group('fields') if mobj else ''
e978789f 1152 value, replacement, default = None, None, na
7c37ff97 1153 while mobj:
e625be0d 1154 mobj = mobj.groupdict()
7c37ff97 1155 default = mobj['default'] if mobj['default'] is not None else default
752cda38 1156 value = get_value(mobj)
e978789f 1157 replacement = mobj['replacement']
7c37ff97 1158 if value is None and mobj['alternate']:
1159 mobj = re.match(INTERNAL_FORMAT_RE, mobj['alternate'][1:])
1160 else:
1161 break
752cda38 1162
b868936c 1163 fmt = outer_mobj.group('format')
752cda38 1164 if fmt == 's' and value is not None and key in field_size_compat_map.keys():
1165 fmt = '0{:d}d'.format(field_size_compat_map[key])
1166
e978789f 1167 value = default if value is None else value if replacement is None else replacement
752cda38 1168
4476d2c7 1169 flags = outer_mobj.group('conversion') or ''
7d1eb38a 1170 str_fmt = f'{fmt[:-1]}s'
524e2e4f 1171 if fmt[-1] == 'l': # list
4476d2c7 1172 delim = '\n' if '#' in flags else ', '
9e907ebd 1173 value, fmt = delim.join(map(str, variadic(value, allowed_types=(str, bytes)))), str_fmt
524e2e4f 1174 elif fmt[-1] == 'j': # json
4476d2c7 1175 value, fmt = json.dumps(value, default=_dumpjson_default, indent=4 if '#' in flags else None), str_fmt
524e2e4f 1176 elif fmt[-1] == 'q': # quoted
4476d2c7 1177 value = map(str, variadic(value) if '#' in flags else [value])
1178 value, fmt = ' '.join(map(compat_shlex_quote, value)), str_fmt
524e2e4f 1179 elif fmt[-1] == 'B': # bytes
f5aa5cfb 1180 value = f'%{str_fmt}'.encode('utf-8') % str(value).encode('utf-8')
1181 value, fmt = value.decode('utf-8', 'ignore'), 's'
524e2e4f 1182 elif fmt[-1] == 'U': # unicode normalized
524e2e4f 1183 value, fmt = unicodedata.normalize(
1184 # "+" = compatibility equivalence, "#" = NFD
4476d2c7 1185 'NF%s%s' % ('K' if '+' in flags else '', 'D' if '#' in flags else 'C'),
524e2e4f 1186 value), str_fmt
e0fd9573 1187 elif fmt[-1] == 'D': # decimal suffix
abbeeebc 1188 num_fmt, fmt = fmt[:-1].replace('#', ''), 's'
1189 value = format_decimal_suffix(value, f'%{num_fmt}f%s' if num_fmt else '%d%s',
1190 factor=1024 if '#' in flags else 1000)
37893bb0 1191 elif fmt[-1] == 'S': # filename sanitization
e0fd9573 1192 value, fmt = filename_sanitizer(initial_field, value, restricted='#' in flags), str_fmt
7d1eb38a 1193 elif fmt[-1] == 'c':
524e2e4f 1194 if value:
1195 value = str(value)[0]
76a264ac 1196 else:
524e2e4f 1197 fmt = str_fmt
76a264ac 1198 elif fmt[-1] not in 'rs': # numeric
a439a3a4 1199 value = float_or_none(value)
752cda38 1200 if value is None:
1201 value, fmt = default, 's'
901130bb 1202
752cda38 1203 if sanitize:
1204 if fmt[-1] == 'r':
1205 # If value is an object, sanitize might convert it to a string
1206 # So we convert it to repr first
7d1eb38a 1207 value, fmt = repr(value), str_fmt
639f1cea 1208 if fmt[-1] in 'csr':
e0fd9573 1209 value = sanitizer(initial_field, value)
901130bb 1210
b868936c 1211 key = '%s\0%s' % (key.replace('%', '%\0'), outer_mobj.group('format'))
385a27fa 1212 TMPL_DICT[key] = value
b868936c 1213 return '{prefix}%({key}){fmt}'.format(key=key, fmt=fmt, prefix=outer_mobj.group('prefix'))
752cda38 1214
385a27fa 1215 return EXTERNAL_FORMAT_RE.sub(create_key, outtmpl), TMPL_DICT
143db31d 1216
819e0531 1217 def evaluate_outtmpl(self, outtmpl, info_dict, *args, **kwargs):
1218 outtmpl, info_dict = self.prepare_outtmpl(outtmpl, info_dict, *args, **kwargs)
1219 return self.escape_outtmpl(outtmpl) % info_dict
1220
de6000d9 1221 def _prepare_filename(self, info_dict, tmpl_type='default'):
8222d8de 1222 try:
b836dc94 1223 outtmpl = self._outtmpl_expandpath(self.outtmpl_dict.get(tmpl_type, self.outtmpl_dict['default']))
e0fd9573 1224 filename = self.evaluate_outtmpl(outtmpl, info_dict, True)
6a0546e3 1225 if not filename:
1226 return None
15da37c7 1227
6a0546e3 1228 if tmpl_type in ('default', 'temp'):
1229 final_ext, ext = self.params.get('final_ext'), info_dict.get('ext')
1230 if final_ext and ext and final_ext != ext and filename.endswith(f'.{final_ext}'):
1231 filename = replace_extension(filename, ext, final_ext)
1232 else:
1233 force_ext = OUTTMPL_TYPES[tmpl_type]
1234 if force_ext:
1235 filename = replace_extension(filename, force_ext, info_dict.get('ext'))
de6000d9 1236
bdc3fd2f
U
1237 # https://github.com/blackjack4494/youtube-dlc/issues/85
1238 trim_file_name = self.params.get('trim_file_name', False)
1239 if trim_file_name:
5c22c63d 1240 no_ext, *ext = filename.rsplit('.', 2)
1241 filename = join_nonempty(no_ext[:trim_file_name], *ext, delim='.')
bdc3fd2f 1242
0202b52a 1243 return filename
8222d8de 1244 except ValueError as err:
6febd1c1 1245 self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')')
8222d8de
JMF
1246 return None
1247
de6000d9 1248 def prepare_filename(self, info_dict, dir_type='', warn=False):
1249 """Generate the output filename."""
21cd8fae 1250
de6000d9 1251 filename = self._prepare_filename(info_dict, dir_type or 'default')
80c03fa9 1252 if not filename and dir_type not in ('', 'temp'):
1253 return ''
de6000d9 1254
c84aeac6 1255 if warn:
21cd8fae 1256 if not self.params.get('paths'):
de6000d9 1257 pass
1258 elif filename == '-':
c84aeac6 1259 self.report_warning('--paths is ignored when an outputting to stdout', only_once=True)
de6000d9 1260 elif os.path.isabs(filename):
c84aeac6 1261 self.report_warning('--paths is ignored since an absolute path is given in output template', only_once=True)
de6000d9 1262 if filename == '-' or not filename:
1263 return filename
1264
21cd8fae 1265 return self.get_output_path(dir_type, filename)
0202b52a 1266
120fe513 1267 def _match_entry(self, info_dict, incomplete=False, silent=False):
ecdec191 1268 """ Returns None if the file should be downloaded """
8222d8de 1269
c77495e3 1270 video_title = info_dict.get('title', info_dict.get('id', 'video'))
1271
8b0d7497 1272 def check_filter():
8b0d7497 1273 if 'title' in info_dict:
1274 # This can happen when we're just evaluating the playlist
1275 title = info_dict['title']
1276 matchtitle = self.params.get('matchtitle', False)
1277 if matchtitle:
1278 if not re.search(matchtitle, title, re.IGNORECASE):
1279 return '"' + title + '" title did not match pattern "' + matchtitle + '"'
1280 rejecttitle = self.params.get('rejecttitle', False)
1281 if rejecttitle:
1282 if re.search(rejecttitle, title, re.IGNORECASE):
1283 return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
1284 date = info_dict.get('upload_date')
1285 if date is not None:
1286 dateRange = self.params.get('daterange', DateRange())
1287 if date not in dateRange:
1288 return '%s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange)
1289 view_count = info_dict.get('view_count')
1290 if view_count is not None:
1291 min_views = self.params.get('min_views')
1292 if min_views is not None and view_count < min_views:
1293 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
1294 max_views = self.params.get('max_views')
1295 if max_views is not None and view_count > max_views:
1296 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
1297 if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')):
1298 return 'Skipping "%s" because it is age restricted' % video_title
8b0d7497 1299
8f18aca8 1300 match_filter = self.params.get('match_filter')
1301 if match_filter is not None:
1302 try:
1303 ret = match_filter(info_dict, incomplete=incomplete)
1304 except TypeError:
1305 # For backward compatibility
1306 ret = None if incomplete else match_filter(info_dict)
1307 if ret is not None:
1308 return ret
8b0d7497 1309 return None
1310
c77495e3 1311 if self.in_download_archive(info_dict):
1312 reason = '%s has already been recorded in the archive' % video_title
1313 break_opt, break_err = 'break_on_existing', ExistingVideoReached
1314 else:
1315 reason = check_filter()
1316 break_opt, break_err = 'break_on_reject', RejectedVideoReached
8b0d7497 1317 if reason is not None:
120fe513 1318 if not silent:
1319 self.to_screen('[download] ' + reason)
c77495e3 1320 if self.params.get(break_opt, False):
1321 raise break_err()
8b0d7497 1322 return reason
fe7e0c98 1323
b6c45014
JMF
1324 @staticmethod
1325 def add_extra_info(info_dict, extra_info):
1326 '''Set the keys from extra_info in info dict if they are missing'''
1327 for key, value in extra_info.items():
1328 info_dict.setdefault(key, value)
1329
409e1828 1330 def extract_info(self, url, download=True, ie_key=None, extra_info=None,
61aa5ba3 1331 process=True, force_generic_extractor=False):
41d1cca3 1332 """
1333 Return a list with a dictionary for each video extracted.
1334
1335 Arguments:
1336 url -- URL to extract
1337
1338 Keyword arguments:
1339 download -- whether to download videos during extraction
1340 ie_key -- extractor key hint
1341 extra_info -- dictionary containing the extra values to add to each result
1342 process -- whether to resolve all unresolved references (URLs, playlist items),
1343 must be True for download to work.
1344 force_generic_extractor -- force using the generic extractor
1345 """
fe7e0c98 1346
409e1828 1347 if extra_info is None:
1348 extra_info = {}
1349
61aa5ba3 1350 if not ie_key and force_generic_extractor:
d22dec74
S
1351 ie_key = 'Generic'
1352
8222d8de 1353 if ie_key:
8b7491c8 1354 ies = {ie_key: self._get_info_extractor_class(ie_key)}
8222d8de
JMF
1355 else:
1356 ies = self._ies
1357
8b7491c8 1358 for ie_key, ie in ies.items():
8222d8de
JMF
1359 if not ie.suitable(url):
1360 continue
1361
1362 if not ie.working():
6febd1c1
PH
1363 self.report_warning('The program functionality for this site has been marked as broken, '
1364 'and will probably not work.')
8222d8de 1365
1151c407 1366 temp_id = ie.get_temp_id(url)
a0566bbf 1367 if temp_id is not None and self.in_download_archive({'id': temp_id, 'ie_key': ie_key}):
5e5be0c0 1368 self.to_screen(f'[{ie_key}] {temp_id}: has already been recorded in the archive')
1369 if self.params.get('break_on_existing', False):
1370 raise ExistingVideoReached()
a0566bbf 1371 break
8b7491c8 1372 return self.__extract_info(url, self.get_info_extractor(ie_key), download, extra_info, process)
a0566bbf 1373 else:
1374 self.report_error('no suitable InfoExtractor for URL %s' % url)
1375
8e5fecc8 1376 def __handle_extraction_exceptions(func):
b5ae35ee 1377 @functools.wraps(func)
a0566bbf 1378 def wrapper(self, *args, **kwargs):
6da22e7d 1379 while True:
1380 try:
1381 return func(self, *args, **kwargs)
1382 except (DownloadCancelled, LazyList.IndexError, PagedList.IndexError):
8222d8de 1383 raise
6da22e7d 1384 except ReExtractInfo as e:
1385 if e.expected:
1386 self.to_screen(f'{e}; Re-extracting data')
1387 else:
1388 self.to_stderr('\r')
1389 self.report_warning(f'{e}; Re-extracting data')
1390 continue
1391 except GeoRestrictedError as e:
1392 msg = e.msg
1393 if e.countries:
1394 msg += '\nThis video is available in %s.' % ', '.join(
1395 map(ISO3166Utils.short2full, e.countries))
1396 msg += '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1397 self.report_error(msg)
1398 except ExtractorError as e: # An error we somewhat expected
1399 self.report_error(str(e), e.format_traceback())
1400 except Exception as e:
1401 if self.params.get('ignoreerrors'):
1402 self.report_error(str(e), tb=encode_compat_str(traceback.format_exc()))
1403 else:
1404 raise
1405 break
a0566bbf 1406 return wrapper
1407
f2ebc5c7 1408 def _wait_for_video(self, ie_result):
1409 if (not self.params.get('wait_for_video')
1410 or ie_result.get('_type', 'video') != 'video'
1411 or ie_result.get('formats') or ie_result.get('url')):
1412 return
1413
1414 format_dur = lambda dur: '%02d:%02d:%02d' % timetuple_from_msec(dur * 1000)[:-1]
1415 last_msg = ''
1416
1417 def progress(msg):
1418 nonlocal last_msg
1419 self.to_screen(msg + ' ' * (len(last_msg) - len(msg)) + '\r', skip_eol=True)
1420 last_msg = msg
1421
1422 min_wait, max_wait = self.params.get('wait_for_video')
1423 diff = try_get(ie_result, lambda x: x['release_timestamp'] - time.time())
1424 if diff is None and ie_result.get('live_status') == 'is_upcoming':
38d79fd1 1425 diff = random.randrange(min_wait, max_wait) if (max_wait and min_wait) else (max_wait or min_wait)
f2ebc5c7 1426 self.report_warning('Release time of video is not known')
1427 elif (diff or 0) <= 0:
1428 self.report_warning('Video should already be available according to extracted info')
38d79fd1 1429 diff = min(max(diff or 0, min_wait or 0), max_wait or float('inf'))
f2ebc5c7 1430 self.to_screen(f'[wait] Waiting for {format_dur(diff)} - Press Ctrl+C to try now')
1431
1432 wait_till = time.time() + diff
1433 try:
1434 while True:
1435 diff = wait_till - time.time()
1436 if diff <= 0:
1437 progress('')
1438 raise ReExtractInfo('[wait] Wait period ended', expected=True)
1439 progress(f'[wait] Remaining time until next attempt: {self._format_screen(format_dur(diff), self.Styles.EMPHASIS)}')
1440 time.sleep(1)
1441 except KeyboardInterrupt:
1442 progress('')
1443 raise ReExtractInfo('[wait] Interrupted by user', expected=True)
1444 except BaseException as e:
1445 if not isinstance(e, ReExtractInfo):
1446 self.to_screen('')
1447 raise
1448
a0566bbf 1449 @__handle_extraction_exceptions
58f197b7 1450 def __extract_info(self, url, ie, download, extra_info, process):
a0566bbf 1451 ie_result = ie.extract(url)
1452 if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
1453 return
1454 if isinstance(ie_result, list):
1455 # Backwards compatibility: old IE result format
1456 ie_result = {
1457 '_type': 'compat_list',
1458 'entries': ie_result,
1459 }
e37d0efb 1460 if extra_info.get('original_url'):
1461 ie_result.setdefault('original_url', extra_info['original_url'])
a0566bbf 1462 self.add_default_extra_info(ie_result, ie, url)
1463 if process:
f2ebc5c7 1464 self._wait_for_video(ie_result)
a0566bbf 1465 return self.process_ie_result(ie_result, download, extra_info)
8222d8de 1466 else:
a0566bbf 1467 return ie_result
fe7e0c98 1468
ea38e55f 1469 def add_default_extra_info(self, ie_result, ie, url):
6033d980 1470 if url is not None:
1471 self.add_extra_info(ie_result, {
1472 'webpage_url': url,
1473 'original_url': url,
57ebfca3 1474 })
1475 webpage_url = ie_result.get('webpage_url')
1476 if webpage_url:
1477 self.add_extra_info(ie_result, {
1478 'webpage_url_basename': url_basename(webpage_url),
1479 'webpage_url_domain': get_domain(webpage_url),
6033d980 1480 })
1481 if ie is not None:
1482 self.add_extra_info(ie_result, {
1483 'extractor': ie.IE_NAME,
1484 'extractor_key': ie.ie_key(),
1485 })
ea38e55f 1486
58adec46 1487 def process_ie_result(self, ie_result, download=True, extra_info=None):
8222d8de
JMF
1488 """
1489 Take the result of the ie(may be modified) and resolve all unresolved
1490 references (URLs, playlist items).
