]> jfr.im git - yt-dlp.git/blame - yt_dlp/YoutubeDL.py
[cleanup] Misc cleanup and refactor (#2173)
[yt-dlp.git] / yt_dlp / YoutubeDL.py
CommitLineData
cc52de43 1#!/usr/bin/env python3
26e63931 2import collections
31bd3925 3import contextlib
9d2ecdbc 4import datetime
c1c9a79c 5import errno
31bd3925 6import fileinput
b5ae35ee 7import functools
8222d8de 8import io
b82f815f 9import itertools
8694c600 10import json
62fec3b2 11import locale
083c9df9 12import operator
8222d8de 13import os
dca08720 14import platform
f8271158 15import random
8222d8de
JMF
16import re
17import shutil
dca08720 18import subprocess
8222d8de 19import sys
21cd8fae 20import tempfile
8222d8de 21import time
67134eab 22import tokenize
8222d8de 23import traceback
524e2e4f 24import unicodedata
f9934b96 25import urllib.request
961ea474
S
26from string import ascii_letters
27
f8271158 28from .cache import Cache
8c25f81b 29from .compat import (
4390d5ec 30 compat_brotli,
003c69a8 31 compat_get_terminal_size,
e9c0cdd3 32 compat_os_name,
edf65256 33 compat_pycrypto_AES,
7d1eb38a 34 compat_shlex_quote,
ce02ed60
PH
35 compat_str,
36 compat_urllib_error,
37 compat_urllib_request,
819e0531 38 windows_enable_vt_mode,
8c25f81b 39)
982ee69a 40from .cookies import load_cookies
f8271158 41from .downloader import FFmpegFD, get_suitable_downloader, shorten_protocol_name
42from .downloader.rtmp import rtmpdump_version
43from .extractor import _LAZY_LOADER
44from .extractor import _PLUGIN_CLASSES as plugin_extractors
45from .extractor import gen_extractor_classes, get_info_extractor
46from .extractor.openload import PhantomJSwrapper
47from .minicurses import format_text
48from .postprocessor import _PLUGIN_CLASSES as plugin_postprocessors
49from .postprocessor import (
50 EmbedThumbnailPP,
51 FFmpegFixupDuplicateMoovPP,
52 FFmpegFixupDurationPP,
53 FFmpegFixupM3u8PP,
54 FFmpegFixupM4aPP,
55 FFmpegFixupStretchedPP,
56 FFmpegFixupTimestampPP,
57 FFmpegMergerPP,
58 FFmpegPostProcessor,
59 MoveFilesAfterDownloadPP,
60 get_postprocessor,
61)
62from .update import detect_variant
8c25f81b 63from .utils import (
f8271158 64 DEFAULT_OUTTMPL,
65 LINK_TEMPLATES,
66 NO_DEFAULT,
67 OUTTMPL_TYPES,
68 POSTPROCESS_WHEN,
69 STR_FORMAT_RE_TMPL,
70 STR_FORMAT_TYPES,
71 ContentTooShortError,
72 DateRange,
73 DownloadCancelled,
74 DownloadError,
75 EntryNotInPlaylist,
76 ExistingVideoReached,
77 ExtractorError,
78 GeoRestrictedError,
79 HEADRequest,
80 InAdvancePagedList,
81 ISO3166Utils,
82 LazyList,
83 MaxDownloadsReached,
19a03940 84 Namespace,
f8271158 85 PagedList,
86 PerRequestProxyHandler,
87 Popen,
88 PostProcessingError,
89 ReExtractInfo,
90 RejectedVideoReached,
91 SameFileError,
92 UnavailableVideoError,
93 YoutubeDLCookieProcessor,
94 YoutubeDLHandler,
95 YoutubeDLRedirectHandler,
eedb7ba5
S
96 age_restricted,
97 args_to_str,
ce02ed60 98 date_from_str,
ce02ed60 99 determine_ext,
b5559424 100 determine_protocol,
c0384f22 101 encode_compat_str,
ce02ed60 102 encodeFilename,
a06916d9 103 error_to_compat_str,
590bc6f6 104 expand_path,
90137ca4 105 filter_dict,
e29663c6 106 float_or_none,
02dbf93f 107 format_bytes,
e0fd9573 108 format_decimal_suffix,
f8271158 109 format_field,
525ef922 110 formatSeconds,
0bb322b9 111 get_domain,
d5820461 112 has_certifi,
c9969434 113 int_or_none,
732044af 114 iri_to_uri,
34921b43 115 join_nonempty,
ce02ed60 116 locked_file,
0202b52a 117 make_dir,
dca08720 118 make_HTTPS_handler,
8b7539d2 119 merge_headers,
3158150c 120 network_exceptions,
ec11a9f4 121 number_of_digits,
cd6fc19e 122 orderedSet,
083c9df9 123 parse_filesize,
dca08720 124 platform_name,
ce02ed60 125 preferredencoding,
eedb7ba5 126 prepend_extension,
51fb4995 127 register_socks_protocols,
3efb96a6 128 remove_terminal_sequences,
cfb56d1a 129 render_table,
eedb7ba5 130 replace_extension,
ce02ed60 131 sanitize_filename,
1bb5c511 132 sanitize_path,
dcf77cf1 133 sanitize_url,
67dda517 134 sanitized_Request,
e5660ee6 135 std_headers,
1211bb6d 136 str_or_none,
e29663c6 137 strftime_or_none,
ce02ed60 138 subtitles_filename,
819e0531 139 supports_terminal_sequences,
f2ebc5c7 140 timetuple_from_msec,
732044af 141 to_high_limit_path,
324ad820 142 traverse_obj,
6033d980 143 try_get,
29eb5174 144 url_basename,
7d1eb38a 145 variadic,
58b1f00d 146 version_tuple,
ce02ed60
PH
147 write_json_file,
148 write_string,
4f026faf 149)
f8271158 150from .version import RELEASE_GIT_HEAD, __version__
8222d8de 151
e9c0cdd3
YCH
152if compat_os_name == 'nt':
153 import ctypes
154
2459b6e1 155
86e5f3ed 156class YoutubeDL:
8222d8de
JMF
157 """YoutubeDL class.
158
159 YoutubeDL objects are the ones responsible of downloading the
160 actual video file and writing it to disk if the user has requested
161 it, among some other tasks. In most cases there should be one per
162 program. As, given a video URL, the downloader doesn't know how to
163 extract all the needed information, task that InfoExtractors do, it
164 has to pass the URL to one of them.
165
166 For this, YoutubeDL objects have a method that allows
167 InfoExtractors to be registered in a given order. When it is passed
168 a URL, the YoutubeDL object handles it to the first InfoExtractor it
169 finds that reports being able to handle it. The InfoExtractor extracts
170 all the information about the video or videos the URL refers to, and
171 YoutubeDL process the extracted information, possibly using a File
172 Downloader to download the video.
173
174 YoutubeDL objects accept a lot of parameters. In order not to saturate
175 the object constructor with arguments, it receives a dictionary of
176 options instead. These options are available through the params
177 attribute for the InfoExtractors to use. The YoutubeDL also
178 registers itself as the downloader in charge for the InfoExtractors
179 that are added to it, so this is a "mutual registration".
180
181 Available options:
182
183 username: Username for authentication purposes.
184 password: Password for authentication purposes.
180940e0 185 videopassword: Password for accessing a video.
1da50aa3
S
186 ap_mso: Adobe Pass multiple-system operator identifier.
187 ap_username: Multiple-system operator account username.
188 ap_password: Multiple-system operator account password.
8222d8de
JMF
189 usenetrc: Use netrc for authentication instead.
190 verbose: Print additional info to stdout.
191 quiet: Do not print messages to stdout.
ad8915b7 192 no_warnings: Do not print out anything for warnings.
bb66c247 193 forceprint: A dict with keys WHEN mapped to a list of templates to
194 print to stdout. The allowed keys are video or any of the
195 items in utils.POSTPROCESS_WHEN.
ca30f449 196 For compatibility, a single list is also accepted
bb66c247 197 print_to_file: A dict with keys WHEN (same as forceprint) mapped to
198 a list of tuples with (template, filename)
53c18592 199 forceurl: Force printing final URL. (Deprecated)
200 forcetitle: Force printing title. (Deprecated)
201 forceid: Force printing ID. (Deprecated)
202 forcethumbnail: Force printing thumbnail URL. (Deprecated)
203 forcedescription: Force printing description. (Deprecated)
204 forcefilename: Force printing final filename. (Deprecated)
205 forceduration: Force printing duration. (Deprecated)
8694c600 206 forcejson: Force printing info_dict as JSON.
63e0be34
PH
207 dump_single_json: Force printing the info_dict of the whole playlist
208 (or video) as a single JSON line.
c25228e5 209 force_write_download_archive: Force writing download archive regardless
210 of 'skip_download' or 'simulate'.
b7b04c78 211 simulate: Do not download the video files. If unset (or None),
212 simulate only if listsubtitles, listformats or list_thumbnails is used
eb8a4433 213 format: Video format code. see "FORMAT SELECTION" for more details.
093a1710 214 You can also pass a function. The function takes 'ctx' as
215 argument and returns the formats to download.
216 See "build_format_selector" for an implementation
63ad4d43 217 allow_unplayable_formats: Allow unplayable formats to be extracted and downloaded.
b7da73eb 218 ignore_no_formats_error: Ignore "No video formats" error. Usefull for
219 extracting metadata even if the video is not actually
220 available for download (experimental)
0930b11f 221 format_sort: A list of fields by which to sort the video formats.
222 See "Sorting Formats" for more details.
c25228e5 223 format_sort_force: Force the given format_sort. see "Sorting Formats"
224 for more details.
08d30158 225 prefer_free_formats: Whether to prefer video formats with free containers
226 over non-free ones of same quality.
c25228e5 227 allow_multiple_video_streams: Allow multiple video streams to be merged
228 into a single file
229 allow_multiple_audio_streams: Allow multiple audio streams to be merged
230 into a single file
0ba692ac 231 check_formats Whether to test if the formats are downloadable.
9f1a1c36 232 Can be True (check all), False (check none),
233 'selected' (check selected formats),
0ba692ac 234 or None (check only if requested by extractor)
4524baf0 235 paths: Dictionary of output paths. The allowed keys are 'home'
236 'temp' and the keys of OUTTMPL_TYPES (in utils.py)
de6000d9 237 outtmpl: Dictionary of templates for output names. Allowed keys
4524baf0 238 are 'default' and the keys of OUTTMPL_TYPES (in utils.py).
34488702 239 For compatibility with youtube-dl, a single string can also be used
a820dc72
RA
240 outtmpl_na_placeholder: Placeholder for unavailable meta fields.
241 restrictfilenames: Do not allow "&" and spaces in file names
242 trim_file_name: Limit length of filename (extension excluded)
4524baf0 243 windowsfilenames: Force the filenames to be windows compatible
b1940459 244 ignoreerrors: Do not stop on download/postprocessing errors.
245 Can be 'only_download' to ignore only download errors.
246 Default is 'only_download' for CLI, but False for API
26e2805c 247 skip_playlist_after_errors: Number of allowed failures until the rest of
248 the playlist is skipped
d22dec74 249 force_generic_extractor: Force downloader to use the generic extractor
0c3d0f51 250 overwrites: Overwrite all video and metadata files if True,
251 overwrite only non-video files if None
252 and don't overwrite any file if False
34488702 253 For compatibility with youtube-dl,
254 "nooverwrites" may also be used instead
8222d8de
JMF
255 playliststart: Playlist item to start at.
256 playlistend: Playlist item to end at.
c14e88f0 257 playlist_items: Specific indices of playlist to download.
ff815fe6 258 playlistreverse: Download playlist items in reverse order.
75822ca7 259 playlistrandom: Download playlist items in random order.
8222d8de
JMF
260 matchtitle: Download only matching titles.
261 rejecttitle: Reject downloads for matching titles.
8bf9319e 262 logger: Log messages to a logging.Logger instance.
8222d8de 263 logtostderr: Log messages to stderr instead of stdout.
819e0531 264 consoletitle: Display progress in console window's titlebar.
8222d8de
JMF
265 writedescription: Write the video description to a .description file
266 writeinfojson: Write the video description to a .info.json file
75d43ca0 267 clean_infojson: Remove private fields from the infojson
34488702 268 getcomments: Extract video comments. This will not be written to disk
06167fbb 269 unless writeinfojson is also given
1fb07d10 270 writeannotations: Write the video annotations to a .annotations.xml file
8222d8de 271 writethumbnail: Write the thumbnail image to a file
c25228e5 272 allow_playlist_files: Whether to write playlists' description, infojson etc
273 also to disk when using the 'write*' options
ec82d85a 274 write_all_thumbnails: Write all thumbnail formats to files
732044af 275 writelink: Write an internet shortcut file, depending on the
276 current platform (.url/.webloc/.desktop)
277 writeurllink: Write a Windows internet shortcut file (.url)
278 writewebloclink: Write a macOS internet shortcut file (.webloc)
279 writedesktoplink: Write a Linux internet shortcut file (.desktop)
8222d8de 280 writesubtitles: Write the video subtitles to a file
741dd8ea 281 writeautomaticsub: Write the automatically generated subtitles to a file
245524e6 282 allsubtitles: Deprecated - Use subtitleslangs = ['all']
c32b0aab 283 Downloads all the subtitles of the video
0b7f3118 284 (requires writesubtitles or writeautomaticsub)
8222d8de 285 listsubtitles: Lists all available subtitles for the video
a504ced0 286 subtitlesformat: The format code for subtitles
c32b0aab 287 subtitleslangs: List of languages of the subtitles to download (can be regex).
288 The list may contain "all" to refer to all the available
289 subtitles. The language can be prefixed with a "-" to
290 exclude it from the requested languages. Eg: ['all', '-live_chat']
8222d8de
JMF
291 keepvideo: Keep the video file after post-processing
292 daterange: A DateRange object, download only if the upload_date is in the range.
293 skip_download: Skip the actual download of the video file
c35f9e72 294 cachedir: Location of the cache files in the filesystem.
a0e07d31 295 False to disable filesystem cache.
47192f92 296 noplaylist: Download single video instead of a playlist if in doubt.
8dbe9899
PH
297 age_limit: An integer representing the user's age in years.
298 Unsuitable videos for the given age are skipped.
5fe18bdb
PH
299 min_views: An integer representing the minimum view count the video
300 must have in order to not be skipped.
301 Videos without view count information are always
302 downloaded. None for no limit.
303 max_views: An integer representing the maximum view count.
304 Videos that are more popular than that are not
305 downloaded.
306 Videos without view count information are always
307 downloaded. None for no limit.
308 download_archive: File name of a file where all downloads are recorded.
c1c9a79c
PH
309 Videos already present in the file are not downloaded
310 again.
8a51f564 311 break_on_existing: Stop the download process after attempting to download a
312 file that is in the archive.
313 break_on_reject: Stop the download process when encountering a video that
314 has been filtered out.
b222c271 315 break_per_url: Whether break_on_reject and break_on_existing
316 should act on each input URL as opposed to for the entire queue
8a51f564 317 cookiefile: File name where cookies should be read from and dumped to
f59f5ef8
MB
318 cookiesfrombrowser: A tuple containing the name of the browser, the profile
319 name/pathfrom where cookies are loaded, and the name of the
320 keyring. Eg: ('chrome', ) or ('vivaldi', 'default', 'BASICTEXT')
f81c62a6 321 legacyserverconnect: Explicitly allow HTTPS connection to servers that do not
322 support RFC 5746 secure renegotiation
f59f5ef8 323 nocheckcertificate: Do not verify SSL certificates
7e8c0af0
PH
324 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
325 At the moment, this is only supported by YouTube.
8b7539d2 326 http_headers: A dictionary of custom headers to be used for all requests
a1ee09e8 327 proxy: URL of the proxy server to use
38cce791 328 geo_verification_proxy: URL of the proxy to use for IP address verification
504f20dd 329 on geo-restricted sites.
e344693b 330 socket_timeout: Time to wait for unresponsive hosts, in seconds
0783b09b
PH
331 bidi_workaround: Work around buggy terminals without bidirectional text
332 support, using fridibi
a0ddb8a2 333 debug_printtraffic:Print out sent and received HTTP traffic
91f071af 334 include_ads: Download ads as well (deprecated)
04b4d394
PH
335 default_search: Prepend this string if an input url is not valid.
336 'auto' for elaborate guessing
62fec3b2 337 encoding: Use this encoding instead of the system-specified.
e8ee972c 338 extract_flat: Do not resolve URLs, return the immediate result.
057a5206
PH
339 Pass in 'in_playlist' to only show this behavior for
340 playlist items.
f2ebc5c7 341 wait_for_video: If given, wait for scheduled streams to become available.
342 The value should be a tuple containing the range
343 (min_secs, max_secs) to wait between retries
4f026faf 344 postprocessors: A list of dictionaries, each with an entry
71b640cc 345 * key: The name of the postprocessor. See
7a5c1cfe 346 yt_dlp/postprocessor/__init__.py for a list.
bb66c247 347 * when: When to run the postprocessor. Allowed values are
348 the entries of utils.POSTPROCESS_WHEN
56d868db 349 Assumed to be 'post_process' if not given
b5ae35ee 350 post_hooks: Deprecated - Register a custom postprocessor instead
351 A list of functions that get called as the final step
ab8e5e51
AM
352 for each video file, after all postprocessors have been
353 called. The filename will be passed as the only argument.
71b640cc
PH
354 progress_hooks: A list of functions that get called on download
355 progress, with a dictionary with the entries
5cda4eda 356 * status: One of "downloading", "error", or "finished".
ee69b99a 357 Check this first and ignore unknown values.
3ba7740d 358 * info_dict: The extracted info_dict
71b640cc 359
5cda4eda 360 If status is one of "downloading", or "finished", the
ee69b99a
PH
361 following properties may also be present:
362 * filename: The final filename (always present)
5cda4eda 363 * tmpfilename: The filename we're currently writing to
71b640cc
PH
364 * downloaded_bytes: Bytes on disk
365 * total_bytes: Size of the whole file, None if unknown
5cda4eda
PH
366 * total_bytes_estimate: Guess of the eventual file size,
367 None if unavailable.
368 * elapsed: The number of seconds since download started.
71b640cc
PH
369 * eta: The estimated time in seconds, None if unknown
370 * speed: The download speed in bytes/second, None if
371 unknown
5cda4eda
PH
372 * fragment_index: The counter of the currently
373 downloaded video fragment.
374 * fragment_count: The number of fragments (= individual
375 files that will be merged)
71b640cc
PH
376
377 Progress hooks are guaranteed to be called at least once
378 (with status "finished") if the download is successful.
819e0531 379 postprocessor_hooks: A list of functions that get called on postprocessing
380 progress, with a dictionary with the entries
381 * status: One of "started", "processing", or "finished".
382 Check this first and ignore unknown values.
383 * postprocessor: Name of the postprocessor
384 * info_dict: The extracted info_dict
385
386 Progress hooks are guaranteed to be called at least twice
387 (with status "started" and "finished") if the processing is successful.
45598f15 388 merge_output_format: Extension to use when merging formats.
6b591b29 389 final_ext: Expected final extension; used to detect when the file was
59a7a13e 390 already downloaded and converted
6271f1ca
PH
391 fixup: Automatically correct known faults of the file.
392 One of:
393 - "never": do nothing
394 - "warn": only emit a warning
395 - "detect_or_warn": check whether we can do anything
62cd676c 396 about it, warn otherwise (default)
504f20dd 397 source_address: Client-side IP address to bind to.
6ec6cb4e 398 call_home: Boolean, true iff we are allowed to contact the
7a5c1cfe 399 yt-dlp servers for debugging. (BROKEN)
1cf376f5 400 sleep_interval_requests: Number of seconds to sleep between requests
401 during extraction
7aa589a5
S
402 sleep_interval: Number of seconds to sleep before each download when
403 used alone or a lower bound of a range for randomized
404 sleep before each download (minimum possible number
405 of seconds to sleep) when used along with
406 max_sleep_interval.
407 max_sleep_interval:Upper bound of a range for randomized sleep before each
408 download (maximum possible number of seconds to sleep).
409 Must only be used along with sleep_interval.
410 Actual sleep time will be a random float from range
411 [sleep_interval; max_sleep_interval].
1cf376f5 412 sleep_interval_subtitles: Number of seconds to sleep before each subtitle download
cfb56d1a
PH
413 listformats: Print an overview of available video formats and exit.
414 list_thumbnails: Print a table of all thumbnails and exit.
347de493
PH
415 match_filter: A function that gets called with the info_dict of
416 every video.
417 If it returns a message, the video is ignored.
418 If it returns None, the video is downloaded.
419 match_filter_func in utils.py is one example for this.
7e5db8c9 420 no_color: Do not emit color codes in output.
