]> jfr.im git - yt-dlp.git/blame - yt_dlp/YoutubeDL.py
[veo] Fix `_VALID_URL`
[yt-dlp.git] / yt_dlp / YoutubeDL.py
CommitLineData
cc52de43 1#!/usr/bin/env python3
dcdb292f 2# coding: utf-8
8222d8de 3
6febd1c1 4from __future__ import absolute_import, unicode_literals
8222d8de 5
26e63931 6import collections
31bd3925 7import contextlib
9d2ecdbc 8import datetime
c1c9a79c 9import errno
31bd3925 10import fileinput
b5ae35ee 11import functools
8222d8de 12import io
b82f815f 13import itertools
8694c600 14import json
62fec3b2 15import locale
083c9df9 16import operator
8222d8de 17import os
dca08720 18import platform
8222d8de
JMF
19import re
20import shutil
dca08720 21import subprocess
8222d8de 22import sys
21cd8fae 23import tempfile
8222d8de 24import time
67134eab 25import tokenize
8222d8de 26import traceback
75822ca7 27import random
524e2e4f 28import unicodedata
8222d8de 29
ec11a9f4 30from enum import Enum
961ea474
S
31from string import ascii_letters
32
8c25f81b 33from .compat import (
82d8a8b6 34 compat_basestring,
4390d5ec 35 compat_brotli,
003c69a8 36 compat_get_terminal_size,
4f026faf 37 compat_kwargs,
d0d9ade4 38 compat_numeric_types,
e9c0cdd3 39 compat_os_name,
edf65256 40 compat_pycrypto_AES,
7d1eb38a 41 compat_shlex_quote,
ce02ed60 42 compat_str,
67134eab 43 compat_tokenize_tokenize,
ce02ed60
PH
44 compat_urllib_error,
45 compat_urllib_request,
8b172c2e 46 compat_urllib_request_DataHandler,
819e0531 47 windows_enable_vt_mode,
8c25f81b 48)
982ee69a 49from .cookies import load_cookies
8c25f81b 50from .utils import (
eedb7ba5
S
51 age_restricted,
52 args_to_str,
ce02ed60
PH
53 ContentTooShortError,
54 date_from_str,
55 DateRange,
acd69589 56 DEFAULT_OUTTMPL,
ce02ed60 57 determine_ext,
b5559424 58 determine_protocol,
48f79687 59 DownloadCancelled,
ce02ed60 60 DownloadError,
c0384f22 61 encode_compat_str,
ce02ed60 62 encodeFilename,
498f5606 63 EntryNotInPlaylist,
a06916d9 64 error_to_compat_str,
8b0d7497 65 ExistingVideoReached,
590bc6f6 66 expand_path,
ce02ed60 67 ExtractorError,
e29663c6 68 float_or_none,
02dbf93f 69 format_bytes,
76d321f6 70 format_field,
e0fd9573 71 format_decimal_suffix,
525ef922 72 formatSeconds,
773f291d 73 GeoRestrictedError,
0bb322b9 74 get_domain,
b0249bca 75 HEADRequest,
d37707bd 76 InAdvancePagedList,
c9969434 77 int_or_none,
732044af 78 iri_to_uri,
773f291d 79 ISO3166Utils,
34921b43 80 join_nonempty,
56a8fb4f 81 LazyList,
08438d2c 82 LINK_TEMPLATES,
ce02ed60 83 locked_file,
0202b52a 84 make_dir,
dca08720 85 make_HTTPS_handler,
ce02ed60 86 MaxDownloadsReached,
8b7539d2 87 merge_headers,
3158150c 88 network_exceptions,
ec11a9f4 89 number_of_digits,
cd6fc19e 90 orderedSet,
a06916d9 91 OUTTMPL_TYPES,
b7ab0590 92 PagedList,
083c9df9 93 parse_filesize,
91410c9b 94 PerRequestProxyHandler,
dca08720 95 platform_name,
d3c93ec2 96 Popen,
1e43a6f7 97 POSTPROCESS_WHEN,
eedb7ba5 98 PostProcessingError,
ce02ed60 99 preferredencoding,
eedb7ba5 100 prepend_extension,
f2ebc5c7 101 ReExtractInfo,
51fb4995 102 register_socks_protocols,
a06916d9 103 RejectedVideoReached,
3efb96a6 104 remove_terminal_sequences,
cfb56d1a 105 render_table,
eedb7ba5 106 replace_extension,
ce02ed60
PH
107 SameFileError,
108 sanitize_filename,
1bb5c511 109 sanitize_path,
dcf77cf1 110 sanitize_url,
67dda517 111 sanitized_Request,
e5660ee6 112 std_headers,
819e0531 113 STR_FORMAT_RE_TMPL,
114 STR_FORMAT_TYPES,
1211bb6d 115 str_or_none,
e29663c6 116 strftime_or_none,
ce02ed60 117 subtitles_filename,
819e0531 118 supports_terminal_sequences,
f2ebc5c7 119 timetuple_from_msec,
732044af 120 to_high_limit_path,
324ad820 121 traverse_obj,
6033d980 122 try_get,
ce02ed60 123 UnavailableVideoError,
29eb5174 124 url_basename,
7d1eb38a 125 variadic,
58b1f00d 126 version_tuple,
ce02ed60
PH
127 write_json_file,
128 write_string,
6a3f4c3f 129 YoutubeDLCookieProcessor,
dca08720 130 YoutubeDLHandler,
fca6dba8 131 YoutubeDLRedirectHandler,
ce02ed60 132)
a0e07d31 133from .cache import Cache
ec11a9f4 134from .minicurses import format_text
52a8a1e1 135from .extractor import (
136 gen_extractor_classes,
137 get_info_extractor,
138 _LAZY_LOADER,
3ae5e797 139 _PLUGIN_CLASSES as plugin_extractors
52a8a1e1 140)
4c54b89e 141from .extractor.openload import PhantomJSwrapper
52a8a1e1 142from .downloader import (
dbf5416a 143 FFmpegFD,
52a8a1e1 144 get_suitable_downloader,
145 shorten_protocol_name
146)
4c83c967 147from .downloader.rtmp import rtmpdump_version
4f026faf 148from .postprocessor import (
e36d50c5 149 get_postprocessor,
4e3b637d 150 EmbedThumbnailPP,
adbc4ec4 151 FFmpegFixupDuplicateMoovPP,
e36d50c5 152 FFmpegFixupDurationPP,
f17f8651 153 FFmpegFixupM3u8PP,
62cd676c 154 FFmpegFixupM4aPP,
6271f1ca 155 FFmpegFixupStretchedPP,
e36d50c5 156 FFmpegFixupTimestampPP,
4f026faf
PH
157 FFmpegMergerPP,
158 FFmpegPostProcessor,
0202b52a 159 MoveFilesAfterDownloadPP,
3ae5e797 160 _PLUGIN_CLASSES as plugin_postprocessors
4f026faf 161)
4c88ff87 162from .update import detect_variant
36eaf303 163from .version import __version__, RELEASE_GIT_HEAD
8222d8de 164
e9c0cdd3
YCH
165if compat_os_name == 'nt':
166 import ctypes
167
2459b6e1 168
8222d8de
JMF
169class YoutubeDL(object):
170 """YoutubeDL class.
171
172 YoutubeDL objects are the ones responsible of downloading the
173 actual video file and writing it to disk if the user has requested
174 it, among some other tasks. In most cases there should be one per
175 program. As, given a video URL, the downloader doesn't know how to
176 extract all the needed information, task that InfoExtractors do, it
177 has to pass the URL to one of them.
178
179 For this, YoutubeDL objects have a method that allows
180 InfoExtractors to be registered in a given order. When it is passed
181 a URL, the YoutubeDL object handles it to the first InfoExtractor it
182 finds that reports being able to handle it. The InfoExtractor extracts
183 all the information about the video or videos the URL refers to, and
184 YoutubeDL process the extracted information, possibly using a File
185 Downloader to download the video.
186
187 YoutubeDL objects accept a lot of parameters. In order not to saturate
188 the object constructor with arguments, it receives a dictionary of
189 options instead. These options are available through the params
190 attribute for the InfoExtractors to use. The YoutubeDL also
191 registers itself as the downloader in charge for the InfoExtractors
192 that are added to it, so this is a "mutual registration".
193
194 Available options:
195
196 username: Username for authentication purposes.
197 password: Password for authentication purposes.
180940e0 198 videopassword: Password for accessing a video.
1da50aa3
S
199 ap_mso: Adobe Pass multiple-system operator identifier.