1491
1492 It will also download the videos if 'download'.
1493 Returns the resolved ie_result.
1494 """
58adec46 1495 if extra_info is None:
1496 extra_info = {}
e8ee972c
PH
1497 result_type = ie_result.get('_type', 'video')
1498
057a5206 1499 if result_type in ('url', 'url_transparent'):
134c6ea8 1500 ie_result['url'] = sanitize_url(ie_result['url'])
e37d0efb 1501 if ie_result.get('original_url'):
1502 extra_info.setdefault('original_url', ie_result['original_url'])
1503
057a5206 1504 extract_flat = self.params.get('extract_flat', False)
3089bc74
S
1505 if ((extract_flat == 'in_playlist' and 'playlist' in extra_info)
1506 or extract_flat is True):
ecb54191 1507 info_copy = ie_result.copy()
6033d980 1508 ie = try_get(ie_result.get('ie_key'), self.get_info_extractor)
360167b9 1509 if ie and not ie_result.get('id'):
4614bc22 1510 info_copy['id'] = ie.get_temp_id(ie_result['url'])
6033d980 1511 self.add_default_extra_info(info_copy, ie, ie_result['url'])
4614bc22 1512 self.add_extra_info(info_copy, extra_info)
b5475f11 1513 info_copy, _ = self.pre_process(info_copy)
ecb54191 1514 self.__forced_printings(info_copy, self.prepare_filename(info_copy), incomplete=True)
4614bc22 1515 if self.params.get('force_write_download_archive', False):
1516 self.record_download_archive(info_copy)
e8ee972c
PH
1517 return ie_result
1518
8222d8de 1519 if result_type == 'video':
b6c45014 1520 self.add_extra_info(ie_result, extra_info)
9c2b75b5 1521 ie_result = self.process_video_result(ie_result, download=download)
28b0eb0f 1522 additional_urls = (ie_result or {}).get('additional_urls')
9c2b75b5 1523 if additional_urls:
e9f4ccd1 1524 # TODO: Improve MetadataParserPP to allow setting a list
9c2b75b5 1525 if isinstance(additional_urls, compat_str):
1526 additional_urls = [additional_urls]
1527 self.to_screen(
1528 '[info] %s: %d additional URL(s) requested' % (ie_result['id'], len(additional_urls)))
1529 self.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls))
1530 ie_result['additional_entries'] = [
1531 self.extract_info(
b69fd25c 1532 url, download, extra_info=extra_info,
9c2b75b5 1533 force_generic_extractor=self.params.get('force_generic_extractor'))
1534 for url in additional_urls
1535 ]
1536 return ie_result
8222d8de
JMF
1537 elif result_type == 'url':
1538 # We have to add extra_info to the results because it may be
1539 # contained in a playlist
07cce701 1540 return self.extract_info(
1541 ie_result['url'], download,
1542 ie_key=ie_result.get('ie_key'),
1543 extra_info=extra_info)
7fc3fa05
PH
1544 elif result_type == 'url_transparent':
1545 # Use the information from the embedding page
1546 info = self.extract_info(
1547 ie_result['url'], ie_key=ie_result.get('ie_key'),
1548 extra_info=extra_info, download=False, process=False)
1549
1640eb09
S
1550 # extract_info may return None when ignoreerrors is enabled and
1551 # extraction failed with an error, don't crash and return early
1552 # in this case
1553 if not info:
1554 return info
1555
412c617d
PH
1556 force_properties = dict(
1557 (k, v) for k, v in ie_result.items() if v is not None)
0396806f 1558 for f in ('_type', 'url', 'id', 'extractor', 'extractor_key', 'ie_key'):
412c617d
PH
1559 if f in force_properties:
1560 del force_properties[f]
1561 new_result = info.copy()
1562 new_result.update(force_properties)
7fc3fa05 1563
0563f7ac
S
1564 # Extracted info may not be a video result (i.e.
1565 # info.get('_type', 'video') != video) but rather an url or
1566 # url_transparent. In such cases outer metadata (from ie_result)
1567 # should be propagated to inner one (info). For this to happen
1568 # _type of info should be overridden with url_transparent. This
067aa17e 1569 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
0563f7ac
S
1570 if new_result.get('_type') == 'url':
1571 new_result['_type'] = 'url_transparent'
7fc3fa05
PH
1572
1573 return self.process_ie_result(
1574 new_result, download=download, extra_info=extra_info)
40fcba5e 1575 elif result_type in ('playlist', 'multi_video'):
30a074c2 1576 # Protect from infinite recursion due to recursively nested playlists
1577 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
1578 webpage_url = ie_result['webpage_url']
1579 if webpage_url in self._playlist_urls:
7e85e872 1580 self.to_screen(
30a074c2 1581 '[download] Skipping already downloaded playlist: %s'
1582 % ie_result.get('title') or ie_result.get('id'))
1583 return
7e85e872 1584
30a074c2 1585 self._playlist_level += 1
1586 self._playlist_urls.add(webpage_url)
03f83004 1587 self._fill_common_fields(ie_result, False)
bc516a3f 1588 self._sanitize_thumbnails(ie_result)
30a074c2 1589 try:
1590 return self.__process_playlist(ie_result, download)
1591 finally:
1592 self._playlist_level -= 1
1593 if not self._playlist_level:
1594 self._playlist_urls.clear()
8222d8de 1595 elif result_type == 'compat_list':
c9bf4114
PH
1596 self.report_warning(
1597 'Extractor %s returned a compat_list result. '
1598 'It needs to be updated.' % ie_result.get('extractor'))
5f6a1245 1599
8222d8de 1600 def _fixup(r):
b868936c 1601 self.add_extra_info(r, {
1602 'extractor': ie_result['extractor'],
1603 'webpage_url': ie_result['webpage_url'],
1604 'webpage_url_basename': url_basename(ie_result['webpage_url']),
0bb322b9 1605 'webpage_url_domain': get_domain(ie_result['webpage_url']),
b868936c 1606 'extractor_key': ie_result['extractor_key'],
1607 })
8222d8de
JMF
1608 return r
1609 ie_result['entries'] = [
b6c45014 1610 self.process_ie_result(_fixup(r), download, extra_info)
8222d8de
JMF
1611 for r in ie_result['entries']
1612 ]
1613 return ie_result
1614 else:
1615 raise Exception('Invalid result type: %s' % result_type)
1616
e92caff5 1617 def _ensure_dir_exists(self, path):
1618 return make_dir(path, self.report_error)
1619
3b603dbd 1620 @staticmethod
1621 def _playlist_infodict(ie_result, **kwargs):
1622 return {
1623 **ie_result,
1624 'playlist': ie_result.get('title') or ie_result.get('id'),
1625 'playlist_id': ie_result.get('id'),
1626 'playlist_title': ie_result.get('title'),
1627 'playlist_uploader': ie_result.get('uploader'),
1628 'playlist_uploader_id': ie_result.get('uploader_id'),
1629 'playlist_index': 0,
1630 **kwargs,
1631 }
1632
30a074c2 1633 def __process_playlist(self, ie_result, download):
1634 # We process each entry in the playlist
1635 playlist = ie_result.get('title') or ie_result.get('id')
1636 self.to_screen('[download] Downloading playlist: %s' % playlist)
1637
498f5606 1638 if 'entries' not in ie_result:
aa9369a2 1639 raise EntryNotInPlaylist('There are no entries')
7c7f7161 1640
1641 MissingEntry = object()
498f5606 1642 incomplete_entries = bool(ie_result.get('requested_entries'))
1643 if incomplete_entries:
bf5f605e 1644 def fill_missing_entries(entries, indices):
7c7f7161 1645 ret = [MissingEntry] * max(indices)
bf5f605e 1646 for i, entry in zip(indices, entries):
498f5606 1647 ret[i - 1] = entry
1648 return ret
1649 ie_result['entries'] = fill_missing_entries(ie_result['entries'], ie_result['requested_entries'])
02fd60d3 1650
30a074c2 1651 playlist_results = []
1652
56a8fb4f 1653 playliststart = self.params.get('playliststart', 1)
30a074c2 1654 playlistend = self.params.get('playlistend')
1655 # For backwards compatibility, interpret -1 as whole list
1656 if playlistend == -1:
1657 playlistend = None
1658
1659 playlistitems_str = self.params.get('playlist_items')
1660 playlistitems = None
1661 if playlistitems_str is not None:
1662 def iter_playlistitems(format):
1663 for string_segment in format.split(','):
1664 if '-' in string_segment:
1665 start, end = string_segment.split('-')
1666 for item in range(int(start), int(end) + 1):
1667 yield int(item)
1668 else:
1669 yield int(string_segment)
1670 playlistitems = orderedSet(iter_playlistitems(playlistitems_str))
1671
1672 ie_entries = ie_result['entries']
8e5fecc8 1673 if isinstance(ie_entries, list):
ed8d87f9 1674 playlist_count = len(ie_entries)
f0d785d3 1675 msg = f'Collected {playlist_count} videos; downloading %d of them'
1676 ie_result['playlist_count'] = ie_result.get('playlist_count') or playlist_count
1677
8e5fecc8 1678 def get_entry(i):
1679 return ie_entries[i - 1]
1680 else:
f0d785d3 1681 msg = 'Downloading %d videos'
c586f9e8 1682 if not isinstance(ie_entries, (PagedList, LazyList)):
8e5fecc8 1683 ie_entries = LazyList(ie_entries)
d37707bd 1684 elif isinstance(ie_entries, InAdvancePagedList):
1685 if ie_entries._pagesize == 1:
1686 playlist_count = ie_entries._pagecount
8e5fecc8 1687
1688 def get_entry(i):
1689 return YoutubeDL.__handle_extraction_exceptions(
1690 lambda self, i: ie_entries[i - 1]
1691 )(self, i)
50fed816 1692
f0d785d3 1693 entries, broken = [], False
ff1c7fc9 1694 items = playlistitems if playlistitems is not None else itertools.count(playliststart)
1695 for i in items:
1696 if i == 0:
1697 continue
56a8fb4f 1698 if playlistitems is None and playlistend is not None and playlistend < i:
1699 break
1700 entry = None
1701 try:
50fed816 1702 entry = get_entry(i)
7c7f7161 1703 if entry is MissingEntry:
498f5606 1704 raise EntryNotInPlaylist()
56a8fb4f 1705 except (IndexError, EntryNotInPlaylist):
1706 if incomplete_entries:
aa9369a2 1707 raise EntryNotInPlaylist(f'Entry {i} cannot be found')
56a8fb4f 1708 elif not playlistitems:
1709 break
1710 entries.append(entry)
120fe513 1711 try:
1712 if entry is not None:
1713 self._match_entry(entry, incomplete=True, silent=True)
1714 except (ExistingVideoReached, RejectedVideoReached):
f0d785d3 1715 broken = True
120fe513 1716 break
56a8fb4f 1717 ie_result['entries'] = entries
30a074c2 1718
56a8fb4f 1719 # Save playlist_index before re-ordering
1720 entries = [
9e598870 1721 ((playlistitems[i - 1] if playlistitems else i + playliststart - 1), entry)
56a8fb4f 1722 for i, entry in enumerate(entries, 1)
1723 if entry is not None]
1724 n_entries = len(entries)
498f5606 1725
f0d785d3 1726 if not (ie_result.get('playlist_count') or broken or playlistitems or playlistend):
1727 ie_result['playlist_count'] = n_entries
1728
e08a85d8 1729 if not playlistitems and (playliststart != 1 or playlistend):
56a8fb4f 1730 playlistitems = list(range(playliststart, playliststart + n_entries))
498f5606 1731 ie_result['requested_entries'] = playlistitems
1732
e08a85d8 1733 _infojson_written = False
0bfc53d0 1734 write_playlist_files = self.params.get('allow_playlist_files', True)
1735 if write_playlist_files and self.params.get('list_thumbnails'):
1736 self.list_thumbnails(ie_result)
1737 if write_playlist_files and not self.params.get('simulate'):
3b603dbd 1738 ie_copy = self._playlist_infodict(ie_result, n_entries=n_entries)
e08a85d8 1739 _infojson_written = self._write_info_json(
1740 'playlist', ie_result, self.prepare_filename(ie_copy, 'pl_infojson'))
1741 if _infojson_written is None:
80c03fa9 1742 return
1743 if self._write_description('playlist', ie_result,
1744 self.prepare_filename(ie_copy, 'pl_description')) is None:
1745 return
681de68e 1746 # TODO: This should be passed to ThumbnailsConvertor if necessary
80c03fa9 1747 self._write_thumbnails('playlist', ie_copy, self.prepare_filename(ie_copy, 'pl_thumbnail'))
30a074c2 1748
1749 if self.params.get('playlistreverse', False):
1750 entries = entries[::-1]
30a074c2 1751 if self.params.get('playlistrandom', False):
1752 random.shuffle(entries)
1753
1754 x_forwarded_for = ie_result.get('__x_forwarded_for_ip')
1755
56a8fb4f 1756 self.to_screen('[%s] playlist %s: %s' % (ie_result['extractor'], playlist, msg % n_entries))
26e2805c 1757 failures = 0
1758 max_failures = self.params.get('skip_playlist_after_errors') or float('inf')
71729754 1759 for i, entry_tuple in enumerate(entries, 1):
1760 playlist_index, entry = entry_tuple
81139999 1761 if 'playlist-index' in self.params.get('compat_opts', []):
1762 playlist_index = playlistitems[i - 1] if playlistitems else i + playliststart - 1
30a074c2 1763 self.to_screen('[download] Downloading video %s of %s' % (i, n_entries))
1764 # This __x_forwarded_for_ip thing is a bit ugly but requires
1765 # minimal changes
1766 if x_forwarded_for:
1767 entry['__x_forwarded_for_ip'] = x_forwarded_for
1768 extra = {
1769 'n_entries': n_entries,
f59ae581 1770 '_last_playlist_index': max(playlistitems) if playlistitems else (playlistend or n_entries),
f0d785d3 1771 'playlist_count': ie_result.get('playlist_count'),
71729754 1772 'playlist_index': playlist_index,
1773 'playlist_autonumber': i,
30a074c2 1774 'playlist': playlist,
1775 'playlist_id': ie_result.get('id'),
1776 'playlist_title': ie_result.get('title'),
1777 'playlist_uploader': ie_result.get('uploader'),
1778 'playlist_uploader_id': ie_result.get('uploader_id'),
30a074c2 1779 'extractor': ie_result['extractor'],
1780 'webpage_url': ie_result['webpage_url'],
1781 'webpage_url_basename': url_basename(ie_result['webpage_url']),
0bb322b9 1782 'webpage_url_domain': get_domain(ie_result['webpage_url']),
30a074c2 1783 'extractor_key': ie_result['extractor_key'],
1784 }
1785
1786 if self._match_entry(entry, incomplete=True) is not None:
1787 continue
1788
1789 entry_result = self.__process_iterable_entry(entry, download, extra)
26e2805c 1790 if not entry_result:
1791 failures += 1
1792 if failures >= max_failures:
1793 self.report_error(
1794 'Skipping the remaining entries in playlist "%s" since %d items failed extraction' % (playlist, failures))
1795 break
30a074c2 1796 playlist_results.append(entry_result)
1797 ie_result['entries'] = playlist_results
e08a85d8 1798
1799 # Write the updated info to json
1800 if _infojson_written and self._write_info_json(
1801 'updated playlist', ie_result,
1802 self.prepare_filename(ie_copy, 'pl_infojson'), overwrite=True) is None:
1803 return
ca30f449 1804
ed5835b4 1805 ie_result = self.run_all_pps('playlist', ie_result)
1806 self.to_screen(f'[download] Finished downloading playlist: {playlist}')
30a074c2 1807 return ie_result
1808
a0566bbf 1809 @__handle_extraction_exceptions
1810 def __process_iterable_entry(self, entry, download, extra_info):
1811 return self.process_ie_result(
1812 entry, download=download, extra_info=extra_info)
1813
67134eab
JMF
1814 def _build_format_filter(self, filter_spec):
1815 " Returns a function to filter the formats according to the filter_spec "
083c9df9
PH
1816
1817 OPERATORS = {
1818 '<': operator.lt,
1819 '<=': operator.le,
1820 '>': operator.gt,
1821 '>=': operator.ge,
1822 '=': operator.eq,
1823 '!=': operator.ne,
1824 }
67134eab 1825 operator_rex = re.compile(r'''(?x)\s*
187986a8 1826 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)\s*
1827 (?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1828 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)\s*
083c9df9 1829 ''' % '|'.join(map(re.escape, OPERATORS.keys())))
187986a8 1830 m = operator_rex.fullmatch(filter_spec)
9ddb6925
S
1831 if m:
1832 try:
1833 comparison_value = int(m.group('value'))
1834 except ValueError:
1835 comparison_value = parse_filesize(m.group('value'))
1836 if comparison_value is None:
1837 comparison_value = parse_filesize(m.group('value') + 'B')
1838 if comparison_value is None:
1839 raise ValueError(
1840 'Invalid value %r in format specification %r' % (
67134eab 1841 m.group('value'), filter_spec))
9ddb6925
S
1842 op = OPERATORS[m.group('op')]
1843
083c9df9 1844 if not m:
9ddb6925
S
1845 STR_OPERATORS = {
1846 '=': operator.eq,
10d33b34
YCH
1847 '^=': lambda attr, value: attr.startswith(value),
1848 '$=': lambda attr, value: attr.endswith(value),
1849 '*=': lambda attr, value: value in attr,
1ce9a3cb 1850 '~=': lambda attr, value: value.search(attr) is not None
9ddb6925 1851 }
187986a8 1852 str_operator_rex = re.compile(r'''(?x)\s*
1853 (?P<key>[a-zA-Z0-9._-]+)\s*
1ce9a3cb
LF
1854 (?P<negation>!\s*)?(?P<op>%s)\s*(?P<none_inclusive>\?\s*)?