0a840f58 421 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
504f20dd 422 HTTP header
0a840f58 423 geo_bypass_country:
773f291d
S
424 Two-letter ISO 3166-2 country code that will be used for
425 explicit geographic restriction bypassing via faking
504f20dd 426 X-Forwarded-For HTTP header
5f95927a
S
427 geo_bypass_ip_block:
428 IP range in CIDR notation that will be used similarly to
504f20dd 429 geo_bypass_country
71b640cc 430
85729c51 431 The following options determine which downloader is picked:
52a8a1e1 432 external_downloader: A dictionary of protocol keys and the executable of the
433 external downloader to use for it. The allowed protocols
434 are default|http|ftp|m3u8|dash|rtsp|rtmp|mms.
435 Set the value to 'native' to use the native downloader
436 hls_prefer_native: Deprecated - Use external_downloader = {'m3u8': 'native'}
437 or {'m3u8': 'ffmpeg'} instead.
438 Use the native HLS downloader instead of ffmpeg/avconv
bf09af3a
S
439 if True, otherwise use ffmpeg/avconv if False, otherwise
440 use downloader suggested by extractor if None.
53ed7066 441 compat_opts: Compatibility options. See "Differences in default behavior".
3acf6d38 442 The following options do not work when used through the API:
b5ae35ee 443 filename, abort-on-error, multistreams, no-live-chat, format-sort
dac5df5a 444 no-clean-infojson, no-playlist-metafiles, no-keep-subs, no-attach-info-json.
e4f02757 445 Refer __init__.py for their implementation
819e0531 446 progress_template: Dictionary of templates for progress outputs.
447 Allowed keys are 'download', 'postprocess',
448 'download-title' (console title) and 'postprocess-title'.
449 The template is mapped on a dictionary with keys 'progress' and 'info'
fe7e0c98 450
8222d8de 451 The following parameters are not used by YoutubeDL itself, they are used by
7a5c1cfe 452 the downloader (see yt_dlp/downloader/common.py):
51d9739f 453 nopart, updatetime, buffersize, ratelimit, throttledratelimit, min_filesize,
205a0654
EH
454 max_filesize, test, noresizebuffer, retries, file_access_retries, fragment_retries,
455 continuedl, noprogress, xattr_set_filesize, hls_use_mpegts, http_chunk_size,
59a7a13e 456 external_downloader_args, concurrent_fragment_downloads.
76b1bd67
JMF
457
458 The following options are used by the post processors:
d4a24f40 459 prefer_ffmpeg: If False, use avconv instead of ffmpeg if both are available,
e4172ac9 460 otherwise prefer ffmpeg. (avconv support is deprecated)
c0b7d117
S
461 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
462 to the binary or its containing directory.
43820c03 463 postprocessor_args: A dictionary of postprocessor/executable keys (in lower case)
34488702 464 and a list of additional command-line arguments for the
465 postprocessor/executable. The dict can also have "PP+EXE" keys
466 which are used when the given exe is used by the given PP.
467 Use 'default' as the name for arguments to passed to all PP
468 For compatibility with youtube-dl, a single list of args
469 can also be used
e409895f 470
471 The following options are used by the extractors:
62bff2c1 472 extractor_retries: Number of times to retry for known errors
473 dynamic_mpd: Whether to process dynamic DASH manifests (default: True)
e409895f 474 hls_split_discontinuity: Split HLS playlists to different formats at
62bff2c1 475 discontinuities such as ad breaks (default: False)
5d3a0e79 476 extractor_args: A dictionary of arguments to be passed to the extractors.
477 See "EXTRACTOR ARGUMENTS" for details.
478 Eg: {'youtube': {'skip': ['dash', 'hls']}}
88f23a18 479 mark_watched: Mark videos watched (even with --simulate). Only for YouTube
5d3a0e79 480 youtube_include_dash_manifest: Deprecated - Use extractor_args instead.
481 If True (default), DASH manifests and related
62bff2c1 482 data will be downloaded and processed by extractor.
483 You can reduce network I/O by disabling it if you don't
484 care about DASH. (only for youtube)
5d3a0e79 485 youtube_include_hls_manifest: Deprecated - Use extractor_args instead.
486 If True (default), HLS manifests and related
62bff2c1 487 data will be downloaded and processed by extractor.
488 You can reduce network I/O by disabling it if you don't
489 care about HLS. (only for youtube)
8222d8de
JMF
490 """
491
86e5f3ed 492 _NUMERIC_FIELDS = {
c9969434 493 'width', 'height', 'tbr', 'abr', 'asr', 'vbr', 'fps', 'filesize', 'filesize_approx',
e6f21b3d 494 'timestamp', 'release_timestamp',
c9969434
S
495 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
496 'average_rating', 'comment_count', 'age_limit',
497 'start_time', 'end_time',
498 'chapter_number', 'season_number', 'episode_number',
499 'track_number', 'disc_number', 'release_year',
86e5f3ed 500 }
c9969434 501
6db9c4d5 502 _format_fields = {
503 # NB: Keep in sync with the docstring of extractor/common.py
a44ca5a4 504 'url', 'manifest_url', 'manifest_stream_number', 'ext', 'format', 'format_id', 'format_note',
6db9c4d5 505 'width', 'height', 'resolution', 'dynamic_range', 'tbr', 'abr', 'acodec', 'asr',
506 'vbr', 'fps', 'vcodec', 'container', 'filesize', 'filesize_approx',
507 'player_url', 'protocol', 'fragment_base_url', 'fragments', 'is_from_start',
508 'preference', 'language', 'language_preference', 'quality', 'source_preference',
509 'http_headers', 'stretched_ratio', 'no_resume', 'has_drm', 'downloader_options',
510 'page_url', 'app', 'play_path', 'tc_url', 'flash_version', 'rtmp_live', 'rtmp_conn', 'rtmp_protocol', 'rtmp_real_time'
511 }
48ee10ee 512 _format_selection_exts = {
513 'audio': {'m4a', 'mp3', 'ogg', 'aac'},
514 'video': {'mp4', 'flv', 'webm', '3gp'},
515 'storyboards': {'mhtml'},
516 }
517
3511266b 518 def __init__(self, params=None, auto_init=True):
883d4b1e 519 """Create a FileDownloader object with the given options.
520 @param auto_init Whether to load the default extractors and print header (if verbose).
49a57e70 521 Set to 'no_verbose_header' to not print the header
883d4b1e 522 """
e9f9a10f
JMF
523 if params is None:
524 params = {}
592b7485 525 self.params = params
8b7491c8 526 self._ies = {}
56c73665 527 self._ies_instances = {}
1e43a6f7 528 self._pps = {k: [] for k in POSTPROCESS_WHEN}
b35496d8 529 self._printed_messages = set()
1cf376f5 530 self._first_webpage_request = True
ab8e5e51 531 self._post_hooks = []
933605d7 532 self._progress_hooks = []
819e0531 533 self._postprocessor_hooks = []
8222d8de
JMF
534 self._download_retcode = 0
535 self._num_downloads = 0
9c906919 536 self._num_videos = 0
592b7485 537 self._playlist_level = 0
538 self._playlist_urls = set()
a0e07d31 539 self.cache = Cache(self)
34308b30 540
819e0531 541 windows_enable_vt_mode()
cf4f42cb 542 self._out_files = {
543 'error': sys.stderr,
544 'print': sys.stderr if self.params.get('logtostderr') else sys.stdout,
545 'console': None if compat_os_name == 'nt' else next(
546 filter(supports_terminal_sequences, (sys.stderr, sys.stdout)), None)
547 }
548 self._out_files['screen'] = sys.stderr if self.params.get('quiet') else self._out_files['print']
ec11a9f4 549 self._allow_colors = {
cf4f42cb 550 type_: not self.params.get('no_color') and supports_terminal_sequences(self._out_files[type_])
551 for type_ in ('screen', 'error')
ec11a9f4 552 }
819e0531 553
a61f4b28 554 if sys.version_info < (3, 6):
555 self.report_warning(
0181adef 556 'Python version %d.%d is not supported! Please update to Python 3.6 or above' % sys.version_info[:2])
a61f4b28 557
88acdbc2 558 if self.params.get('allow_unplayable_formats'):
559 self.report_warning(
ec11a9f4 560 f'You have asked for {self._format_err("UNPLAYABLE", self.Styles.EMPHASIS)} formats to be listed/downloaded. '
819e0531 561 'This is a developer option intended for debugging. \n'
562 ' If you experience any issues while using this option, '
ec11a9f4 563 f'{self._format_err("DO NOT", self.Styles.ERROR)} open a bug report')
88acdbc2 564
be5df5ee
S
565 def check_deprecated(param, option, suggestion):
566 if self.params.get(param) is not None:
86e5f3ed 567 self.report_warning(f'{option} is deprecated. Use {suggestion} instead')
be5df5ee
S
568 return True
569 return False
570
571 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
38cce791
YCH
572 if self.params.get('geo_verification_proxy') is None:
573 self.params['geo_verification_proxy'] = self.params['cn_verification_proxy']
574
0d1bb027 575 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
576 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
53ed7066 577 check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"')
0d1bb027 578
49a57e70 579 for msg in self.params.get('_warnings', []):
0d1bb027 580 self.report_warning(msg)
ee8dd27a 581 for msg in self.params.get('_deprecation_warnings', []):
582 self.deprecation_warning(msg)
0d1bb027 583
ec11a9f4 584 if 'list-formats' in self.params.get('compat_opts', []):
585 self.params['listformats_table'] = False
586
b5ae35ee 587 if 'overwrites' not in self.params and self.params.get('nooverwrites') is not None:
b868936c 588 # nooverwrites was unnecessarily changed to overwrites
589 # in 0c3d0f51778b153f65c21906031c2e091fcfb641
590 # This ensures compatibility with both keys
591 self.params['overwrites'] = not self.params['nooverwrites']
b5ae35ee 592 elif self.params.get('overwrites') is None:
593 self.params.pop('overwrites', None)
b868936c 594 else:
595 self.params['nooverwrites'] = not self.params['overwrites']
b9d973be 596
455a15e2 597 self.params.setdefault('forceprint', {})
598 self.params.setdefault('print_to_file', {})
bb66c247 599
600 # Compatibility with older syntax
ca30f449 601 if not isinstance(params['forceprint'], dict):
455a15e2 602 self.params['forceprint'] = {'video': params['forceprint']}
ca30f449 603
455a15e2 604 if self.params.get('bidi_workaround', False):
1c088fa8
PH
605 try:
606 import pty
607 master, slave = pty.openpty()
003c69a8 608 width = compat_get_terminal_size().columns
1c088fa8
PH
609 if width is None:
610 width_args = []
611 else:
612 width_args = ['-w', str(width)]
5d681e96 613 sp_kwargs = dict(
1c088fa8
PH
614 stdin=subprocess.PIPE,
615 stdout=slave,
cf4f42cb 616 stderr=self._out_files['error'])
5d681e96 617 try:
d3c93ec2 618 self._output_process = Popen(['bidiv'] + width_args, **sp_kwargs)
5d681e96 619 except OSError:
d3c93ec2 620 self._output_process = Popen(['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
5d681e96 621 self._output_channel = os.fdopen(master, 'rb')
1c088fa8 622 except OSError as ose:
66e7ace1 623 if ose.errno == errno.ENOENT:
49a57e70 624 self.report_warning(
625 'Could not find fribidi executable, ignoring --bidi-workaround. '
626 'Make sure that fribidi is an executable file in one of the directories in your $PATH.')
1c088fa8
PH
627 else:
628 raise
0783b09b 629
97ec5bc5 630 if auto_init:
631 if auto_init != 'no_verbose_header':
632 self.print_debug_header()
633 self.add_default_info_extractors()
634
3089bc74
S
635 if (sys.platform != 'win32'
636 and sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
455a15e2 637 and not self.params.get('restrictfilenames', False)):
e9137224 638 # Unicode filesystem API will throw errors (#1474, #13027)
34308b30 639 self.report_warning(
6febd1c1 640 'Assuming --restrict-filenames since file system encoding '
1b725173 641 'cannot encode all characters. '
6febd1c1 642 'Set the LC_ALL environment variable to fix this.')
4a98cdbf 643 self.params['restrictfilenames'] = True
34308b30 644
de6000d9 645 self.outtmpl_dict = self.parse_outtmpl()
486dd09e 646
187986a8 647 # Creating format selector here allows us to catch syntax errors before the extraction
648 self.format_selector = (
fa9f30b8 649 self.params.get('format') if self.params.get('format') in (None, '-')
093a1710 650 else self.params['format'] if callable(self.params['format'])
187986a8 651 else self.build_format_selector(self.params['format']))
652
8b7539d2 653 # Set http_headers defaults according to std_headers
654 self.params['http_headers'] = merge_headers(std_headers, self.params.get('http_headers', {}))
655
013b50b7 656 hooks = {
657 'post_hooks': self.add_post_hook,
658 'progress_hooks': self.add_progress_hook,
659 'postprocessor_hooks': self.add_postprocessor_hook,
660 }
661 for opt, fn in hooks.items():
662 for ph in self.params.get(opt, []):
663 fn(ph)
71b640cc 664
5bfc8bee 665 for pp_def_raw in self.params.get('postprocessors', []):
666 pp_def = dict(pp_def_raw)
667 when = pp_def.pop('when', 'post_process')
668 self.add_post_processor(
f9934b96 669 get_postprocessor(pp_def.pop('key'))(self, **pp_def),
5bfc8bee 670 when=when)
671
97ec5bc5 672 self._setup_opener()
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())
86e5f3ed 684 except OSError as ioe:
ed39cac5 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
cf4f42cb 783 def to_stdout(self, message, skip_eol=False, quiet=None):
0760b0a7 784 """Print message to stdout"""
cf4f42cb 785 if quiet is not None:
ae6a1b95 786 self.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument quiet. Use "YoutubeDL.to_screen" instead')
cf4f42cb 787 self._write_string(
788 '%s%s' % (self._bidi_workaround(message), ('' if skip_eol else '\n')),
789 self._out_files['print'])
790
791 def to_screen(self, message, skip_eol=False, quiet=None):
792 """Print message to screen if not in quiet mode"""
8bf9319e 793 if self.params.get('logger'):
43afe285 794 self.params['logger'].debug(message)
cf4f42cb 795 return
796 if (self.params.get('quiet') if quiet is None else quiet) and not self.params.get('verbose'):
797 return
798 self._write_string(
799 '%s%s' % (self._bidi_workaround(message), ('' if skip_eol else '\n')),
800 self._out_files['screen'])
8222d8de 801
b35496d8 802 def to_stderr(self, message, only_once=False):
0760b0a7 803 """Print message to stderr"""
11b85ce6 804 assert isinstance(message, compat_str)
8bf9319e 805 if self.params.get('logger'):
43afe285
IB
806 self.params['logger'].error(message)
807 else:
cf4f42cb 808 self._write_string('%s\n' % self._bidi_workaround(message), self._out_files['error'], only_once=only_once)
809
810 def _send_console_code(self, code):
811 if compat_os_name == 'nt' or not self._out_files['console']:
812 return
813 self._write_string(code, self._out_files['console'])
8222d8de 814
1e5b9a95
PH
815 def to_console_title(self, message):
816 if not self.params.get('consoletitle', False):
817 return
3efb96a6 818 message = remove_terminal_sequences(message)
4bede0d8
C
819 if compat_os_name == 'nt':
820 if ctypes.windll.kernel32.GetConsoleWindow():
821 # c_wchar_p() might not be necessary if `message` is
822 # already of type unicode()
823 ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
cf4f42cb 824 else:
825 self._send_console_code(f'\033]0;{message}\007')
1e5b9a95 826
bdde425c 827 def save_console_title(self):
cf4f42cb 828 if not self.params.get('consoletitle') or self.params.get('simulate'):
bdde425c 829 return
592b7485 830 self._send_console_code('\033[22;0t') # Save the title on stack
bdde425c
PH
831
832 def restore_console_title(self):
cf4f42cb 833 if not self.params.get('consoletitle') or self.params.get('simulate'):
bdde425c 834 return
592b7485 835 self._send_console_code('\033[23;0t') # Restore the title from stack
bdde425c
PH
836
837 def __enter__(self):
838 self.save_console_title()
839 return self
840
841 def __exit__(self, *args):
842 self.restore_console_title()
f89197d7 843
dca08720 844 if self.params.get('cookiefile') is not None:
1bab3437 845 self.cookiejar.save(ignore_discard=True, ignore_expires=True)
bdde425c 846
fa9f30b8 847 def trouble(self, message=None, tb=None, is_error=True):
8222d8de
JMF
848 """Determine action to take when a download problem appears.
849
850 Depending on if the downloader has been configured to ignore
851 download errors or not, this method may throw an exception or
852 not when errors are found, after printing the message.
853
fa9f30b8 854 @param tb If given, is additional traceback information
855 @param is_error Whether to raise error according to ignorerrors
8222d8de
JMF
856 """
857 if message is not None:
858 self.to_stderr(message)
859 if self.params.get('verbose'):
860 if tb is None:
861 if sys.exc_info()[0]: # if .trouble has been called from an except block
6febd1c1 862 tb = ''
8222d8de 863 if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
6febd1c1 864 tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
c0384f22 865 tb += encode_compat_str(traceback.format_exc())
8222d8de
JMF
866 else:
867 tb_data = traceback.format_list(traceback.extract_stack())
6febd1c1 868 tb = ''.join(tb_data)
c19bc311 869 if tb:
870 self.to_stderr(tb)
fa9f30b8 871 if not is_error:
872 return
b1940459 873 if not self.params.get('ignoreerrors'):
8222d8de
JMF
874 if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
875 exc_info = sys.exc_info()[1].exc_info
876 else:
877 exc_info = sys.exc_info()
878 raise DownloadError(message, exc_info)
879 self._download_retcode = 1
880
19a03940 881 Styles = Namespace(
882 HEADERS='yellow',
883 EMPHASIS='light blue',
884 ID='green',
885 DELIM='blue',
886 ERROR='red',
887 WARNING='yellow',
888 SUPPRESS='light black',
889 )
ec11a9f4 890
7578d77d 891 def _format_text(self, handle, allow_colors, text, f, fallback=None, *, test_encoding=False):
e5a998f3 892 text = str(text)
ec11a9f4 893 if test_encoding:
894 original_text = text
5c104538 895 # handle.encoding can be None. See https://github.com/yt-dlp/yt-dlp/issues/2711
896 encoding = self.params.get('encoding') or getattr(handle, 'encoding', None) or 'ascii'
ec11a9f4 897 text = text.encode(encoding, 'ignore').decode(encoding)
898 if fallback is not None and text != original_text:
899 text = fallback
7578d77d 900 return format_text(text, f) if allow_colors else text if fallback is None else fallback
ec11a9f4 901
902 def _format_screen(self, *args, **kwargs):
7578d77d 903 return self._format_text(
cf4f42cb 904 self._out_files['screen'], self._allow_colors['screen'], *args, **kwargs)
ec11a9f4 905
906 def _format_err(self, *args, **kwargs):
7578d77d 907 return self._format_text(
cf4f42cb 908 self._out_files['error'], self._allow_colors['error'], *args, **kwargs)
819e0531 909
c84aeac6 910 def report_warning(self, message, only_once=False):
8222d8de
JMF
911 '''
912 Print the message to stderr, it will be prefixed with 'WARNING:'
913 If stderr is a tty file the 'WARNING:' will be colored
914 '''
6d07ce01
JMF
915 if self.params.get('logger') is not None:
916 self.params['logger'].warning(message)
8222d8de 917 else:
ad8915b7
PH
918 if self.params.get('no_warnings'):
919 return
ec11a9f4 920 self.to_stderr(f'{self._format_err("WARNING:", self.Styles.WARNING)} {message}', only_once)
8222d8de 921
ee8dd27a 922 def deprecation_warning(self, message):
923 if self.params.get('logger') is not None:
a44ca5a4 924 self.params['logger'].warning(f'DeprecationWarning: {message}')
ee8dd27a 925 else:
926 self.to_stderr(f'{self._format_err("DeprecationWarning:", self.Styles.ERROR)} {message}', True)
927
fa9f30b8 928 def report_error(self, message, *args, **kwargs):
8222d8de
JMF
929 '''
930 Do the same as trouble, but prefixes the message with 'ERROR:', colored
931 in red if stderr is a tty file.
932 '''
fa9f30b8 933 self.trouble(f'{self._format_err("ERROR:", self.Styles.ERROR)} {message}', *args, **kwargs)
8222d8de 934
b35496d8 935 def write_debug(self, message, only_once=False):
0760b0a7 936 '''Log debug message or Print message to stderr'''
937 if not self.params.get('verbose', False):
938 return
939 message = '[debug] %s' % message
940 if self.params.get('logger'):
941 self.params['logger'].debug(message)
942 else:
b35496d8 943 self.to_stderr(message, only_once)
0760b0a7 944
8222d8de
JMF
945 def report_file_already_downloaded(self, file_name):
946 """Report file has already been fully downloaded."""