200 ap_username: Multiple-system operator account username.
201 ap_password: Multiple-system operator account password.
8222d8de
JMF
202 usenetrc: Use netrc for authentication instead.
203 verbose: Print additional info to stdout.
204 quiet: Do not print messages to stdout.
ad8915b7 205 no_warnings: Do not print out anything for warnings.
bb66c247 206 forceprint: A dict with keys WHEN mapped to a list of templates to
207 print to stdout. The allowed keys are video or any of the
208 items in utils.POSTPROCESS_WHEN.
ca30f449 209 For compatibility, a single list is also accepted
bb66c247 210 print_to_file: A dict with keys WHEN (same as forceprint) mapped to
211 a list of tuples with (template, filename)
53c18592 212 forceurl: Force printing final URL. (Deprecated)
213 forcetitle: Force printing title. (Deprecated)
214 forceid: Force printing ID. (Deprecated)
215 forcethumbnail: Force printing thumbnail URL. (Deprecated)
216 forcedescription: Force printing description. (Deprecated)
217 forcefilename: Force printing final filename. (Deprecated)
218 forceduration: Force printing duration. (Deprecated)
8694c600 219 forcejson: Force printing info_dict as JSON.
63e0be34
PH
220 dump_single_json: Force printing the info_dict of the whole playlist
221 (or video) as a single JSON line.
c25228e5 222 force_write_download_archive: Force writing download archive regardless
223 of 'skip_download' or 'simulate'.
b7b04c78 224 simulate: Do not download the video files. If unset (or None),
225 simulate only if listsubtitles, listformats or list_thumbnails is used
eb8a4433 226 format: Video format code. see "FORMAT SELECTION" for more details.
093a1710 227 You can also pass a function. The function takes 'ctx' as
228 argument and returns the formats to download.
229 See "build_format_selector" for an implementation
63ad4d43 230 allow_unplayable_formats: Allow unplayable formats to be extracted and downloaded.
b7da73eb 231 ignore_no_formats_error: Ignore "No video formats" error. Usefull for
232 extracting metadata even if the video is not actually
233 available for download (experimental)
0930b11f 234 format_sort: A list of fields by which to sort the video formats.
235 See "Sorting Formats" for more details.
c25228e5 236 format_sort_force: Force the given format_sort. see "Sorting Formats"
237 for more details.
08d30158 238 prefer_free_formats: Whether to prefer video formats with free containers
239 over non-free ones of same quality.
c25228e5 240 allow_multiple_video_streams: Allow multiple video streams to be merged
241 into a single file
242 allow_multiple_audio_streams: Allow multiple audio streams to be merged
243 into a single file
0ba692ac 244 check_formats Whether to test if the formats are downloadable.
9f1a1c36 245 Can be True (check all), False (check none),
246 'selected' (check selected formats),
0ba692ac 247 or None (check only if requested by extractor)
4524baf0 248 paths: Dictionary of output paths. The allowed keys are 'home'
249 'temp' and the keys of OUTTMPL_TYPES (in utils.py)
de6000d9 250 outtmpl: Dictionary of templates for output names. Allowed keys
4524baf0 251 are 'default' and the keys of OUTTMPL_TYPES (in utils.py).
34488702 252 For compatibility with youtube-dl, a single string can also be used
a820dc72
RA
253 outtmpl_na_placeholder: Placeholder for unavailable meta fields.
254 restrictfilenames: Do not allow "&" and spaces in file names
255 trim_file_name: Limit length of filename (extension excluded)
4524baf0 256 windowsfilenames: Force the filenames to be windows compatible
b1940459 257 ignoreerrors: Do not stop on download/postprocessing errors.
258 Can be 'only_download' to ignore only download errors.
259 Default is 'only_download' for CLI, but False for API
26e2805c 260 skip_playlist_after_errors: Number of allowed failures until the rest of
261 the playlist is skipped
d22dec74 262 force_generic_extractor: Force downloader to use the generic extractor
0c3d0f51 263 overwrites: Overwrite all video and metadata files if True,
264 overwrite only non-video files if None
265 and don't overwrite any file if False
34488702 266 For compatibility with youtube-dl,
267 "nooverwrites" may also be used instead
8222d8de
JMF
268 playliststart: Playlist item to start at.
269 playlistend: Playlist item to end at.
c14e88f0 270 playlist_items: Specific indices of playlist to download.
ff815fe6 271 playlistreverse: Download playlist items in reverse order.
75822ca7 272 playlistrandom: Download playlist items in random order.
8222d8de
JMF
273 matchtitle: Download only matching titles.
274 rejecttitle: Reject downloads for matching titles.
8bf9319e 275 logger: Log messages to a logging.Logger instance.
8222d8de 276 logtostderr: Log messages to stderr instead of stdout.
819e0531 277 consoletitle: Display progress in console window's titlebar.
8222d8de
JMF
278 writedescription: Write the video description to a .description file
279 writeinfojson: Write the video description to a .info.json file
75d43ca0 280 clean_infojson: Remove private fields from the infojson
34488702 281 getcomments: Extract video comments. This will not be written to disk
06167fbb 282 unless writeinfojson is also given
1fb07d10 283 writeannotations: Write the video annotations to a .annotations.xml file
8222d8de 284 writethumbnail: Write the thumbnail image to a file
c25228e5 285 allow_playlist_files: Whether to write playlists' description, infojson etc
286 also to disk when using the 'write*' options
ec82d85a 287 write_all_thumbnails: Write all thumbnail formats to files
732044af 288 writelink: Write an internet shortcut file, depending on the
289 current platform (.url/.webloc/.desktop)
290 writeurllink: Write a Windows internet shortcut file (.url)
291 writewebloclink: Write a macOS internet shortcut file (.webloc)
292 writedesktoplink: Write a Linux internet shortcut file (.desktop)
8222d8de 293 writesubtitles: Write the video subtitles to a file
741dd8ea 294 writeautomaticsub: Write the automatically generated subtitles to a file
245524e6 295 allsubtitles: Deprecated - Use subtitleslangs = ['all']
c32b0aab 296 Downloads all the subtitles of the video
0b7f3118 297 (requires writesubtitles or writeautomaticsub)
8222d8de 298 listsubtitles: Lists all available subtitles for the video
a504ced0 299 subtitlesformat: The format code for subtitles
c32b0aab 300 subtitleslangs: List of languages of the subtitles to download (can be regex).
301 The list may contain "all" to refer to all the available
302 subtitles. The language can be prefixed with a "-" to
303 exclude it from the requested languages. Eg: ['all', '-live_chat']
8222d8de
JMF
304 keepvideo: Keep the video file after post-processing
305 daterange: A DateRange object, download only if the upload_date is in the range.
306 skip_download: Skip the actual download of the video file
c35f9e72 307 cachedir: Location of the cache files in the filesystem.
a0e07d31 308 False to disable filesystem cache.
47192f92 309 noplaylist: Download single video instead of a playlist if in doubt.
8dbe9899
PH
310 age_limit: An integer representing the user's age in years.
311 Unsuitable videos for the given age are skipped.
5fe18bdb
PH
312 min_views: An integer representing the minimum view count the video
313 must have in order to not be skipped.
314 Videos without view count information are always
315 downloaded. None for no limit.
316 max_views: An integer representing the maximum view count.
317 Videos that are more popular than that are not
318 downloaded.
319 Videos without view count information are always
320 downloaded. None for no limit.
321 download_archive: File name of a file where all downloads are recorded.
c1c9a79c
PH
322 Videos already present in the file are not downloaded
323 again.
8a51f564 324 break_on_existing: Stop the download process after attempting to download a
325 file that is in the archive.
326 break_on_reject: Stop the download process when encountering a video that
327 has been filtered out.
b222c271 328 break_per_url: Whether break_on_reject and break_on_existing
329 should act on each input URL as opposed to for the entire queue
8a51f564 330 cookiefile: File name where cookies should be read from and dumped to
f59f5ef8
MB
331 cookiesfrombrowser: A tuple containing the name of the browser, the profile
332 name/pathfrom where cookies are loaded, and the name of the
333 keyring. Eg: ('chrome', ) or ('vivaldi', 'default', 'BASICTEXT')
f81c62a6 334 legacyserverconnect: Explicitly allow HTTPS connection to servers that do not
335 support RFC 5746 secure renegotiation
f59f5ef8 336 nocheckcertificate: Do not verify SSL certificates
7e8c0af0
PH
337 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
338 At the moment, this is only supported by YouTube.
8b7539d2 339 http_headers: A dictionary of custom headers to be used for all requests
a1ee09e8 340 proxy: URL of the proxy server to use
38cce791 341 geo_verification_proxy: URL of the proxy to use for IP address verification
504f20dd 342 on geo-restricted sites.
e344693b 343 socket_timeout: Time to wait for unresponsive hosts, in seconds
0783b09b
PH
344 bidi_workaround: Work around buggy terminals without bidirectional text
345 support, using fridibi
a0ddb8a2 346 debug_printtraffic:Print out sent and received HTTP traffic
91f071af 347 include_ads: Download ads as well (deprecated)
04b4d394
PH
348 default_search: Prepend this string if an input url is not valid.