1855 (?P<quote>["'])?
1856 (?P<value>(?(quote)(?:(?!(?P=quote))[^\\]|\\.)+|[\w.-]+))
1857 (?(quote)(?P=quote))\s*
9ddb6925 1858 ''' % '|'.join(map(re.escape, STR_OPERATORS.keys())))
187986a8 1859 m = str_operator_rex.fullmatch(filter_spec)
9ddb6925 1860 if m:
1ce9a3cb
LF
1861 if m.group('op') == '~=':
1862 comparison_value = re.compile(m.group('value'))
1863 else:
1864 comparison_value = re.sub(r'''\\([\\"'])''', r'\1', m.group('value'))
2cc779f4
S
1865 str_op = STR_OPERATORS[m.group('op')]
1866 if m.group('negation'):
e118a879 1867 op = lambda attr, value: not str_op(attr, value)
2cc779f4
S
1868 else:
1869 op = str_op
083c9df9 1870
9ddb6925 1871 if not m:
187986a8 1872 raise SyntaxError('Invalid filter specification %r' % filter_spec)
083c9df9
PH
1873
1874 def _filter(f):
1875 actual_value = f.get(m.group('key'))
1876 if actual_value is None:
1877 return m.group('none_inclusive')
1878 return op(actual_value, comparison_value)
67134eab
JMF
1879 return _filter
1880
9f1a1c36 1881 def _check_formats(self, formats):
1882 for f in formats:
1883 self.to_screen('[info] Testing format %s' % f['format_id'])
75689fe5 1884 path = self.get_output_path('temp')
1885 if not self._ensure_dir_exists(f'{path}/'):
1886 continue
1887 temp_file = tempfile.NamedTemporaryFile(suffix='.tmp', delete=False, dir=path or None)
9f1a1c36 1888 temp_file.close()
1889 try:
1890 success, _ = self.dl(temp_file.name, f, test=True)
1891 except (DownloadError, IOError, OSError, ValueError) + network_exceptions:
1892 success = False
1893 finally:
1894 if os.path.exists(temp_file.name):
1895 try:
1896 os.remove(temp_file.name)
1897 except OSError:
1898 self.report_warning('Unable to delete temporary file "%s"' % temp_file.name)
1899 if success:
1900 yield f
1901 else:
1902 self.to_screen('[info] Unable to download format %s. Skipping...' % f['format_id'])
1903
0017d9ad 1904 def _default_format_spec(self, info_dict, download=True):
0017d9ad 1905
af0f7428
S
1906 def can_merge():
1907 merger = FFmpegMergerPP(self)
1908 return merger.available and merger.can_merge()
1909
91ebc640 1910 prefer_best = (
b7b04c78 1911 not self.params.get('simulate')
91ebc640 1912 and download
1913 and (
1914 not can_merge()
19807826 1915 or info_dict.get('is_live', False)
de6000d9 1916 or self.outtmpl_dict['default'] == '-'))
53ed7066 1917 compat = (
1918 prefer_best
1919 or self.params.get('allow_multiple_audio_streams', False)
1920 or 'format-spec' in self.params.get('compat_opts', []))
91ebc640 1921
1922 return (
53ed7066 1923 'best/bestvideo+bestaudio' if prefer_best
1924 else 'bestvideo*+bestaudio/best' if not compat
91ebc640 1925 else 'bestvideo+bestaudio/best')
0017d9ad 1926
67134eab
JMF
1927 def build_format_selector(self, format_spec):
1928 def syntax_error(note, start):
1929 message = (
1930 'Invalid format specification: '
1931 '{0}\n\t{1}\n\t{2}^'.format(note, format_spec, ' ' * start[1]))
1932 return SyntaxError(message)
1933
1934 PICKFIRST = 'PICKFIRST'
1935 MERGE = 'MERGE'
1936 SINGLE = 'SINGLE'
0130afb7 1937 GROUP = 'GROUP'
67134eab
JMF
1938 FormatSelector = collections.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
1939
91ebc640 1940 allow_multiple_streams = {'audio': self.params.get('allow_multiple_audio_streams', False),
1941 'video': self.params.get('allow_multiple_video_streams', False)}
909d24dd 1942
9f1a1c36 1943 check_formats = self.params.get('check_formats') == 'selected'
e8e73840 1944
67134eab
JMF
1945 def _parse_filter(tokens):
1946 filter_parts = []
1947 for type, string, start, _, _ in tokens:
1948 if type == tokenize.OP and string == ']':
1949 return ''.join(filter_parts)
1950 else:
1951 filter_parts.append(string)
1952
232541df 1953 def _remove_unused_ops(tokens):
17cc1534 1954 # Remove operators that we don't use and join them with the surrounding strings
232541df
JMF
1955 # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
1956 ALLOWED_OPS = ('/', '+', ',', '(', ')')
1957 last_string, last_start, last_end, last_line = None, None, None, None
1958 for type, string, start, end, line in tokens:
1959 if type == tokenize.OP and string == '[':
1960 if last_string:
1961 yield tokenize.NAME, last_string, last_start, last_end, last_line
1962 last_string = None
1963 yield type, string, start, end, line
1964 # everything inside brackets will be handled by _parse_filter
1965 for type, string, start, end, line in tokens:
1966 yield type, string, start, end, line
1967 if type == tokenize.OP and string == ']':
1968 break
1969 elif type == tokenize.OP and string in ALLOWED_OPS:
1970 if last_string:
1971 yield tokenize.NAME, last_string, last_start, last_end, last_line
1972 last_string = None
1973 yield type, string, start, end, line
1974 elif type in [tokenize.NAME, tokenize.NUMBER, tokenize.OP]:
1975 if not last_string:
1976 last_string = string
1977 last_start = start
1978 last_end = end
1979 else:
1980 last_string += string
1981 if last_string:
1982 yield tokenize.NAME, last_string, last_start, last_end, last_line
1983
cf2ac6df 1984 def _parse_format_selection(tokens, inside_merge=False, inside_choice=False, inside_group=False):
67134eab
JMF
1985 selectors = []
1986 current_selector = None
1987 for type, string, start, _, _ in tokens:
1988 # ENCODING is only defined in python 3.x
1989 if type == getattr(tokenize, 'ENCODING', None):
1990 continue
1991 elif type in [tokenize.NAME, tokenize.NUMBER]:
1992 current_selector = FormatSelector(SINGLE, string, [])
1993 elif type == tokenize.OP:
cf2ac6df
JMF
1994 if string == ')':
1995 if not inside_group:
1996 # ')' will be handled by the parentheses group
1997 tokens.restore_last_token()
67134eab 1998 break
cf2ac6df 1999 elif inside_merge and string in ['/', ',']:
0130afb7
JMF
2000 tokens.restore_last_token()
2001 break
cf2ac6df
JMF
2002 elif inside_choice and string == ',':
2003 tokens.restore_last_token()
2004 break
2005 elif string == ',':
0a31a350
JMF
2006 if not current_selector:
2007 raise syntax_error('"," must follow a format selector', start)
67134eab
JMF
2008 selectors.append(current_selector)
2009 current_selector = None
2010 elif string == '/':
d96d604e
JMF
2011 if not current_selector:
2012 raise syntax_error('"/" must follow a format selector', start)
67134eab 2013 first_choice = current_selector
cf2ac6df 2014 second_choice = _parse_format_selection(tokens, inside_choice=True)
f5f4a27a 2015 current_selector = FormatSelector(PICKFIRST, (first_choice, second_choice), [])
67134eab
JMF
2016 elif string == '[':
2017 if not current_selector:
2018 current_selector = FormatSelector(SINGLE, 'best', [])
2019 format_filter = _parse_filter(tokens)
2020 current_selector.filters.append(format_filter)
0130afb7
JMF
2021 elif string == '(':
2022 if current_selector:
2023 raise syntax_error('Unexpected "("', start)
cf2ac6df
JMF
2024 group = _parse_format_selection(tokens, inside_group=True)
2025 current_selector = FormatSelector(GROUP, group, [])
67134eab 2026 elif string == '+':
d03cfdce 2027 if not current_selector:
2028 raise syntax_error('Unexpected "+"', start)
2029 selector_1 = current_selector
2030 selector_2 = _parse_format_selection(tokens, inside_merge=True)
2031 if not selector_2:
2032 raise syntax_error('Expected a selector', start)
2033 current_selector = FormatSelector(MERGE, (selector_1, selector_2), [])
67134eab
JMF
2034 else:
2035 raise syntax_error('Operator not recognized: "{0}"'.format(string), start)
2036 elif type == tokenize.ENDMARKER:
2037 break
2038 if current_selector:
2039 selectors.append(current_selector)
2040 return selectors
2041
f8d4ad9a 2042 def _merge(formats_pair):
2043 format_1, format_2 = formats_pair
2044
2045 formats_info = []
2046 formats_info.extend(format_1.get('requested_formats', (format_1,)))
2047 formats_info.extend(format_2.get('requested_formats', (format_2,)))
2048
2049 if not allow_multiple_streams['video'] or not allow_multiple_streams['audio']:
551f9388 2050 get_no_more = {'video': False, 'audio': False}
f8d4ad9a 2051 for (i, fmt_info) in enumerate(formats_info):
551f9388 2052 if fmt_info.get('acodec') == fmt_info.get('vcodec') == 'none':
2053 formats_info.pop(i)
2054 continue
2055 for aud_vid in ['audio', 'video']:
f8d4ad9a 2056 if not allow_multiple_streams[aud_vid] and fmt_info.get(aud_vid[0] + 'codec') != 'none':
2057 if get_no_more[aud_vid]:
2058 formats_info.pop(i)
f5510afe 2059 break
f8d4ad9a 2060 get_no_more[aud_vid] = True
2061
2062 if len(formats_info) == 1:
2063 return formats_info[0]
2064
2065 video_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('vcodec') != 'none']
2066 audio_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('acodec') != 'none']
2067
2068 the_only_video = video_fmts[0] if len(video_fmts) == 1 else None
2069 the_only_audio = audio_fmts[0] if len(audio_fmts) == 1 else None
2070
2071 output_ext = self.params.get('merge_output_format')
2072 if not output_ext:
2073 if the_only_video:
2074 output_ext = the_only_video['ext']
2075 elif the_only_audio and not video_fmts:
2076 output_ext = the_only_audio['ext']
2077 else:
2078 output_ext = 'mkv'
2079
975a0d0d 2080 filtered = lambda *keys: filter(None, (traverse_obj(fmt, *keys) for fmt in formats_info))
2081
f8d4ad9a 2082 new_dict = {
2083 'requested_formats': formats_info,
975a0d0d 2084 'format': '+'.join(filtered('format')),
2085 'format_id': '+'.join(filtered('format_id')),
f8d4ad9a 2086 'ext': output_ext,
975a0d0d 2087 'protocol': '+'.join(map(determine_protocol, formats_info)),
093a1710 2088 'language': '+'.join(orderedSet(filtered('language'))) or None,
2089 'format_note': '+'.join(orderedSet(filtered('format_note'))) or None,
2090 'filesize_approx': sum(filtered('filesize', 'filesize_approx')) or None,
975a0d0d 2091 'tbr': sum(filtered('tbr', 'vbr', 'abr')),
f8d4ad9a 2092 }
2093
2094 if the_only_video:
2095 new_dict.update({
2096 'width': the_only_video.get('width'),
2097 'height': the_only_video.get('height'),
2098 'resolution': the_only_video.get('resolution') or self.format_resolution(the_only_video),
2099 'fps': the_only_video.get('fps'),
49a57e70 2100 'dynamic_range': the_only_video.get('dynamic_range'),
f8d4ad9a 2101 'vcodec': the_only_video.get('vcodec'),
2102 'vbr': the_only_video.get('vbr'),
2103 'stretched_ratio': the_only_video.get('stretched_ratio'),
2104 })
2105
2106 if the_only_audio:
2107 new_dict.update({
2108 'acodec': the_only_audio.get('acodec'),
2109 'abr': the_only_audio.get('abr'),
975a0d0d 2110 'asr': the_only_audio.get('asr'),
f8d4ad9a 2111 })
2112
2113 return new_dict
2114
e8e73840 2115 def _check_formats(formats):
981052c9 2116 if not check_formats:
2117 yield from formats
b5ac45b1 2118 return
9f1a1c36 2119 yield from self._check_formats(formats)
e8e73840 2120
67134eab 2121 def _build_selector_function(selector):
909d24dd 2122 if isinstance(selector, list): # ,
67134eab
JMF
2123 fs = [_build_selector_function(s) for s in selector]
2124
317f7ab6 2125 def selector_function(ctx):
67134eab 2126 for f in fs:
981052c9 2127 yield from f(ctx)
67134eab 2128 return selector_function
909d24dd 2129
2130 elif selector.type == GROUP: # ()
0130afb7 2131 selector_function = _build_selector_function(selector.selector)
909d24dd 2132
2133 elif selector.type == PICKFIRST: # /
67134eab
JMF
2134 fs = [_build_selector_function(s) for s in selector.selector]
2135
317f7ab6 2136 def selector_function(ctx):
67134eab 2137 for f in fs:
317f7ab6 2138 picked_formats = list(f(ctx))
67134eab
JMF
2139 if picked_formats:
2140 return picked_formats
2141 return []
67134eab 2142
981052c9 2143 elif selector.type == MERGE: # +
2144 selector_1, selector_2 = map(_build_selector_function, selector.selector)
2145
2146 def selector_function(ctx):
adbc4ec4 2147 for pair in itertools.product(selector_1(ctx), selector_2(ctx)):
981052c9 2148 yield _merge(pair)
2149
909d24dd 2150 elif selector.type == SINGLE: # atom
598d185d 2151 format_spec = selector.selector or 'best'
909d24dd 2152
f8d4ad9a 2153 # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
909d24dd 2154 if format_spec == 'all':
2155 def selector_function(ctx):
9222c381 2156 yield from _check_formats(ctx['formats'][::-1])
f8d4ad9a 2157 elif format_spec == 'mergeall':
2158 def selector_function(ctx):
dd2a987d 2159 formats = list(_check_formats(ctx['formats']))
e01d6aa4 2160 if not formats:
2161 return
921b76ca 2162 merged_format = formats[-1]
2163 for f in formats[-2::-1]:
f8d4ad9a 2164 merged_format = _merge((merged_format, f))
2165 yield merged_format
909d24dd 2166
2167 else:
e8e73840 2168 format_fallback, format_reverse, format_idx = False, True, 1
eff63539 2169 mobj = re.match(
2170 r'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
2171 format_spec)
2172 if mobj is not None:
2173 format_idx = int_or_none(mobj.group('n'), default=1)
e8e73840 2174 format_reverse = mobj.group('bw')[0] == 'b'
eff63539 2175 format_type = (mobj.group('type') or [None])[0]
2176 not_format_type = {'v': 'a', 'a': 'v'}.get(format_type)
2177 format_modified = mobj.group('mod') is not None
909d24dd 2178
2179 format_fallback = not format_type and not format_modified # for b, w
8326b00a 2180 _filter_f = (
eff63539 2181 (lambda f: f.get('%scodec' % format_type) != 'none')
2182 if format_type and format_modified # bv*, ba*, wv*, wa*
2183 else (lambda f: f.get('%scodec' % not_format_type) == 'none')
2184 if format_type # bv, ba, wv, wa
2185 else (lambda f: f.get('vcodec') != 'none' and f.get('acodec') != 'none')
2186 if not format_modified # b, w
8326b00a 2187 else lambda f: True) # b*, w*
2188 filter_f = lambda f: _filter_f(f) and (
2189 f.get('vcodec') != 'none' or f.get('acodec') != 'none')
67134eab 2190 else:
48ee10ee 2191 if format_spec in self._format_selection_exts['audio']:
b11c04a8 2192 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none'
48ee10ee 2193 elif format_spec in self._format_selection_exts['video']:
b11c04a8 2194 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none' and f.get('vcodec') != 'none'
48ee10ee 2195 elif format_spec in self._format_selection_exts['storyboards']:
b11c04a8 2196 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') == 'none' and f.get('vcodec') == 'none'
2197 else:
b5ae35ee 2198 filter_f = lambda f: f.get('format_id') == format_spec # id
909d24dd 2199
2200 def selector_function(ctx):
2201 formats = list(ctx['formats'])
909d24dd 2202 matches = list(filter(filter_f, formats)) if filter_f is not None else formats
e8e73840 2203 if format_fallback and ctx['incomplete_formats'] and not matches:
909d24dd 2204 # for extractors with incomplete formats (audio only (soundcloud)
2205 # or video only (imgur)) best/worst will fallback to
2206 # best/worst {video,audio}-only format
e8e73840 2207 matches = formats
981052c9 2208 matches = LazyList(_check_formats(matches[::-1 if format_reverse else 1]))
2209 try:
e8e73840 2210 yield matches[format_idx - 1]
981052c9 2211 except IndexError:
2212 return
083c9df9 2213
67134eab 2214 filters = [self._build_format_filter(f) for f in selector.filters]
083c9df9 2215
317f7ab6 2216 def final_selector(ctx):
adbc4ec4 2217 ctx_copy = dict(ctx)
67134eab 2218 for _filter in filters:
317f7ab6
S
2219 ctx_copy['formats'] = list(filter(_filter, ctx_copy['formats']))
2220 return selector_function(ctx_copy)
67134eab 2221 return final_selector
083c9df9 2222
67134eab 2223 stream = io.BytesIO(format_spec.encode('utf-8'))
0130afb7 2224 try:
232541df 2225 tokens = list(_remove_unused_ops(compat_tokenize_tokenize(stream.readline)))
0130afb7
JMF
2226 except tokenize.TokenError:
2227 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec)))
2228
2229 class TokenIterator(object):
2230 def __init__(self, tokens):
2231 self.tokens = tokens
2232 self.counter = 0
2233
2234 def __iter__(self):
2235 return self
2236
2237 def __next__(self):
2238 if self.counter >= len(self.tokens):
2239 raise StopIteration()
2240 value = self.tokens[self.counter]
2241 self.counter += 1
2242 return value
2243
2244 next = __next__
2245
2246 def restore_last_token(self):
2247 self.counter -= 1
2248
2249 parsed_selector = _parse_format_selection(iter(TokenIterator(tokens)))
67134eab 2250 return _build_selector_function(parsed_selector)
a9c58ad9 2251
e5660ee6
JMF
2252 def _calc_headers(self, info_dict):
2253 res = std_headers.copy()
455a15e2 2254 res.update(info_dict.get('http_headers') or {})
e5660ee6
JMF
2255
2256 cookies = self._calc_cookies(info_dict)
2257 if cookies:
2258 res['Cookie'] = cookies
2259
0016b84e
S
2260 if 'X-Forwarded-For' not in res:
2261 x_forwarded_for_ip = info_dict.get('__x_forwarded_for_ip')
2262 if x_forwarded_for_ip:
2263 res['X-Forwarded-For'] = x_forwarded_for_ip
2264
e5660ee6
JMF
2265 return res
2266
2267 def _calc_cookies(self, info_dict):
5c2266df 2268 pr = sanitized_Request(info_dict['url'])
e5660ee6 2269 self.cookiejar.add_cookie_header(pr)
662435f7 2270 return pr.get_header('Cookie')
e5660ee6 2271
9f1a1c36 2272 def _sort_thumbnails(self, thumbnails):
2273 thumbnails.sort(key=lambda t: (
2274 t.get('preference') if t.get('preference') is not None else -1,
2275 t.get('width') if t.get('width') is not None else -1,
2276 t.get('height') if t.get('height') is not None else -1,
2277 t.get('id') if t.get('id') is not None else '',
2278 t.get('url')))
2279
b0249bca 2280 def _sanitize_thumbnails(self, info_dict):
bc516a3f 2281 thumbnails = info_dict.get('thumbnails')
2282 if thumbnails is None:
2283 thumbnail = info_dict.get('thumbnail')
2284 if thumbnail:
2285 info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}]
9f1a1c36 2286 if not thumbnails:
2287 return
2288
2289 def check_thumbnails(thumbnails):
2290 for t in thumbnails:
2291 self.to_screen(f'[info] Testing thumbnail {t["id"]}')
2292 try:
2293 self.urlopen(HEADRequest(t['url']))
2294 except network_exceptions as err:
2295 self.to_screen(f'[info] Unable to connect to thumbnail {t["id"]} URL {t["url"]!r} - {err}. Skipping...')