947 try:
6febd1c1 948 self.to_screen('[download] %s has already been downloaded' % file_name)
ce02ed60 949 except UnicodeEncodeError:
6febd1c1 950 self.to_screen('[download] The file has already been downloaded')
8222d8de 951
0c3d0f51 952 def report_file_delete(self, file_name):
953 """Report that existing file will be deleted."""
954 try:
c25228e5 955 self.to_screen('Deleting existing file %s' % file_name)
0c3d0f51 956 except UnicodeEncodeError:
c25228e5 957 self.to_screen('Deleting existing file')
0c3d0f51 958
319b6059 959 def raise_no_formats(self, info, forced=False, *, msg=None):
1151c407 960 has_drm = info.get('__has_drm')
319b6059 961 ignored, expected = self.params.get('ignore_no_formats_error'), bool(msg)
962 msg = msg or has_drm and 'This video is DRM protected' or 'No video formats found!'
963 if forced or not ignored:
1151c407 964 raise ExtractorError(msg, video_id=info['id'], ie=info['extractor'],
319b6059 965 expected=has_drm or ignored or expected)
88acdbc2 966 else:
967 self.report_warning(msg)
968
de6000d9 969 def parse_outtmpl(self):
970 outtmpl_dict = self.params.get('outtmpl', {})
971 if not isinstance(outtmpl_dict, dict):
972 outtmpl_dict = {'default': outtmpl_dict}
71ce444a 973 # Remove spaces in the default template
974 if self.params.get('restrictfilenames'):
975 sanitize = lambda x: x.replace(' - ', ' ').replace(' ', '-')
976 else:
977 sanitize = lambda x: x
de6000d9 978 outtmpl_dict.update({
71ce444a 979 k: sanitize(v) for k, v in DEFAULT_OUTTMPL.items()
80c03fa9 980 if outtmpl_dict.get(k) is None})
86e5f3ed 981 for _, val in outtmpl_dict.items():
de6000d9 982 if isinstance(val, bytes):
86e5f3ed 983 self.report_warning('Parameter outtmpl is bytes, but should be a unicode string')
de6000d9 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 '')
21cd8fae 993 return sanitize_path(path, force=self.params.get('windowsfilenames'))
994
76a264ac 995 @staticmethod
901130bb 996 def _outtmpl_expandpath(outtmpl):
997 # expand_path translates '%%' into '%' and '$$' into '$'
998 # correspondingly that is not what we want since we need to keep
999 # '%%' intact for template dict substitution step. Working around
1000 # with boundary-alike separator hack.
1001 sep = ''.join([random.choice(ascii_letters) for _ in range(32)])
86e5f3ed 1002 outtmpl = outtmpl.replace('%%', f'%{sep}%').replace('$$', f'${sep}$')
901130bb 1003
1004 # outtmpl should be expand_path'ed before template dict substitution
1005 # because meta fields may contain env variables we don't want to
1006 # be expanded. For example, for outtmpl "%(title)s.%(ext)s" and
1007 # title "Hello $PATH", we don't want `$PATH` to be expanded.
1008 return expand_path(outtmpl).replace(sep, '')
1009
1010 @staticmethod
1011 def escape_outtmpl(outtmpl):
1012 ''' Escape any remaining strings like %s, %abc% etc. '''
1013 return re.sub(
1014 STR_FORMAT_RE_TMPL.format('', '(?![%(\0])'),
1015 lambda mobj: ('' if mobj.group('has_key') else '%') + mobj.group(0),
1016 outtmpl)
1017
1018 @classmethod
1019 def validate_outtmpl(cls, outtmpl):
76a264ac 1020 ''' @return None or Exception object '''
7d1eb38a 1021 outtmpl = re.sub(
37893bb0 1022 STR_FORMAT_RE_TMPL.format('[^)]*', '[ljqBUDS]'),
7d1eb38a 1023 lambda mobj: f'{mobj.group(0)[:-1]}s',
1024 cls._outtmpl_expandpath(outtmpl))
76a264ac 1025 try:
7d1eb38a 1026 cls.escape_outtmpl(outtmpl) % collections.defaultdict(int)
76a264ac 1027 return None
1028 except ValueError as err:
1029 return err
1030
03b4de72 1031 @staticmethod
1032 def _copy_infodict(info_dict):
1033 info_dict = dict(info_dict)
09b49e1f 1034 info_dict.pop('__postprocessors', None)
03b4de72 1035 return info_dict
1036
e0fd9573 1037 def prepare_outtmpl(self, outtmpl, info_dict, sanitize=False):
1038 """ Make the outtmpl and info_dict suitable for substitution: ydl.escape_outtmpl(outtmpl) % info_dict
1039 @param sanitize Whether to sanitize the output as a filename.
1040 For backward compatibility, a function can also be passed
1041 """
1042
6e84b215 1043 info_dict.setdefault('epoch', int(time.time())) # keep epoch consistent once set
143db31d 1044
03b4de72 1045 info_dict = self._copy_infodict(info_dict)
752cda38 1046 info_dict['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
53c18592 1047 formatSeconds(info_dict['duration'], '-' if sanitize else ':')
143db31d 1048 if info_dict.get('duration', None) is not None
1049 else None)
752cda38 1050 info_dict['autonumber'] = self.params.get('autonumber_start', 1) - 1 + self._num_downloads
9c906919 1051 info_dict['video_autonumber'] = self._num_videos
752cda38 1052 if info_dict.get('resolution') is None:
1053 info_dict['resolution'] = self.format_resolution(info_dict, default=None)
143db31d 1054
e6f21b3d 1055 # For fields playlist_index, playlist_autonumber and autonumber convert all occurrences
143db31d 1056 # of %(field)s to %(field)0Nd for backward compatibility
1057 field_size_compat_map = {
ec11a9f4 1058 'playlist_index': number_of_digits(info_dict.get('_last_playlist_index') or 0),
1059 'playlist_autonumber': number_of_digits(info_dict.get('n_entries') or 0),
752cda38 1060 'autonumber': self.params.get('autonumber_size') or 5,
143db31d 1061 }
752cda38 1062
385a27fa 1063 TMPL_DICT = {}
37893bb0 1064 EXTERNAL_FORMAT_RE = re.compile(STR_FORMAT_RE_TMPL.format('[^)]*', f'[{STR_FORMAT_TYPES}ljqBUDS]'))
385a27fa 1065 MATH_FUNCTIONS = {
1066 '+': float.__add__,
1067 '-': float.__sub__,
1068 }
e625be0d 1069 # Field is of the form key1.key2...
1070 # where keys (except first) can be string, int or slice
2b8a2973 1071 FIELD_RE = r'\w*(?:\.(?:\w+|{num}|{num}?(?::{num}?){{1,2}}))*'.format(num=r'(?:-?\d+)')
e0fd9573 1072 MATH_FIELD_RE = r'''(?:{field}|{num})'''.format(field=FIELD_RE, num=r'-?\d+(?:.\d+)?')
385a27fa 1073 MATH_OPERATORS_RE = r'(?:%s)' % '|'.join(map(re.escape, MATH_FUNCTIONS.keys()))
e625be0d 1074 INTERNAL_FORMAT_RE = re.compile(r'''(?x)
1075 (?P<negate>-)?
385a27fa 1076 (?P<fields>{field})
1077 (?P<maths>(?:{math_op}{math_field})*)
e625be0d 1078 (?:>(?P<strf_format>.+?))?
34baa9fd 1079 (?P<remaining>
1080 (?P<alternate>(?<!\\),[^|&)]+)?
1081 (?:&(?P<replacement>.*?))?
1082 (?:\|(?P<default>.*?))?
1083 )$'''.format(field=FIELD_RE, math_op=MATH_OPERATORS_RE, math_field=MATH_FIELD_RE))
752cda38 1084
2b8a2973 1085 def _traverse_infodict(k):
1086 k = k.split('.')
1087 if k[0] == '':
1088 k.pop(0)
1089 return traverse_obj(info_dict, k, is_user_input=True, traverse_string=True)
76a264ac 1090
752cda38 1091 def get_value(mdict):
1092 # Object traversal
2b8a2973 1093 value = _traverse_infodict(mdict['fields'])
752cda38 1094 # Negative
1095 if mdict['negate']:
1096 value = float_or_none(value)
1097 if value is not None:
1098 value *= -1
1099 # Do maths
385a27fa 1100 offset_key = mdict['maths']
1101 if offset_key:
752cda38 1102 value = float_or_none(value)
1103 operator = None
385a27fa 1104 while offset_key:
1105 item = re.match(
1106 MATH_FIELD_RE if operator else MATH_OPERATORS_RE,
1107 offset_key).group(0)
1108 offset_key = offset_key[len(item):]
1109 if operator is None:
752cda38 1110 operator = MATH_FUNCTIONS[item]
385a27fa 1111 continue
1112 item, multiplier = (item[1:], -1) if item[0] == '-' else (item, 1)
1113 offset = float_or_none(item)
1114 if offset is None:
2b8a2973 1115 offset = float_or_none(_traverse_infodict(item))
385a27fa 1116 try:
1117 value = operator(value, multiplier * offset)
1118 except (TypeError, ZeroDivisionError):
1119 return None
1120 operator = None
752cda38 1121 # Datetime formatting
1122 if mdict['strf_format']:
7c37ff97 1123 value = strftime_or_none(value, mdict['strf_format'].replace('\\,', ','))
752cda38 1124
1125 return value
1126
b868936c 1127 na = self.params.get('outtmpl_na_placeholder', 'NA')
1128
e0fd9573 1129 def filename_sanitizer(key, value, restricted=self.params.get('restrictfilenames')):
5c3895ff 1130 return sanitize_filename(str(value), restricted=restricted, is_id=(
1131 bool(re.search(r'(^|[_.])id(\.|$)', key))
1132 if 'filename-sanitization' in self.params.get('compat_opts', [])
1133 else NO_DEFAULT))
e0fd9573 1134
1135 sanitizer = sanitize if callable(sanitize) else filename_sanitizer
1136 sanitize = bool(sanitize)
1137
6e84b215 1138 def _dumpjson_default(obj):
1139 if isinstance(obj, (set, LazyList)):
1140 return list(obj)
adbc4ec4 1141 return repr(obj)
6e84b215 1142
752cda38 1143 def create_key(outer_mobj):
1144 if not outer_mobj.group('has_key'):
b836dc94 1145 return outer_mobj.group(0)
752cda38 1146 key = outer_mobj.group('key')
752cda38 1147 mobj = re.match(INTERNAL_FORMAT_RE, key)
e0fd9573 1148 initial_field = mobj.group('fields') if mobj else ''
e978789f 1149 value, replacement, default = None, None, na
7c37ff97 1150 while mobj:
e625be0d 1151 mobj = mobj.groupdict()
7c37ff97 1152 default = mobj['default'] if mobj['default'] is not None else default
752cda38 1153 value = get_value(mobj)
e978789f 1154 replacement = mobj['replacement']
7c37ff97 1155 if value is None and mobj['alternate']:
34baa9fd 1156 mobj = re.match(INTERNAL_FORMAT_RE, mobj['remaining'][1:])
7c37ff97 1157 else:
1158 break
752cda38 1159
b868936c 1160 fmt = outer_mobj.group('format')
752cda38 1161 if fmt == 's' and value is not None and key in field_size_compat_map.keys():
86e5f3ed 1162 fmt = f'0{field_size_compat_map[key]:d}d'
752cda38 1163
e978789f 1164 value = default if value is None else value if replacement is None else replacement
752cda38 1165
4476d2c7 1166 flags = outer_mobj.group('conversion') or ''
7d1eb38a 1167 str_fmt = f'{fmt[:-1]}s'
524e2e4f 1168 if fmt[-1] == 'l': # list
4476d2c7 1169 delim = '\n' if '#' in flags else ', '
9e907ebd 1170 value, fmt = delim.join(map(str, variadic(value, allowed_types=(str, bytes)))), str_fmt
524e2e4f 1171 elif fmt[-1] == 'j': # json
4476d2c7 1172 value, fmt = json.dumps(value, default=_dumpjson_default, indent=4 if '#' in flags else None), str_fmt
524e2e4f 1173 elif fmt[-1] == 'q': # quoted
4476d2c7 1174 value = map(str, variadic(value) if '#' in flags else [value])
1175 value, fmt = ' '.join(map(compat_shlex_quote, value)), str_fmt
524e2e4f 1176 elif fmt[-1] == 'B': # bytes
86e5f3ed 1177 value = f'%{str_fmt}'.encode() % str(value).encode('utf-8')
f5aa5cfb 1178 value, fmt = value.decode('utf-8', 'ignore'), 's'
524e2e4f 1179 elif fmt[-1] == 'U': # unicode normalized
524e2e4f 1180 value, fmt = unicodedata.normalize(
1181 # "+" = compatibility equivalence, "#" = NFD
4476d2c7 1182 'NF%s%s' % ('K' if '+' in flags else '', 'D' if '#' in flags else 'C'),
524e2e4f 1183 value), str_fmt
e0fd9573 1184 elif fmt[-1] == 'D': # decimal suffix
abbeeebc 1185 num_fmt, fmt = fmt[:-1].replace('#', ''), 's'
1186 value = format_decimal_suffix(value, f'%{num_fmt}f%s' if num_fmt else '%d%s',
1187 factor=1024 if '#' in flags else 1000)
37893bb0 1188 elif fmt[-1] == 'S': # filename sanitization
e0fd9573 1189 value, fmt = filename_sanitizer(initial_field, value, restricted='#' in flags), str_fmt
7d1eb38a 1190 elif fmt[-1] == 'c':
524e2e4f 1191 if value:
1192 value = str(value)[0]
76a264ac 1193 else:
524e2e4f 1194 fmt = str_fmt
76a264ac 1195 elif fmt[-1] not in 'rs': # numeric
a439a3a4 1196 value = float_or_none(value)
752cda38 1197 if value is None:
1198 value, fmt = default, 's'
901130bb 1199
752cda38 1200 if sanitize:
1201 if fmt[-1] == 'r':
1202 # If value is an object, sanitize might convert it to a string
1203 # So we convert it to repr first
7d1eb38a 1204 value, fmt = repr(value), str_fmt
639f1cea 1205 if fmt[-1] in 'csr':
e0fd9573 1206 value = sanitizer(initial_field, value)
901130bb 1207
b868936c 1208 key = '%s\0%s' % (key.replace('%', '%\0'), outer_mobj.group('format'))
385a27fa 1209 TMPL_DICT[key] = value
b868936c 1210 return '{prefix}%({key}){fmt}'.format(key=key, fmt=fmt, prefix=outer_mobj.group('prefix'))
752cda38 1211
385a27fa 1212 return EXTERNAL_FORMAT_RE.sub(create_key, outtmpl), TMPL_DICT
143db31d 1213
819e0531 1214 def evaluate_outtmpl(self, outtmpl, info_dict, *args, **kwargs):
1215 outtmpl, info_dict = self.prepare_outtmpl(outtmpl, info_dict, *args, **kwargs)
1216 return self.escape_outtmpl(outtmpl) % info_dict
1217
5127e92a 1218 def _prepare_filename(self, info_dict, *, outtmpl=None, tmpl_type=None):
1219 assert None in (outtmpl, tmpl_type), 'outtmpl and tmpl_type are mutually exclusive'
1220 if outtmpl is None:
1221 outtmpl = self.outtmpl_dict.get(tmpl_type or 'default', self.outtmpl_dict['default'])
8222d8de 1222 try:
5127e92a 1223 outtmpl = self._outtmpl_expandpath(outtmpl)
e0fd9573 1224 filename = self.evaluate_outtmpl(outtmpl, info_dict, True)
6a0546e3 1225 if not filename:
1226 return None
15da37c7 1227
5127e92a 1228 if tmpl_type in ('', 'temp'):
6a0546e3 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)
5127e92a 1232 elif tmpl_type:
6a0546e3 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
5127e92a 1248 def prepare_filename(self, info_dict, dir_type='', *, outtmpl=None, warn=False):
1249 """Generate the output filename"""
1250 if outtmpl:
1251 assert not dir_type, 'outtmpl and dir_type are mutually exclusive'
1252 dir_type = None
1253 filename = self._prepare_filename(info_dict, tmpl_type=dir_type, outtmpl=outtmpl)
80c03fa9 1254 if not filename and dir_type not in ('', 'temp'):
1255 return ''
de6000d9 1256
c84aeac6 1257 if warn:
21cd8fae 1258 if not self.params.get('paths'):
de6000d9 1259 pass
1260 elif filename == '-':
c84aeac6 1261 self.report_warning('--paths is ignored when an outputting to stdout', only_once=True)
de6000d9 1262 elif os.path.isabs(filename):
c84aeac6 1263 self.report_warning('--paths is ignored since an absolute path is given in output template', only_once=True)
de6000d9 1264 if filename == '-' or not filename:
1265 return filename
1266
21cd8fae 1267 return self.get_output_path(dir_type, filename)
0202b52a 1268
120fe513 1269 def _match_entry(self, info_dict, incomplete=False, silent=False):
ecdec191 1270 """ Returns None if the file should be downloaded """
8222d8de 1271
c77495e3 1272 video_title = info_dict.get('title', info_dict.get('id', 'video'))
1273
8b0d7497 1274 def check_filter():
8b0d7497 1275 if 'title' in info_dict:
1276 # This can happen when we're just evaluating the playlist
1277 title = info_dict['title']
1278 matchtitle = self.params.get('matchtitle', False)
1279 if matchtitle:
1280 if not re.search(matchtitle, title, re.IGNORECASE):
1281 return '"' + title + '" title did not match pattern "' + matchtitle + '"'
1282 rejecttitle = self.params.get('rejecttitle', False)
1283 if rejecttitle:
1284 if re.search(rejecttitle, title, re.IGNORECASE):
1285 return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
1286 date = info_dict.get('upload_date')
1287 if date is not None:
1288 dateRange = self.params.get('daterange', DateRange())
1289 if date not in dateRange:
86e5f3ed 1290 return f'{date_from_str(date).isoformat()} upload date is not in range {dateRange}'
8b0d7497 1291 view_count = info_dict.get('view_count')
1292 if view_count is not None:
1293 min_views = self.params.get('min_views')
1294 if min_views is not None and view_count < min_views:
1295 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
1296 max_views = self.params.get('max_views')
1297 if max_views is not None and view_count > max_views:
1298 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
1299 if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')):
1300 return 'Skipping "%s" because it is age restricted' % video_title
8b0d7497 1301
8f18aca8 1302 match_filter = self.params.get('match_filter')
1303 if match_filter is not None:
1304 try:
1305 ret = match_filter(info_dict, incomplete=incomplete)
1306 except TypeError:
1307 # For backward compatibility
1308 ret = None if incomplete else match_filter(info_dict)
1309 if ret is not None:
1310 return ret
8b0d7497 1311 return None
1312
c77495e3 1313 if self.in_download_archive(info_dict):
1314 reason = '%s has already been recorded in the archive' % video_title
1315 break_opt, break_err = 'break_on_existing', ExistingVideoReached
1316 else:
1317 reason = check_filter()
1318 break_opt, break_err = 'break_on_reject', RejectedVideoReached
8b0d7497 1319 if reason is not None:
120fe513 1320 if not silent:
1321 self.to_screen('[download] ' + reason)
c77495e3 1322 if self.params.get(break_opt, False):
1323 raise break_err()
8b0d7497 1324 return reason
fe7e0c98 1325
b6c45014
JMF
1326 @staticmethod
1327 def add_extra_info(info_dict, extra_info):
1328 '''Set the keys from extra_info in info dict if they are missing'''
1329 for key, value in extra_info.items():
1330 info_dict.setdefault(key, value)
1331
409e1828 1332 def extract_info(self, url, download=True, ie_key=None, extra_info=None,
61aa5ba3 1333 process=True, force_generic_extractor=False):
41d1cca3 1334 """
1335 Return a list with a dictionary for each video extracted.
1336
1337 Arguments:
1338 url -- URL to extract
1339
1340 Keyword arguments:
1341 download -- whether to download videos during extraction
1342 ie_key -- extractor key hint
1343 extra_info -- dictionary containing the extra values to add to each result
1344 process -- whether to resolve all unresolved references (URLs, playlist items),
1345 must be True for download to work.
1346 force_generic_extractor -- force using the generic extractor
1347 """
fe7e0c98 1348
409e1828 1349 if extra_info is None:
1350 extra_info = {}
1351
61aa5ba3 1352 if not ie_key and force_generic_extractor:
d22dec74
S
1353 ie_key = 'Generic'
1354
8222d8de 1355 if ie_key:
8b7491c8 1356 ies = {ie_key: self._get_info_extractor_class(ie_key)}
8222d8de
JMF
1357 else:
1358 ies = self._ies
1359
8b7491c8 1360 for ie_key, ie in ies.items():
8222d8de
JMF
1361 if not ie.suitable(url):
1362 continue
1363
1364 if not ie.working():
6febd1c1
PH
1365 self.report_warning('The program functionality for this site has been marked as broken, '
1366 'and will probably not work.')