349 'auto' for elaborate guessing
62fec3b2 350 encoding: Use this encoding instead of the system-specified.
e8ee972c 351 extract_flat: Do not resolve URLs, return the immediate result.
057a5206
PH
352 Pass in 'in_playlist' to only show this behavior for
353 playlist items.
f2ebc5c7 354 wait_for_video: If given, wait for scheduled streams to become available.
355 The value should be a tuple containing the range
356 (min_secs, max_secs) to wait between retries
4f026faf 357 postprocessors: A list of dictionaries, each with an entry
71b640cc 358 * key: The name of the postprocessor. See
7a5c1cfe 359 yt_dlp/postprocessor/__init__.py for a list.
bb66c247 360 * when: When to run the postprocessor. Allowed values are
361 the entries of utils.POSTPROCESS_WHEN
56d868db 362 Assumed to be 'post_process' if not given
b5ae35ee 363 post_hooks: Deprecated - Register a custom postprocessor instead
364 A list of functions that get called as the final step
ab8e5e51
AM
365 for each video file, after all postprocessors have been
366 called. The filename will be passed as the only argument.
71b640cc
PH
367 progress_hooks: A list of functions that get called on download
368 progress, with a dictionary with the entries
5cda4eda 369 * status: One of "downloading", "error", or "finished".
ee69b99a 370 Check this first and ignore unknown values.
3ba7740d 371 * info_dict: The extracted info_dict
71b640cc 372
5cda4eda 373 If status is one of "downloading", or "finished", the
ee69b99a
PH
374 following properties may also be present:
375 * filename: The final filename (always present)
5cda4eda 376 * tmpfilename: The filename we're currently writing to
71b640cc
PH
377 * downloaded_bytes: Bytes on disk
378 * total_bytes: Size of the whole file, None if unknown
5cda4eda
PH
379 * total_bytes_estimate: Guess of the eventual file size,
380 None if unavailable.
381 * elapsed: The number of seconds since download started.
71b640cc
PH
382 * eta: The estimated time in seconds, None if unknown
383 * speed: The download speed in bytes/second, None if
384 unknown
5cda4eda
PH
385 * fragment_index: The counter of the currently
386 downloaded video fragment.
387 * fragment_count: The number of fragments (= individual
388 files that will be merged)
71b640cc
PH
389
390 Progress hooks are guaranteed to be called at least once
391 (with status "finished") if the download is successful.
819e0531 392 postprocessor_hooks: A list of functions that get called on postprocessing
393 progress, with a dictionary with the entries
394 * status: One of "started", "processing", or "finished".
395 Check this first and ignore unknown values.
396 * postprocessor: Name of the postprocessor
397 * info_dict: The extracted info_dict
398
399 Progress hooks are guaranteed to be called at least twice
400 (with status "started" and "finished") if the processing is successful.
45598f15 401 merge_output_format: Extension to use when merging formats.
6b591b29 402 final_ext: Expected final extension; used to detect when the file was
59a7a13e 403 already downloaded and converted
6271f1ca
PH
404 fixup: Automatically correct known faults of the file.
405 One of:
406 - "never": do nothing
407 - "warn": only emit a warning
408 - "detect_or_warn": check whether we can do anything
62cd676c 409 about it, warn otherwise (default)
504f20dd 410 source_address: Client-side IP address to bind to.
6ec6cb4e 411 call_home: Boolean, true iff we are allowed to contact the
7a5c1cfe 412 yt-dlp servers for debugging. (BROKEN)
1cf376f5 413 sleep_interval_requests: Number of seconds to sleep between requests
414 during extraction
7aa589a5
S
415 sleep_interval: Number of seconds to sleep before each download when
416 used alone or a lower bound of a range for randomized
417 sleep before each download (minimum possible number
418 of seconds to sleep) when used along with
419 max_sleep_interval.
420 max_sleep_interval:Upper bound of a range for randomized sleep before each
421 download (maximum possible number of seconds to sleep).
422 Must only be used along with sleep_interval.
423 Actual sleep time will be a random float from range
424 [sleep_interval; max_sleep_interval].
1cf376f5 425 sleep_interval_subtitles: Number of seconds to sleep before each subtitle download
cfb56d1a
PH
426 listformats: Print an overview of available video formats and exit.
427 list_thumbnails: Print a table of all thumbnails and exit.
347de493
PH
428 match_filter: A function that gets called with the info_dict of
429 every video.
430 If it returns a message, the video is ignored.
431 If it returns None, the video is downloaded.
432 match_filter_func in utils.py is one example for this.
7e5db8c9 433 no_color: Do not emit color codes in output.
0a840f58 434 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
504f20dd 435 HTTP header
0a840f58 436 geo_bypass_country:
773f291d
S
437 Two-letter ISO 3166-2 country code that will be used for
438 explicit geographic restriction bypassing via faking
504f20dd 439 X-Forwarded-For HTTP header
5f95927a
S
440 geo_bypass_ip_block:
441 IP range in CIDR notation that will be used similarly to
504f20dd 442 geo_bypass_country
71b640cc 443
85729c51 444 The following options determine which downloader is picked:
52a8a1e1 445 external_downloader: A dictionary of protocol keys and the executable of the
446 external downloader to use for it. The allowed protocols
447 are default|http|ftp|m3u8|dash|rtsp|rtmp|mms.
448 Set the value to 'native' to use the native downloader
449 hls_prefer_native: Deprecated - Use external_downloader = {'m3u8': 'native'}
450 or {'m3u8': 'ffmpeg'} instead.
451 Use the native HLS downloader instead of ffmpeg/avconv
bf09af3a
S
452 if True, otherwise use ffmpeg/avconv if False, otherwise
453 use downloader suggested by extractor if None.
53ed7066 454 compat_opts: Compatibility options. See "Differences in default behavior".
3acf6d38 455 The following options do not work when used through the API:
b5ae35ee 456 filename, abort-on-error, multistreams, no-live-chat, format-sort
dac5df5a 457 no-clean-infojson, no-playlist-metafiles, no-keep-subs, no-attach-info-json.
e4f02757 458 Refer __init__.py for their implementation
819e0531 459 progress_template: Dictionary of templates for progress outputs.
460 Allowed keys are 'download', 'postprocess',
461 'download-title' (console title) and 'postprocess-title'.
462 The template is mapped on a dictionary with keys 'progress' and 'info'
fe7e0c98 463
8222d8de 464 The following parameters are not used by YoutubeDL itself, they are used by
7a5c1cfe 465 the downloader (see yt_dlp/downloader/common.py):
51d9739f 466 nopart, updatetime, buffersize, ratelimit, throttledratelimit, min_filesize,
205a0654
EH
467 max_filesize, test, noresizebuffer, retries, file_access_retries, fragment_retries,
468 continuedl, noprogress, xattr_set_filesize, hls_use_mpegts, http_chunk_size,
59a7a13e 469 external_downloader_args, concurrent_fragment_downloads.
76b1bd67
JMF
470
471 The following options are used by the post processors:
d4a24f40 472 prefer_ffmpeg: If False, use avconv instead of ffmpeg if both are available,
e4172ac9 473 otherwise prefer ffmpeg. (avconv support is deprecated)
c0b7d117
S
474 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
475 to the binary or its containing directory.
43820c03 476 postprocessor_args: A dictionary of postprocessor/executable keys (in lower case)
34488702 477 and a list of additional command-line arguments for the
478 postprocessor/executable. The dict can also have "PP+EXE" keys
479 which are used when the given exe is used by the given PP.
480 Use 'default' as the name for arguments to passed to all PP
481 For compatibility with youtube-dl, a single list of args
482 can also be used
e409895f 483
484 The following options are used by the extractors:
62bff2c1 485 extractor_retries: Number of times to retry for known errors
486 dynamic_mpd: Whether to process dynamic DASH manifests (default: True)
e409895f 487 hls_split_discontinuity: Split HLS playlists to different formats at
62bff2c1 488 discontinuities such as ad breaks (default: False)
5d3a0e79 489 extractor_args: A dictionary of arguments to be passed to the extractors.