2296 continue
2297 yield t
2298
2299 self._sort_thumbnails(thumbnails)
2300 for i, t in enumerate(thumbnails):
2301 if t.get('id') is None:
2302 t['id'] = '%d' % i
2303 if t.get('width') and t.get('height'):
2304 t['resolution'] = '%dx%d' % (t['width'], t['height'])
2305 t['url'] = sanitize_url(t['url'])
2306
2307 if self.params.get('check_formats') is True:
282f5709 2308 info_dict['thumbnails'] = LazyList(check_thumbnails(thumbnails[::-1]), reverse=True)
9f1a1c36 2309 else:
2310 info_dict['thumbnails'] = thumbnails
bc516a3f 2311
03f83004
LNO
2312 def _fill_common_fields(self, info_dict, is_video=True):
2313 # TODO: move sanitization here
2314 if is_video:
2315 # playlists are allowed to lack "title"
2316 info_dict['fulltitle'] = info_dict.get('title')
2317 if 'title' not in info_dict:
2318 raise ExtractorError('Missing "title" field in extractor result',
2319 video_id=info_dict['id'], ie=info_dict['extractor'])
2320 elif not info_dict.get('title'):
2321 self.report_warning('Extractor failed to obtain "title". Creating a generic title instead')
2322 info_dict['title'] = f'{info_dict["extractor"]} video #{info_dict["id"]}'
2323
2324 if info_dict.get('duration') is not None:
2325 info_dict['duration_string'] = formatSeconds(info_dict['duration'])
2326
2327 for ts_key, date_key in (
2328 ('timestamp', 'upload_date'),
2329 ('release_timestamp', 'release_date'),
2330 ('modified_timestamp', 'modified_date'),
2331 ):
2332 if info_dict.get(date_key) is None and info_dict.get(ts_key) is not None:
2333 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
2334 # see http://bugs.python.org/issue1646728)
2335 try:
2336 upload_date = datetime.datetime.utcfromtimestamp(info_dict[ts_key])
2337 info_dict[date_key] = upload_date.strftime('%Y%m%d')
2338 except (ValueError, OverflowError, OSError):
2339 pass
2340
2341 live_keys = ('is_live', 'was_live')
2342 live_status = info_dict.get('live_status')
2343 if live_status is None:
2344 for key in live_keys:
2345 if info_dict.get(key) is False:
2346 continue
2347 if info_dict.get(key):
2348 live_status = key
2349 break
2350 if all(info_dict.get(key) is False for key in live_keys):
2351 live_status = 'not_live'
2352 if live_status:
2353 info_dict['live_status'] = live_status
2354 for key in live_keys:
2355 if info_dict.get(key) is None:
2356 info_dict[key] = (live_status == key)
2357
2358 # Auto generate title fields corresponding to the *_number fields when missing
2359 # in order to always have clean titles. This is very common for TV series.
2360 for field in ('chapter', 'season', 'episode'):
2361 if info_dict.get('%s_number' % field) is not None and not info_dict.get(field):
2362 info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field])
2363
dd82ffea
JMF
2364 def process_video_result(self, info_dict, download=True):
2365 assert info_dict.get('_type', 'video') == 'video'
9c906919 2366 self._num_videos += 1
dd82ffea 2367
bec1fad2 2368 if 'id' not in info_dict:
fc08bdd6 2369 raise ExtractorError('Missing "id" field in extractor result', ie=info_dict['extractor'])
2370 elif not info_dict.get('id'):
2371 raise ExtractorError('Extractor failed to obtain "id"', ie=info_dict['extractor'])
455a15e2 2372
c9969434
S
2373 def report_force_conversion(field, field_not, conversion):
2374 self.report_warning(
2375 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
2376 % (field, field_not, conversion))
2377
2378 def sanitize_string_field(info, string_field):
2379 field = info.get(string_field)
2380 if field is None or isinstance(field, compat_str):
2381 return
2382 report_force_conversion(string_field, 'a string', 'string')
2383 info[string_field] = compat_str(field)
2384
2385 def sanitize_numeric_fields(info):
2386 for numeric_field in self._NUMERIC_FIELDS:
2387 field = info.get(numeric_field)
2388 if field is None or isinstance(field, compat_numeric_types):
2389 continue
2390 report_force_conversion(numeric_field, 'numeric', 'int')
2391 info[numeric_field] = int_or_none(field)
2392
2393 sanitize_string_field(info_dict, 'id')
2394 sanitize_numeric_fields(info_dict)
be6217b2 2395
dd82ffea
JMF
2396 if 'playlist' not in info_dict:
2397 # It isn't part of a playlist
2398 info_dict['playlist'] = None
2399 info_dict['playlist_index'] = None
2400
bc516a3f 2401 self._sanitize_thumbnails(info_dict)
d5519808 2402
536a55da 2403 thumbnail = info_dict.get('thumbnail')
bc516a3f 2404 thumbnails = info_dict.get('thumbnails')
536a55da
S
2405 if thumbnail:
2406 info_dict['thumbnail'] = sanitize_url(thumbnail)
2407 elif thumbnails:
d5519808
PH
2408 info_dict['thumbnail'] = thumbnails[-1]['url']
2409
ae30b840 2410 if info_dict.get('display_id') is None and 'id' in info_dict:
0afef30b
PH
2411 info_dict['display_id'] = info_dict['id']
2412
03f83004 2413 self._fill_common_fields(info_dict)
33d2fc2f 2414
05108a49
S
2415 for cc_kind in ('subtitles', 'automatic_captions'):
2416 cc = info_dict.get(cc_kind)
2417 if cc:
2418 for _, subtitle in cc.items():
2419 for subtitle_format in subtitle:
2420 if subtitle_format.get('url'):
2421 subtitle_format['url'] = sanitize_url(subtitle_format['url'])
2422 if subtitle_format.get('ext') is None:
2423 subtitle_format['ext'] = determine_ext(subtitle_format['url']).lower()
2424
2425 automatic_captions = info_dict.get('automatic_captions')
4bba3716 2426 subtitles = info_dict.get('subtitles')
4bba3716 2427
360e1ca5 2428 info_dict['requested_subtitles'] = self.process_subtitles(
05108a49 2429 info_dict['id'], subtitles, automatic_captions)
a504ced0 2430
dd82ffea
JMF
2431 if info_dict.get('formats') is None:
2432 # There's only one format available
2433 formats = [info_dict]
2434 else:
2435 formats = info_dict['formats']
2436
e0493e90 2437 info_dict['__has_drm'] = any(f.get('has_drm') for f in formats)
88acdbc2 2438 if not self.params.get('allow_unplayable_formats'):
2439 formats = [f for f in formats if not f.get('has_drm')]
88acdbc2 2440
adbc4ec4
THD
2441 if info_dict.get('is_live'):
2442 get_from_start = bool(self.params.get('live_from_start'))
2443 formats = [f for f in formats if bool(f.get('is_from_start')) == get_from_start]
4c922dd3 2444 if not get_from_start:
2445 info_dict['title'] += ' ' + datetime.datetime.now().strftime('%Y-%m-%d %H:%M')
adbc4ec4 2446
db95dc13 2447 if not formats:
1151c407 2448 self.raise_no_formats(info_dict)
db95dc13 2449
73af5cc8
S
2450 def is_wellformed(f):
2451 url = f.get('url')
a5ac0c47 2452 if not url:
73af5cc8
S
2453 self.report_warning(
2454 '"url" field is missing or empty - skipping format, '
2455 'there is an error in extractor')
a5ac0c47
S
2456 return False
2457 if isinstance(url, bytes):
2458 sanitize_string_field(f, 'url')
2459 return True
73af5cc8
S
2460
2461 # Filter out malformed formats for better extraction robustness
2462 formats = list(filter(is_wellformed, formats))
2463
181c7053
S
2464 formats_dict = {}
2465
dd82ffea 2466 # We check that all the formats have the format and format_id fields
db95dc13 2467 for i, format in enumerate(formats):
c9969434
S
2468 sanitize_string_field(format, 'format_id')
2469 sanitize_numeric_fields(format)
dcf77cf1 2470 format['url'] = sanitize_url(format['url'])
e74e3b63 2471 if not format.get('format_id'):
8016c922 2472 format['format_id'] = compat_str(i)
e2effb08
S
2473 else:
2474 # Sanitize format_id from characters used in format selector expression
ec85ded8 2475 format['format_id'] = re.sub(r'[\s,/+\[\]()]', '_', format['format_id'])
181c7053
S
2476 format_id = format['format_id']
2477 if format_id not in formats_dict:
2478 formats_dict[format_id] = []
2479 formats_dict[format_id].append(format)
2480
2481 # Make sure all formats have unique format_id
03b4de72 2482 common_exts = set(itertools.chain(*self._format_selection_exts.values()))
181c7053 2483 for format_id, ambiguous_formats in formats_dict.items():
48ee10ee 2484 ambigious_id = len(ambiguous_formats) > 1
2485 for i, format in enumerate(ambiguous_formats):
2486 if ambigious_id:
181c7053 2487 format['format_id'] = '%s-%d' % (format_id, i)
48ee10ee 2488 if format.get('ext') is None:
2489 format['ext'] = determine_ext(format['url']).lower()
2490 # Ensure there is no conflict between id and ext in format selection
2491 # See https://github.com/yt-dlp/yt-dlp/issues/1282
2492 if format['format_id'] != format['ext'] and format['format_id'] in common_exts:
2493 format['format_id'] = 'f%s' % format['format_id']
181c7053
S
2494
2495 for i, format in enumerate(formats):
8c51aa65 2496 if format.get('format') is None:
6febd1c1 2497 format['format'] = '{id} - {res}{note}'.format(
8c51aa65
JMF
2498 id=format['format_id'],
2499 res=self.format_resolution(format),
b868936c 2500 note=format_field(format, 'format_note', ' (%s)'),
8c51aa65 2501 )
6f0be937 2502 if format.get('protocol') is None:
b5559424 2503 format['protocol'] = determine_protocol(format)
239df021 2504 if format.get('resolution') is None:
2505 format['resolution'] = self.format_resolution(format, default=None)
176f1866 2506 if format.get('dynamic_range') is None and format.get('vcodec') != 'none':
2507 format['dynamic_range'] = 'SDR'
f2fe69c7 2508 if (info_dict.get('duration') and format.get('tbr')
2509 and not format.get('filesize') and not format.get('filesize_approx')):
2510 format['filesize_approx'] = info_dict['duration'] * format['tbr'] * (1024 / 8)
2511
e5660ee6
JMF
2512 # Add HTTP headers, so that external programs can use them from the
2513 # json output
2514 full_format_info = info_dict.copy()
2515 full_format_info.update(format)
2516 format['http_headers'] = self._calc_headers(full_format_info)
0016b84e
S
2517 # Remove private housekeeping stuff
2518 if '__x_forwarded_for_ip' in info_dict:
2519 del info_dict['__x_forwarded_for_ip']
dd82ffea 2520
9f1a1c36 2521 if self.params.get('check_formats') is True:
282f5709 2522 formats = LazyList(self._check_formats(formats[::-1]), reverse=True)
9f1a1c36 2523
88acdbc2 2524 if not formats or formats[0] is not info_dict:
b3d9ef88
JMF
2525 # only set the 'formats' fields if the original info_dict list them
2526 # otherwise we end up with a circular reference, the first (and unique)
f89197d7 2527 # element in the 'formats' field in info_dict is info_dict itself,
dfb1b146 2528 # which can't be exported to json
b3d9ef88 2529 info_dict['formats'] = formats
4ec82a72 2530
2531 info_dict, _ = self.pre_process(info_dict)
2532
09b49e1f 2533 if self._match_entry(info_dict) is not None:
2534 return info_dict
2535
2536 self.post_extract(info_dict)
2537 info_dict, _ = self.pre_process(info_dict, 'after_filter')
2538
093a1710 2539 # The pre-processors may have modified the formats
2540 formats = info_dict.get('formats', [info_dict])
2541
fa9f30b8 2542 list_only = self.params.get('simulate') is None and (
2543 self.params.get('list_thumbnails') or self.params.get('listformats') or self.params.get('listsubtitles'))
2544 interactive_format_selection = not list_only and self.format_selector == '-'
b7b04c78 2545 if self.params.get('list_thumbnails'):
2546 self.list_thumbnails(info_dict)
b7b04c78 2547 if self.params.get('listsubtitles'):
2548 if 'automatic_captions' in info_dict:
2549 self.list_subtitles(
2550 info_dict['id'], automatic_captions, 'automatic captions')
2551 self.list_subtitles(info_dict['id'], subtitles, 'subtitles')
fa9f30b8 2552 if self.params.get('listformats') or interactive_format_selection:
b69fd25c 2553 self.list_formats(info_dict)
169dbde9 2554 if list_only:
b7b04c78 2555 # Without this printing, -F --print-json will not work
169dbde9 2556 self.__forced_printings(info_dict, self.prepare_filename(info_dict), incomplete=True)
bfaae0a7 2557 return
2558
187986a8 2559 format_selector = self.format_selector
2560 if format_selector is None:
0017d9ad 2561 req_format = self._default_format_spec(info_dict, download=download)
0760b0a7 2562 self.write_debug('Default format spec: %s' % req_format)
187986a8 2563 format_selector = self.build_format_selector(req_format)
317f7ab6 2564
fa9f30b8 2565 while True:
2566 if interactive_format_selection:
2567 req_format = input(
2568 self._format_screen('\nEnter format selector: ', self.Styles.EMPHASIS))
2569 try:
2570 format_selector = self.build_format_selector(req_format)
2571 except SyntaxError as err:
2572 self.report_error(err, tb=False, is_error=False)
2573 continue
2574
2575 # While in format selection we may need to have an access to the original
2576 # format set in order to calculate some metrics or do some processing.