8222d8de 1367
1151c407 1368 temp_id = ie.get_temp_id(url)
a0566bbf 1369 if temp_id is not None and self.in_download_archive({'id': temp_id, 'ie_key': ie_key}):
5e5be0c0 1370 self.to_screen(f'[{ie_key}] {temp_id}: has already been recorded in the archive')
1371 if self.params.get('break_on_existing', False):
1372 raise ExistingVideoReached()
a0566bbf 1373 break
8b7491c8 1374 return self.__extract_info(url, self.get_info_extractor(ie_key), download, extra_info, process)
a0566bbf 1375 else:
1376 self.report_error('no suitable InfoExtractor for URL %s' % url)
1377
8e5fecc8 1378 def __handle_extraction_exceptions(func):
b5ae35ee 1379 @functools.wraps(func)
a0566bbf 1380 def wrapper(self, *args, **kwargs):
6da22e7d 1381 while True:
1382 try:
1383 return func(self, *args, **kwargs)
1384 except (DownloadCancelled, LazyList.IndexError, PagedList.IndexError):
8222d8de 1385 raise
6da22e7d 1386 except ReExtractInfo as e:
1387 if e.expected:
1388 self.to_screen(f'{e}; Re-extracting data')
1389 else:
1390 self.to_stderr('\r')
1391 self.report_warning(f'{e}; Re-extracting data')
1392 continue
1393 except GeoRestrictedError as e:
1394 msg = e.msg
1395 if e.countries:
1396 msg += '\nThis video is available in %s.' % ', '.join(
1397 map(ISO3166Utils.short2full, e.countries))
1398 msg += '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1399 self.report_error(msg)
1400 except ExtractorError as e: # An error we somewhat expected
1401 self.report_error(str(e), e.format_traceback())
1402 except Exception as e:
1403 if self.params.get('ignoreerrors'):
1404 self.report_error(str(e), tb=encode_compat_str(traceback.format_exc()))
1405 else:
1406 raise
1407 break
a0566bbf 1408 return wrapper
1409
f2ebc5c7 1410 def _wait_for_video(self, ie_result):
1411 if (not self.params.get('wait_for_video')
1412 or ie_result.get('_type', 'video') != 'video'
1413 or ie_result.get('formats') or ie_result.get('url')):
1414 return
1415
1416 format_dur = lambda dur: '%02d:%02d:%02d' % timetuple_from_msec(dur * 1000)[:-1]
1417 last_msg = ''
1418
1419 def progress(msg):
1420 nonlocal last_msg
1421 self.to_screen(msg + ' ' * (len(last_msg) - len(msg)) + '\r', skip_eol=True)
1422 last_msg = msg
1423
1424 min_wait, max_wait = self.params.get('wait_for_video')
1425 diff = try_get(ie_result, lambda x: x['release_timestamp'] - time.time())
1426 if diff is None and ie_result.get('live_status') == 'is_upcoming':
16c620bc 1427 diff = round(random.uniform(min_wait, max_wait) if (max_wait and min_wait) else (max_wait or min_wait), 0)
f2ebc5c7 1428 self.report_warning('Release time of video is not known')
1429 elif (diff or 0) <= 0:
1430 self.report_warning('Video should already be available according to extracted info')
38d79fd1 1431 diff = min(max(diff or 0, min_wait or 0), max_wait or float('inf'))
f2ebc5c7 1432 self.to_screen(f'[wait] Waiting for {format_dur(diff)} - Press Ctrl+C to try now')
1433
1434 wait_till = time.time() + diff
1435 try:
1436 while True:
1437 diff = wait_till - time.time()
1438 if diff <= 0:
1439 progress('')
1440 raise ReExtractInfo('[wait] Wait period ended', expected=True)
1441 progress(f'[wait] Remaining time until next attempt: {self._format_screen(format_dur(diff), self.Styles.EMPHASIS)}')
1442 time.sleep(1)
1443 except KeyboardInterrupt:
1444 progress('')
1445 raise ReExtractInfo('[wait] Interrupted by user', expected=True)
1446 except BaseException as e:
1447 if not isinstance(e, ReExtractInfo):
1448 self.to_screen('')
1449 raise
1450
a0566bbf 1451 @__handle_extraction_exceptions
58f197b7 1452 def __extract_info(self, url, ie, download, extra_info, process):
a0566bbf 1453 ie_result = ie.extract(url)
1454 if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
1455 return
1456 if isinstance(ie_result, list):
1457 # Backwards compatibility: old IE result format
1458 ie_result = {
1459 '_type': 'compat_list',
1460 'entries': ie_result,
1461 }
e37d0efb 1462 if extra_info.get('original_url'):
1463 ie_result.setdefault('original_url', extra_info['original_url'])
a0566bbf 1464 self.add_default_extra_info(ie_result, ie, url)
1465 if process:
f2ebc5c7 1466 self._wait_for_video(ie_result)
a0566bbf 1467 return self.process_ie_result(ie_result, download, extra_info)
8222d8de 1468 else:
a0566bbf 1469 return ie_result
fe7e0c98 1470
ea38e55f 1471 def add_default_extra_info(self, ie_result, ie, url):
6033d980 1472 if url is not None:
1473 self.add_extra_info(ie_result, {
1474 'webpage_url': url,
1475 'original_url': url,
57ebfca3 1476 })
1477 webpage_url = ie_result.get('webpage_url')
1478 if webpage_url:
1479 self.add_extra_info(ie_result, {
1480 'webpage_url_basename': url_basename(webpage_url),
1481 'webpage_url_domain': get_domain(webpage_url),
6033d980 1482 })
1483 if ie is not None:
1484 self.add_extra_info(ie_result, {
1485 'extractor': ie.IE_NAME,
1486 'extractor_key': ie.ie_key(),
1487 })
ea38e55f 1488
58adec46 1489 def process_ie_result(self, ie_result, download=True, extra_info=None):
8222d8de
JMF
1490 """
1491 Take the result of the ie(may be modified) and resolve all unresolved
1492 references (URLs, playlist items).
1493
1494 It will also download the videos if 'download'.
1495 Returns the resolved ie_result.
1496 """
58adec46 1497 if extra_info is None:
1498 extra_info = {}
e8ee972c
PH
1499 result_type = ie_result.get('_type', 'video')
1500
057a5206 1501 if result_type in ('url', 'url_transparent'):
134c6ea8 1502 ie_result['url'] = sanitize_url(ie_result['url'])
e37d0efb 1503 if ie_result.get('original_url'):
1504 extra_info.setdefault('original_url', ie_result['original_url'])
1505
057a5206 1506 extract_flat = self.params.get('extract_flat', False)
3089bc74
S
1507 if ((extract_flat == 'in_playlist' and 'playlist' in extra_info)
1508 or extract_flat is True):
ecb54191 1509 info_copy = ie_result.copy()
6033d980 1510 ie = try_get(ie_result.get('ie_key'), self.get_info_extractor)
360167b9 1511 if ie and not ie_result.get('id'):
4614bc22 1512 info_copy['id'] = ie.get_temp_id(ie_result['url'])
6033d980 1513 self.add_default_extra_info(info_copy, ie, ie_result['url'])
4614bc22 1514 self.add_extra_info(info_copy, extra_info)
b5475f11 1515 info_copy, _ = self.pre_process(info_copy)
ecb54191 1516 self.__forced_printings(info_copy, self.prepare_filename(info_copy), incomplete=True)
4614bc22 1517 if self.params.get('force_write_download_archive', False):
1518 self.record_download_archive(info_copy)
e8ee972c
PH
1519 return ie_result
1520
8222d8de 1521 if result_type == 'video':
b6c45014 1522 self.add_extra_info(ie_result, extra_info)
9c2b75b5 1523 ie_result = self.process_video_result(ie_result, download=download)
28b0eb0f 1524 additional_urls = (ie_result or {}).get('additional_urls')
9c2b75b5 1525 if additional_urls:
e9f4ccd1 1526 # TODO: Improve MetadataParserPP to allow setting a list
9c2b75b5 1527 if isinstance(additional_urls, compat_str):
1528 additional_urls = [additional_urls]
1529 self.to_screen(
1530 '[info] %s: %d additional URL(s) requested' % (ie_result['id'], len(additional_urls)))
1531 self.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls))
1532 ie_result['additional_entries'] = [
1533 self.extract_info(
b69fd25c 1534 url, download, extra_info=extra_info,
9c2b75b5 1535 force_generic_extractor=self.params.get('force_generic_extractor'))
1536 for url in additional_urls
1537 ]
1538 return ie_result
8222d8de
JMF
1539 elif result_type == 'url':
1540 # We have to add extra_info to the results because it may be
1541 # contained in a playlist
07cce701 1542 return self.extract_info(
1543 ie_result['url'], download,
1544 ie_key=ie_result.get('ie_key'),
1545 extra_info=extra_info)
7fc3fa05
PH
1546 elif result_type == 'url_transparent':
1547 # Use the information from the embedding page
1548 info = self.extract_info(
1549 ie_result['url'], ie_key=ie_result.get('ie_key'),
1550 extra_info=extra_info, download=False, process=False)
1551
1640eb09
S
1552 # extract_info may return None when ignoreerrors is enabled and
1553 # extraction failed with an error, don't crash and return early
1554 # in this case
1555 if not info:
1556 return info
1557
412c617d 1558 new_result = info.copy()
90137ca4 1559 new_result.update(filter_dict(ie_result, lambda k, v: (
1560 v is not None and k not in {'_type', 'url', 'id', 'extractor', 'extractor_key', 'ie_key'})))
7fc3fa05 1561
0563f7ac
S
1562 # Extracted info may not be a video result (i.e.
1563 # info.get('_type', 'video') != video) but rather an url or
1564 # url_transparent. In such cases outer metadata (from ie_result)
1565 # should be propagated to inner one (info). For this to happen
1566 # _type of info should be overridden with url_transparent. This
067aa17e 1567 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
0563f7ac
S
1568 if new_result.get('_type') == 'url':
1569 new_result['_type'] = 'url_transparent'
7fc3fa05
PH
1570
1571 return self.process_ie_result(
1572 new_result, download=download, extra_info=extra_info)
40fcba5e 1573 elif result_type in ('playlist', 'multi_video'):
30a074c2 1574 # Protect from infinite recursion due to recursively nested playlists
1575 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
1576 webpage_url = ie_result['webpage_url']
1577 if webpage_url in self._playlist_urls:
7e85e872 1578 self.to_screen(
30a074c2 1579 '[download] Skipping already downloaded playlist: %s'
1580 % ie_result.get('title') or ie_result.get('id'))
1581 return
7e85e872 1582
30a074c2 1583 self._playlist_level += 1
1584 self._playlist_urls.add(webpage_url)
03f83004 1585 self._fill_common_fields(ie_result, False)
bc516a3f 1586 self._sanitize_thumbnails(ie_result)
30a074c2 1587 try:
1588 return self.__process_playlist(ie_result, download)
1589 finally:
1590 self._playlist_level -= 1
1591 if not self._playlist_level:
1592 self._playlist_urls.clear()
8222d8de 1593 elif result_type == 'compat_list':
c9bf4114
PH
1594 self.report_warning(
1595 'Extractor %s returned a compat_list result. '
1596 'It needs to be updated.' % ie_result.get('extractor'))
5f6a1245 1597
8222d8de 1598 def _fixup(r):
b868936c 1599 self.add_extra_info(r, {
1600 'extractor': ie_result['extractor'],
1601 'webpage_url': ie_result['webpage_url'],
1602 'webpage_url_basename': url_basename(ie_result['webpage_url']),
0bb322b9 1603 'webpage_url_domain': get_domain(ie_result['webpage_url']),
b868936c 1604 'extractor_key': ie_result['extractor_key'],
1605 })
8222d8de
JMF
1606 return r
1607 ie_result['entries'] = [
b6c45014 1608 self.process_ie_result(_fixup(r), download, extra_info)
8222d8de
JMF
1609 for r in ie_result['entries']
1610 ]
1611 return ie_result
1612 else:
1613 raise Exception('Invalid result type: %s' % result_type)
1614
e92caff5 1615 def _ensure_dir_exists(self, path):
1616 return make_dir(path, self.report_error)
1617
3b603dbd 1618 @staticmethod
1619 def _playlist_infodict(ie_result, **kwargs):
1620 return {
1621 **ie_result,
1622 'playlist': ie_result.get('title') or ie_result.get('id'),
1623 'playlist_id': ie_result.get('id'),
1624 'playlist_title': ie_result.get('title'),
1625 'playlist_uploader': ie_result.get('uploader'),
1626 'playlist_uploader_id': ie_result.get('uploader_id'),
1627 'playlist_index': 0,
1628 **kwargs,
1629 }
1630
30a074c2 1631 def __process_playlist(self, ie_result, download):
1632 # We process each entry in the playlist
1633 playlist = ie_result.get('title') or ie_result.get('id')
1634 self.to_screen('[download] Downloading playlist: %s' % playlist)
1635
498f5606 1636 if 'entries' not in ie_result:
aa9369a2 1637 raise EntryNotInPlaylist('There are no entries')
7c7f7161 1638
1639 MissingEntry = object()
498f5606 1640 incomplete_entries = bool(ie_result.get('requested_entries'))
1641 if incomplete_entries:
bf5f605e 1642 def fill_missing_entries(entries, indices):
7c7f7161 1643 ret = [MissingEntry] * max(indices)
bf5f605e 1644 for i, entry in zip(indices, entries):
498f5606 1645 ret[i - 1] = entry
1646 return ret
1647 ie_result['entries'] = fill_missing_entries(ie_result['entries'], ie_result['requested_entries'])
02fd60d3 1648
30a074c2 1649 playlist_results = []
1650
56a8fb4f 1651 playliststart = self.params.get('playliststart', 1)
30a074c2 1652 playlistend = self.params.get('playlistend')
1653 # For backwards compatibility, interpret -1 as whole list
1654 if playlistend == -1:
1655 playlistend = None
1656
1657 playlistitems_str = self.params.get('playlist_items')
1658 playlistitems = None
1659 if playlistitems_str is not None:
1660 def iter_playlistitems(format):
1661 for string_segment in format.split(','):
1662 if '-' in string_segment:
1663 start, end = string_segment.split('-')
1664 for item in range(int(start), int(end) + 1):
1665 yield int(item)
1666 else:
1667 yield int(string_segment)
1668 playlistitems = orderedSet(iter_playlistitems(playlistitems_str))
1669
1670 ie_entries = ie_result['entries']
8e5fecc8 1671 if isinstance(ie_entries, list):
ed8d87f9 1672 playlist_count = len(ie_entries)
f0d785d3 1673 msg = f'Collected {playlist_count} videos; downloading %d of them'
1674 ie_result['playlist_count'] = ie_result.get('playlist_count') or playlist_count
1675
8e5fecc8 1676 def get_entry(i):
1677 return ie_entries[i - 1]
1678 else:
f0d785d3 1679 msg = 'Downloading %d videos'
c586f9e8 1680 if not isinstance(ie_entries, (PagedList, LazyList)):
8e5fecc8 1681 ie_entries = LazyList(ie_entries)
d37707bd 1682 elif isinstance(ie_entries, InAdvancePagedList):
1683 if ie_entries._pagesize == 1:
1684 playlist_count = ie_entries._pagecount
8e5fecc8 1685
1686 def get_entry(i):
1687 return YoutubeDL.__handle_extraction_exceptions(
1688 lambda self, i: ie_entries[i - 1]
1689 )(self, i)
50fed816 1690
f0d785d3 1691 entries, broken = [], False
ff1c7fc9 1692 items = playlistitems if playlistitems is not None else itertools.count(playliststart)
1693 for i in items:
1694 if i == 0:
1695 continue
56a8fb4f 1696 if playlistitems is None and playlistend is not None and playlistend < i:
1697 break
1698 entry = None
1699 try:
50fed816 1700 entry = get_entry(i)
7c7f7161 1701 if entry is MissingEntry:
498f5606 1702 raise EntryNotInPlaylist()
56a8fb4f 1703 except (IndexError, EntryNotInPlaylist):
1704 if incomplete_entries:
aa9369a2 1705 raise EntryNotInPlaylist(f'Entry {i} cannot be found')
56a8fb4f 1706 elif not playlistitems:
1707 break
1708 entries.append(entry)
120fe513 1709 try:
1710 if entry is not None:
e5a998f3 1711 # TODO: Add auto-generated fields
120fe513 1712 self._match_entry(entry, incomplete=True, silent=True)
1713 except (ExistingVideoReached, RejectedVideoReached):
f0d785d3 1714 broken = True
120fe513 1715 break
56a8fb4f 1716 ie_result['entries'] = entries
30a074c2 1717
56a8fb4f 1718 # Save playlist_index before re-ordering
1719 entries = [
9e598870 1720 ((playlistitems[i - 1] if playlistitems else i + playliststart - 1), entry)
56a8fb4f 1721 for i, entry in enumerate(entries, 1)
1722 if entry is not None]
1723 n_entries = len(entries)
498f5606 1724
f0d785d3 1725 if not (ie_result.get('playlist_count') or broken or playlistitems or playlistend):
1726 ie_result['playlist_count'] = n_entries
1727
e08a85d8 1728 if not playlistitems and (playliststart != 1 or playlistend):
56a8fb4f 1729 playlistitems = list(range(playliststart, playliststart + n_entries))
498f5606 1730 ie_result['requested_entries'] = playlistitems
1731
e08a85d8 1732 _infojson_written = False
0bfc53d0 1733 write_playlist_files = self.params.get('allow_playlist_files', True)
1734 if write_playlist_files and self.params.get('list_thumbnails'):
1735 self.list_thumbnails(ie_result)
1736 if write_playlist_files and not self.params.get('simulate'):
3b603dbd 1737 ie_copy = self._playlist_infodict(ie_result, n_entries=n_entries)
e08a85d8 1738 _infojson_written = self._write_info_json(
1739 'playlist', ie_result, self.prepare_filename(ie_copy, 'pl_infojson'))
1740 if _infojson_written is None:
80c03fa9 1741 return
1742 if self._write_description('playlist', ie_result,
1743 self.prepare_filename(ie_copy, 'pl_description')) is None:
1744 return
681de68e 1745 # TODO: This should be passed to ThumbnailsConvertor if necessary
80c03fa9 1746 self._write_thumbnails('playlist', ie_copy, self.prepare_filename(ie_copy, 'pl_thumbnail'))
30a074c2 1747
1748 if self.params.get('playlistreverse', False):
1749 entries = entries[::-1]
30a074c2 1750 if self.params.get('playlistrandom', False):
1751 random.shuffle(entries)
1752
1753 x_forwarded_for = ie_result.get('__x_forwarded_for_ip')
1754
86e5f3ed 1755 self.to_screen(f'[{ie_result["extractor"]}] playlist {playlist}: {msg % n_entries}')
26e2805c 1756 failures = 0
1757 max_failures = self.params.get('skip_playlist_after_errors') or float('inf')
71729754 1758 for i, entry_tuple in enumerate(entries, 1):
1759 playlist_index, entry = entry_tuple
81139999 1760 if 'playlist-index' in self.params.get('compat_opts', []):
1761 playlist_index = playlistitems[i - 1] if playlistitems else i + playliststart - 1
19a03940 1762 self.to_screen('[download] Downloading video %s of %s' % (
1763 self._format_screen(i, self.Styles.ID), self._format_screen(n_entries, self.Styles.EMPHASIS)))
30a074c2 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
cb96c5be 1800 if _infojson_written is True and self._write_info_json(
e08a85d8 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: '
86e5f3ed 1931 '{}\n\t{}\n\t{}^'.format(note, format_spec, ' ' * start[1]))
67134eab
JMF
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 2034 else:
86e5f3ed 2035 raise syntax_error(f'Operator not recognized: "{string}"', start)
67134eab
JMF
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):
316f2650 2159 formats = list(_check_formats(
2160 f for f in ctx['formats'] if f.get('vcodec') != 'none' or f.get('acodec') != 'none'))
e01d6aa4 2161 if not formats:
2162 return
921b76ca 2163 merged_format = formats[-1]
2164 for f in formats[-2::-1]:
f8d4ad9a 2165 merged_format = _merge((merged_format, f))
2166 yield merged_format
909d24dd 2167
2168 else:
85e801a9 2169 format_fallback, seperate_fallback, format_reverse, format_idx = False, None, True, 1
eff63539 2170 mobj = re.match(
2171 r'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
2172 format_spec)
2173 if mobj is not None:
2174 format_idx = int_or_none(mobj.group('n'), default=1)
e8e73840 2175 format_reverse = mobj.group('bw')[0] == 'b'
eff63539 2176 format_type = (mobj.group('type') or [None])[0]
2177 not_format_type = {'v': 'a', 'a': 'v'}.get(format_type)
2178 format_modified = mobj.group('mod') is not None
909d24dd 2179
2180 format_fallback = not format_type and not format_modified # for b, w
8326b00a 2181 _filter_f = (
eff63539 2182 (lambda f: f.get('%scodec' % format_type) != 'none')
2183 if format_type and format_modified # bv*, ba*, wv*, wa*
2184 else (lambda f: f.get('%scodec' % not_format_type) == 'none')
2185 if format_type # bv, ba, wv, wa
2186 else (lambda f: f.get('vcodec') != 'none' and f.get('acodec') != 'none')
2187 if not format_modified # b, w
8326b00a 2188 else lambda f: True) # b*, w*
2189 filter_f = lambda f: _filter_f(f) and (
2190 f.get('vcodec') != 'none' or f.get('acodec') != 'none')
67134eab 2191 else:
48ee10ee 2192 if format_spec in self._format_selection_exts['audio']:
b11c04a8 2193 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none'
48ee10ee 2194 elif format_spec in self._format_selection_exts['video']:
b11c04a8 2195 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none' and f.get('vcodec') != 'none'
85e801a9 2196 seperate_fallback = lambda f: f.get('ext') == format_spec and f.get('vcodec') != 'none'
48ee10ee 2197 elif format_spec in self._format_selection_exts['storyboards']:
b11c04a8 2198 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') == 'none' and f.get('vcodec') == 'none'
2199 else:
b5ae35ee 2200 filter_f = lambda f: f.get('format_id') == format_spec # id
909d24dd 2201
2202 def selector_function(ctx):
2203 formats = list(ctx['formats'])
909d24dd 2204 matches = list(filter(filter_f, formats)) if filter_f is not None else formats
85e801a9 2205 if not matches:
2206 if format_fallback and ctx['incomplete_formats']:
2207 # for extractors with incomplete formats (audio only (soundcloud)
2208 # or video only (imgur)) best/worst will fallback to
2209 # best/worst {video,audio}-only format
2210 matches = formats
2211 elif seperate_fallback and not ctx['has_merged_format']:
2212 # for compatibility with youtube-dl when there is no pre-merged format
2213 matches = list(filter(seperate_fallback, formats))
981052c9 2214 matches = LazyList(_check_formats(matches[::-1 if format_reverse else 1]))
2215 try:
e8e73840 2216 yield matches[format_idx - 1]
4abea8ca 2217 except LazyList.IndexError:
981052c9 2218 return
083c9df9 2219
67134eab 2220 filters = [self._build_format_filter(f) for f in selector.filters]
083c9df9 2221
317f7ab6 2222 def final_selector(ctx):
adbc4ec4 2223 ctx_copy = dict(ctx)
67134eab 2224 for _filter in filters:
317f7ab6
S
2225 ctx_copy['formats'] = list(filter(_filter, ctx_copy['formats']))
2226 return selector_function(ctx_copy)
67134eab 2227 return final_selector
083c9df9 2228
67134eab 2229 stream = io.BytesIO(format_spec.encode('utf-8'))
0130afb7 2230 try:
f9934b96 2231 tokens = list(_remove_unused_ops(tokenize.tokenize(stream.readline)))
0130afb7
JMF
2232 except tokenize.TokenError:
2233 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec)))
2234
86e5f3ed 2235 class TokenIterator:
0130afb7
JMF
2236 def __init__(self, tokens):
2237 self.tokens = tokens
2238 self.counter = 0
2239
2240 def __iter__(self):
2241 return self
2242
2243 def __next__(self):
2244 if self.counter >= len(self.tokens):
2245 raise StopIteration()
2246 value = self.tokens[self.counter]
2247 self.counter += 1
2248 return value
2249
2250 next = __next__
2251
2252 def restore_last_token(self):
2253 self.counter -= 1
2254
2255 parsed_selector = _parse_format_selection(iter(TokenIterator(tokens)))
67134eab 2256 return _build_selector_function(parsed_selector)
a9c58ad9 2257
e5660ee6 2258 def _calc_headers(self, info_dict):
8b7539d2 2259 res = merge_headers(self.params['http_headers'], info_dict.get('http_headers') or {})
e5660ee6
JMF
2260
2261 cookies = self._calc_cookies(info_dict)
2262 if cookies:
2263 res['Cookie'] = cookies
2264
0016b84e
S
2265 if 'X-Forwarded-For' not in res:
2266 x_forwarded_for_ip = info_dict.get('__x_forwarded_for_ip')
2267 if x_forwarded_for_ip:
2268 res['X-Forwarded-For'] = x_forwarded_for_ip
2269
e5660ee6
JMF
2270 return res
2271
2272 def _calc_cookies(self, info_dict):
5c2266df 2273 pr = sanitized_Request(info_dict['url'])
e5660ee6 2274 self.cookiejar.add_cookie_header(pr)
662435f7 2275 return pr.get_header('Cookie')
e5660ee6 2276
9f1a1c36 2277 def _sort_thumbnails(self, thumbnails):
2278 thumbnails.sort(key=lambda t: (
2279 t.get('preference') if t.get('preference') is not None else -1,
2280 t.get('width') if t.get('width') is not None else -1,
2281 t.get('height') if t.get('height') is not None else -1,
2282 t.get('id') if t.get('id') is not None else '',
2283 t.get('url')))
2284
b0249bca 2285 def _sanitize_thumbnails(self, info_dict):
bc516a3f 2286 thumbnails = info_dict.get('thumbnails')
2287 if thumbnails is None:
2288 thumbnail = info_dict.get('thumbnail')
2289 if thumbnail:
2290 info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}]
9f1a1c36 2291 if not thumbnails:
2292 return
2293
2294 def check_thumbnails(thumbnails):
2295 for t in thumbnails:
2296 self.to_screen(f'[info] Testing thumbnail {t["id"]}')
2297 try:
2298 self.urlopen(HEADRequest(t['url']))
2299 except network_exceptions as err:
2300 self.to_screen(f'[info] Unable to connect to thumbnail {t["id"]} URL {t["url"]!r} - {err}. Skipping...')