490 See "EXTRACTOR ARGUMENTS" for details.
491 Eg: {'youtube': {'skip': ['dash', 'hls']}}
88f23a18 492 mark_watched: Mark videos watched (even with --simulate). Only for YouTube
5d3a0e79 493 youtube_include_dash_manifest: Deprecated - Use extractor_args instead.
494 If True (default), DASH manifests and related
62bff2c1 495 data will be downloaded and processed by extractor.
496 You can reduce network I/O by disabling it if you don't
497 care about DASH. (only for youtube)
5d3a0e79 498 youtube_include_hls_manifest: Deprecated - Use extractor_args instead.
499 If True (default), HLS manifests and related
62bff2c1 500 data will be downloaded and processed by extractor.
501 You can reduce network I/O by disabling it if you don't
502 care about HLS. (only for youtube)
8222d8de
JMF
503 """
504
c9969434
S
505 _NUMERIC_FIELDS = set((
506 'width', 'height', 'tbr', 'abr', 'asr', 'vbr', 'fps', 'filesize', 'filesize_approx',
e6f21b3d 507 'timestamp', 'release_timestamp',
c9969434
S
508 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
509 'average_rating', 'comment_count', 'age_limit',
510 'start_time', 'end_time',
511 'chapter_number', 'season_number', 'episode_number',
512 'track_number', 'disc_number', 'release_year',
c9969434
S
513 ))
514
48ee10ee 515 _format_selection_exts = {
516 'audio': {'m4a', 'mp3', 'ogg', 'aac'},
517 'video': {'mp4', 'flv', 'webm', '3gp'},
518 'storyboards': {'mhtml'},
519 }
520
3511266b 521 def __init__(self, params=None, auto_init=True):
883d4b1e 522 """Create a FileDownloader object with the given options.
523 @param auto_init Whether to load the default extractors and print header (if verbose).
49a57e70 524 Set to 'no_verbose_header' to not print the header
883d4b1e 525 """
e9f9a10f
JMF
526 if params is None:
527 params = {}
592b7485 528 self.params = params
8b7491c8 529 self._ies = {}
56c73665 530 self._ies_instances = {}
1e43a6f7 531 self._pps = {k: [] for k in POSTPROCESS_WHEN}
b35496d8 532 self._printed_messages = set()
1cf376f5 533 self._first_webpage_request = True
ab8e5e51 534 self._post_hooks = []
933605d7 535 self._progress_hooks = []
819e0531 536 self._postprocessor_hooks = []
8222d8de
JMF
537 self._download_retcode = 0
538 self._num_downloads = 0
9c906919 539 self._num_videos = 0
592b7485 540 self._playlist_level = 0
541 self._playlist_urls = set()
a0e07d31 542 self.cache = Cache(self)
34308b30 543
819e0531 544 windows_enable_vt_mode()
cf4f42cb 545 self._out_files = {
546 'error': sys.stderr,
547 'print': sys.stderr if self.params.get('logtostderr') else sys.stdout,
548 'console': None if compat_os_name == 'nt' else next(
549 filter(supports_terminal_sequences, (sys.stderr, sys.stdout)), None)
550 }
551 self._out_files['screen'] = sys.stderr if self.params.get('quiet') else self._out_files['print']
ec11a9f4 552 self._allow_colors = {
cf4f42cb 553 type_: not self.params.get('no_color') and supports_terminal_sequences(self._out_files[type_])
554 for type_ in ('screen', 'error')
ec11a9f4 555 }
819e0531 556
a61f4b28 557 if sys.version_info < (3, 6):
558 self.report_warning(
0181adef 559 'Python version %d.%d is not supported! Please update to Python 3.6 or above' % sys.version_info[:2])
a61f4b28 560
88acdbc2 561 if self.params.get('allow_unplayable_formats'):
562 self.report_warning(
ec11a9f4 563 f'You have asked for {self._format_err("UNPLAYABLE", self.Styles.EMPHASIS)} formats to be listed/downloaded. '
819e0531 564 'This is a developer option intended for debugging. \n'
565 ' If you experience any issues while using this option, '
ec11a9f4 566 f'{self._format_err("DO NOT", self.Styles.ERROR)} open a bug report')
88acdbc2 567
be5df5ee
S
568 def check_deprecated(param, option, suggestion):
569 if self.params.get(param) is not None:
53ed7066 570 self.report_warning('%s is deprecated. Use %s instead' % (option, suggestion))
be5df5ee
S
571 return True
572 return False
573
574 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
38cce791
YCH
575 if self.params.get('geo_verification_proxy') is None:
576 self.params['geo_verification_proxy'] = self.params['cn_verification_proxy']
577
0d1bb027 578 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
579 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
53ed7066 580 check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"')
0d1bb027 581
49a57e70 582 for msg in self.params.get('_warnings', []):
0d1bb027 583 self.report_warning(msg)
ee8dd27a 584 for msg in self.params.get('_deprecation_warnings', []):
585 self.deprecation_warning(msg)
0d1bb027 586
ec11a9f4 587 if 'list-formats' in self.params.get('compat_opts', []):
588 self.params['listformats_table'] = False
589
b5ae35ee 590 if 'overwrites' not in self.params and self.params.get('nooverwrites') is not None:
b868936c 591 # nooverwrites was unnecessarily changed to overwrites
592 # in 0c3d0f51778b153f65c21906031c2e091fcfb641
593 # This ensures compatibility with both keys
594 self.params['overwrites'] = not self.params['nooverwrites']
b5ae35ee 595 elif self.params.get('overwrites') is None:
596 self.params.pop('overwrites', None)
b868936c 597 else:
598 self.params['nooverwrites'] = not self.params['overwrites']
b9d973be 599
455a15e2 600 self.params.setdefault('forceprint', {})
601 self.params.setdefault('print_to_file', {})
bb66c247 602
603 # Compatibility with older syntax
ca30f449 604 if not isinstance(params['forceprint'], dict):
455a15e2 605 self.params['forceprint'] = {'video': params['forceprint']}
ca30f449 606
455a15e2 607 if self.params.get('bidi_workaround', False):
1c088fa8
PH
608 try:
609 import pty
610 master, slave = pty.openpty()
003c69a8 611 width = compat_get_terminal_size().columns
1c088fa8
PH
612 if width is None:
613 width_args = []
614 else:
615 width_args = ['-w', str(width)]
5d681e96 616 sp_kwargs = dict(
1c088fa8
PH
617 stdin=subprocess.PIPE,
618 stdout=slave,
cf4f42cb 619 stderr=self._out_files['error'])
5d681e96 620 try:
d3c93ec2 621 self._output_process = Popen(['bidiv'] + width_args, **sp_kwargs)
5d681e96 622 except OSError:
d3c93ec2 623 self._output_process = Popen(['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
5d681e96 624 self._output_channel = os.fdopen(master, 'rb')
1c088fa8 625 except OSError as ose:
66e7ace1 626 if ose.errno == errno.ENOENT:
49a57e70 627 self.report_warning(
628 'Could not find fribidi executable, ignoring --bidi-workaround. '
629 'Make sure that fribidi is an executable file in one of the directories in your $PATH.')
1c088fa8
PH
630 else:
631 raise
0783b09b 632
3089bc74
S
633 if (sys.platform != 'win32'
634 and sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
455a15e2 635 and not self.params.get('restrictfilenames', False)):
e9137224 636 # Unicode filesystem API will throw errors (#1474, #13027)
34308b30 637 self.report_warning(
6febd1c1 638 'Assuming --restrict-filenames since file system encoding '
1b725173 639 'cannot encode all characters. '
6febd1c1 640 'Set the LC_ALL environment variable to fix this.')