2577 # For now we need to be able to guess whether original formats provided
2578 # by extractor are incomplete or not (i.e. whether extractor provides only
2579 # video-only or audio-only formats) for proper formats selection for
2580 # extractors with such incomplete formats (see
2581 # https://github.com/ytdl-org/youtube-dl/pull/5556).
2582 # Since formats may be filtered during format selection and may not match
2583 # the original formats the results may be incorrect. Thus original formats
2584 # or pre-calculated metrics should be passed to format selection routines
2585 # as well.
2586 # We will pass a context object containing all necessary additional data
2587 # instead of just formats.
2588 # This fixes incorrect format selection issue (see
2589 # https://github.com/ytdl-org/youtube-dl/issues/10083).
2590 incomplete_formats = (
2591 # All formats are video-only or
2592 all(f.get('vcodec') != 'none' and f.get('acodec') == 'none' for f in formats)
2593 # all formats are audio-only
2594 or all(f.get('vcodec') == 'none' and f.get('acodec') != 'none' for f in formats))
2595
2596 ctx = {
2597 'formats': formats,
2598 'incomplete_formats': incomplete_formats,
2599 }
2600
2601 formats_to_download = list(format_selector(ctx))
2602 if interactive_format_selection and not formats_to_download:
2603 self.report_error('Requested format is not available', tb=False, is_error=False)
2604 continue
2605 break
317f7ab6 2606
dd82ffea 2607 if not formats_to_download:
b7da73eb 2608 if not self.params.get('ignore_no_formats_error'):
1151c407 2609 raise ExtractorError('Requested format is not available', expected=True,
2610 video_id=info_dict['id'], ie=info_dict['extractor'])
b62fa6d7 2611 self.report_warning('Requested format is not available')
2612 # Process what we can, even without any available formats.
2613 formats_to_download = [{}]
a13e6848 2614
b62fa6d7 2615 best_format = formats_to_download[-1]
2616 if download:
2617 if best_format:
2618 self.to_screen(
2619 f'[info] {info_dict["id"]}: Downloading {len(formats_to_download)} format(s): '
2620 + ', '.join([f['format_id'] for f in formats_to_download]))
a13e6848 2621 max_downloads_reached = False
f46e2f9d 2622 for i, fmt in enumerate(formats_to_download):
09b49e1f 2623 formats_to_download[i] = new_info = self._copy_infodict(info_dict)
b7da73eb 2624 new_info.update(fmt)
a13e6848 2625 try:
2626 self.process_info(new_info)
2627 except MaxDownloadsReached:
2628 max_downloads_reached = True
f46e2f9d 2629 # Remove copied info
2630 for key, val in tuple(new_info.items()):
2631 if info_dict.get(key) == val:
2632 new_info.pop(key)
a13e6848 2633 if max_downloads_reached:
2634 break
ebed8b37 2635
9e907ebd 2636 write_archive = set(f.get('__write_download_archive', False) for f in formats_to_download)
a13e6848 2637 assert write_archive.issubset({True, False, 'ignore'})
2638 if True in write_archive and False not in write_archive:
2639 self.record_download_archive(info_dict)
be72c624 2640
2641 info_dict['requested_downloads'] = formats_to_download
ed5835b4 2642 info_dict = self.run_all_pps('after_video', info_dict)
a13e6848 2643 if max_downloads_reached:
2644 raise MaxDownloadsReached()
ebed8b37 2645
49a57e70 2646 # We update the info dict with the selected best quality format (backwards compatibility)
be72c624 2647 info_dict.update(best_format)
dd82ffea
JMF
2648 return info_dict
2649
98c70d6f 2650 def process_subtitles(self, video_id, normal_subtitles, automatic_captions):
a504ced0 2651 """Select the requested subtitles and their format"""
98c70d6f
JMF
2652 available_subs = {}
2653 if normal_subtitles and self.params.get('writesubtitles'):
2654 available_subs.update(normal_subtitles)
2655 if automatic_captions and self.params.get('writeautomaticsub'):
2656 for lang, cap_info in automatic_captions.items():
360e1ca5
JMF
2657 if lang not in available_subs:
2658 available_subs[lang] = cap_info
2659
4d171848
JMF
2660 if (not self.params.get('writesubtitles') and not
2661 self.params.get('writeautomaticsub') or not
2662 available_subs):
2663 return None
a504ced0 2664
c32b0aab 2665 all_sub_langs = available_subs.keys()
a504ced0 2666 if self.params.get('allsubtitles', False):
c32b0aab 2667 requested_langs = all_sub_langs
2668 elif self.params.get('subtitleslangs', False):
77c4a9ef 2669 # A list is used so that the order of languages will be the same as
2670 # given in subtitleslangs. See https://github.com/yt-dlp/yt-dlp/issues/1041
2671 requested_langs = []
2672 for lang_re in self.params.get('subtitleslangs'):
77c4a9ef 2673 discard = lang_re[0] == '-'
c32b0aab 2674 if discard:
77c4a9ef 2675 lang_re = lang_re[1:]
3aa91540 2676 if lang_re == 'all':
2677 if discard:
2678 requested_langs = []
2679 else:
2680 requested_langs.extend(all_sub_langs)
2681 continue
77c4a9ef 2682 current_langs = filter(re.compile(lang_re + '$').match, all_sub_langs)
c32b0aab 2683 if discard:
2684 for lang in current_langs:
77c4a9ef 2685 while lang in requested_langs:
2686 requested_langs.remove(lang)
c32b0aab 2687 else:
77c4a9ef 2688 requested_langs.extend(current_langs)
2689 requested_langs = orderedSet(requested_langs)
c32b0aab 2690 elif 'en' in available_subs:
2691 requested_langs = ['en']
a504ced0 2692 else:
c32b0aab 2693 requested_langs = [list(all_sub_langs)[0]]
ad3dc496 2694 if requested_langs:
2695 self.write_debug('Downloading subtitles: %s' % ', '.join(requested_langs))
a504ced0
JMF
2696
2697 formats_query = self.params.get('subtitlesformat', 'best')
2698 formats_preference = formats_query.split('/') if formats_query else []
2699 subs = {}
2700 for lang in requested_langs:
2701 formats = available_subs.get(lang)
2702 if formats is None:
2703 self.report_warning('%s subtitles not available for %s' % (lang, video_id))
2704 continue
a504ced0
JMF
2705 for ext in formats_preference:
2706 if ext == 'best':
2707 f = formats[-1]
2708 break
2709 matches = list(filter(lambda f: f['ext'] == ext, formats))
2710 if matches:
2711 f = matches[-1]
2712 break
2713 else:
2714 f = formats[-1]
2715 self.report_warning(
2716 'No subtitle format found matching "%s" for language %s, '
2717 'using %s' % (formats_query, lang, f['ext']))
2718 subs[lang] = f
2719 return subs
2720
bb66c247 2721 def _forceprint(self, key, info_dict):
2722 if info_dict is None:
2723 return
2724 info_copy = info_dict.copy()
2725 info_copy['formats_table'] = self.render_formats_table(info_dict)
2726 info_copy['thumbnails_table'] = self.render_thumbnails_table(info_dict)
2727 info_copy['subtitles_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('subtitles'))
2728 info_copy['automatic_captions_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('automatic_captions'))
2729
2730 def format_tmpl(tmpl):
2731 mobj = re.match(r'\w+(=?)$', tmpl)
2732 if mobj and mobj.group(1):
2733 return f'{tmpl[:-1]} = %({tmpl[:-1]})r'
2734 elif mobj:
2735 return f'%({tmpl})s'
2736 return tmpl
8130779d 2737
bb66c247 2738 for tmpl in self.params['forceprint'].get(key, []):
2739 self.to_stdout(self.evaluate_outtmpl(format_tmpl(tmpl), info_copy))
2740
2741 for tmpl, file_tmpl in self.params['print_to_file'].get(key, []):
2742 filename = self.evaluate_outtmpl(file_tmpl, info_dict)
2743 tmpl = format_tmpl(tmpl)
2744 self.to_screen(f'[info] Writing {tmpl!r} to: {filename}')
8d93e69d 2745 if self._ensure_dir_exists(filename):
2746 with io.open(filename, 'a', encoding='utf-8') as f:
2747 f.write(self.evaluate_outtmpl(tmpl, info_copy) + '\n')
ca30f449 2748
d06daf23 2749 def __forced_printings(self, info_dict, filename, incomplete):
53c18592 2750 def print_mandatory(field, actual_field=None):
2751 if actual_field is None:
2752 actual_field = field
d06daf23 2753 if (self.params.get('force%s' % field, False)
53c18592 2754 and (not incomplete or info_dict.get(actual_field) is not None)):
2755 self.to_stdout(info_dict[actual_field])
d06daf23
S
2756
2757 def print_optional(field):
2758 if (self.params.get('force%s' % field, False)
2759 and info_dict.get(field) is not None):
2760 self.to_stdout(info_dict[field])
2761
53c18592 2762 info_dict = info_dict.copy()
2763 if filename is not None:
2764 info_dict['filename'] = filename
2765 if info_dict.get('requested_formats') is not None:
2766 # For RTMP URLs, also include the playpath
2767 info_dict['urls'] = '\n'.join(f['url'] + f.get('play_path', '') for f in info_dict['requested_formats'])
2768 elif 'url' in info_dict:
2769 info_dict['urls'] = info_dict['url'] + info_dict.get('play_path', '')
2770
bb66c247 2771 if (self.params.get('forcejson')
2772 or self.params['forceprint'].get('video')
2773 or self.params['print_to_file'].get('video')):
2b8a2973 2774 self.post_extract(info_dict)
bb66c247 2775 self._forceprint('video', info_dict)
53c18592 2776
d06daf23
S
2777 print_mandatory('title')
2778 print_mandatory('id')
53c18592 2779 print_mandatory('url', 'urls')
d06daf23
S
2780 print_optional('thumbnail')
2781 print_optional('description')
53c18592 2782 print_optional('filename')
b868936c 2783 if self.params.get('forceduration') and info_dict.get('duration') is not None:
d06daf23
S
2784 self.to_stdout(formatSeconds(info_dict['duration']))
2785 print_mandatory('format')
53c18592 2786
2b8a2973 2787 if self.params.get('forcejson'):
6e84b215 2788 self.to_stdout(json.dumps(self.sanitize_info(info_dict)))
d06daf23 2789
e8e73840 2790 def dl(self, name, info, subtitle=False, test=False):
88acdbc2 2791 if not info.get('url'):
1151c407 2792 self.raise_no_formats(info, True)
e8e73840 2793
2794 if test:
2795 verbose = self.params.get('verbose')
2796 params = {
2797 'test': True,
a169858f 2798 'quiet': self.params.get('quiet') or not verbose,
e8e73840 2799 'verbose': verbose,
2800 'noprogress': not verbose,
2801 'nopart': True,
2802 'skip_unavailable_fragments': False,
2803 'keep_fragments': False,
2804 'overwrites': True,
2805 '_no_ytdl_file': True,
2806 }
2807 else:
2808 params = self.params
96fccc10 2809 fd = get_suitable_downloader(info, params, to_stdout=(name == '-'))(self, params)
e8e73840 2810 if not test:
2811 for ph in self._progress_hooks:
2812 fd.add_progress_hook(ph)
42676437
M
2813 urls = '", "'.join(
2814 (f['url'].split(',')[0] + ',<data>' if f['url'].startswith('data:') else f['url'])
2815 for f in info.get('requested_formats', []) or [info])
18e674b4 2816 self.write_debug('Invoking downloader on "%s"' % urls)
03b4de72 2817
adbc4ec4
THD
2818 # Note: Ideally info should be a deep-copied so that hooks cannot modify it.
2819 # But it may contain objects that are not deep-copyable
2820 new_info = self._copy_infodict(info)
e8e73840 2821 if new_info.get('http_headers') is None:
2822 new_info['http_headers'] = self._calc_headers(new_info)
2823 return fd.download(name, new_info, subtitle)
2824
e04938ab 2825 def existing_file(self, filepaths, *, default_overwrite=True):
2826 existing_files = list(filter(os.path.exists, orderedSet(filepaths)))
2827 if existing_files and not self.params.get('overwrites', default_overwrite):
2828 return existing_files[0]
2829
2830 for file in existing_files:
2831 self.report_file_delete(file)
2832 os.remove(file)
2833 return None
2834
8222d8de 2835 def process_info(self, info_dict):
09b49e1f 2836 """Process a single resolved IE result. (Modifies it in-place)"""
8222d8de
JMF
2837
2838 assert info_dict.get('_type', 'video') == 'video'
f46e2f9d 2839 original_infodict = info_dict
fd288278 2840
4513a41a 2841 if 'format' not in info_dict and 'ext' in info_dict:
8222d8de
JMF
2842 info_dict['format'] = info_dict['ext']
2843
09b49e1f 2844 # This is mostly just for backward compatibility of process_info
2845 # As a side-effect, this allows for format-specific filters
c77495e3 2846 if self._match_entry(info_dict) is not None:
9e907ebd 2847 info_dict['__write_download_archive'] = 'ignore'
8222d8de
JMF
2848 return
2849
09b49e1f 2850 # Does nothing under normal operation - for backward compatibility of process_info
277d6ff5 2851 self.post_extract(info_dict)
8222d8de 2852
dcf64d43 2853 # info_dict['_filename'] needs to be set for backward compatibility
de6000d9 2854 info_dict['_filename'] = full_filename = self.prepare_filename(info_dict, warn=True)
2855 temp_filename = self.prepare_filename(info_dict, 'temp')
0202b52a 2856 files_to_move = {}
8222d8de 2857
09b49e1f 2858 self._num_downloads += 1
2859
8222d8de 2860 # Forced printings
4513a41a 2861 self.__forced_printings(info_dict, full_filename, incomplete=('format' not in info_dict))
8222d8de 2862
b7b04c78 2863 if self.params.get('simulate'):
9e907ebd 2864 info_dict['__write_download_archive'] = self.params.get('force_write_download_archive')
8222d8de
JMF
2865 return
2866
de6000d9 2867 if full_filename is None:
8222d8de 2868 return
e92caff5 2869 if not self._ensure_dir_exists(encodeFilename(full_filename)):
0202b52a 2870 return
e92caff5 2871 if not self._ensure_dir_exists(encodeFilename(temp_filename)):
8222d8de
JMF
2872 return
2873
80c03fa9 2874 if self._write_description('video', info_dict,
2875 self.prepare_filename(info_dict, 'description')) is None:
2876 return
2877
2878 sub_files = self._write_subtitles(info_dict, temp_filename)
2879 if sub_files is None:
2880 return
2881 files_to_move.update(dict(sub_files))
2882
2883 thumb_files = self._write_thumbnails(
2884 'video', info_dict, temp_filename, self.prepare_filename(info_dict, 'thumbnail'))
2885 if thumb_files is None:
2886 return
2887 files_to_move.update(dict(thumb_files))
8222d8de 2888
80c03fa9 2889 infofn = self.prepare_filename(info_dict, 'infojson')
2890 _infojson_written = self._write_info_json('video', info_dict, infofn)
2891 if _infojson_written:
dac5df5a 2892 info_dict['infojson_filename'] = infofn
e75bb0d6 2893 # For backward compatibility, even though it was a private field
80c03fa9 2894 info_dict['__infojson_filename'] = infofn
2895 elif _infojson_written is None:
2896 return
2897
2898 # Note: Annotations are deprecated
2899 annofn = None
1fb07d10 2900 if self.params.get('writeannotations', False):
de6000d9 2901 annofn = self.prepare_filename(info_dict, 'annotation')
80c03fa9 2902 if annofn:
e92caff5 2903 if not self._ensure_dir_exists(encodeFilename(annofn)):
0202b52a 2904 return
0c3d0f51 2905 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(annofn)):
6febd1c1 2906 self.to_screen('[info] Video annotations are already present')
ffddb112
RA
2907 elif not info_dict.get('annotations'):
2908 self.report_warning('There are no annotations to write.')