2301 continue
2302 yield t
2303
2304 self._sort_thumbnails(thumbnails)
2305 for i, t in enumerate(thumbnails):
2306 if t.get('id') is None:
2307 t['id'] = '%d' % i
2308 if t.get('width') and t.get('height'):
2309 t['resolution'] = '%dx%d' % (t['width'], t['height'])
2310 t['url'] = sanitize_url(t['url'])
2311
2312 if self.params.get('check_formats') is True:
282f5709 2313 info_dict['thumbnails'] = LazyList(check_thumbnails(thumbnails[::-1]), reverse=True)
9f1a1c36 2314 else:
2315 info_dict['thumbnails'] = thumbnails
bc516a3f 2316
03f83004
LNO
2317 def _fill_common_fields(self, info_dict, is_video=True):
2318 # TODO: move sanitization here
2319 if is_video:
2320 # playlists are allowed to lack "title"
2321 info_dict['fulltitle'] = info_dict.get('title')
2322 if 'title' not in info_dict:
2323 raise ExtractorError('Missing "title" field in extractor result',
2324 video_id=info_dict['id'], ie=info_dict['extractor'])
2325 elif not info_dict.get('title'):
2326 self.report_warning('Extractor failed to obtain "title". Creating a generic title instead')
2327 info_dict['title'] = f'{info_dict["extractor"]} video #{info_dict["id"]}'
2328
2329 if info_dict.get('duration') is not None:
2330 info_dict['duration_string'] = formatSeconds(info_dict['duration'])
2331
2332 for ts_key, date_key in (
2333 ('timestamp', 'upload_date'),
2334 ('release_timestamp', 'release_date'),
2335 ('modified_timestamp', 'modified_date'),
2336 ):
2337 if info_dict.get(date_key) is None and info_dict.get(ts_key) is not None:
2338 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
2339 # see http://bugs.python.org/issue1646728)
19a03940 2340 with contextlib.suppress(ValueError, OverflowError, OSError):
03f83004
LNO
2341 upload_date = datetime.datetime.utcfromtimestamp(info_dict[ts_key])
2342 info_dict[date_key] = upload_date.strftime('%Y%m%d')
03f83004
LNO
2343
2344 live_keys = ('is_live', 'was_live')
2345 live_status = info_dict.get('live_status')
2346 if live_status is None:
2347 for key in live_keys:
2348 if info_dict.get(key) is False:
2349 continue
2350 if info_dict.get(key):
2351 live_status = key
2352 break
2353 if all(info_dict.get(key) is False for key in live_keys):
2354 live_status = 'not_live'
2355 if live_status:
2356 info_dict['live_status'] = live_status
2357 for key in live_keys:
2358 if info_dict.get(key) is None:
2359 info_dict[key] = (live_status == key)
2360
2361 # Auto generate title fields corresponding to the *_number fields when missing
2362 # in order to always have clean titles. This is very common for TV series.
2363 for field in ('chapter', 'season', 'episode'):
2364 if info_dict.get('%s_number' % field) is not None and not info_dict.get(field):
2365 info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field])
2366
dd82ffea
JMF
2367 def process_video_result(self, info_dict, download=True):
2368 assert info_dict.get('_type', 'video') == 'video'
9c906919 2369 self._num_videos += 1
dd82ffea 2370
bec1fad2 2371 if 'id' not in info_dict:
fc08bdd6 2372 raise ExtractorError('Missing "id" field in extractor result', ie=info_dict['extractor'])
2373 elif not info_dict.get('id'):
2374 raise ExtractorError('Extractor failed to obtain "id"', ie=info_dict['extractor'])
455a15e2 2375
c9969434
S
2376 def report_force_conversion(field, field_not, conversion):
2377 self.report_warning(
2378 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
2379 % (field, field_not, conversion))
2380
2381 def sanitize_string_field(info, string_field):
2382 field = info.get(string_field)
2383 if field is None or isinstance(field, compat_str):
2384 return
2385 report_force_conversion(string_field, 'a string', 'string')
2386 info[string_field] = compat_str(field)
2387
2388 def sanitize_numeric_fields(info):
2389 for numeric_field in self._NUMERIC_FIELDS:
2390 field = info.get(numeric_field)
f9934b96 2391 if field is None or isinstance(field, (int, float)):
c9969434
S
2392 continue
2393 report_force_conversion(numeric_field, 'numeric', 'int')
2394 info[numeric_field] = int_or_none(field)
2395
2396 sanitize_string_field(info_dict, 'id')
2397 sanitize_numeric_fields(info_dict)
4c3f8c3f 2398 if (info_dict.get('duration') or 0) <= 0 and info_dict.pop('duration', None):
50e93e03 2399 self.report_warning('"duration" field is negative, there is an error in extractor')
be6217b2 2400
dd82ffea
JMF
2401 if 'playlist' not in info_dict:
2402 # It isn't part of a playlist
2403 info_dict['playlist'] = None
2404 info_dict['playlist_index'] = None
2405
bc516a3f 2406 self._sanitize_thumbnails(info_dict)
d5519808 2407
536a55da 2408 thumbnail = info_dict.get('thumbnail')
bc516a3f 2409 thumbnails = info_dict.get('thumbnails')
536a55da
S
2410 if thumbnail:
2411 info_dict['thumbnail'] = sanitize_url(thumbnail)
2412 elif thumbnails:
d5519808
PH
2413 info_dict['thumbnail'] = thumbnails[-1]['url']
2414
ae30b840 2415 if info_dict.get('display_id') is None and 'id' in info_dict:
0afef30b
PH
2416 info_dict['display_id'] = info_dict['id']
2417
03f83004 2418 self._fill_common_fields(info_dict)
33d2fc2f 2419
05108a49
S
2420 for cc_kind in ('subtitles', 'automatic_captions'):
2421 cc = info_dict.get(cc_kind)
2422 if cc:
2423 for _, subtitle in cc.items():
2424 for subtitle_format in subtitle:
2425 if subtitle_format.get('url'):
2426 subtitle_format['url'] = sanitize_url(subtitle_format['url'])
2427 if subtitle_format.get('ext') is None:
2428 subtitle_format['ext'] = determine_ext(subtitle_format['url']).lower()
2429
2430 automatic_captions = info_dict.get('automatic_captions')
4bba3716 2431 subtitles = info_dict.get('subtitles')
4bba3716 2432
360e1ca5 2433 info_dict['requested_subtitles'] = self.process_subtitles(
05108a49 2434 info_dict['id'], subtitles, automatic_captions)
a504ced0 2435
dd82ffea
JMF
2436 if info_dict.get('formats') is None:
2437 # There's only one format available
2438 formats = [info_dict]
2439 else:
2440 formats = info_dict['formats']
2441
e0493e90 2442 info_dict['__has_drm'] = any(f.get('has_drm') for f in formats)
88acdbc2 2443 if not self.params.get('allow_unplayable_formats'):
2444 formats = [f for f in formats if not f.get('has_drm')]
c0b6e5c7 2445 if info_dict['__has_drm'] and all(
2446 f.get('acodec') == f.get('vcodec') == 'none' for f in formats):
2447 self.report_warning(
2448 'This video is DRM protected and only images are available for download. '
2449 'Use --list-formats to see them')
88acdbc2 2450
319b6059 2451 get_from_start = not info_dict.get('is_live') or bool(self.params.get('live_from_start'))
2452 if not get_from_start:
2453 info_dict['title'] += ' ' + datetime.datetime.now().strftime('%Y-%m-%d %H:%M')
2454 if info_dict.get('is_live') and formats:
adbc4ec4 2455 formats = [f for f in formats if bool(f.get('is_from_start')) == get_from_start]
319b6059 2456 if get_from_start and not formats:
a44ca5a4 2457 self.raise_no_formats(info_dict, msg=(
2458 '--live-from-start is passed, but there are no formats that can be downloaded from the start. '
2459 'If you want to download from the current time, use --no-live-from-start'))
adbc4ec4 2460
db95dc13 2461 if not formats:
1151c407 2462 self.raise_no_formats(info_dict)
db95dc13 2463
73af5cc8
S
2464 def is_wellformed(f):
2465 url = f.get('url')
a5ac0c47 2466 if not url:
73af5cc8
S
2467 self.report_warning(
2468 '"url" field is missing or empty - skipping format, '
2469 'there is an error in extractor')
a5ac0c47
S
2470 return False
2471 if isinstance(url, bytes):
2472 sanitize_string_field(f, 'url')
2473 return True
73af5cc8
S
2474
2475 # Filter out malformed formats for better extraction robustness
2476 formats = list(filter(is_wellformed, formats))
2477
181c7053
S
2478 formats_dict = {}
2479
dd82ffea 2480 # We check that all the formats have the format and format_id fields
db95dc13 2481 for i, format in enumerate(formats):
c9969434
S
2482 sanitize_string_field(format, 'format_id')
2483 sanitize_numeric_fields(format)
dcf77cf1 2484 format['url'] = sanitize_url(format['url'])
e74e3b63 2485 if not format.get('format_id'):
8016c922 2486 format['format_id'] = compat_str(i)
e2effb08
S
2487 else:
2488 # Sanitize format_id from characters used in format selector expression
ec85ded8 2489 format['format_id'] = re.sub(r'[\s,/+\[\]()]', '_', format['format_id'])
181c7053
S
2490 format_id = format['format_id']
2491 if format_id not in formats_dict:
2492 formats_dict[format_id] = []
2493 formats_dict[format_id].append(format)
2494
2495 # Make sure all formats have unique format_id
03b4de72 2496 common_exts = set(itertools.chain(*self._format_selection_exts.values()))
181c7053 2497 for format_id, ambiguous_formats in formats_dict.items():
48ee10ee 2498 ambigious_id = len(ambiguous_formats) > 1
2499 for i, format in enumerate(ambiguous_formats):
2500 if ambigious_id:
181c7053 2501 format['format_id'] = '%s-%d' % (format_id, i)
48ee10ee 2502 if format.get('ext') is None:
2503 format['ext'] = determine_ext(format['url']).lower()
2504 # Ensure there is no conflict between id and ext in format selection
2505 # See https://github.com/yt-dlp/yt-dlp/issues/1282
2506 if format['format_id'] != format['ext'] and format['format_id'] in common_exts:
2507 format['format_id'] = 'f%s' % format['format_id']
181c7053
S
2508
2509 for i, format in enumerate(formats):
8c51aa65 2510 if format.get('format') is None:
6febd1c1 2511 format['format'] = '{id} - {res}{note}'.format(
8c51aa65
JMF
2512 id=format['format_id'],
2513 res=self.format_resolution(format),
b868936c 2514 note=format_field(format, 'format_note', ' (%s)'),
8c51aa65 2515 )
6f0be937 2516 if format.get('protocol') is None:
b5559424 2517 format['protocol'] = determine_protocol(format)
239df021 2518 if format.get('resolution') is None:
2519 format['resolution'] = self.format_resolution(format, default=None)
176f1866 2520 if format.get('dynamic_range') is None and format.get('vcodec') != 'none':
2521 format['dynamic_range'] = 'SDR'
f2fe69c7 2522 if (info_dict.get('duration') and format.get('tbr')
2523 and not format.get('filesize') and not format.get('filesize_approx')):
2524 format['filesize_approx'] = info_dict['duration'] * format['tbr'] * (1024 / 8)
2525
e5660ee6
JMF
2526 # Add HTTP headers, so that external programs can use them from the
2527 # json output
2528 full_format_info = info_dict.copy()
2529 full_format_info.update(format)
2530 format['http_headers'] = self._calc_headers(full_format_info)
0016b84e
S
2531 # Remove private housekeeping stuff
2532 if '__x_forwarded_for_ip' in info_dict:
2533 del info_dict['__x_forwarded_for_ip']
dd82ffea 2534
9f1a1c36 2535 if self.params.get('check_formats') is True:
282f5709 2536 formats = LazyList(self._check_formats(formats[::-1]), reverse=True)
9f1a1c36 2537
88acdbc2 2538 if not formats or formats[0] is not info_dict:
b3d9ef88
JMF
2539 # only set the 'formats' fields if the original info_dict list them
2540 # otherwise we end up with a circular reference, the first (and unique)
f89197d7 2541 # element in the 'formats' field in info_dict is info_dict itself,
dfb1b146 2542 # which can't be exported to json
b3d9ef88 2543 info_dict['formats'] = formats
4ec82a72 2544
2545 info_dict, _ = self.pre_process(info_dict)
2546
6db9c4d5 2547 if self._match_entry(info_dict, incomplete=self._format_fields) is not None:
09b49e1f 2548 return info_dict
2549
2550 self.post_extract(info_dict)
2551 info_dict, _ = self.pre_process(info_dict, 'after_filter')
2552
093a1710 2553 # The pre-processors may have modified the formats
2554 formats = info_dict.get('formats', [info_dict])
2555
fa9f30b8 2556 list_only = self.params.get('simulate') is None and (
2557 self.params.get('list_thumbnails') or self.params.get('listformats') or self.params.get('listsubtitles'))
2558 interactive_format_selection = not list_only and self.format_selector == '-'
b7b04c78 2559 if self.params.get('list_thumbnails'):
2560 self.list_thumbnails(info_dict)
b7b04c78 2561 if self.params.get('listsubtitles'):
2562 if 'automatic_captions' in info_dict:
2563 self.list_subtitles(
2564 info_dict['id'], automatic_captions, 'automatic captions')
2565 self.list_subtitles(info_dict['id'], subtitles, 'subtitles')
fa9f30b8 2566 if self.params.get('listformats') or interactive_format_selection:
b69fd25c 2567 self.list_formats(info_dict)
169dbde9 2568 if list_only:
b7b04c78 2569 # Without this printing, -F --print-json will not work
169dbde9 2570 self.__forced_printings(info_dict, self.prepare_filename(info_dict), incomplete=True)
bfaae0a7 2571 return
2572
187986a8 2573 format_selector = self.format_selector
2574 if format_selector is None:
0017d9ad 2575 req_format = self._default_format_spec(info_dict, download=download)
0760b0a7 2576 self.write_debug('Default format spec: %s' % req_format)
187986a8 2577 format_selector = self.build_format_selector(req_format)
317f7ab6 2578
fa9f30b8 2579 while True:
2580 if interactive_format_selection:
2581 req_format = input(
2582 self._format_screen('\nEnter format selector: ', self.Styles.EMPHASIS))
2583 try:
2584 format_selector = self.build_format_selector(req_format)
2585 except SyntaxError as err:
2586 self.report_error(err, tb=False, is_error=False)
2587 continue
2588
85e801a9 2589 formats_to_download = list(format_selector({
fa9f30b8 2590 'formats': formats,
85e801a9 2591 'has_merged_format': any('none' not in (f.get('acodec'), f.get('vcodec')) for f in formats),
2592 'incomplete_formats': (
2593 # All formats are video-only or
2594 all(f.get('vcodec') != 'none' and f.get('acodec') == 'none' for f in formats)
2595 # all formats are audio-only
2596 or all(f.get('vcodec') == 'none' and f.get('acodec') != 'none' for f in formats)),
2597 }))
fa9f30b8 2598 if interactive_format_selection and not formats_to_download:
2599 self.report_error('Requested format is not available', tb=False, is_error=False)
2600 continue
2601 break
317f7ab6 2602
dd82ffea 2603 if not formats_to_download:
b7da73eb 2604 if not self.params.get('ignore_no_formats_error'):
c0b6e5c7 2605 raise ExtractorError(
2606 'Requested format is not available. Use --list-formats for a list of available formats',
2607 expected=True, video_id=info_dict['id'], ie=info_dict['extractor'])
b62fa6d7 2608 self.report_warning('Requested format is not available')
2609 # Process what we can, even without any available formats.