4a98cdbf 641 self.params['restrictfilenames'] = True
34308b30 642
de6000d9 643 self.outtmpl_dict = self.parse_outtmpl()
486dd09e 644
187986a8 645 # Creating format selector here allows us to catch syntax errors before the extraction
646 self.format_selector = (
fa9f30b8 647 self.params.get('format') if self.params.get('format') in (None, '-')
093a1710 648 else self.params['format'] if callable(self.params['format'])
187986a8 649 else self.build_format_selector(self.params['format']))
650
8b7539d2 651 # Set http_headers defaults according to std_headers
652 self.params['http_headers'] = merge_headers(std_headers, self.params.get('http_headers', {}))
653
dca08720
PH
654 self._setup_opener()
655
3511266b 656 if auto_init:
883d4b1e 657 if auto_init != 'no_verbose_header':
658 self.print_debug_header()
3511266b
PH
659 self.add_default_info_extractors()
660
013b50b7 661 hooks = {
662 'post_hooks': self.add_post_hook,
663 'progress_hooks': self.add_progress_hook,
664 'postprocessor_hooks': self.add_postprocessor_hook,
665 }
666 for opt, fn in hooks.items():
667 for ph in self.params.get(opt, []):
668 fn(ph)
71b640cc 669
5bfc8bee 670 for pp_def_raw in self.params.get('postprocessors', []):
671 pp_def = dict(pp_def_raw)
672 when = pp_def.pop('when', 'post_process')
673 self.add_post_processor(
674 get_postprocessor(pp_def.pop('key'))(self, **compat_kwargs(pp_def)),
675 when=when)
676
51fb4995
YCH
677 register_socks_protocols()
678
ed39cac5 679 def preload_download_archive(fn):
680 """Preload the archive, if any is specified"""
681 if fn is None:
682 return False
49a57e70 683 self.write_debug(f'Loading archive file {fn!r}')
ed39cac5 684 try:
685 with locked_file(fn, 'r', encoding='utf-8') as archive_file:
686 for line in archive_file:
687 self.archive.add(line.strip())
688 except IOError as ioe:
689 if ioe.errno != errno.ENOENT:
690 raise
691 return False
692 return True
693
694 self.archive = set()
695 preload_download_archive(self.params.get('download_archive'))
696
7d4111ed
PH
697 def warn_if_short_id(self, argv):
698 # short YouTube ID starting with dash?
699 idxs = [
700 i for i, a in enumerate(argv)
701 if re.match(r'^-[0-9A-Za-z_-]{10}$', a)]
702 if idxs:
703 correct_argv = (
7a5c1cfe 704 ['yt-dlp']
3089bc74
S
705 + [a for i, a in enumerate(argv) if i not in idxs]
706 + ['--'] + [argv[i] for i in idxs]
7d4111ed
PH
707 )
708 self.report_warning(
709 'Long argument string detected. '
49a57e70 710 'Use -- to separate parameters and URLs, like this:\n%s' %
7d4111ed
PH
711 args_to_str(correct_argv))
712
8222d8de
JMF
713 def add_info_extractor(self, ie):
714 """Add an InfoExtractor object to the end of the list."""
8b7491c8 715 ie_key = ie.ie_key()
716 self._ies[ie_key] = ie
e52d7f85 717 if not isinstance(ie, type):
8b7491c8 718 self._ies_instances[ie_key] = ie
e52d7f85 719 ie.set_downloader(self)
8222d8de 720
8b7491c8 721 def _get_info_extractor_class(self, ie_key):
722 ie = self._ies.get(ie_key)
723 if ie is None:
724 ie = get_info_extractor(ie_key)
725 self.add_info_extractor(ie)
726 return ie
727
56c73665
JMF
728 def get_info_extractor(self, ie_key):
729 """
730 Get an instance of an IE with name ie_key, it will try to get one from
731 the _ies list, if there's no instance it will create a new one and add
732 it to the extractor list.
733 """
734 ie = self._ies_instances.get(ie_key)
735 if ie is None:
736 ie = get_info_extractor(ie_key)()
737 self.add_info_extractor(ie)
738 return ie
739
023fa8c4
JMF
740 def add_default_info_extractors(self):
741 """
742 Add the InfoExtractors returned by gen_extractors to the end of the list
743 """
e52d7f85 744 for ie in gen_extractor_classes():
023fa8c4
JMF
745 self.add_info_extractor(ie)
746
56d868db 747 def add_post_processor(self, pp, when='post_process'):
8222d8de 748 """Add a PostProcessor object to the end of the chain."""
5bfa4862 749 self._pps[when].append(pp)
8222d8de
JMF
750 pp.set_downloader(self)
751
ab8e5e51
AM
752 def add_post_hook(self, ph):
753 """Add the post hook"""
754 self._post_hooks.append(ph)
755
933605d7 756 def add_progress_hook(self, ph):
819e0531 757 """Add the download progress hook"""
933605d7 758 self._progress_hooks.append(ph)
8ab470f1 759
819e0531 760 def add_postprocessor_hook(self, ph):
761 """Add the postprocessing progress hook"""
762 self._postprocessor_hooks.append(ph)
5bfc8bee 763 for pps in self._pps.values():
764 for pp in pps:
765 pp.add_progress_hook(ph)
819e0531 766
1c088fa8 767 def _bidi_workaround(self, message):
5d681e96 768 if not hasattr(self, '_output_channel'):
1c088fa8
PH
769 return message
770
5d681e96 771 assert hasattr(self, '_output_process')
11b85ce6 772 assert isinstance(message, compat_str)
6febd1c1
PH
773 line_count = message.count('\n') + 1
774 self._output_process.stdin.write((message + '\n').encode('utf-8'))
5d681e96 775 self._output_process.stdin.flush()
6febd1c1 776 res = ''.join(self._output_channel.readline().decode('utf-8')
9e1a5b84 777 for _ in range(line_count))
6febd1c1 778 return res[:-len('\n')]
1c088fa8 779
b35496d8 780 def _write_string(self, message, out=None, only_once=False):
781 if only_once:
782 if message in self._printed_messages:
783 return
784 self._printed_messages.add(message)
785 write_string(message, out=out, encoding=self.params.get('encoding'))
734f90bb 786
cf4f42cb 787 def to_stdout(self, message, skip_eol=False, quiet=None):
0760b0a7 788 """Print message to stdout"""
cf4f42cb 789 if quiet is not None:
790 self.deprecation_warning('"ydl.to_stdout" no longer accepts the argument quiet. Use "ydl.to_screen" instead')
791 self._write_string(
792 '%s%s' % (self._bidi_workaround(message), ('' if skip_eol else '\n')),
793 self._out_files['print'])
794
795 def to_screen(self, message, skip_eol=False, quiet=None):
796 """Print message to screen if not in quiet mode"""
8bf9319e 797 if self.params.get('logger'):
43afe285 798 self.params['logger'].debug(message)
cf4f42cb 799 return
800 if (self.params.get('quiet') if quiet is None else quiet) and not self.params.get('verbose'):
801 return
802 self._write_string(
803 '%s%s' % (self._bidi_workaround(message), ('' if skip_eol else '\n')),
804 self._out_files['screen'])
8222d8de 805
b35496d8 806 def to_stderr(self, message, only_once=False):
0760b0a7 807 """Print message to stderr"""
11b85ce6 808 assert isinstance(message, compat_str)
8bf9319e 809 if self.params.get('logger'):
43afe285
IB
810 self.params['logger'].error(message)
811 else:
cf4f42cb 812 self._write_string('%s\n' % self._bidi_workaround(message), self._out_files['error'], only_once=only_once)
813
814 def _send_console_code(self, code):
815 if compat_os_name == 'nt' or not self._out_files['console']:
816 return
817 self._write_string(code, self._out_files['console'])
8222d8de 818
1e5b9a95
PH
819 def to_console_title(self, message):
820 if not self.params.get('consoletitle', False):
821 return
3efb96a6 822 message = remove_terminal_sequences(message)
4bede0d8
C
823 if compat_os_name == 'nt':
824 if ctypes.windll.kernel32.GetConsoleWindow():
825 # c_wchar_p() might not be necessary if `message` is
826 # already of type unicode()
827 ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
cf4f42cb 828 else:
829 self._send_console_code(f'\033]0;{message}\007')
1e5b9a95 830
bdde425c 831 def save_console_title(self):
cf4f42cb 832 if not self.params.get('consoletitle') or self.params.get('simulate'):
bdde425c 833 return
592b7485 834 self._send_console_code('\033[22;0t') # Save the title on stack
bdde425c
PH
835
836 def restore_console_title(self):
cf4f42cb 837 if not self.params.get('consoletitle') or self.params.get('simulate'):
bdde425c 838 return
592b7485 839 self._send_console_code('\033[23;0t') # Restore the title from stack
bdde425c
PH
840
841 def __enter__(self):
842 self.save_console_title()
843 return self
844
845 def __exit__(self, *args):
846 self.restore_console_title()
f89197d7 847
dca08720 848 if self.params.get('cookiefile') is not None:
1bab3437 849 self.cookiejar.save(ignore_discard=True, ignore_expires=True)
bdde425c 850
fa9f30b8 851 def trouble(self, message=None, tb=None, is_error=True):
8222d8de
JMF
852 """Determine action to take when a download problem appears.