7b6fefc9
PH
2909 else:
2910 try:
6febd1c1 2911 self.to_screen('[info] Writing video annotations to: ' + annofn)
7b6fefc9
PH
2912 with io.open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
2913 annofile.write(info_dict['annotations'])
2914 except (KeyError, TypeError):
6febd1c1 2915 self.report_warning('There are no annotations to write.')
7b6fefc9 2916 except (OSError, IOError):
6febd1c1 2917 self.report_error('Cannot write annotations file: ' + annofn)
7b6fefc9 2918 return
1fb07d10 2919
732044af 2920 # Write internet shortcut files
08438d2c 2921 def _write_link_file(link_type):
60f3e995 2922 url = try_get(info_dict['webpage_url'], iri_to_uri)
2923 if not url:
2924 self.report_warning(
2925 f'Cannot write internet shortcut file because the actual URL of "{info_dict["webpage_url"]}" is unknown')
2926 return True
08438d2c 2927 linkfn = replace_extension(self.prepare_filename(info_dict, 'link'), link_type, info_dict.get('ext'))
0e6b018a
Z
2928 if not self._ensure_dir_exists(encodeFilename(linkfn)):
2929 return False
10e3742e 2930 if self.params.get('overwrites', True) and os.path.exists(encodeFilename(linkfn)):
08438d2c 2931 self.to_screen(f'[info] Internet shortcut (.{link_type}) is already present')
2932 return True
2933 try:
2934 self.to_screen(f'[info] Writing internet shortcut (.{link_type}) to: {linkfn}')
2935 with io.open(encodeFilename(to_high_limit_path(linkfn)), 'w', encoding='utf-8',
2936 newline='\r\n' if link_type == 'url' else '\n') as linkfile:
60f3e995 2937 template_vars = {'url': url}
08438d2c 2938 if link_type == 'desktop':
2939 template_vars['filename'] = linkfn[:-(len(link_type) + 1)]
2940 linkfile.write(LINK_TEMPLATES[link_type] % template_vars)
2941 except (OSError, IOError):
2942 self.report_error(f'Cannot write internet shortcut {linkfn}')
2943 return False
732044af 2944 return True
2945
08438d2c 2946 write_links = {
2947 'url': self.params.get('writeurllink'),
2948 'webloc': self.params.get('writewebloclink'),
2949 'desktop': self.params.get('writedesktoplink'),
2950 }
2951 if self.params.get('writelink'):
2952 link_type = ('webloc' if sys.platform == 'darwin'
2953 else 'desktop' if sys.platform.startswith('linux')
2954 else 'url')
2955 write_links[link_type] = True
2956
2957 if any(should_write and not _write_link_file(link_type)
2958 for link_type, should_write in write_links.items()):
2959 return
732044af 2960
f46e2f9d 2961 def replace_info_dict(new_info):
2962 nonlocal info_dict
2963 if new_info == info_dict:
2964 return
2965 info_dict.clear()
2966 info_dict.update(new_info)
2967
56d868db 2968 try:
f46e2f9d 2969 new_info, files_to_move = self.pre_process(info_dict, 'before_dl', files_to_move)
2970 replace_info_dict(new_info)
56d868db 2971 except PostProcessingError as err:
2972 self.report_error('Preprocessing: %s' % str(err))
2973 return
2974
a13e6848 2975 if self.params.get('skip_download'):
56d868db 2976 info_dict['filepath'] = temp_filename
2977 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
2978 info_dict['__files_to_move'] = files_to_move
f46e2f9d 2979 replace_info_dict(self.run_pp(MoveFilesAfterDownloadPP(self, False), info_dict))
9e907ebd 2980 info_dict['__write_download_archive'] = self.params.get('force_write_download_archive')
56d868db 2981 else:
2982 # Download
b868936c 2983 info_dict.setdefault('__postprocessors', [])
4340deca 2984 try:
0202b52a 2985
e04938ab 2986 def existing_video_file(*filepaths):
6b591b29 2987 ext = info_dict.get('ext')
e04938ab 2988 converted = lambda file: replace_extension(file, self.params.get('final_ext') or ext, ext)
2989 file = self.existing_file(itertools.chain(*zip(map(converted, filepaths), filepaths)),
2990 default_overwrite=False)
2991 if file:
2992 info_dict['ext'] = os.path.splitext(file)[1][1:]
2993 return file
0202b52a 2994
2995 success = True
4340deca 2996 if info_dict.get('requested_formats') is not None:
81cd954a
S
2997
2998 def compatible_formats(formats):
d03cfdce 2999 # TODO: some formats actually allow this (mkv, webm, ogg, mp4), but not all of them.
3000 video_formats = [format for format in formats if format.get('vcodec') != 'none']
3001 audio_formats = [format for format in formats if format.get('acodec') != 'none']
3002 if len(video_formats) > 2 or len(audio_formats) > 2:
3003 return False
3004
81cd954a 3005 # Check extension
d03cfdce 3006 exts = set(format.get('ext') for format in formats)
3007 COMPATIBLE_EXTS = (
3008 set(('mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma')),
3009 set(('webm',)),
3010 )
3011 for ext_sets in COMPATIBLE_EXTS:
3012 if ext_sets.issuperset(exts):
3013 return True
81cd954a
S
3014 # TODO: Check acodec/vcodec
3015 return False
3016
3017 requested_formats = info_dict['requested_formats']
0202b52a 3018 old_ext = info_dict['ext']
4e3b637d 3019 if self.params.get('merge_output_format') is None:
3020 if not compatible_formats(requested_formats):
3021 info_dict['ext'] = 'mkv'
3022 self.report_warning(
3023 'Requested formats are incompatible for merge and will be merged into mkv')
3024 if (info_dict['ext'] == 'webm'
3025 and info_dict.get('thumbnails')
3026 # check with type instead of pp_key, __name__, or isinstance
3027 # since we dont want any custom PPs to trigger this
3028 and any(type(pp) == EmbedThumbnailPP for pp in self._pps['post_process'])):
3029 info_dict['ext'] = 'mkv'
3030 self.report_warning(
3031 'webm doesn\'t support embedding a thumbnail, mkv will be used')
124bc071 3032 new_ext = info_dict['ext']
0202b52a 3033
124bc071 3034 def correct_ext(filename, ext=new_ext):
96fccc10 3035 if filename == '-':
3036 return filename
0202b52a 3037 filename_real_ext = os.path.splitext(filename)[1][1:]
3038 filename_wo_ext = (
3039 os.path.splitext(filename)[0]
124bc071 3040 if filename_real_ext in (old_ext, new_ext)
0202b52a 3041 else filename)
124bc071 3042 return '%s.%s' % (filename_wo_ext, ext)
0202b52a 3043
38c6902b 3044 # Ensure filename always has a correct extension for successful merge
0202b52a 3045 full_filename = correct_ext(full_filename)
3046 temp_filename = correct_ext(temp_filename)
e04938ab 3047 dl_filename = existing_video_file(full_filename, temp_filename)
1ea24129 3048 info_dict['__real_download'] = False
18e674b4 3049
adbc4ec4
THD
3050 downloaded = []
3051 merger = FFmpegMergerPP(self)
3052
3053 fd = get_suitable_downloader(info_dict, self.params, to_stdout=temp_filename == '-')
dbf5416a 3054 if dl_filename is not None:
6c7274ec 3055 self.report_file_already_downloaded(dl_filename)
adbc4ec4
THD
3056 elif fd:
3057 for f in requested_formats if fd != FFmpegFD else []:
3058 f['filepath'] = fname = prepend_extension(
3059 correct_ext(temp_filename, info_dict['ext']),
3060 'f%s' % f['format_id'], info_dict['ext'])
3061 downloaded.append(fname)
dbf5416a 3062 info_dict['url'] = '\n'.join(f['url'] for f in requested_formats)
3063 success, real_download = self.dl(temp_filename, info_dict)
3064 info_dict['__real_download'] = real_download
18e674b4 3065 else:
18e674b4 3066 if self.params.get('allow_unplayable_formats'):
3067 self.report_warning(
3068 'You have requested merging of multiple formats '
3069 'while also allowing unplayable formats to be downloaded. '
3070 'The formats won\'t be merged to prevent data corruption.')
3071 elif not merger.available:
e8969bda 3072 msg = 'You have requested merging of multiple formats but ffmpeg is not installed'
3073 if not self.params.get('ignoreerrors'):
3074 self.report_error(f'{msg}. Aborting due to --abort-on-error')
3075 return
3076 self.report_warning(f'{msg}. The formats won\'t be merged')
18e674b4 3077
96fccc10 3078 if temp_filename == '-':
adbc4ec4 3079 reason = ('using a downloader other than ffmpeg' if FFmpegFD.can_merge_formats(info_dict, self.params)
96fccc10 3080 else 'but the formats are incompatible for simultaneous download' if merger.available
3081 else 'but ffmpeg is not installed')
3082 self.report_warning(
3083 f'You have requested downloading multiple formats to stdout {reason}. '
3084 'The formats will be streamed one after the other')
3085 fname = temp_filename
dbf5416a 3086 for f in requested_formats:
3087 new_info = dict(info_dict)
3088 del new_info['requested_formats']
3089 new_info.update(f)
96fccc10 3090 if temp_filename != '-':
124bc071 3091 fname = prepend_extension(
3092 correct_ext(temp_filename, new_info['ext']),
3093 'f%s' % f['format_id'], new_info['ext'])
96fccc10 3094 if not self._ensure_dir_exists(fname):
3095 return
a21e0ab1 3096 f['filepath'] = fname
96fccc10 3097 downloaded.append(fname)
dbf5416a 3098 partial_success, real_download = self.dl(fname, new_info)
3099 info_dict['__real_download'] = info_dict['__real_download'] or real_download
3100 success = success and partial_success
adbc4ec4
THD
3101
3102 if downloaded and merger.available and not self.params.get('allow_unplayable_formats'):
3103 info_dict['__postprocessors'].append(merger)
3104 info_dict['__files_to_merge'] = downloaded
3105 # Even if there were no downloads, it is being merged only now
3106 info_dict['__real_download'] = True
3107 else:
3108 for file in downloaded:
3109 files_to_move[file] = None
4340deca
P
3110 else:
3111 # Just a single file
e04938ab 3112 dl_filename = existing_video_file(full_filename, temp_filename)
6c7274ec 3113 if dl_filename is None or dl_filename == temp_filename:
3114 # dl_filename == temp_filename could mean that the file was partially downloaded with --no-part.
3115 # So we should try to resume the download
e8e73840 3116 success, real_download = self.dl(temp_filename, info_dict)
0202b52a 3117 info_dict['__real_download'] = real_download
6c7274ec 3118 else:
3119 self.report_file_already_downloaded(dl_filename)
0202b52a 3120
0202b52a 3121 dl_filename = dl_filename or temp_filename
c571435f 3122 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
0202b52a 3123
3158150c 3124 except network_exceptions as err:
7960b056 3125 self.report_error('unable to download video data: %s' % error_to_compat_str(err))
4340deca
P
3126 return
3127 except (OSError, IOError) as err:
3128 raise UnavailableVideoError(err)
3129 except (ContentTooShortError, ) as err:
3130 self.report_error('content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded))
3131 return
8222d8de 3132
de6000d9 3133 if success and full_filename != '-':
f17f8651 3134
fd7cfb64 3135 def fixup():
3136 do_fixup = True
3137 fixup_policy = self.params.get('fixup')
3138 vid = info_dict['id']
3139
3140 if fixup_policy in ('ignore', 'never'):
3141 return
3142 elif fixup_policy == 'warn':
3143 do_fixup = False
f89b3e2d 3144 elif fixup_policy != 'force':
3145 assert fixup_policy in ('detect_or_warn', None)
3146 if not info_dict.get('__real_download'):
3147 do_fixup = False
fd7cfb64 3148
3149 def ffmpeg_fixup(cndn, msg, cls):
3150 if not cndn:
3151 return
3152 if not do_fixup:
3153 self.report_warning(f'{vid}: {msg}')
3154 return
3155 pp = cls(self)
3156 if pp.available:
3157 info_dict['__postprocessors'].append(pp)
3158 else:
3159 self.report_warning(f'{vid}: {msg}. Install ffmpeg to fix this automatically')
3160
3161 stretched_ratio = info_dict.get('stretched_ratio')
3162 ffmpeg_fixup(
3163 stretched_ratio not in (1, None),
3164 f'Non-uniform pixel ratio {stretched_ratio}',
3165 FFmpegFixupStretchedPP)
3166
3167 ffmpeg_fixup(
3168 (info_dict.get('requested_formats') is None
3169 and info_dict.get('container') == 'm4a_dash'
3170 and info_dict.get('ext') == 'm4a'),
3171 'writing DASH m4a. Only some players support this container',
3172 FFmpegFixupM4aPP)
3173
993191c0 3174 downloader = get_suitable_downloader(info_dict, self.params) if 'protocol' in info_dict else None
3175 downloader = downloader.__name__ if downloader else None
adbc4ec4
THD
3176
3177 if info_dict.get('requested_formats') is None: # Not necessary if doing merger
3178 ffmpeg_fixup(downloader == 'HlsFD',
3179 'Possible MPEG-TS in MP4 container or malformed AAC timestamps',
3180 FFmpegFixupM3u8PP)
3181 ffmpeg_fixup(info_dict.get('is_live') and downloader == 'DashSegmentsFD',
3182 'Possible duplicate MOOV atoms', FFmpegFixupDuplicateMoovPP)
3183
e04b003e 3184 ffmpeg_fixup(downloader == 'WebSocketFragmentFD', 'Malformed timestamps detected', FFmpegFixupTimestampPP)
3185 ffmpeg_fixup(downloader == 'WebSocketFragmentFD', 'Malformed duration detected', FFmpegFixupDurationPP)
fd7cfb64 3186
3187 fixup()
8222d8de 3188 try:
f46e2f9d 3189 replace_info_dict(self.post_process(dl_filename, info_dict, files_to_move))
af819c21 3190 except PostProcessingError as err:
3191 self.report_error('Postprocessing: %s' % str(err))
8222d8de 3192 return
ab8e5e51
AM
3193 try:
3194 for ph in self._post_hooks:
23c1a667 3195 ph(info_dict['filepath'])
ab8e5e51
AM
3196 except Exception as err:
3197 self.report_error('post hooks: %s' % str(err))
3198 return
9e907ebd 3199 info_dict['__write_download_archive'] = True
2d30509f 3200
a13e6848 3201 if self.params.get('force_write_download_archive'):
9e907ebd 3202 info_dict['__write_download_archive'] = True
a13e6848 3203
3204 # Make sure the info_dict was modified in-place
f46e2f9d 3205 assert info_dict is original_infodict
a13e6848 3206
c3e6ffba 3207 max_downloads = self.params.get('max_downloads')
3208 if max_downloads is not None and self._num_downloads >= int(max_downloads):
3209 raise MaxDownloadsReached()
8222d8de 3210
aa9369a2 3211 def __download_wrapper(self, func):
3212 @functools.wraps(func)
3213 def wrapper(*args, **kwargs):
3214 try:
3215 res = func(*args, **kwargs)
3216 except UnavailableVideoError as e:
3217 self.report_error(e)
b222c271 3218 except MaxDownloadsReached as e:
aa9369a2 3219 self.to_screen(f'[info] {e}')
3220 raise
b222c271 3221 except DownloadCancelled as e:
3222 self.to_screen(f'[info] {e}')
3223 if not self.params.get('break_per_url'):
3224 raise
aa9369a2 3225 else:
3226 if self.params.get('dump_single_json', False):
3227 self.post_extract(res)
3228 self.to_stdout(json.dumps(self.sanitize_info(res)))
3229 return wrapper
3230
8222d8de
JMF
3231 def download(self, url_list):
3232 """Download a given list of URLs."""