2610 formats_to_download = [{}]
a13e6848 2611
b62fa6d7 2612 best_format = formats_to_download[-1]
2613 if download:
2614 if best_format:
2615 self.to_screen(
2616 f'[info] {info_dict["id"]}: Downloading {len(formats_to_download)} format(s): '
2617 + ', '.join([f['format_id'] for f in formats_to_download]))
a13e6848 2618 max_downloads_reached = False
f46e2f9d 2619 for i, fmt in enumerate(formats_to_download):
09b49e1f 2620 formats_to_download[i] = new_info = self._copy_infodict(info_dict)
b7da73eb 2621 new_info.update(fmt)
a13e6848 2622 try:
2623 self.process_info(new_info)
2624 except MaxDownloadsReached:
2625 max_downloads_reached = True
f46e2f9d 2626 # Remove copied info
2627 for key, val in tuple(new_info.items()):
2628 if info_dict.get(key) == val:
2629 new_info.pop(key)
a13e6848 2630 if max_downloads_reached:
2631 break
ebed8b37 2632
86e5f3ed 2633 write_archive = {f.get('__write_download_archive', False) for f in formats_to_download}
a13e6848 2634 assert write_archive.issubset({True, False, 'ignore'})
2635 if True in write_archive and False not in write_archive:
2636 self.record_download_archive(info_dict)
be72c624 2637
2638 info_dict['requested_downloads'] = formats_to_download
ed5835b4 2639 info_dict = self.run_all_pps('after_video', info_dict)
a13e6848 2640 if max_downloads_reached:
2641 raise MaxDownloadsReached()
ebed8b37 2642
49a57e70 2643 # We update the info dict with the selected best quality format (backwards compatibility)
be72c624 2644 info_dict.update(best_format)
dd82ffea
JMF
2645 return info_dict
2646
98c70d6f 2647 def process_subtitles(self, video_id, normal_subtitles, automatic_captions):
a504ced0 2648 """Select the requested subtitles and their format"""
d8a58ddc 2649 available_subs, normal_sub_langs = {}, []
98c70d6f
JMF
2650 if normal_subtitles and self.params.get('writesubtitles'):
2651 available_subs.update(normal_subtitles)
d8a58ddc 2652 normal_sub_langs = tuple(normal_subtitles.keys())
98c70d6f
JMF
2653 if automatic_captions and self.params.get('writeautomaticsub'):
2654 for lang, cap_info in automatic_captions.items():
360e1ca5
JMF
2655 if lang not in available_subs:
2656 available_subs[lang] = cap_info
2657
4d171848
JMF
2658 if (not self.params.get('writesubtitles') and not
2659 self.params.get('writeautomaticsub') or not
2660 available_subs):
2661 return None
a504ced0 2662
d8a58ddc 2663 all_sub_langs = tuple(available_subs.keys())
a504ced0 2664 if self.params.get('allsubtitles', False):
c32b0aab 2665 requested_langs = all_sub_langs
2666 elif self.params.get('subtitleslangs', False):
77c4a9ef 2667 # A list is used so that the order of languages will be the same as
2668 # given in subtitleslangs. See https://github.com/yt-dlp/yt-dlp/issues/1041
2669 requested_langs = []
2670 for lang_re in self.params.get('subtitleslangs'):
77c4a9ef 2671 discard = lang_re[0] == '-'
c32b0aab 2672 if discard:
77c4a9ef 2673 lang_re = lang_re[1:]
3aa91540 2674 if lang_re == 'all':
2675 if discard:
2676 requested_langs = []
2677 else:
2678 requested_langs.extend(all_sub_langs)
2679 continue
77c4a9ef 2680 current_langs = filter(re.compile(lang_re + '$').match, all_sub_langs)
c32b0aab 2681 if discard:
2682 for lang in current_langs:
77c4a9ef 2683 while lang in requested_langs:
2684 requested_langs.remove(lang)
c32b0aab 2685 else:
77c4a9ef 2686 requested_langs.extend(current_langs)
2687 requested_langs = orderedSet(requested_langs)
d8a58ddc 2688 elif normal_sub_langs:
2689 requested_langs = ['en'] if 'en' in normal_sub_langs else normal_sub_langs[:1]
a504ced0 2690 else:
d8a58ddc 2691 requested_langs = ['en'] if 'en' in all_sub_langs else all_sub_langs[:1]
ad3dc496 2692 if requested_langs:
2693 self.write_debug('Downloading subtitles: %s' % ', '.join(requested_langs))
a504ced0
JMF
2694
2695 formats_query = self.params.get('subtitlesformat', 'best')
2696 formats_preference = formats_query.split('/') if formats_query else []
2697 subs = {}
2698 for lang in requested_langs:
2699 formats = available_subs.get(lang)
2700 if formats is None:
86e5f3ed 2701 self.report_warning(f'{lang} subtitles not available for {video_id}')
a504ced0 2702 continue
a504ced0
JMF
2703 for ext in formats_preference:
2704 if ext == 'best':
2705 f = formats[-1]
2706 break
2707 matches = list(filter(lambda f: f['ext'] == ext, formats))
2708 if matches:
2709 f = matches[-1]
2710 break
2711 else:
2712 f = formats[-1]
2713 self.report_warning(
2714 'No subtitle format found matching "%s" for language %s, '
2715 'using %s' % (formats_query, lang, f['ext']))
2716 subs[lang] = f
2717 return subs
2718
bb66c247 2719 def _forceprint(self, key, info_dict):
2720 if info_dict is None:
2721 return
2722 info_copy = info_dict.copy()
2723 info_copy['formats_table'] = self.render_formats_table(info_dict)
2724 info_copy['thumbnails_table'] = self.render_thumbnails_table(info_dict)
2725 info_copy['subtitles_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('subtitles'))
2726 info_copy['automatic_captions_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('automatic_captions'))
2727
2728 def format_tmpl(tmpl):
2729 mobj = re.match(r'\w+(=?)$', tmpl)
2730 if mobj and mobj.group(1):
2731 return f'{tmpl[:-1]} = %({tmpl[:-1]})r'
2732 elif mobj:
2733 return f'%({tmpl})s'
2734 return tmpl
8130779d 2735
bb66c247 2736 for tmpl in self.params['forceprint'].get(key, []):
2737 self.to_stdout(self.evaluate_outtmpl(format_tmpl(tmpl), info_copy))
2738
2739 for tmpl, file_tmpl in self.params['print_to_file'].get(key, []):
5127e92a 2740 filename = self.prepare_filename(info_dict, outtmpl=file_tmpl)
bb66c247 2741 tmpl = format_tmpl(tmpl)
2742 self.to_screen(f'[info] Writing {tmpl!r} to: {filename}')
8d93e69d 2743 if self._ensure_dir_exists(filename):
86e5f3ed 2744 with open(filename, 'a', encoding='utf-8') as f:
8d93e69d 2745 f.write(self.evaluate_outtmpl(tmpl, info_copy) + '\n')
ca30f449 2746
d06daf23 2747 def __forced_printings(self, info_dict, filename, incomplete):
53c18592 2748 def print_mandatory(field, actual_field=None):
2749 if actual_field is None:
2750 actual_field = field
d06daf23 2751 if (self.params.get('force%s' % field, False)
53c18592 2752 and (not incomplete or info_dict.get(actual_field) is not None)):
2753 self.to_stdout(info_dict[actual_field])
d06daf23
S
2754
2755 def print_optional(field):
2756 if (self.params.get('force%s' % field, False)
2757 and info_dict.get(field) is not None):
2758 self.to_stdout(info_dict[field])
2759
53c18592 2760 info_dict = info_dict.copy()
2761 if filename is not None:
2762 info_dict['filename'] = filename
2763 if info_dict.get('requested_formats') is not None:
2764 # For RTMP URLs, also include the playpath
2765 info_dict['urls'] = '\n'.join(f['url'] + f.get('play_path', '') for f in info_dict['requested_formats'])
10331a26 2766 elif info_dict.get('url'):
53c18592 2767 info_dict['urls'] = info_dict['url'] + info_dict.get('play_path', '')
2768
bb66c247 2769 if (self.params.get('forcejson')
2770 or self.params['forceprint'].get('video')
2771 or self.params['print_to_file'].get('video')):
2b8a2973 2772 self.post_extract(info_dict)
bb66c247 2773 self._forceprint('video', info_dict)
53c18592 2774
d06daf23
S
2775 print_mandatory('title')
2776 print_mandatory('id')
53c18592 2777 print_mandatory('url', 'urls')
d06daf23
S
2778 print_optional('thumbnail')
2779 print_optional('description')
53c18592 2780 print_optional('filename')
b868936c 2781 if self.params.get('forceduration') and info_dict.get('duration') is not None:
d06daf23
S
2782 self.to_stdout(formatSeconds(info_dict['duration']))
2783 print_mandatory('format')
53c18592 2784
2b8a2973 2785 if self.params.get('forcejson'):
6e84b215 2786 self.to_stdout(json.dumps(self.sanitize_info(info_dict)))
d06daf23 2787
e8e73840 2788 def dl(self, name, info, subtitle=False, test=False):
88acdbc2 2789 if not info.get('url'):
1151c407 2790 self.raise_no_formats(info, True)
e8e73840 2791
2792 if test:
2793 verbose = self.params.get('verbose')
2794 params = {
2795 'test': True,
a169858f 2796 'quiet': self.params.get('quiet') or not verbose,
e8e73840 2797 'verbose': verbose,
2798 'noprogress': not verbose,
2799 'nopart': True,
2800 'skip_unavailable_fragments': False,
2801 'keep_fragments': False,
2802 'overwrites': True,
2803 '_no_ytdl_file': True,
2804 }
2805 else:
2806 params = self.params
96fccc10 2807 fd = get_suitable_downloader(info, params, to_stdout=(name == '-'))(self, params)
e8e73840 2808 if not test:
2809 for ph in self._progress_hooks:
2810 fd.add_progress_hook(ph)
42676437
M
2811 urls = '", "'.join(
2812 (f['url'].split(',')[0] + ',<data>' if f['url'].startswith('data:') else f['url'])
2813 for f in info.get('requested_formats', []) or [info])
18e674b4 2814 self.write_debug('Invoking downloader on "%s"' % urls)
03b4de72 2815
adbc4ec4
THD
2816 # Note: Ideally info should be a deep-copied so that hooks cannot modify it.
2817 # But it may contain objects that are not deep-copyable
2818 new_info = self._copy_infodict(info)
e8e73840 2819 if new_info.get('http_headers') is None:
2820 new_info['http_headers'] = self._calc_headers(new_info)
2821 return fd.download(name, new_info, subtitle)
2822
e04938ab 2823 def existing_file(self, filepaths, *, default_overwrite=True):
2824 existing_files = list(filter(os.path.exists, orderedSet(filepaths)))
2825 if existing_files and not self.params.get('overwrites', default_overwrite):
2826 return existing_files[0]
2827
2828 for file in existing_files:
2829 self.report_file_delete(file)
2830 os.remove(file)
2831 return None
2832
8222d8de 2833 def process_info(self, info_dict):
09b49e1f 2834 """Process a single resolved IE result. (Modifies it in-place)"""
8222d8de
JMF
2835
2836 assert info_dict.get('_type', 'video') == 'video'
f46e2f9d 2837 original_infodict = info_dict
fd288278 2838
4513a41a 2839 if 'format' not in info_dict and 'ext' in info_dict:
8222d8de
JMF
2840 info_dict['format'] = info_dict['ext']
2841
09b49e1f 2842 # This is mostly just for backward compatibility of process_info
2843 # As a side-effect, this allows for format-specific filters
c77495e3 2844 if self._match_entry(info_dict) is not None:
9e907ebd 2845 info_dict['__write_download_archive'] = 'ignore'
8222d8de
JMF
2846 return
2847
09b49e1f 2848 # Does nothing under normal operation - for backward compatibility of process_info
277d6ff5 2849 self.post_extract(info_dict)
0c14d66a 2850 self._num_downloads += 1
8222d8de 2851
dcf64d43 2852 # info_dict['_filename'] needs to be set for backward compatibility
de6000d9 2853 info_dict['_filename'] = full_filename = self.prepare_filename(info_dict, warn=True)
2854 temp_filename = self.prepare_filename(info_dict, 'temp')
0202b52a 2855 files_to_move = {}
8222d8de
JMF
2856
2857 # Forced printings
4513a41a 2858 self.__forced_printings(info_dict, full_filename, incomplete=('format' not in info_dict))
8222d8de 2859
b7b04c78 2860 if self.params.get('simulate'):
9e907ebd 2861 info_dict['__write_download_archive'] = self.params.get('force_write_download_archive')
8222d8de
JMF
2862 return
2863
de6000d9 2864 if full_filename is None:
8222d8de 2865 return
e92caff5 2866 if not self._ensure_dir_exists(encodeFilename(full_filename)):
0202b52a 2867 return
e92caff5 2868 if not self._ensure_dir_exists(encodeFilename(temp_filename)):
8222d8de
JMF
2869 return
2870
80c03fa9 2871 if self._write_description('video', info_dict,
2872 self.prepare_filename(info_dict, 'description')) is None:
2873 return
2874
2875 sub_files = self._write_subtitles(info_dict, temp_filename)
2876 if sub_files is None:
2877 return
2878 files_to_move.update(dict(sub_files))
2879
2880 thumb_files = self._write_thumbnails(
2881 'video', info_dict, temp_filename, self.prepare_filename(info_dict, 'thumbnail'))
2882 if thumb_files is None:
2883 return
2884 files_to_move.update(dict(thumb_files))
8222d8de 2885
80c03fa9 2886 infofn = self.prepare_filename(info_dict, 'infojson')
2887 _infojson_written = self._write_info_json('video', info_dict, infofn)
2888 if _infojson_written:
dac5df5a 2889 info_dict['infojson_filename'] = infofn
e75bb0d6 2890 # For backward compatibility, even though it was a private field
80c03fa9 2891 info_dict['__infojson_filename'] = infofn
2892 elif _infojson_written is None:
2893 return
2894
2895 # Note: Annotations are deprecated
2896 annofn = None
1fb07d10 2897 if self.params.get('writeannotations', False):
de6000d9 2898 annofn = self.prepare_filename(info_dict, 'annotation')
80c03fa9 2899 if annofn:
e92caff5 2900 if not self._ensure_dir_exists(encodeFilename(annofn)):
0202b52a 2901 return
0c3d0f51 2902 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(annofn)):
6febd1c1 2903 self.to_screen('[info] Video annotations are already present')
ffddb112
RA
2904 elif not info_dict.get('annotations'):
2905 self.report_warning('There are no annotations to write.')
7b6fefc9
PH
2906 else:
2907 try:
6febd1c1 2908 self.to_screen('[info] Writing video annotations to: ' + annofn)
86e5f3ed 2909 with open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
7b6fefc9
PH
2910 annofile.write(info_dict['annotations'])
2911 except (KeyError, TypeError):
6febd1c1 2912 self.report_warning('There are no annotations to write.')
86e5f3ed 2913 except OSError:
6febd1c1 2914 self.report_error('Cannot write annotations file: ' + annofn)
7b6fefc9 2915 return
1fb07d10 2916
732044af 2917 # Write internet shortcut files
08438d2c 2918 def _write_link_file(link_type):
60f3e995 2919 url = try_get(info_dict['webpage_url'], iri_to_uri)
2920 if not url:
2921 self.report_warning(
2922 f'Cannot write internet shortcut file because the actual URL of "{info_dict["webpage_url"]}" is unknown')
2923 return True
08438d2c 2924 linkfn = replace_extension(self.prepare_filename(info_dict, 'link'), link_type, info_dict.get('ext'))
0e6b018a
Z
2925 if not self._ensure_dir_exists(encodeFilename(linkfn)):
2926 return False
10e3742e 2927 if self.params.get('overwrites', True) and os.path.exists(encodeFilename(linkfn)):
08438d2c 2928 self.to_screen(f'[info] Internet shortcut (.{link_type}) is already present')
2929 return True
2930 try:
2931 self.to_screen(f'[info] Writing internet shortcut (.{link_type}) to: {linkfn}')
86e5f3ed 2932 with open(encodeFilename(to_high_limit_path(linkfn)), 'w', encoding='utf-8',
2933 newline='\r\n' if link_type == 'url' else '\n') as linkfile:
60f3e995 2934 template_vars = {'url': url}
08438d2c 2935 if link_type == 'desktop':
2936 template_vars['filename'] = linkfn[:-(len(link_type) + 1)]
2937 linkfile.write(LINK_TEMPLATES[link_type] % template_vars)
86e5f3ed 2938 except OSError:
08438d2c 2939 self.report_error(f'Cannot write internet shortcut {linkfn}')
2940 return False
732044af 2941 return True
2942
08438d2c 2943 write_links = {
2944 'url': self.params.get('writeurllink'),
2945 'webloc': self.params.get('writewebloclink'),
2946 'desktop': self.params.get('writedesktoplink'),
2947 }
2948 if self.params.get('writelink'):
2949 link_type = ('webloc' if sys.platform == 'darwin'
2950 else 'desktop' if sys.platform.startswith('linux')
2951 else 'url')
2952 write_links[link_type] = True
2953
2954 if any(should_write and not _write_link_file(link_type)
2955 for link_type, should_write in write_links.items()):
2956 return
732044af 2957
f46e2f9d 2958 def replace_info_dict(new_info):
2959 nonlocal info_dict
2960 if new_info == info_dict:
2961 return
2962 info_dict.clear()
2963 info_dict.update(new_info)
2964
56d868db 2965 try:
f46e2f9d 2966 new_info, files_to_move = self.pre_process(info_dict, 'before_dl', files_to_move)
2967 replace_info_dict(new_info)
56d868db 2968 except PostProcessingError as err:
2969 self.report_error('Preprocessing: %s' % str(err))
2970 return
2971
a13e6848 2972 if self.params.get('skip_download'):
56d868db 2973 info_dict['filepath'] = temp_filename
2974 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
2975 info_dict['__files_to_move'] = files_to_move
f46e2f9d 2976 replace_info_dict(self.run_pp(MoveFilesAfterDownloadPP(self, False), info_dict))
9e907ebd 2977 info_dict['__write_download_archive'] = self.params.get('force_write_download_archive')
56d868db 2978 else:
2979 # Download
b868936c 2980 info_dict.setdefault('__postprocessors', [])
4340deca 2981 try:
0202b52a 2982
e04938ab 2983 def existing_video_file(*filepaths):
6b591b29 2984 ext = info_dict.get('ext')
e04938ab 2985 converted = lambda file: replace_extension(file, self.params.get('final_ext') or ext, ext)
2986 file = self.existing_file(itertools.chain(*zip(map(converted, filepaths), filepaths)),
2987 default_overwrite=False)
2988 if file:
2989 info_dict['ext'] = os.path.splitext(file)[1][1:]
2990 return file
0202b52a 2991
2992 success = True
4340deca 2993 if info_dict.get('requested_formats') is not None:
81cd954a
S
2994
2995 def compatible_formats(formats):
d03cfdce 2996 # TODO: some formats actually allow this (mkv, webm, ogg, mp4), but not all of them.