853
854 Depending on if the downloader has been configured to ignore
855 download errors or not, this method may throw an exception or
856 not when errors are found, after printing the message.
857
fa9f30b8 858 @param tb If given, is additional traceback information
859 @param is_error Whether to raise error according to ignorerrors
8222d8de
JMF
860 """
861 if message is not None:
862 self.to_stderr(message)
863 if self.params.get('verbose'):
864 if tb is None:
865 if sys.exc_info()[0]: # if .trouble has been called from an except block
6febd1c1 866 tb = ''
8222d8de 867 if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
6febd1c1 868 tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
c0384f22 869 tb += encode_compat_str(traceback.format_exc())
8222d8de
JMF
870 else:
871 tb_data = traceback.format_list(traceback.extract_stack())
6febd1c1 872 tb = ''.join(tb_data)
c19bc311 873 if tb:
874 self.to_stderr(tb)
fa9f30b8 875 if not is_error:
876 return
b1940459 877 if not self.params.get('ignoreerrors'):
8222d8de
JMF
878 if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
879 exc_info = sys.exc_info()[1].exc_info
880 else:
881 exc_info = sys.exc_info()
882 raise DownloadError(message, exc_info)
883 self._download_retcode = 1
884
ec11a9f4 885 class Styles(Enum):
886 HEADERS = 'yellow'
f304da8a 887 EMPHASIS = 'light blue'
ec11a9f4 888 ID = 'green'
889 DELIM = 'blue'
890 ERROR = 'red'
891 WARNING = 'yellow'
ff51ed58 892 SUPPRESS = 'light black'
ec11a9f4 893
7578d77d 894 def _format_text(self, handle, allow_colors, text, f, fallback=None, *, test_encoding=False):
ec11a9f4 895 if test_encoding:
896 original_text = text
5c104538 897 # handle.encoding can be None. See https://github.com/yt-dlp/yt-dlp/issues/2711
898 encoding = self.params.get('encoding') or getattr(handle, 'encoding', None) or 'ascii'
ec11a9f4 899 text = text.encode(encoding, 'ignore').decode(encoding)
900 if fallback is not None and text != original_text:
901 text = fallback
902 if isinstance(f, self.Styles):
f304da8a 903 f = f.value
7578d77d 904 return format_text(text, f) if allow_colors else text if fallback is None else fallback
ec11a9f4 905
906 def _format_screen(self, *args, **kwargs):
7578d77d 907 return self._format_text(
cf4f42cb 908 self._out_files['screen'], self._allow_colors['screen'], *args, **kwargs)
ec11a9f4 909
910 def _format_err(self, *args, **kwargs):
7578d77d 911 return self._format_text(
cf4f42cb 912 self._out_files['error'], self._allow_colors['error'], *args, **kwargs)
819e0531 913
c84aeac6 914 def report_warning(self, message, only_once=False):
8222d8de
JMF
915 '''
916 Print the message to stderr, it will be prefixed with 'WARNING:'
917 If stderr is a tty file the 'WARNING:' will be colored
918 '''
6d07ce01
JMF
919 if self.params.get('logger') is not None:
920 self.params['logger'].warning(message)
8222d8de 921 else:
ad8915b7
PH
922 if self.params.get('no_warnings'):
923 return
ec11a9f4 924 self.to_stderr(f'{self._format_err("WARNING:", self.Styles.WARNING)} {message}', only_once)
8222d8de 925
ee8dd27a 926 def deprecation_warning(self, message):
927 if self.params.get('logger') is not None:
928 self.params['logger'].warning('DeprecationWarning: {message}')
929 else:
930 self.to_stderr(f'{self._format_err("DeprecationWarning:", self.Styles.ERROR)} {message}', True)
931
fa9f30b8 932 def report_error(self, message, *args, **kwargs):
8222d8de
JMF
933 '''
934 Do the same as trouble, but prefixes the message with 'ERROR:', colored
935 in red if stderr is a tty file.
936 '''
fa9f30b8 937 self.trouble(f'{self._format_err("ERROR:", self.Styles.ERROR)} {message}', *args, **kwargs)
8222d8de 938
b35496d8 939 def write_debug(self, message, only_once=False):
0760b0a7 940 '''Log debug message or Print message to stderr'''
941 if not self.params.get('verbose', False):
942 return
943 message = '[debug] %s' % message
944 if self.params.get('logger'):
945 self.params['logger'].debug(message)
946 else:
b35496d8 947 self.to_stderr(message, only_once)
0760b0a7 948
8222d8de
JMF
949 def report_file_already_downloaded(self, file_name):
950 """Report file has already been fully downloaded."""
951 try:
6febd1c1 952 self.to_screen('[download] %s has already been downloaded' % file_name)
ce02ed60 953 except UnicodeEncodeError:
6febd1c1 954 self.to_screen('[download] The file has already been downloaded')
8222d8de 955
0c3d0f51 956 def report_file_delete(self, file_name):
957 """Report that existing file will be deleted."""
958 try:
c25228e5 959 self.to_screen('Deleting existing file %s' % file_name)
0c3d0f51 960 except UnicodeEncodeError:
c25228e5 961 self.to_screen('Deleting existing file')
0c3d0f51 962
319b6059 963 def raise_no_formats(self, info, forced=False, *, msg=None):
1151c407 964 has_drm = info.get('__has_drm')
319b6059 965 ignored, expected = self.params.get('ignore_no_formats_error'), bool(msg)
966 msg = msg or has_drm and 'This video is DRM protected' or 'No video formats found!'
967 if forced or not ignored:
1151c407 968 raise ExtractorError(msg, video_id=info['id'], ie=info['extractor'],
319b6059 969 expected=has_drm or ignored or expected)
88acdbc2 970 else:
971 self.report_warning(msg)
972
de6000d9 973 def parse_outtmpl(self):
974 outtmpl_dict = self.params.get('outtmpl', {})
975 if not isinstance(outtmpl_dict, dict):
976 outtmpl_dict = {'default': outtmpl_dict}
71ce444a 977 # Remove spaces in the default template
978 if self.params.get('restrictfilenames'):
979 sanitize = lambda x: x.replace(' - ', ' ').replace(' ', '-')
980 else:
981 sanitize = lambda x: x
de6000d9 982 outtmpl_dict.update({
71ce444a 983 k: sanitize(v) for k, v in DEFAULT_OUTTMPL.items()
80c03fa9 984 if outtmpl_dict.get(k) is None})
de6000d9 985 for key, val in outtmpl_dict.items():
986 if isinstance(val, bytes):
987 self.report_warning(
988 'Parameter outtmpl is bytes, but should be a unicode string. '
989 'Put from __future__ import unicode_literals at the top of your code file or consider switching to Python 3.x.')
990 return outtmpl_dict
991
21cd8fae 992 def get_output_path(self, dir_type='', filename=None):
993 paths = self.params.get('paths', {})
994 assert isinstance(paths, dict)
995 path = os.path.join(
996 expand_path(paths.get('home', '').strip()),
997 expand_path(paths.get(dir_type, '').strip()) if dir_type else '',
998 filename or '')
999
1000 # Temporary fix for #4787
1001 # 'Treat' all problem characters by passing filename through preferredencoding
1002 # to workaround encoding issues with subprocess on python2 @ Windows
1003 if sys.version_info < (3, 0) and sys.platform == 'win32':
1004 path = encodeFilename(path, True).decode(preferredencoding())
1005 return sanitize_path(path, force=self.params.get('windowsfilenames'))
1006
76a264ac 1007 @staticmethod
901130bb 1008 def _outtmpl_expandpath(outtmpl):
1009 # expand_path translates '%%' into '%' and '$$' into '$'
1010 # correspondingly that is not what we want since we need to keep
1011 # '%%' intact for template dict substitution step. Working around
1012 # with boundary-alike separator hack.
1013 sep = ''.join([random.choice(ascii_letters) for _ in range(32)])
1014 outtmpl = outtmpl.replace('%%', '%{0}%'.format(sep)).replace('$$', '${0}$'.format(sep))
1015
1016 # outtmpl should be expand_path'ed before template dict substitution
1017 # because meta fields may contain env variables we don't want to
1018 # be expanded. For example, for outtmpl "%(title)s.%(ext)s" and
1019 # title "Hello $PATH", we don't want `$PATH` to be expanded.