aa9369a2 3233 url_list = variadic(url_list) # Passing a single URL is a common mistake
de6000d9 3234 outtmpl = self.outtmpl_dict['default']
3089bc74
S
3235 if (len(url_list) > 1
3236 and outtmpl != '-'
3237 and '%' not in outtmpl
3238 and self.params.get('max_downloads') != 1):
acd69589 3239 raise SameFileError(outtmpl)
8222d8de
JMF
3240
3241 for url in url_list:
aa9369a2 3242 self.__download_wrapper(self.extract_info)(
3243 url, force_generic_extractor=self.params.get('force_generic_extractor', False))
8222d8de
JMF
3244
3245 return self._download_retcode
3246
1dcc4c0c 3247 def download_with_info_file(self, info_filename):
31bd3925
JMF
3248 with contextlib.closing(fileinput.FileInput(
3249 [info_filename], mode='r',
3250 openhook=fileinput.hook_encoded('utf-8'))) as f:
3251 # FileInput doesn't have a read method, we can't call json.load
8012d892 3252 info = self.sanitize_info(json.loads('\n'.join(f)), self.params.get('clean_infojson', True))
d4943898 3253 try:
aa9369a2 3254 self.__download_wrapper(self.process_ie_result)(info, download=True)
f2ebc5c7 3255 except (DownloadError, EntryNotInPlaylist, ReExtractInfo) as e:
bf5f605e 3256 if not isinstance(e, EntryNotInPlaylist):
3257 self.to_stderr('\r')
d4943898
JMF
3258 webpage_url = info.get('webpage_url')
3259 if webpage_url is not None:
aa9369a2 3260 self.report_warning(f'The info failed to download: {e}; trying with URL {webpage_url}')
d4943898
JMF
3261 return self.download([webpage_url])
3262 else:
3263 raise
3264 return self._download_retcode
1dcc4c0c 3265
cb202fd2 3266 @staticmethod
8012d892 3267 def sanitize_info(info_dict, remove_private_keys=False):
3268 ''' Sanitize the infodict for converting to json '''
3ad56b42 3269 if info_dict is None:
3270 return info_dict
6e84b215 3271 info_dict.setdefault('epoch', int(time.time()))
6a5a30f9 3272 info_dict.setdefault('_type', 'video')
09b49e1f 3273
8012d892 3274 if remove_private_keys:
09b49e1f 3275 reject = lambda k, v: v is None or (k.startswith('_') and k != '_type') or k in {
f46e2f9d 3276 'requested_downloads', 'requested_formats', 'requested_subtitles', 'requested_entries',
3277 'entries', 'filepath', 'infojson_filename', 'original_url', 'playlist_autonumber',
6e84b215 3278 }
ae8f99e6 3279 else:
09b49e1f 3280 reject = lambda k, v: False
adbc4ec4
THD
3281
3282 def filter_fn(obj):
3283 if isinstance(obj, dict):
3284 return {k: filter_fn(v) for k, v in obj.items() if not reject(k, v)}
3285 elif isinstance(obj, (list, tuple, set, LazyList)):
3286 return list(map(filter_fn, obj))
3287 elif obj is None or isinstance(obj, (str, int, float, bool)):
3288 return obj
3289 else:
3290 return repr(obj)
3291
5226731e 3292 return filter_fn(info_dict)
cb202fd2 3293
8012d892 3294 @staticmethod
3295 def filter_requested_info(info_dict, actually_filter=True):
3296 ''' Alias of sanitize_info for backward compatibility '''
3297 return YoutubeDL.sanitize_info(info_dict, actually_filter)
3298
ed5835b4 3299 @staticmethod
3300 def post_extract(info_dict):
3301 def actual_post_extract(info_dict):
3302 if info_dict.get('_type') in ('playlist', 'multi_video'):
3303 for video_dict in info_dict.get('entries', {}):
3304 actual_post_extract(video_dict or {})
3305 return
3306
09b49e1f 3307 post_extractor = info_dict.pop('__post_extractor', None) or (lambda: {})
3308 info_dict.update(post_extractor())
ed5835b4 3309
3310 actual_post_extract(info_dict or {})
3311
dcf64d43 3312 def run_pp(self, pp, infodict):
5bfa4862 3313 files_to_delete = []
dcf64d43 3314 if '__files_to_move' not in infodict:
3315 infodict['__files_to_move'] = {}
b1940459 3316 try:
3317 files_to_delete, infodict = pp.run(infodict)
3318 except PostProcessingError as e:
3319 # Must be True and not 'only_download'
3320 if self.params.get('ignoreerrors') is True:
3321 self.report_error(e)
3322 return infodict
3323 raise
3324
5bfa4862 3325 if not files_to_delete:
dcf64d43 3326 return infodict
5bfa4862 3327 if self.params.get('keepvideo', False):
3328 for f in files_to_delete:
dcf64d43 3329 infodict['__files_to_move'].setdefault(f, '')
5bfa4862 3330 else:
3331 for old_filename in set(files_to_delete):
3332 self.to_screen('Deleting original file %s (pass -k to keep)' % old_filename)
3333 try:
3334 os.remove(encodeFilename(old_filename))
3335 except (IOError, OSError):
3336 self.report_warning('Unable to remove downloaded original file')
dcf64d43 3337 if old_filename in infodict['__files_to_move']:
3338 del infodict['__files_to_move'][old_filename]
3339 return infodict
5bfa4862 3340
ed5835b4 3341 def run_all_pps(self, key, info, *, additional_pps=None):
bb66c247 3342 self._forceprint(key, info)
ed5835b4 3343 for pp in (additional_pps or []) + self._pps[key]:
dc5f409c 3344 info = self.run_pp(pp, info)
ed5835b4 3345 return info
277d6ff5 3346
56d868db 3347 def pre_process(self, ie_info, key='pre_process', files_to_move=None):
5bfa4862 3348 info = dict(ie_info)
56d868db 3349 info['__files_to_move'] = files_to_move or {}
ed5835b4 3350 info = self.run_all_pps(key, info)
56d868db 3351 return info, info.pop('__files_to_move', None)
5bfa4862 3352
f46e2f9d 3353 def post_process(self, filename, info, files_to_move=None):
8222d8de 3354 """Run all the postprocessors on the given file."""
8222d8de 3355 info['filepath'] = filename
dcf64d43 3356 info['__files_to_move'] = files_to_move or {}
ed5835b4 3357 info = self.run_all_pps('post_process', info, additional_pps=info.get('__postprocessors'))
dcf64d43 3358 info = self.run_pp(MoveFilesAfterDownloadPP(self), info)
3359 del info['__files_to_move']
ed5835b4 3360 return self.run_all_pps('after_move', info)
c1c9a79c 3361
5db07df6 3362 def _make_archive_id(self, info_dict):
e9fef7ee
S
3363 video_id = info_dict.get('id')
3364 if not video_id:
3365 return
5db07df6
PH
3366 # Future-proof against any change in case
3367 # and backwards compatibility with prior versions
e9fef7ee 3368 extractor = info_dict.get('extractor_key') or info_dict.get('ie_key') # key in a playlist
7012b23c 3369 if extractor is None:
1211bb6d
S
3370 url = str_or_none(info_dict.get('url'))
3371 if not url:
3372 return
e9fef7ee 3373 # Try to find matching extractor for the URL and take its ie_key
8b7491c8 3374 for ie_key, ie in self._ies.items():
1211bb6d 3375 if ie.suitable(url):
8b7491c8 3376 extractor = ie_key
e9fef7ee
S
3377 break
3378 else:
3379 return
d0757229 3380 return '%s %s' % (extractor.lower(), video_id)
5db07df6
PH
3381
3382 def in_download_archive(self, info_dict):
3383 fn = self.params.get('download_archive')
3384 if fn is None:
3385 return False
3386
3387 vid_id = self._make_archive_id(info_dict)
e9fef7ee 3388 if not vid_id:
7012b23c 3389 return False # Incomplete video information
5db07df6 3390
a45e8619 3391 return vid_id in self.archive
c1c9a79c
PH
3392
3393 def record_download_archive(self, info_dict):
3394 fn = self.params.get('download_archive')
3395 if fn is None:
3396 return
5db07df6
PH
3397 vid_id = self._make_archive_id(info_dict)
3398 assert vid_id
a13e6848 3399 self.write_debug(f'Adding to archive: {vid_id}')
c1c9a79c 3400 with locked_file(fn, 'a', encoding='utf-8') as archive_file:
6febd1c1 3401 archive_file.write(vid_id + '\n')
a45e8619 3402 self.archive.add(vid_id)
dd82ffea 3403
8c51aa65 3404 @staticmethod
8abeeb94 3405 def format_resolution(format, default='unknown'):
9359f3d4 3406 if format.get('vcodec') == 'none' and format.get('acodec') != 'none':
fb04e403 3407 return 'audio only'
f49d89ee
PH
3408 if format.get('resolution') is not None:
3409 return format['resolution']
35615307 3410 if format.get('width') and format.get('height'):
ff51ed58 3411 return '%dx%d' % (format['width'], format['height'])
35615307 3412 elif format.get('height'):
ff51ed58 3413 return '%sp' % format['height']
35615307 3414 elif format.get('width'):
ff51ed58 3415 return '%dx?' % format['width']
3416 return default
8c51aa65 3417
8130779d 3418 def _list_format_headers(self, *headers):
3419 if self.params.get('listformats_table', True) is not False:
3420 return [self._format_screen(header, self.Styles.HEADERS) for header in headers]
3421 return headers
3422
c57f7757
PH
3423 def _format_note(self, fdict):
3424 res = ''
3425 if fdict.get('ext') in ['f4f', 'f4m']:
f304da8a 3426 res += '(unsupported)'
32f90364
PH
3427 if fdict.get('language'):
3428 if res:
3429 res += ' '
f304da8a 3430 res += '[%s]' % fdict['language']
c57f7757 3431 if fdict.get('format_note') is not None:
f304da8a 3432 if res:
3433 res += ' '
3434 res += fdict['format_note']
c57f7757 3435 if fdict.get('tbr') is not None:
f304da8a 3436 if res:
3437 res += ', '
3438 res += '%4dk' % fdict['tbr']
c57f7757
PH
3439 if fdict.get('container') is not None:
3440 if res:
3441 res += ', '
3442 res += '%s container' % fdict['container']
3089bc74
S
3443 if (fdict.get('vcodec') is not None
3444 and fdict.get('vcodec') != 'none'):
c57f7757
PH
3445 if res:
3446 res += ', '
3447 res += fdict['vcodec']
91c7271a 3448 if fdict.get('vbr') is not None:
c57f7757
PH
3449 res += '@'
3450 elif fdict.get('vbr') is not None and fdict.get('abr') is not None:
3451 res += 'video@'
3452 if fdict.get('vbr') is not None:
3453 res += '%4dk' % fdict['vbr']
fbb21cf5 3454 if fdict.get('fps') is not None:
5d583bdf
S
3455 if res:
3456 res += ', '
3457 res += '%sfps' % fdict['fps']
c57f7757
PH
3458 if fdict.get('acodec') is not None:
3459 if res:
3460 res += ', '
3461 if fdict['acodec'] == 'none':
3462 res += 'video only'
3463 else:
3464 res += '%-5s' % fdict['acodec']
3465 elif fdict.get('abr') is not None:
3466 if res:
3467 res += ', '
3468 res += 'audio'
3469 if fdict.get('abr') is not None:
3470 res += '@%3dk' % fdict['abr']
3471 if fdict.get('asr') is not None:
3472 res += ' (%5dHz)' % fdict['asr']
3473 if fdict.get('filesize') is not None:
3474 if res:
3475 res += ', '
3476 res += format_bytes(fdict['filesize'])
9732d77e
PH
3477 elif fdict.get('filesize_approx') is not None:
3478 if res:
3479 res += ', '
3480 res += '~' + format_bytes(fdict['filesize_approx'])
c57f7757 3481 return res
91c7271a 3482
8130779d 3483 def render_formats_table(self, info_dict):
b69fd25c 3484 if not info_dict.get('formats') and not info_dict.get('url'):
8130779d 3485 return None
b69fd25c 3486
94badb25 3487 formats = info_dict.get('formats', [info_dict])
8130779d 3488 if not self.params.get('listformats_table', True) is not False:
76d321f6 3489 table = [
3490 [
3491 format_field(f, 'format_id'),
3492 format_field(f, 'ext'),
3493 self.format_resolution(f),
8130779d 3494 self._format_note(f)
3495 ] for f in formats if f.get('preference') is None or f['preference'] >= -1000]
3496 return render_table(['format code', 'extension', 'resolution', 'note'], table, extra_gap=1)
3497
3498 delim = self._format_screen('\u2502', self.Styles.DELIM, '|', test_encoding=True)
3499 table = [
3500 [
3501 self._format_screen(format_field(f, 'format_id'), self.Styles.ID),
3502 format_field(f, 'ext'),
3503 format_field(f, func=self.format_resolution, ignore=('audio only', 'images')),
3504 format_field(f, 'fps', '\t%d'),
3505 format_field(f, 'dynamic_range', '%s', ignore=(None, 'SDR')).replace('HDR', ''),
3506 delim,
3507 format_field(f, 'filesize', ' \t%s', func=format_bytes) + format_field(f, 'filesize_approx', '~\t%s', func=format_bytes),
3508 format_field(f, 'tbr', '\t%dk'),
3509 shorten_protocol_name(f.get('protocol', '')),
3510 delim,
3511 format_field(f, 'vcodec', default='unknown').replace(
3512 'none', 'images' if f.get('acodec') == 'none'
3513 else self._format_screen('audio only', self.Styles.SUPPRESS)),
3514 format_field(f, 'vbr', '\t%dk'),
3515 format_field(f, 'acodec', default='unknown').replace(
3516 'none', '' if f.get('vcodec') == 'none'
3517 else self._format_screen('video only', self.Styles.SUPPRESS)),
3518 format_field(f, 'abr', '\t%dk'),
3519 format_field(f, 'asr', '\t%dHz'),
3520 join_nonempty(
3521 self._format_screen('UNSUPPORTED', 'light red') if f.get('ext') in ('f4f', 'f4m') else None,
3522 format_field(f, 'language', '[%s]'),
3523 join_nonempty(format_field(f, 'format_note'),
3524 format_field(f, 'container', ignore=(None, f.get('ext'))),
3525 delim=', '),
3526 delim=' '),
3527 ] for f in formats if f.get('preference') is None or f['preference'] >= -1000]
3528 header_line = self._list_format_headers(
3529 'ID', 'EXT', 'RESOLUTION', '\tFPS', 'HDR', delim, '\tFILESIZE', '\tTBR', 'PROTO',
3530 delim, 'VCODEC', '\tVBR', 'ACODEC', '\tABR', '\tASR', 'MORE INFO')
3531
3532 return render_table(
3533 header_line, table, hide_empty=True,
3534 delim=self._format_screen('\u2500', self.Styles.DELIM, '-', test_encoding=True))
3535
3536 def render_thumbnails_table(self, info_dict):
88f23a18 3537 thumbnails = list(info_dict.get('thumbnails') or [])
cfb56d1a 3538 if not thumbnails:
8130779d 3539 return None
3540 return render_table(
ec11a9f4 3541 self._list_format_headers('ID', 'Width', 'Height', 'URL'),
6970b600 3542 [[t.get('id'), t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails])
2412044c 3543
8130779d 3544 def render_subtitles_table(self, video_id, subtitles):
2412044c 3545 def _row(lang, formats):
49c258e1 3546 exts, names = zip(*((f['ext'], f.get('name') or 'unknown') for f in reversed(formats)))
2412044c 3547 if len(set(names)) == 1:
7aee40c1 3548 names = [] if names[0] == 'unknown' else names[:1]
2412044c 3549 return [lang, ', '.join(names), ', '.join(exts)]
3550
8130779d 3551 if not subtitles:
3552 return None
3553 return render_table(
ec11a9f4 3554 self._list_format_headers('Language', 'Name', 'Formats'),
2412044c 3555 [_row(lang, formats) for lang, formats in subtitles.items()],
8130779d 3556 hide_empty=True)
3557
3558 def __list_table(self, video_id, name, func, *args):
3559 table = func(*args)
3560 if not table:
3561 self.to_screen(f'{video_id} has no {name}')
3562 return
3563 self.to_screen(f'[info] Available {name} for {video_id}:')
3564 self.to_stdout(table)
3565
3566 def list_formats(self, info_dict):
3567 self.__list_table(info_dict['id'], 'formats', self.render_formats_table, info_dict)
3568
3569 def list_thumbnails(self, info_dict):
3570 self.__list_table(info_dict['id'], 'thumbnails', self.render_thumbnails_table, info_dict)
3571
3572 def list_subtitles(self, video_id, subtitles, name='subtitles'):
3573 self.__list_table(video_id, name, self.render_subtitles_table, video_id, subtitles)
a504ced0 3574
dca08720
PH
3575 def urlopen(self, req):
3576 """ Start an HTTP download """
82d8a8b6 3577 if isinstance(req, compat_basestring):
67dda517 3578 req = sanitized_Request(req)
19a41fc6 3579 return self._opener.open(req, timeout=self._socket_timeout)
dca08720
PH
3580
3581 def print_debug_header(self):
3582 if not self.params.get('verbose'):
3583 return