2997 video_formats = [format for format in formats if format.get('vcodec') != 'none']
2998 audio_formats = [format for format in formats if format.get('acodec') != 'none']
2999 if len(video_formats) > 2 or len(audio_formats) > 2:
3000 return False
3001
81cd954a 3002 # Check extension
86e5f3ed 3003 exts = {format.get('ext') for format in formats}
d03cfdce 3004 COMPATIBLE_EXTS = (
86e5f3ed 3005 {'mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma'},
3006 {'webm'},
d03cfdce 3007 )
3008 for ext_sets in COMPATIBLE_EXTS:
3009 if ext_sets.issuperset(exts):
3010 return True
81cd954a
S
3011 # TODO: Check acodec/vcodec
3012 return False
3013
3014 requested_formats = info_dict['requested_formats']
0202b52a 3015 old_ext = info_dict['ext']
4e3b637d 3016 if self.params.get('merge_output_format') is None:
3017 if not compatible_formats(requested_formats):
3018 info_dict['ext'] = 'mkv'
3019 self.report_warning(
3020 'Requested formats are incompatible for merge and will be merged into mkv')
3021 if (info_dict['ext'] == 'webm'
3022 and info_dict.get('thumbnails')
3023 # check with type instead of pp_key, __name__, or isinstance
3024 # since we dont want any custom PPs to trigger this
3025 and any(type(pp) == EmbedThumbnailPP for pp in self._pps['post_process'])):
3026 info_dict['ext'] = 'mkv'
3027 self.report_warning(
3028 'webm doesn\'t support embedding a thumbnail, mkv will be used')
124bc071 3029 new_ext = info_dict['ext']
0202b52a 3030
124bc071 3031 def correct_ext(filename, ext=new_ext):
96fccc10 3032 if filename == '-':
3033 return filename
0202b52a 3034 filename_real_ext = os.path.splitext(filename)[1][1:]
3035 filename_wo_ext = (
3036 os.path.splitext(filename)[0]
124bc071 3037 if filename_real_ext in (old_ext, new_ext)
0202b52a 3038 else filename)
86e5f3ed 3039 return f'{filename_wo_ext}.{ext}'
0202b52a 3040
38c6902b 3041 # Ensure filename always has a correct extension for successful merge
0202b52a 3042 full_filename = correct_ext(full_filename)
3043 temp_filename = correct_ext(temp_filename)
e04938ab 3044 dl_filename = existing_video_file(full_filename, temp_filename)
1ea24129 3045 info_dict['__real_download'] = False
18e674b4 3046
adbc4ec4
THD
3047 downloaded = []
3048 merger = FFmpegMergerPP(self)
3049
3050 fd = get_suitable_downloader(info_dict, self.params, to_stdout=temp_filename == '-')
dbf5416a 3051 if dl_filename is not None:
6c7274ec 3052 self.report_file_already_downloaded(dl_filename)
adbc4ec4
THD
3053 elif fd:
3054 for f in requested_formats if fd != FFmpegFD else []:
3055 f['filepath'] = fname = prepend_extension(
3056 correct_ext(temp_filename, info_dict['ext']),
3057 'f%s' % f['format_id'], info_dict['ext'])
3058 downloaded.append(fname)
dbf5416a 3059 info_dict['url'] = '\n'.join(f['url'] for f in requested_formats)
3060 success, real_download = self.dl(temp_filename, info_dict)
3061 info_dict['__real_download'] = real_download
18e674b4 3062 else:
18e674b4 3063 if self.params.get('allow_unplayable_formats'):
3064 self.report_warning(
3065 'You have requested merging of multiple formats '
3066 'while also allowing unplayable formats to be downloaded. '
3067 'The formats won\'t be merged to prevent data corruption.')
3068 elif not merger.available:
e8969bda 3069 msg = 'You have requested merging of multiple formats but ffmpeg is not installed'
3070 if not self.params.get('ignoreerrors'):
3071 self.report_error(f'{msg}. Aborting due to --abort-on-error')
3072 return
3073 self.report_warning(f'{msg}. The formats won\'t be merged')
18e674b4 3074
96fccc10 3075 if temp_filename == '-':
adbc4ec4 3076 reason = ('using a downloader other than ffmpeg' if FFmpegFD.can_merge_formats(info_dict, self.params)
96fccc10 3077 else 'but the formats are incompatible for simultaneous download' if merger.available
3078 else 'but ffmpeg is not installed')
3079 self.report_warning(
3080 f'You have requested downloading multiple formats to stdout {reason}. '
3081 'The formats will be streamed one after the other')
3082 fname = temp_filename
dbf5416a 3083 for f in requested_formats:
3084 new_info = dict(info_dict)
3085 del new_info['requested_formats']
3086 new_info.update(f)
96fccc10 3087 if temp_filename != '-':
124bc071 3088 fname = prepend_extension(
3089 correct_ext(temp_filename, new_info['ext']),
3090 'f%s' % f['format_id'], new_info['ext'])
96fccc10 3091 if not self._ensure_dir_exists(fname):
3092 return
a21e0ab1 3093 f['filepath'] = fname
96fccc10 3094 downloaded.append(fname)
dbf5416a 3095 partial_success, real_download = self.dl(fname, new_info)
3096 info_dict['__real_download'] = info_dict['__real_download'] or real_download
3097 success = success and partial_success
adbc4ec4
THD
3098
3099 if downloaded and merger.available and not self.params.get('allow_unplayable_formats'):
3100 info_dict['__postprocessors'].append(merger)
3101 info_dict['__files_to_merge'] = downloaded
3102 # Even if there were no downloads, it is being merged only now
3103 info_dict['__real_download'] = True
3104 else:
3105 for file in downloaded:
3106 files_to_move[file] = None
4340deca
P
3107 else:
3108 # Just a single file
e04938ab 3109 dl_filename = existing_video_file(full_filename, temp_filename)
6c7274ec 3110 if dl_filename is None or dl_filename == temp_filename:
3111 # dl_filename == temp_filename could mean that the file was partially downloaded with --no-part.
3112 # So we should try to resume the download
e8e73840 3113 success, real_download = self.dl(temp_filename, info_dict)
0202b52a 3114 info_dict['__real_download'] = real_download
6c7274ec 3115 else:
3116 self.report_file_already_downloaded(dl_filename)
0202b52a 3117
0202b52a 3118 dl_filename = dl_filename or temp_filename
c571435f 3119 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
0202b52a 3120
3158150c 3121 except network_exceptions as err:
7960b056 3122 self.report_error('unable to download video data: %s' % error_to_compat_str(err))
4340deca 3123 return
86e5f3ed 3124 except OSError as err:
4340deca
P
3125 raise UnavailableVideoError(err)
3126 except (ContentTooShortError, ) as err:
86e5f3ed 3127 self.report_error(f'content too short (expected {err.expected} bytes and served {err.downloaded})')
4340deca 3128 return
8222d8de 3129
de6000d9 3130 if success and full_filename != '-':
f17f8651 3131
fd7cfb64 3132 def fixup():
3133 do_fixup = True
3134 fixup_policy = self.params.get('fixup')
3135 vid = info_dict['id']
3136
3137 if fixup_policy in ('ignore', 'never'):
3138 return
3139 elif fixup_policy == 'warn':
3140 do_fixup = False
f89b3e2d 3141 elif fixup_policy != 'force':
3142 assert fixup_policy in ('detect_or_warn', None)
3143 if not info_dict.get('__real_download'):
3144 do_fixup = False
fd7cfb64 3145
3146 def ffmpeg_fixup(cndn, msg, cls):
3147 if not cndn:
3148 return
3149 if not do_fixup:
3150 self.report_warning(f'{vid}: {msg}')
3151 return
3152 pp = cls(self)
3153 if pp.available:
3154 info_dict['__postprocessors'].append(pp)
3155 else:
3156 self.report_warning(f'{vid}: {msg}. Install ffmpeg to fix this automatically')
3157
3158 stretched_ratio = info_dict.get('stretched_ratio')
3159 ffmpeg_fixup(
3160 stretched_ratio not in (1, None),
3161 f'Non-uniform pixel ratio {stretched_ratio}',
3162 FFmpegFixupStretchedPP)
3163
3164 ffmpeg_fixup(
3165 (info_dict.get('requested_formats') is None
3166 and info_dict.get('container') == 'm4a_dash'
3167 and info_dict.get('ext') == 'm4a'),
3168 'writing DASH m4a. Only some players support this container',
3169 FFmpegFixupM4aPP)
3170
993191c0 3171 downloader = get_suitable_downloader(info_dict, self.params) if 'protocol' in info_dict else None
3172 downloader = downloader.__name__ if downloader else None
adbc4ec4
THD
3173
3174 if info_dict.get('requested_formats') is None: # Not necessary if doing merger
3175 ffmpeg_fixup(downloader == 'HlsFD',
3176 'Possible MPEG-TS in MP4 container or malformed AAC timestamps',
3177 FFmpegFixupM3u8PP)
3178 ffmpeg_fixup(info_dict.get('is_live') and downloader == 'DashSegmentsFD',
3179 'Possible duplicate MOOV atoms', FFmpegFixupDuplicateMoovPP)
3180
e04b003e 3181 ffmpeg_fixup(downloader == 'WebSocketFragmentFD', 'Malformed timestamps detected', FFmpegFixupTimestampPP)
3182 ffmpeg_fixup(downloader == 'WebSocketFragmentFD', 'Malformed duration detected', FFmpegFixupDurationPP)
fd7cfb64 3183
3184 fixup()
8222d8de 3185 try:
f46e2f9d 3186 replace_info_dict(self.post_process(dl_filename, info_dict, files_to_move))
af819c21 3187 except PostProcessingError as err:
3188 self.report_error('Postprocessing: %s' % str(err))
8222d8de 3189 return
ab8e5e51
AM
3190 try:
3191 for ph in self._post_hooks:
23c1a667 3192 ph(info_dict['filepath'])
ab8e5e51
AM
3193 except Exception as err:
3194 self.report_error('post hooks: %s' % str(err))
3195 return
9e907ebd 3196 info_dict['__write_download_archive'] = True
2d30509f 3197
a13e6848 3198 if self.params.get('force_write_download_archive'):
9e907ebd 3199 info_dict['__write_download_archive'] = True
a13e6848 3200
3201 # Make sure the info_dict was modified in-place
f46e2f9d 3202 assert info_dict is original_infodict
a13e6848 3203
c3e6ffba 3204 max_downloads = self.params.get('max_downloads')
3205 if max_downloads is not None and self._num_downloads >= int(max_downloads):
3206 raise MaxDownloadsReached()
8222d8de 3207
aa9369a2 3208 def __download_wrapper(self, func):
3209 @functools.wraps(func)
3210 def wrapper(*args, **kwargs):
3211 try:
3212 res = func(*args, **kwargs)
3213 except UnavailableVideoError as e:
3214 self.report_error(e)
b222c271 3215 except MaxDownloadsReached as e:
aa9369a2 3216 self.to_screen(f'[info] {e}')
3217 raise
b222c271 3218 except DownloadCancelled as e:
3219 self.to_screen(f'[info] {e}')
3220 if not self.params.get('break_per_url'):
3221 raise
aa9369a2 3222 else:
3223 if self.params.get('dump_single_json', False):
3224 self.post_extract(res)
3225 self.to_stdout(json.dumps(self.sanitize_info(res)))
3226 return wrapper
3227
8222d8de
JMF
3228 def download(self, url_list):
3229 """Download a given list of URLs."""
aa9369a2 3230 url_list = variadic(url_list) # Passing a single URL is a common mistake
de6000d9 3231 outtmpl = self.outtmpl_dict['default']
3089bc74
S
3232 if (len(url_list) > 1
3233 and outtmpl != '-'
3234 and '%' not in outtmpl
3235 and self.params.get('max_downloads') != 1):
acd69589 3236 raise SameFileError(outtmpl)
8222d8de
JMF
3237
3238 for url in url_list:
aa9369a2 3239 self.__download_wrapper(self.extract_info)(
3240 url, force_generic_extractor=self.params.get('force_generic_extractor', False))
8222d8de
JMF
3241
3242 return self._download_retcode
3243
1dcc4c0c 3244 def download_with_info_file(self, info_filename):
31bd3925
JMF
3245 with contextlib.closing(fileinput.FileInput(
3246 [info_filename], mode='r',
3247 openhook=fileinput.hook_encoded('utf-8'))) as f:
3248 # FileInput doesn't have a read method, we can't call json.load
8012d892 3249 info = self.sanitize_info(json.loads('\n'.join(f)), self.params.get('clean_infojson', True))
d4943898 3250 try:
aa9369a2 3251 self.__download_wrapper(self.process_ie_result)(info, download=True)
f2ebc5c7 3252 except (DownloadError, EntryNotInPlaylist, ReExtractInfo) as e:
bf5f605e 3253 if not isinstance(e, EntryNotInPlaylist):
3254 self.to_stderr('\r')
d4943898
JMF
3255 webpage_url = info.get('webpage_url')
3256 if webpage_url is not None:
aa9369a2 3257 self.report_warning(f'The info failed to download: {e}; trying with URL {webpage_url}')
d4943898
JMF
3258 return self.download([webpage_url])
3259 else:
3260 raise
3261 return self._download_retcode
1dcc4c0c 3262
cb202fd2 3263 @staticmethod
8012d892 3264 def sanitize_info(info_dict, remove_private_keys=False):
3265 ''' Sanitize the infodict for converting to json '''
3ad56b42 3266 if info_dict is None:
3267 return info_dict
6e84b215 3268 info_dict.setdefault('epoch', int(time.time()))
6a5a30f9 3269 info_dict.setdefault('_type', 'video')
09b49e1f 3270
8012d892 3271 if remove_private_keys:
09b49e1f 3272 reject = lambda k, v: v is None or (k.startswith('_') and k != '_type') or k in {
f46e2f9d 3273 'requested_downloads', 'requested_formats', 'requested_subtitles', 'requested_entries',
3274 'entries', 'filepath', 'infojson_filename', 'original_url', 'playlist_autonumber',
6e84b215 3275 }
ae8f99e6 3276 else:
09b49e1f 3277 reject = lambda k, v: False
adbc4ec4
THD
3278
3279 def filter_fn(obj):
3280 if isinstance(obj, dict):
3281 return {k: filter_fn(v) for k, v in obj.items() if not reject(k, v)}
3282 elif isinstance(obj, (list, tuple, set, LazyList)):
3283 return list(map(filter_fn, obj))
3284 elif obj is None or isinstance(obj, (str, int, float, bool)):
3285 return obj
3286 else:
3287 return repr(obj)
3288
5226731e 3289 return filter_fn(info_dict)
cb202fd2 3290
8012d892 3291 @staticmethod
3292 def filter_requested_info(info_dict, actually_filter=True):
3293 ''' Alias of sanitize_info for backward compatibility '''
3294 return YoutubeDL.sanitize_info(info_dict, actually_filter)
3295
ed5835b4 3296 @staticmethod
3297 def post_extract(info_dict):
3298 def actual_post_extract(info_dict):
3299 if info_dict.get('_type') in ('playlist', 'multi_video'):
3300 for video_dict in info_dict.get('entries', {}):
3301 actual_post_extract(video_dict or {})
3302 return
3303
09b49e1f 3304 post_extractor = info_dict.pop('__post_extractor', None) or (lambda: {})
3305 info_dict.update(post_extractor())
ed5835b4 3306
3307 actual_post_extract(info_dict or {})
3308
dcf64d43 3309 def run_pp(self, pp, infodict):
5bfa4862 3310 files_to_delete = []
dcf64d43 3311 if '__files_to_move' not in infodict:
3312 infodict['__files_to_move'] = {}
b1940459 3313 try:
3314 files_to_delete, infodict = pp.run(infodict)
3315 except PostProcessingError as e:
3316 # Must be True and not 'only_download'
3317 if self.params.get('ignoreerrors') is True:
3318 self.report_error(e)
3319 return infodict
3320 raise
3321
5bfa4862 3322 if not files_to_delete:
dcf64d43 3323 return infodict
5bfa4862 3324 if self.params.get('keepvideo', False):
3325 for f in files_to_delete:
dcf64d43 3326 infodict['__files_to_move'].setdefault(f, '')
5bfa4862 3327 else:
3328 for old_filename in set(files_to_delete):
3329 self.to_screen('Deleting original file %s (pass -k to keep)' % old_filename)
3330 try:
3331 os.remove(encodeFilename(old_filename))
86e5f3ed 3332 except OSError:
5bfa4862 3333 self.report_warning('Unable to remove downloaded original file')
dcf64d43 3334 if old_filename in infodict['__files_to_move']:
3335 del infodict['__files_to_move'][old_filename]
3336 return infodict
5bfa4862 3337
ed5835b4 3338 def run_all_pps(self, key, info, *, additional_pps=None):
bb66c247 3339 self._forceprint(key, info)
ed5835b4 3340 for pp in (additional_pps or []) + self._pps[key]:
dc5f409c 3341 info = self.run_pp(pp, info)
ed5835b4 3342 return info
277d6ff5 3343
56d868db 3344 def pre_process(self, ie_info, key='pre_process', files_to_move=None):
5bfa4862 3345 info = dict(ie_info)
56d868db 3346 info['__files_to_move'] = files_to_move or {}
ed5835b4 3347 info = self.run_all_pps(key, info)
56d868db 3348 return info, info.pop('__files_to_move', None)
5bfa4862 3349
f46e2f9d 3350 def post_process(self, filename, info, files_to_move=None):
8222d8de 3351 """Run all the postprocessors on the given file."""