1020 return expand_path(outtmpl).replace(sep, '')
1021
1022 @staticmethod
1023 def escape_outtmpl(outtmpl):
1024 ''' Escape any remaining strings like %s, %abc% etc. '''
1025 return re.sub(
1026 STR_FORMAT_RE_TMPL.format('', '(?![%(\0])'),
1027 lambda mobj: ('' if mobj.group('has_key') else '%') + mobj.group(0),
1028 outtmpl)
1029
1030 @classmethod
1031 def validate_outtmpl(cls, outtmpl):
76a264ac 1032 ''' @return None or Exception object '''
7d1eb38a 1033 outtmpl = re.sub(
37893bb0 1034 STR_FORMAT_RE_TMPL.format('[^)]*', '[ljqBUDS]'),
7d1eb38a 1035 lambda mobj: f'{mobj.group(0)[:-1]}s',
1036 cls._outtmpl_expandpath(outtmpl))
76a264ac 1037 try:
7d1eb38a 1038 cls.escape_outtmpl(outtmpl) % collections.defaultdict(int)
76a264ac 1039 return None
1040 except ValueError as err:
1041 return err
1042
03b4de72 1043 @staticmethod
1044 def _copy_infodict(info_dict):
1045 info_dict = dict(info_dict)
09b49e1f 1046 info_dict.pop('__postprocessors', None)
03b4de72 1047 return info_dict
1048
e0fd9573 1049 def prepare_outtmpl(self, outtmpl, info_dict, sanitize=False):
1050 """ Make the outtmpl and info_dict suitable for substitution: ydl.escape_outtmpl(outtmpl) % info_dict
1051 @param sanitize Whether to sanitize the output as a filename.
1052 For backward compatibility, a function can also be passed
1053 """
1054
6e84b215 1055 info_dict.setdefault('epoch', int(time.time())) # keep epoch consistent once set
143db31d 1056
03b4de72 1057 info_dict = self._copy_infodict(info_dict)
752cda38 1058 info_dict['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
53c18592 1059 formatSeconds(info_dict['duration'], '-' if sanitize else ':')
143db31d 1060 if info_dict.get('duration', None) is not None
1061 else None)
752cda38 1062 info_dict['autonumber'] = self.params.get('autonumber_start', 1) - 1 + self._num_downloads
9c906919 1063 info_dict['video_autonumber'] = self._num_videos
752cda38 1064 if info_dict.get('resolution') is None:
1065 info_dict['resolution'] = self.format_resolution(info_dict, default=None)
143db31d 1066
e6f21b3d 1067 # For fields playlist_index, playlist_autonumber and autonumber convert all occurrences
143db31d 1068 # of %(field)s to %(field)0Nd for backward compatibility
1069 field_size_compat_map = {
ec11a9f4 1070 'playlist_index': number_of_digits(info_dict.get('_last_playlist_index') or 0),
1071 'playlist_autonumber': number_of_digits(info_dict.get('n_entries') or 0),
752cda38 1072 'autonumber': self.params.get('autonumber_size') or 5,
143db31d 1073 }
752cda38 1074
385a27fa 1075 TMPL_DICT = {}
37893bb0 1076 EXTERNAL_FORMAT_RE = re.compile(STR_FORMAT_RE_TMPL.format('[^)]*', f'[{STR_FORMAT_TYPES}ljqBUDS]'))
385a27fa 1077 MATH_FUNCTIONS = {
1078 '+': float.__add__,
1079 '-': float.__sub__,
1080 }
e625be0d 1081 # Field is of the form key1.key2...
1082 # where keys (except first) can be string, int or slice
2b8a2973 1083 FIELD_RE = r'\w*(?:\.(?:\w+|{num}|{num}?(?::{num}?){{1,2}}))*'.format(num=r'(?:-?\d+)')
e0fd9573 1084 MATH_FIELD_RE = r'''(?:{field}|{num})'''.format(field=FIELD_RE, num=r'-?\d+(?:.\d+)?')
385a27fa 1085 MATH_OPERATORS_RE = r'(?:%s)' % '|'.join(map(re.escape, MATH_FUNCTIONS.keys()))
e625be0d 1086 INTERNAL_FORMAT_RE = re.compile(r'''(?x)
1087 (?P<negate>-)?
385a27fa 1088 (?P<fields>{field})
1089 (?P<maths>(?:{math_op}{math_field})*)
e625be0d 1090 (?:>(?P<strf_format>.+?))?
e978789f
P
1091 (?P<alternate>(?<!\\),[^|&)]+)?
1092 (?:&(?P<replacement>.*?))?
e625be0d 1093 (?:\|(?P<default>.*?))?
385a27fa 1094 $'''.format(field=FIELD_RE, math_op=MATH_OPERATORS_RE, math_field=MATH_FIELD_RE))
752cda38 1095
2b8a2973 1096 def _traverse_infodict(k):
1097 k = k.split('.')
1098 if k[0] == '':
1099 k.pop(0)
1100 return traverse_obj(info_dict, k, is_user_input=True, traverse_string=True)
76a264ac 1101
752cda38 1102 def get_value(mdict):
1103 # Object traversal
2b8a2973 1104 value = _traverse_infodict(mdict['fields'])
752cda38 1105 # Negative
1106 if mdict['negate']:
1107 value = float_or_none(value)
1108 if value is not None:
1109 value *= -1
1110 # Do maths
385a27fa 1111 offset_key = mdict['maths']
1112 if offset_key:
752cda38 1113 value = float_or_none(value)
1114 operator = None
385a27fa 1115 while offset_key:
1116 item = re.match(
1117 MATH_FIELD_RE if operator else MATH_OPERATORS_RE,
1118 offset_key).group(0)
1119 offset_key = offset_key[len(item):]
1120 if operator is None:
752cda38 1121 operator = MATH_FUNCTIONS[item]
385a27fa 1122 continue
1123 item, multiplier = (item[1:], -1) if item[0] == '-' else (item, 1)
1124 offset = float_or_none(item)
1125 if offset is None:
2b8a2973 1126 offset = float_or_none(_traverse_infodict(item))
385a27fa 1127 try:
1128 value = operator(value, multiplier * offset)
1129 except (TypeError, ZeroDivisionError):
1130 return None
1131 operator = None
752cda38 1132 # Datetime formatting
1133 if mdict['strf_format']:
7c37ff97 1134 value = strftime_or_none(value, mdict['strf_format'].replace('\\,', ','))
752cda38 1135
1136 return value
1137
b868936c 1138 na = self.params.get('outtmpl_na_placeholder', 'NA')
1139
e0fd9573 1140 def filename_sanitizer(key, value, restricted=self.params.get('restrictfilenames')):
1141 return sanitize_filename(str(value), restricted=restricted,
1142 is_id=re.search(r'(^|[_.])id(\.|$)', key))
1143
1144 sanitizer = sanitize if callable(sanitize) else filename_sanitizer
1145 sanitize = bool(sanitize)
1146
6e84b215 1147 def _dumpjson_default(obj):
1148 if isinstance(obj, (set, LazyList)):
1149 return list(obj)
adbc4ec4 1150 return repr(obj)
6e84b215 1151
752cda38 1152 def create_key(outer_mobj):
1153 if not outer_mobj.group('has_key'):
b836dc94 1154 return outer_mobj.group(0)
752cda38 1155 key = outer_mobj.group('key')
752cda38 1156 mobj = re.match(INTERNAL_FORMAT_RE, key)
e0fd9573 1157 initial_field = mobj.group('fields') if mobj else ''
e978789f 1158 value, replacement, default = None, None, na
7c37ff97 1159 while mobj:
e625be0d 1160 mobj = mobj.groupdict()
7c37ff97 1161 default = mobj['default'] if mobj['default'] is not None else default
752cda38 1162 value = get_value(mobj)
e978789f 1163 replacement = mobj['replacement']
7c37ff97 1164 if value is None and mobj['alternate']:
1165 mobj = re.match(INTERNAL_FORMAT_RE, mobj['alternate'][1:])
1166 else:
1167 break
752cda38 1168
b868936c 1169 fmt = outer_mobj.group('format')
752cda38 1170 if fmt == 's' and value is not None and key in field_size_compat_map.keys():
1171 fmt = '0{:d}d'.format(field_size_compat_map[key])
1172
e978789f 1173 value = default if value is None else value if replacement is None else replacement
752cda38 1174
4476d2c7 1175 flags = outer_mobj.group('conversion') or ''
7d1eb38a 1176 str_fmt = f'{fmt[:-1]}s'
524e2e4f 1177 if fmt[-1] == 'l': # list
4476d2c7 1178 delim = '\n' if '#' in flags else ', '
9e907ebd 1179 value, fmt = delim.join(map(str, variadic(value, allowed_types=(str, bytes)))), str_fmt
524e2e4f 1180 elif fmt[-1] == 'j': # json
4476d2c7 1181 value, fmt = json.dumps(value, default=_dumpjson_default, indent=4 if '#' in flags else None), str_fmt
524e2e4f 1182 elif fmt[-1] == 'q': # quoted