49a57e70 3584
3585 def get_encoding(stream):
3586 ret = getattr(stream, 'encoding', 'missing (%s)' % type(stream).__name__)
3587 if not supports_terminal_sequences(stream):
e3c7d495 3588 from .compat import WINDOWS_VT_MODE
3589 ret += ' (No VT)' if WINDOWS_VT_MODE is False else ' (No ANSI)'
49a57e70 3590 return ret
3591
3592 encoding_str = 'Encodings: locale %s, fs %s, out %s, err %s, pref %s' % (
3593 locale.getpreferredencoding(),
3594 sys.getfilesystemencoding(),
3595 get_encoding(self._screen_file), get_encoding(self._err_file),
3596 self.get_encoding())
883d4b1e 3597
3598 logger = self.params.get('logger')
3599 if logger:
3600 write_debug = lambda msg: logger.debug(f'[debug] {msg}')
3601 write_debug(encoding_str)
3602 else:
96565c7e 3603 write_string(f'[debug] {encoding_str}\n', encoding=None)
49a57e70 3604 write_debug = lambda msg: self._write_string(f'[debug] {msg}\n')
734f90bb 3605
4c88ff87 3606 source = detect_variant()
36eaf303 3607 write_debug(join_nonempty(
3608 'yt-dlp version', __version__,
3609 f'[{RELEASE_GIT_HEAD}]' if RELEASE_GIT_HEAD else '',
3610 '' if source == 'unknown' else f'({source})',
3611 delim=' '))
6e21fdd2 3612 if not _LAZY_LOADER:
3613 if os.environ.get('YTDLP_NO_LAZY_EXTRACTORS'):
49a57e70 3614 write_debug('Lazy loading extractors is forcibly disabled')
6e21fdd2 3615 else:
49a57e70 3616 write_debug('Lazy loading extractors is disabled')
3ae5e797 3617 if plugin_extractors or plugin_postprocessors:
49a57e70 3618 write_debug('Plugins: %s' % [
3ae5e797 3619 '%s%s' % (klass.__name__, '' if klass.__name__ == name else f' as {name}')
3620 for name, klass in itertools.chain(plugin_extractors.items(), plugin_postprocessors.items())])
53ed7066 3621 if self.params.get('compat_opts'):
49a57e70 3622 write_debug('Compatibility options: %s' % ', '.join(self.params.get('compat_opts')))
36eaf303 3623
3624 if source == 'source':
dca08720 3625 try:
36eaf303 3626 sp = Popen(
3627 ['git', 'rev-parse', '--short', 'HEAD'],
3628 stdout=subprocess.PIPE, stderr=subprocess.PIPE,
3629 cwd=os.path.dirname(os.path.abspath(__file__)))
3630 out, err = sp.communicate_or_kill()
3631 out = out.decode().strip()
3632 if re.match('[0-9a-f]+', out):
3633 write_debug('Git HEAD: %s' % out)
70a1165b 3634 except Exception:
36eaf303 3635 try:
3636 sys.exc_clear()
3637 except Exception:
3638 pass
b300cda4
S
3639
3640 def python_implementation():
3641 impl_name = platform.python_implementation()
3642 if impl_name == 'PyPy' and hasattr(sys, 'pypy_version_info'):
3643 return impl_name + ' version %d.%d.%d' % sys.pypy_version_info[:3]
3644 return impl_name
3645
49a57e70 3646 write_debug('Python version %s (%s %s) - %s' % (
e5813e53 3647 platform.python_version(),
3648 python_implementation(),
3649 platform.architecture()[0],
b300cda4 3650 platform_name()))
d28b5171 3651
8913ef74 3652 exe_versions, ffmpeg_features = FFmpegPostProcessor.get_versions_and_features(self)
3653 ffmpeg_features = {key for key, val in ffmpeg_features.items() if val}
3654 if ffmpeg_features:
a4211baf 3655 exe_versions['ffmpeg'] += ' (%s)' % ','.join(ffmpeg_features)
8913ef74 3656
4c83c967 3657 exe_versions['rtmpdump'] = rtmpdump_version()
feee8d32 3658 exe_versions['phantomjs'] = PhantomJSwrapper._version()
d28b5171 3659 exe_str = ', '.join(
2831b468 3660 f'{exe} {v}' for exe, v in sorted(exe_versions.items()) if v
3661 ) or 'none'
49a57e70 3662 write_debug('exe versions: %s' % exe_str)
dca08720 3663
2831b468 3664 from .downloader.websocket import has_websockets
3665 from .postprocessor.embedthumbnail import has_mutagen
f59f5ef8 3666 from .cookies import SQLITE_AVAILABLE, SECRETSTORAGE_AVAILABLE
2831b468 3667
c586f9e8 3668 lib_str = join_nonempty(
edf65256 3669 compat_pycrypto_AES and compat_pycrypto_AES.__name__.split('.')[0],
f59f5ef8 3670 SECRETSTORAGE_AVAILABLE and 'secretstorage',
2831b468 3671 has_mutagen and 'mutagen',
3672 SQLITE_AVAILABLE and 'sqlite',
c586f9e8 3673 has_websockets and 'websockets',
3674 delim=', ') or 'none'
49a57e70 3675 write_debug('Optional libraries: %s' % lib_str)
2831b468 3676
dca08720
PH
3677 proxy_map = {}
3678 for handler in self._opener.handlers:
3679 if hasattr(handler, 'proxies'):
3680 proxy_map.update(handler.proxies)
49a57e70 3681 write_debug(f'Proxy map: {proxy_map}')
dca08720 3682
49a57e70 3683 # Not implemented
3684 if False and self.params.get('call_home'):
58b1f00d 3685 ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode('utf-8')
49a57e70 3686 write_debug('Public IP address: %s' % ipaddr)
58b1f00d
PH
3687 latest_version = self.urlopen(
3688 'https://yt-dl.org/latest/version').read().decode('utf-8')
3689 if version_tuple(latest_version) > version_tuple(__version__):
3690 self.report_warning(
3691 'You are using an outdated version (newest version: %s)! '
3692 'See https://yt-dl.org/update if you need help updating.' %
3693 latest_version)
3694
e344693b 3695 def _setup_opener(self):
6ad14cab 3696 timeout_val = self.params.get('socket_timeout')
17bddf3e 3697 self._socket_timeout = 20 if timeout_val is None else float(timeout_val)
6ad14cab 3698
982ee69a 3699 opts_cookiesfrombrowser = self.params.get('cookiesfrombrowser')
dca08720
PH
3700 opts_cookiefile = self.params.get('cookiefile')
3701 opts_proxy = self.params.get('proxy')
3702
982ee69a 3703 self.cookiejar = load_cookies(opts_cookiefile, opts_cookiesfrombrowser, self)
dca08720 3704
6a3f4c3f 3705 cookie_processor = YoutubeDLCookieProcessor(self.cookiejar)
dca08720
PH
3706 if opts_proxy is not None:
3707 if opts_proxy == '':
3708 proxies = {}
3709 else:
3710 proxies = {'http': opts_proxy, 'https': opts_proxy}
3711 else:
3712 proxies = compat_urllib_request.getproxies()
067aa17e 3713 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
dca08720
PH
3714 if 'http' in proxies and 'https' not in proxies:
3715 proxies['https'] = proxies['http']
91410c9b 3716 proxy_handler = PerRequestProxyHandler(proxies)
a0ddb8a2
PH
3717
3718 debuglevel = 1 if self.params.get('debug_printtraffic') else 0
be4a824d
PH
3719 https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel)
3720 ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel)
fca6dba8 3721 redirect_handler = YoutubeDLRedirectHandler()
8b172c2e 3722 data_handler = compat_urllib_request_DataHandler()
6240b0a2
JMF
3723
3724 # When passing our own FileHandler instance, build_opener won't add the
3725 # default FileHandler and allows us to disable the file protocol, which
3726 # can be used for malicious purposes (see
067aa17e 3727 # https://github.com/ytdl-org/youtube-dl/issues/8227)
6240b0a2
JMF
3728 file_handler = compat_urllib_request.FileHandler()
3729
3730 def file_open(*args, **kwargs):
7a5c1cfe 3731 raise compat_urllib_error.URLError('file:// scheme is explicitly disabled in yt-dlp for security reasons')
6240b0a2
JMF
3732 file_handler.file_open = file_open
3733
3734 opener = compat_urllib_request.build_opener(
fca6dba8 3735 proxy_handler, https_handler, cookie_processor, ydlh, redirect_handler, data_handler, file_handler)
2461f79d 3736
dca08720
PH
3737 # Delete the default user-agent header, which would otherwise apply in
3738 # cases where our custom HTTP handler doesn't come into play
067aa17e 3739 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
dca08720
PH
3740 opener.addheaders = []
3741 self._opener = opener
62fec3b2
PH
3742
3743 def encode(self, s):
3744 if isinstance(s, bytes):
3745 return s # Already encoded
3746
3747 try:
3748 return s.encode(self.get_encoding())
3749 except UnicodeEncodeError as err:
3750 err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.'
3751 raise
3752
3753 def get_encoding(self):
3754 encoding = self.params.get('encoding')
3755 if encoding is None:
3756 encoding = preferredencoding()
3757 return encoding
ec82d85a 3758
e08a85d8 3759 def _write_info_json(self, label, ie_result, infofn, overwrite=None):
80c03fa9 3760 ''' Write infojson and returns True = written, False = skip, None = error '''
e08a85d8 3761 if overwrite is None:
3762 overwrite = self.params.get('overwrites', True)
80c03fa9 3763 if not self.params.get('writeinfojson'):
3764 return False
3765 elif not infofn:
3766 self.write_debug(f'Skipping writing {label} infojson')
3767 return False
3768 elif not self._ensure_dir_exists(infofn):
3769 return None
e08a85d8 3770 elif not overwrite and os.path.exists(infofn):
80c03fa9 3771 self.to_screen(f'[info] {label.title()} metadata is already present')
3772 else:
3773 self.to_screen(f'[info] Writing {label} metadata as JSON to: {infofn}')
3774 try:
3775 write_json_file(self.sanitize_info(ie_result, self.params.get('clean_infojson', True)), infofn)
3776 except (OSError, IOError):
3777 self.report_error(f'Cannot write {label} metadata to JSON file {infofn}')
3778 return None
3779 return True
3780
3781 def _write_description(self, label, ie_result, descfn):
3782 ''' Write description and returns True = written, False = skip, None = error '''
3783 if not self.params.get('writedescription'):
3784 return False
3785 elif not descfn:
3786 self.write_debug(f'Skipping writing {label} description')
3787 return False
3788 elif not self._ensure_dir_exists(descfn):
3789 return None
3790 elif not self.params.get('overwrites', True) and os.path.exists(descfn):
3791 self.to_screen(f'[info] {label.title()} description is already present')
3792 elif ie_result.get('description') is None:
3793 self.report_warning(f'There\'s no {label} description to write')
3794 return False
3795 else:
3796 try:
3797 self.to_screen(f'[info] Writing {label} description to: {descfn}')
3798 with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
3799 descfile.write(ie_result['description'])
3800 except (OSError, IOError):
3801 self.report_error(f'Cannot write {label} description file {descfn}')
3802 return None
3803 return True
3804
3805 def _write_subtitles(self, info_dict, filename):
3806 ''' Write subtitles to file and return list of (sub_filename, final_sub_filename); or None if error'''
3807 ret = []
3808 subtitles = info_dict.get('requested_subtitles')
3809 if not subtitles or not (self.params.get('writesubtitles') or self.params.get('writeautomaticsub')):
3810 # subtitles download errors are already managed as troubles in relevant IE
3811 # that way it will silently go on when used with unsupporting IE
3812 return ret
3813
3814 sub_filename_base = self.prepare_filename(info_dict, 'subtitle')
3815 if not sub_filename_base:
3816 self.to_screen('[info] Skipping writing video subtitles')
3817 return ret
3818 for sub_lang, sub_info in subtitles.items():
3819 sub_format = sub_info['ext']
3820 sub_filename = subtitles_filename(filename, sub_lang, sub_format, info_dict.get('ext'))
3821 sub_filename_final = subtitles_filename(sub_filename_base, sub_lang, sub_format, info_dict.get('ext'))
e04938ab 3822 existing_sub = self.existing_file((sub_filename_final, sub_filename))
3823 if existing_sub:
80c03fa9 3824 self.to_screen(f'[info] Video subtitle {sub_lang}.{sub_format} is already present')
e04938ab 3825 sub_info['filepath'] = existing_sub
3826 ret.append((existing_sub, sub_filename_final))
80c03fa9 3827 continue
3828
3829 self.to_screen(f'[info] Writing video subtitles to: {sub_filename}')
3830 if sub_info.get('data') is not None:
3831 try:
3832 # Use newline='' to prevent conversion of newline characters
3833 # See https://github.com/ytdl-org/youtube-dl/issues/10268
3834 with io.open(sub_filename, 'w', encoding='utf-8', newline='') as subfile:
3835 subfile.write(sub_info['data'])
3836 sub_info['filepath'] = sub_filename
3837 ret.append((sub_filename, sub_filename_final))
3838 continue
3839 except (OSError, IOError):
3840 self.report_error(f'Cannot write video subtitles file {sub_filename}')
3841 return None
3842
3843 try:
3844 sub_copy = sub_info.copy()
3845 sub_copy.setdefault('http_headers', info_dict.get('http_headers'))
3846 self.dl(sub_filename, sub_copy, subtitle=True)
3847 sub_info['filepath'] = sub_filename
3848 ret.append((sub_filename, sub_filename_final))
6020e05d 3849 except (DownloadError, ExtractorError, IOError, OSError, ValueError) + network_exceptions as err:
3850 if self.params.get('ignoreerrors') is not True: # False or 'only_download'
3851 raise DownloadError(f'Unable to download video subtitles for {sub_lang!r}: {err}', err)
80c03fa9 3852 self.report_warning(f'Unable to download video subtitles for {sub_lang!r}: {err}')
519804a9 3853 return ret
80c03fa9 3854
3855 def _write_thumbnails(self, label, info_dict, filename, thumb_filename_base=None):
3856 ''' Write thumbnails to file and return list of (thumb_filename, final_thumb_filename) '''
6c4fd172 3857 write_all = self.params.get('write_all_thumbnails', False)
80c03fa9 3858 thumbnails, ret = [], []
6c4fd172 3859 if write_all or self.params.get('writethumbnail', False):
0202b52a 3860 thumbnails = info_dict.get('thumbnails') or []
6c4fd172 3861 multiple = write_all and len(thumbnails) > 1
ec82d85a 3862
80c03fa9 3863 if thumb_filename_base is None:
3864 thumb_filename_base = filename
3865 if thumbnails and not thumb_filename_base:
3866 self.write_debug(f'Skipping writing {label} thumbnail')
3867 return ret
3868
dd0228ce 3869 for idx, t in list(enumerate(thumbnails))[::-1]:
80c03fa9 3870 thumb_ext = (f'{t["id"]}.' if multiple else '') + determine_ext(t['url'], 'jpg')
aa9369a2 3871 thumb_display_id = f'{label} thumbnail {t["id"]}'
80c03fa9 3872 thumb_filename = replace_extension(filename, thumb_ext, info_dict.get('ext'))
3873 thumb_filename_final = replace_extension(thumb_filename_base, thumb_ext, info_dict.get('ext'))
ec82d85a 3874
e04938ab 3875 existing_thumb = self.existing_file((thumb_filename_final, thumb_filename))
3876 if existing_thumb:
aa9369a2 3877 self.to_screen('[info] %s is already present' % (
3878 thumb_display_id if multiple else f'{label} thumbnail').capitalize())
e04938ab 3879 t['filepath'] = existing_thumb
3880 ret.append((existing_thumb, thumb_filename_final))
ec82d85a 3881 else:
80c03fa9 3882 self.to_screen(f'[info] Downloading {thumb_display_id} ...')
ec82d85a 3883 try:
297e9952 3884 uf = self.urlopen(sanitized_Request(t['url'], headers=t.get('http_headers', {})))
80c03fa9 3885 self.to_screen(f'[info] Writing {thumb_display_id} to: {thumb_filename}')
d3d89c32 3886 with open(encodeFilename(thumb_filename), 'wb') as thumbf:
ec82d85a 3887 shutil.copyfileobj(uf, thumbf)
80c03fa9 3888 ret.append((thumb_filename, thumb_filename_final))
885cc0b7 3889 t['filepath'] = thumb_filename
3158150c 3890 except network_exceptions as err:
dd0228ce 3891 thumbnails.pop(idx)
80c03fa9 3892 self.report_warning(f'Unable to download {thumb_display_id}: {err}')
6c4fd172 3893 if ret and not write_all:
3894 break
0202b52a 3895 return ret