8222d8de 3352 info['filepath'] = filename
dcf64d43 3353 info['__files_to_move'] = files_to_move or {}
ed5835b4 3354 info = self.run_all_pps('post_process', info, additional_pps=info.get('__postprocessors'))
dcf64d43 3355 info = self.run_pp(MoveFilesAfterDownloadPP(self), info)
3356 del info['__files_to_move']
ed5835b4 3357 return self.run_all_pps('after_move', info)
c1c9a79c 3358
5db07df6 3359 def _make_archive_id(self, info_dict):
e9fef7ee
S
3360 video_id = info_dict.get('id')
3361 if not video_id:
3362 return
5db07df6
PH
3363 # Future-proof against any change in case
3364 # and backwards compatibility with prior versions
e9fef7ee 3365 extractor = info_dict.get('extractor_key') or info_dict.get('ie_key') # key in a playlist
7012b23c 3366 if extractor is None:
1211bb6d
S
3367 url = str_or_none(info_dict.get('url'))
3368 if not url:
3369 return
e9fef7ee 3370 # Try to find matching extractor for the URL and take its ie_key
8b7491c8 3371 for ie_key, ie in self._ies.items():
1211bb6d 3372 if ie.suitable(url):
8b7491c8 3373 extractor = ie_key
e9fef7ee
S
3374 break
3375 else:
3376 return
86e5f3ed 3377 return f'{extractor.lower()} {video_id}'
5db07df6
PH
3378
3379 def in_download_archive(self, info_dict):
3380 fn = self.params.get('download_archive')
3381 if fn is None:
3382 return False
3383
3384 vid_id = self._make_archive_id(info_dict)
e9fef7ee 3385 if not vid_id:
7012b23c 3386 return False # Incomplete video information
5db07df6 3387
a45e8619 3388 return vid_id in self.archive
c1c9a79c
PH
3389
3390 def record_download_archive(self, info_dict):
3391 fn = self.params.get('download_archive')
3392 if fn is None:
3393 return
5db07df6
PH
3394 vid_id = self._make_archive_id(info_dict)
3395 assert vid_id
a13e6848 3396 self.write_debug(f'Adding to archive: {vid_id}')
c1c9a79c 3397 with locked_file(fn, 'a', encoding='utf-8') as archive_file:
6febd1c1 3398 archive_file.write(vid_id + '\n')
a45e8619 3399 self.archive.add(vid_id)
dd82ffea 3400
8c51aa65 3401 @staticmethod
8abeeb94 3402 def format_resolution(format, default='unknown'):
9359f3d4 3403 if format.get('vcodec') == 'none' and format.get('acodec') != 'none':
fb04e403 3404 return 'audio only'
f49d89ee
PH
3405 if format.get('resolution') is not None:
3406 return format['resolution']
35615307 3407 if format.get('width') and format.get('height'):
ff51ed58 3408 return '%dx%d' % (format['width'], format['height'])
35615307 3409 elif format.get('height'):
ff51ed58 3410 return '%sp' % format['height']
35615307 3411 elif format.get('width'):
ff51ed58 3412 return '%dx?' % format['width']
3413 return default
8c51aa65 3414
8130779d 3415 def _list_format_headers(self, *headers):
3416 if self.params.get('listformats_table', True) is not False:
3417 return [self._format_screen(header, self.Styles.HEADERS) for header in headers]
3418 return headers
3419
c57f7757
PH
3420 def _format_note(self, fdict):
3421 res = ''
3422 if fdict.get('ext') in ['f4f', 'f4m']:
f304da8a 3423 res += '(unsupported)'
32f90364
PH
3424 if fdict.get('language'):
3425 if res:
3426 res += ' '
f304da8a 3427 res += '[%s]' % fdict['language']
c57f7757 3428 if fdict.get('format_note') is not None:
f304da8a 3429 if res:
3430 res += ' '
3431 res += fdict['format_note']
c57f7757 3432 if fdict.get('tbr') is not None:
f304da8a 3433 if res:
3434 res += ', '
3435 res += '%4dk' % fdict['tbr']
c57f7757
PH
3436 if fdict.get('container') is not None:
3437 if res:
3438 res += ', '
3439 res += '%s container' % fdict['container']
3089bc74
S
3440 if (fdict.get('vcodec') is not None
3441 and fdict.get('vcodec') != 'none'):
c57f7757
PH
3442 if res:
3443 res += ', '
3444 res += fdict['vcodec']
91c7271a 3445 if fdict.get('vbr') is not None:
c57f7757
PH
3446 res += '@'
3447 elif fdict.get('vbr') is not None and fdict.get('abr') is not None:
3448 res += 'video@'
3449 if fdict.get('vbr') is not None:
3450 res += '%4dk' % fdict['vbr']
fbb21cf5 3451 if fdict.get('fps') is not None:
5d583bdf
S
3452 if res:
3453 res += ', '
3454 res += '%sfps' % fdict['fps']
c57f7757
PH
3455 if fdict.get('acodec') is not None:
3456 if res:
3457 res += ', '
3458 if fdict['acodec'] == 'none':
3459 res += 'video only'
3460 else:
3461 res += '%-5s' % fdict['acodec']
3462 elif fdict.get('abr') is not None:
3463 if res:
3464 res += ', '
3465 res += 'audio'
3466 if fdict.get('abr') is not None:
3467 res += '@%3dk' % fdict['abr']
3468 if fdict.get('asr') is not None:
3469 res += ' (%5dHz)' % fdict['asr']
3470 if fdict.get('filesize') is not None:
3471 if res:
3472 res += ', '
3473 res += format_bytes(fdict['filesize'])
9732d77e
PH
3474 elif fdict.get('filesize_approx') is not None:
3475 if res:
3476 res += ', '
3477 res += '~' + format_bytes(fdict['filesize_approx'])
c57f7757 3478 return res
91c7271a 3479
8130779d 3480 def render_formats_table(self, info_dict):
b69fd25c 3481 if not info_dict.get('formats') and not info_dict.get('url'):
8130779d 3482 return None
b69fd25c 3483
94badb25 3484 formats = info_dict.get('formats', [info_dict])
8130779d 3485 if not self.params.get('listformats_table', True) is not False:
76d321f6 3486 table = [
3487 [
3488 format_field(f, 'format_id'),
3489 format_field(f, 'ext'),
3490 self.format_resolution(f),
8130779d 3491 self._format_note(f)
3492 ] for f in formats if f.get('preference') is None or f['preference'] >= -1000]
3493 return render_table(['format code', 'extension', 'resolution', 'note'], table, extra_gap=1)
3494
3495 delim = self._format_screen('\u2502', self.Styles.DELIM, '|', test_encoding=True)
3496 table = [
3497 [
3498 self._format_screen(format_field(f, 'format_id'), self.Styles.ID),
3499 format_field(f, 'ext'),
3500 format_field(f, func=self.format_resolution, ignore=('audio only', 'images')),
3501 format_field(f, 'fps', '\t%d'),
3502 format_field(f, 'dynamic_range', '%s', ignore=(None, 'SDR')).replace('HDR', ''),
3503 delim,
3504 format_field(f, 'filesize', ' \t%s', func=format_bytes) + format_field(f, 'filesize_approx', '~\t%s', func=format_bytes),
3505 format_field(f, 'tbr', '\t%dk'),
3506 shorten_protocol_name(f.get('protocol', '')),
3507 delim,
3508 format_field(f, 'vcodec', default='unknown').replace(
3509 'none', 'images' if f.get('acodec') == 'none'
3510 else self._format_screen('audio only', self.Styles.SUPPRESS)),
3511 format_field(f, 'vbr', '\t%dk'),
3512 format_field(f, 'acodec', default='unknown').replace(
3513 'none', '' if f.get('vcodec') == 'none'
3514 else self._format_screen('video only', self.Styles.SUPPRESS)),
3515 format_field(f, 'abr', '\t%dk'),
3516 format_field(f, 'asr', '\t%dHz'),
3517 join_nonempty(
3518 self._format_screen('UNSUPPORTED', 'light red') if f.get('ext') in ('f4f', 'f4m') else None,
3519 format_field(f, 'language', '[%s]'),
3520 join_nonempty(format_field(f, 'format_note'),
3521 format_field(f, 'container', ignore=(None, f.get('ext'))),
3522 delim=', '),
3523 delim=' '),
3524 ] for f in formats if f.get('preference') is None or f['preference'] >= -1000]
3525 header_line = self._list_format_headers(
3526 'ID', 'EXT', 'RESOLUTION', '\tFPS', 'HDR', delim, '\tFILESIZE', '\tTBR', 'PROTO',
3527 delim, 'VCODEC', '\tVBR', 'ACODEC', '\tABR', '\tASR', 'MORE INFO')
3528
3529 return render_table(
3530 header_line, table, hide_empty=True,
3531 delim=self._format_screen('\u2500', self.Styles.DELIM, '-', test_encoding=True))
3532
3533 def render_thumbnails_table(self, info_dict):
88f23a18 3534 thumbnails = list(info_dict.get('thumbnails') or [])
cfb56d1a 3535 if not thumbnails:
8130779d 3536 return None
3537 return render_table(
ec11a9f4 3538 self._list_format_headers('ID', 'Width', 'Height', 'URL'),
6970b600 3539 [[t.get('id'), t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails])
2412044c 3540
8130779d 3541 def render_subtitles_table(self, video_id, subtitles):
2412044c 3542 def _row(lang, formats):
49c258e1 3543 exts, names = zip(*((f['ext'], f.get('name') or 'unknown') for f in reversed(formats)))
2412044c 3544 if len(set(names)) == 1:
7aee40c1 3545 names = [] if names[0] == 'unknown' else names[:1]
2412044c 3546 return [lang, ', '.join(names), ', '.join(exts)]
3547
8130779d 3548 if not subtitles:
3549 return None
3550 return render_table(
ec11a9f4 3551 self._list_format_headers('Language', 'Name', 'Formats'),
2412044c 3552 [_row(lang, formats) for lang, formats in subtitles.items()],
8130779d 3553 hide_empty=True)
3554
3555 def __list_table(self, video_id, name, func, *args):
3556 table = func(*args)
3557 if not table:
3558 self.to_screen(f'{video_id} has no {name}')
3559 return
3560 self.to_screen(f'[info] Available {name} for {video_id}:')
3561 self.to_stdout(table)
3562
3563 def list_formats(self, info_dict):
3564 self.__list_table(info_dict['id'], 'formats', self.render_formats_table, info_dict)
3565
3566 def list_thumbnails(self, info_dict):
3567 self.__list_table(info_dict['id'], 'thumbnails', self.render_thumbnails_table, info_dict)
3568
3569 def list_subtitles(self, video_id, subtitles, name='subtitles'):
3570 self.__list_table(video_id, name, self.render_subtitles_table, video_id, subtitles)
a504ced0 3571
dca08720
PH
3572 def urlopen(self, req):
3573 """ Start an HTTP download """
f9934b96 3574 if isinstance(req, str):
67dda517 3575 req = sanitized_Request(req)
19a41fc6 3576 return self._opener.open(req, timeout=self._socket_timeout)
dca08720
PH
3577
3578 def print_debug_header(self):
3579 if not self.params.get('verbose'):
3580 return
49a57e70 3581
3582 def get_encoding(stream):
2a938746 3583 ret = str(getattr(stream, 'encoding', 'missing (%s)' % type(stream).__name__))
49a57e70 3584 if not supports_terminal_sequences(stream):
e3c7d495 3585 from .compat import WINDOWS_VT_MODE
3586 ret += ' (No VT)' if WINDOWS_VT_MODE is False else ' (No ANSI)'
49a57e70 3587 return ret
3588
3589 encoding_str = 'Encodings: locale %s, fs %s, out %s, err %s, pref %s' % (
3590 locale.getpreferredencoding(),
3591 sys.getfilesystemencoding(),
cf4f42cb 3592 get_encoding(self._out_files['screen']), get_encoding(self._out_files['error']),
49a57e70 3593 self.get_encoding())
883d4b1e 3594
3595 logger = self.params.get('logger')
3596 if logger:
3597 write_debug = lambda msg: logger.debug(f'[debug] {msg}')
3598 write_debug(encoding_str)
3599 else:
96565c7e 3600 write_string(f'[debug] {encoding_str}\n', encoding=None)
49a57e70 3601 write_debug = lambda msg: self._write_string(f'[debug] {msg}\n')
734f90bb 3602
4c88ff87 3603 source = detect_variant()
36eaf303 3604 write_debug(join_nonempty(
3605 'yt-dlp version', __version__,
3606 f'[{RELEASE_GIT_HEAD}]' if RELEASE_GIT_HEAD else '',
3607 '' if source == 'unknown' else f'({source})',
3608 delim=' '))
6e21fdd2 3609 if not _LAZY_LOADER:
3610 if os.environ.get('YTDLP_NO_LAZY_EXTRACTORS'):
49a57e70 3611 write_debug('Lazy loading extractors is forcibly disabled')
6e21fdd2 3612 else:
49a57e70 3613 write_debug('Lazy loading extractors is disabled')
3ae5e797 3614 if plugin_extractors or plugin_postprocessors:
49a57e70 3615 write_debug('Plugins: %s' % [
3ae5e797 3616 '%s%s' % (klass.__name__, '' if klass.__name__ == name else f' as {name}')
3617 for name, klass in itertools.chain(plugin_extractors.items(), plugin_postprocessors.items())])
53ed7066 3618 if self.params.get('compat_opts'):
49a57e70 3619 write_debug('Compatibility options: %s' % ', '.join(self.params.get('compat_opts')))
36eaf303 3620
3621 if source == 'source':
dca08720 3622 try:
36eaf303 3623 sp = Popen(
3624 ['git', 'rev-parse', '--short', 'HEAD'],
3625 stdout=subprocess.PIPE, stderr=subprocess.PIPE,
3626 cwd=os.path.dirname(os.path.abspath(__file__)))
3627 out, err = sp.communicate_or_kill()
3628 out = out.decode().strip()
3629 if re.match('[0-9a-f]+', out):
3630 write_debug('Git HEAD: %s' % out)
70a1165b 3631 except Exception:
19a03940 3632 with contextlib.suppress(Exception):
36eaf303 3633 sys.exc_clear()
b300cda4
S
3634
3635 def python_implementation():
3636 impl_name = platform.python_implementation()
3637 if impl_name == 'PyPy' and hasattr(sys, 'pypy_version_info'):
3638 return impl_name + ' version %d.%d.%d' % sys.pypy_version_info[:3]
3639 return impl_name
3640
49a57e70 3641 write_debug('Python version %s (%s %s) - %s' % (
e5813e53 3642 platform.python_version(),
3643 python_implementation(),
3644 platform.architecture()[0],
b300cda4 3645 platform_name()))
d28b5171 3646
8913ef74 3647 exe_versions, ffmpeg_features = FFmpegPostProcessor.get_versions_and_features(self)
3648 ffmpeg_features = {key for key, val in ffmpeg_features.items() if val}
3649 if ffmpeg_features:
19a03940 3650 exe_versions['ffmpeg'] += ' (%s)' % ','.join(sorted(ffmpeg_features))
8913ef74 3651
4c83c967 3652 exe_versions['rtmpdump'] = rtmpdump_version()
feee8d32 3653 exe_versions['phantomjs'] = PhantomJSwrapper._version()
d28b5171 3654 exe_str = ', '.join(
2831b468 3655 f'{exe} {v}' for exe, v in sorted(exe_versions.items()) if v
3656 ) or 'none'
49a57e70 3657 write_debug('exe versions: %s' % exe_str)
dca08720 3658
f8271158 3659 from .cookies import SECRETSTORAGE_AVAILABLE, SQLITE_AVAILABLE
2831b468 3660 from .downloader.websocket import has_websockets
3661 from .postprocessor.embedthumbnail import has_mutagen
2831b468 3662
c586f9e8 3663 lib_str = join_nonempty(
4390d5ec 3664 compat_brotli and compat_brotli.__name__,
d5820461 3665 has_certifi and 'certifi',
edf65256 3666 compat_pycrypto_AES and compat_pycrypto_AES.__name__.split('.')[0],
f59f5ef8 3667 SECRETSTORAGE_AVAILABLE and 'secretstorage',
2831b468 3668 has_mutagen and 'mutagen',
3669 SQLITE_AVAILABLE and 'sqlite',
c586f9e8 3670 has_websockets and 'websockets',
3671 delim=', ') or 'none'
49a57e70 3672 write_debug('Optional libraries: %s' % lib_str)
2831b468 3673
97ec5bc5 3674 self._setup_opener()
dca08720
PH
3675 proxy_map = {}
3676 for handler in self._opener.handlers:
3677 if hasattr(handler, 'proxies'):
3678 proxy_map.update(handler.proxies)
49a57e70 3679 write_debug(f'Proxy map: {proxy_map}')
dca08720 3680
49a57e70 3681 # Not implemented
3682 if False and self.params.get('call_home'):
58b1f00d 3683 ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode('utf-8')
49a57e70 3684 write_debug('Public IP address: %s' % ipaddr)
58b1f00d
PH
3685 latest_version = self.urlopen(
3686 'https://yt-dl.org/latest/version').read().decode('utf-8')
3687 if version_tuple(latest_version) > version_tuple(__version__):
3688 self.report_warning(
3689 'You are using an outdated version (newest version: %s)! '
3690 'See https://yt-dl.org/update if you need help updating.' %
3691 latest_version)
3692
e344693b 3693 def _setup_opener(self):
97ec5bc5 3694 if hasattr(self, '_opener'):
3695 return
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()
f9934b96 3722 data_handler = 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):
cb96c5be 3760 ''' Write infojson and returns True = written, 'exists' = Already exists, 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')
cb96c5be 3772 return 'exists'
3773
3774 self.to_screen(f'[info] Writing {label} metadata as JSON to: {infofn}')
3775 try:
3776 write_json_file(self.sanitize_info(ie_result, self.params.get('clean_infojson', True)), infofn)
3777 return True
86e5f3ed 3778 except OSError:
cb96c5be 3779 self.report_error(f'Cannot write {label} metadata to JSON file {infofn}')
3780 return None
80c03fa9 3781
3782 def _write_description(self, label, ie_result, descfn):
3783 ''' Write description and returns True = written, False = skip, None = error '''
3784 if not self.params.get('writedescription'):
3785 return False
3786 elif not descfn:
3787 self.write_debug(f'Skipping writing {label} description')
3788 return False
3789 elif not self._ensure_dir_exists(descfn):
3790 return None
3791 elif not self.params.get('overwrites', True) and os.path.exists(descfn):
3792 self.to_screen(f'[info] {label.title()} description is already present')
3793 elif ie_result.get('description') is None:
3794 self.report_warning(f'There\'s no {label} description to write')
3795 return False
3796 else:
3797 try:
3798 self.to_screen(f'[info] Writing {label} description to: {descfn}')
86e5f3ed 3799 with open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
80c03fa9 3800 descfile.write(ie_result['description'])
86e5f3ed 3801 except OSError:
80c03fa9 3802 self.report_error(f'Cannot write {label} description file {descfn}')
3803 return None
3804 return True
3805
3806 def _write_subtitles(self, info_dict, filename):
3807 ''' Write subtitles to file and return list of (sub_filename, final_sub_filename); or None if error'''
3808 ret = []
3809 subtitles = info_dict.get('requested_subtitles')
3810 if not subtitles or not (self.params.get('writesubtitles') or self.params.get('writeautomaticsub')):
3811 # subtitles download errors are already managed as troubles in relevant IE
3812 # that way it will silently go on when used with unsupporting IE
3813 return ret
3814
3815 sub_filename_base = self.prepare_filename(info_dict, 'subtitle')
3816 if not sub_filename_base:
3817 self.to_screen('[info] Skipping writing video subtitles')
3818 return ret
3819 for sub_lang, sub_info in subtitles.items():
3820 sub_format = sub_info['ext']
3821 sub_filename = subtitles_filename(filename, sub_lang, sub_format, info_dict.get('ext'))
3822 sub_filename_final = subtitles_filename(sub_filename_base, sub_lang, sub_format, info_dict.get('ext'))
e04938ab 3823 existing_sub = self.existing_file((sub_filename_final, sub_filename))
3824 if existing_sub:
80c03fa9 3825 self.to_screen(f'[info] Video subtitle {sub_lang}.{sub_format} is already present')
e04938ab 3826 sub_info['filepath'] = existing_sub
3827 ret.append((existing_sub, sub_filename_final))
80c03fa9 3828 continue
3829
3830 self.to_screen(f'[info] Writing video subtitles to: {sub_filename}')
3831 if sub_info.get('data') is not None:
3832 try:
3833 # Use newline='' to prevent conversion of newline characters
3834 # See https://github.com/ytdl-org/youtube-dl/issues/10268
86e5f3ed 3835 with open(sub_filename, 'w', encoding='utf-8', newline='') as subfile:
80c03fa9 3836 subfile.write(sub_info['data'])
3837 sub_info['filepath'] = sub_filename
3838 ret.append((sub_filename, sub_filename_final))
3839 continue
86e5f3ed 3840 except OSError:
80c03fa9 3841 self.report_error(f'Cannot write video subtitles file {sub_filename}')
3842 return None
3843
3844 try:
3845 sub_copy = sub_info.copy()
3846 sub_copy.setdefault('http_headers', info_dict.get('http_headers'))
3847 self.dl(sub_filename, sub_copy, subtitle=True)
3848 sub_info['filepath'] = sub_filename
3849 ret.append((sub_filename, sub_filename_final))
6020e05d 3850 except (DownloadError, ExtractorError, IOError, OSError, ValueError) + network_exceptions as err:
c70c418d 3851 msg = f'Unable to download video subtitles for {sub_lang!r}: {err}'
6020e05d 3852 if self.params.get('ignoreerrors') is not True: # False or 'only_download'
c70c418d 3853 if not self.params.get('ignoreerrors'):
3854 self.report_error(msg)
3855 raise DownloadError(msg)
3856 self.report_warning(msg)
519804a9 3857 return ret
80c03fa9 3858
3859 def _write_thumbnails(self, label, info_dict, filename, thumb_filename_base=None):
3860 ''' Write thumbnails to file and return list of (thumb_filename, final_thumb_filename) '''
6c4fd172 3861 write_all = self.params.get('write_all_thumbnails', False)
80c03fa9 3862 thumbnails, ret = [], []
6c4fd172 3863 if write_all or self.params.get('writethumbnail', False):
0202b52a 3864 thumbnails = info_dict.get('thumbnails') or []
6c4fd172 3865 multiple = write_all and len(thumbnails) > 1
ec82d85a 3866
80c03fa9 3867 if thumb_filename_base is None:
3868 thumb_filename_base = filename
3869 if thumbnails and not thumb_filename_base:
3870 self.write_debug(f'Skipping writing {label} thumbnail')
3871 return ret
3872
dd0228ce 3873 for idx, t in list(enumerate(thumbnails))[::-1]:
80c03fa9 3874 thumb_ext = (f'{t["id"]}.' if multiple else '') + determine_ext(t['url'], 'jpg')
aa9369a2 3875 thumb_display_id = f'{label} thumbnail {t["id"]}'
80c03fa9 3876 thumb_filename = replace_extension(filename, thumb_ext, info_dict.get('ext'))
3877 thumb_filename_final = replace_extension(thumb_filename_base, thumb_ext, info_dict.get('ext'))
ec82d85a 3878
e04938ab 3879 existing_thumb = self.existing_file((thumb_filename_final, thumb_filename))
3880 if existing_thumb:
aa9369a2 3881 self.to_screen('[info] %s is already present' % (
3882 thumb_display_id if multiple else f'{label} thumbnail').capitalize())
e04938ab 3883 t['filepath'] = existing_thumb
3884 ret.append((existing_thumb, thumb_filename_final))
ec82d85a 3885 else:
80c03fa9 3886 self.to_screen(f'[info] Downloading {thumb_display_id} ...')
ec82d85a 3887 try:
297e9952 3888 uf = self.urlopen(sanitized_Request(t['url'], headers=t.get('http_headers', {})))
80c03fa9 3889 self.to_screen(f'[info] Writing {thumb_display_id} to: {thumb_filename}')
d3d89c32 3890 with open(encodeFilename(thumb_filename), 'wb') as thumbf:
ec82d85a 3891 shutil.copyfileobj(uf, thumbf)
80c03fa9 3892 ret.append((thumb_filename, thumb_filename_final))
885cc0b7 3893 t['filepath'] = thumb_filename
3158150c 3894 except network_exceptions as err:
dd0228ce 3895 thumbnails.pop(idx)
80c03fa9 3896 self.report_warning(f'Unable to download {thumb_display_id}: {err}')
6c4fd172 3897 if ret and not write_all:
3898 break
0202b52a 3899 return ret