4476d2c7 1183 value = map(str, variadic(value) if '#' in flags else [value])
1184 value, fmt = ' '.join(map(compat_shlex_quote, value)), str_fmt
524e2e4f 1185 elif fmt[-1] == 'B': # bytes
f5aa5cfb 1186 value = f'%{str_fmt}'.encode('utf-8') % str(value).encode('utf-8')
1187 value, fmt = value.decode('utf-8', 'ignore'), 's'
524e2e4f 1188 elif fmt[-1] == 'U': # unicode normalized
524e2e4f 1189 value, fmt = unicodedata.normalize(
1190 # "+" = compatibility equivalence, "#" = NFD
4476d2c7 1191 'NF%s%s' % ('K' if '+' in flags else '', 'D' if '#' in flags else 'C'),
524e2e4f 1192 value), str_fmt
e0fd9573 1193 elif fmt[-1] == 'D': # decimal suffix
abbeeebc 1194 num_fmt, fmt = fmt[:-1].replace('#', ''), 's'
1195 value = format_decimal_suffix(value, f'%{num_fmt}f%s' if num_fmt else '%d%s',
1196 factor=1024 if '#' in flags else 1000)
37893bb0 1197 elif fmt[-1] == 'S': # filename sanitization
e0fd9573 1198 value, fmt = filename_sanitizer(initial_field, value, restricted='#' in flags), str_fmt
7d1eb38a 1199 elif fmt[-1] == 'c':
524e2e4f 1200 if value:
1201 value = str(value)[0]
76a264ac 1202 else:
524e2e4f 1203 fmt = str_fmt
76a264ac 1204 elif fmt[-1] not in 'rs': # numeric
a439a3a4 1205 value = float_or_none(value)
752cda38 1206 if value is None:
1207 value, fmt = default, 's'
901130bb 1208
752cda38 1209 if sanitize:
1210 if fmt[-1] == 'r':
1211 # If value is an object, sanitize might convert it to a string
1212 # So we convert it to repr first
7d1eb38a 1213 value, fmt = repr(value), str_fmt
639f1cea 1214 if fmt[-1] in 'csr':
e0fd9573 1215 value = sanitizer(initial_field, value)
901130bb 1216
b868936c 1217 key = '%s\0%s' % (key.replace('%', '%\0'), outer_mobj.group('format'))
385a27fa 1218 TMPL_DICT[key] = value
b868936c 1219 return '{prefix}%({key}){fmt}'.format(key=key, fmt=fmt, prefix=outer_mobj.group('prefix'))
752cda38 1220
385a27fa 1221 return EXTERNAL_FORMAT_RE.sub(create_key, outtmpl), TMPL_DICT
143db31d 1222
819e0531 1223 def evaluate_outtmpl(self, outtmpl, info_dict, *args, **kwargs):
1224 outtmpl, info_dict = self.prepare_outtmpl(outtmpl, info_dict, *args, **kwargs)
1225 return self.escape_outtmpl(outtmpl) % info_dict
1226
de6000d9 1227 def _prepare_filename(self, info_dict, tmpl_type='default'):
8222d8de 1228 try:
b836dc94 1229 outtmpl = self._outtmpl_expandpath(self.outtmpl_dict.get(tmpl_type, self.outtmpl_dict['default']))
e0fd9573 1230 filename = self.evaluate_outtmpl(outtmpl, info_dict, True)
6a0546e3 1231 if not filename:
1232 return None
15da37c7 1233
6a0546e3 1234 if tmpl_type in ('default', 'temp'):
1235 final_ext, ext = self.params.get('final_ext'), info_dict.get('ext')
1236 if final_ext and ext and final_ext != ext and filename.endswith(f'.{final_ext}'):
1237 filename = replace_extension(filename, ext, final_ext)
1238 else:
1239 force_ext = OUTTMPL_TYPES[tmpl_type]
1240 if force_ext:
1241 filename = replace_extension(filename, force_ext, info_dict.get('ext'))
de6000d9 1242
bdc3fd2f
U
1243 # https://github.com/blackjack4494/youtube-dlc/issues/85
1244 trim_file_name = self.params.get('trim_file_name', False)
1245 if trim_file_name:
5c22c63d 1246 no_ext, *ext = filename.rsplit('.', 2)
1247 filename = join_nonempty(no_ext[:trim_file_name], *ext, delim='.')
bdc3fd2f 1248
0202b52a 1249 return filename
8222d8de 1250 except ValueError as err:
6febd1c1 1251 self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')')
8222d8de
JMF
1252 return None
1253
de6000d9 1254 def prepare_filename(self, info_dict, dir_type='', warn=False):
1255 """Generate the output filename."""
21cd8fae 1256
de6000d9 1257 filename = self._prepare_filename(info_dict, dir_type or 'default')
80c03fa9 1258 if not filename and dir_type not in ('', 'temp'):
1259 return ''
de6000d9 1260
c84aeac6 1261 if warn:
21cd8fae 1262 if not self.params.get('paths'):
de6000d9 1263 pass
1264 elif filename == '-':
c84aeac6 1265 self.report_warning('--paths is ignored when an outputting to stdout', only_once=True)
de6000d9 1266 elif os.path.isabs(filename):
c84aeac6 1267 self.report_warning('--paths is ignored since an absolute path is given in output template', only_once=True)
de6000d9 1268 if filename == '-' or not filename:
1269 return filename
1270
21cd8fae 1271 return self.get_output_path(dir_type, filename)
0202b52a 1272
120fe513 1273 def _match_entry(self, info_dict, incomplete=False, silent=False):
ecdec191 1274 """ Returns None if the file should be downloaded """
8222d8de 1275
c77495e3 1276 video_title = info_dict.get('title', info_dict.get('id', 'video'))
1277
8b0d7497 1278 def check_filter():
8b0d7497 1279 if 'title' in info_dict:
1280 # This can happen when we're just evaluating the playlist
1281 title = info_dict['title']
1282 matchtitle = self.params.get('matchtitle', False)
1283 if matchtitle:
1284 if not re.search(matchtitle, title, re.IGNORECASE):
1285 return '"' + title + '" title did not match pattern "' + matchtitle + '"'
1286 rejecttitle = self.params.get('rejecttitle', False)
1287 if rejecttitle:
1288 if re.search(rejecttitle, title, re.IGNORECASE):
1289 return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
1290 date = info_dict.get('upload_date')
1291 if date is not None:
1292 dateRange = self.params.get('daterange', DateRange())
1293 if date not in dateRange:
1294 return '%s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange)
1295 view_count = info_dict.get('view_count')
1296 if view_count is not None:
1297 min_views = self.params.get('min_views')
1298 if min_views is not None and view_count < min_views:
1299 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
1300 max_views = self.params.get('max_views')
1301 if max_views is not None and view_count > max_views:
1302 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
1303 if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')):
1304 return 'Skipping "%s" because it is age restricted' % video_title
8b0d7497 1305
8f18aca8 1306 match_filter = self.params.get('match_filter')
1307 if match_filter is not None:
1308 try:
1309 ret = match_filter(info_dict, incomplete=incomplete)
1310 except TypeError:
1311 # For backward compatibility
1312 ret = None if incomplete else match_filter(info_dict)
1313 if ret is not None:
1314 return ret
8b0d7497 1315 return None
1316
c77495e3 1317 if self.in_download_archive(info_dict):
1318 reason = '%s has already been recorded in the archive' % video_title
1319 break_opt, break_err = 'break_on_existing', ExistingVideoReached
1320 else:
1321 reason = check_filter()
1322 break_opt, break_err = 'break_on_reject', RejectedVideoReached
8b0d7497 1323 if reason is not None:
120fe513 1324 if not silent:
1325 self.to_screen('[download] ' + reason)
c77495e3 1326 if self.params.get(break_opt, False):
1327 raise break_err()
8b0d7497 1328 return reason
fe7e0c98 1329
b6c45014
JMF
1330 @staticmethod
1331 def add_extra_info(info_dict, extra_info):
1332 '''Set the keys from extra_info in info dict if they are missing'''
1333 for key, value in extra_info.items():
1334 info_dict.setdefault(key, value)
1335
409e1828 1336 def extract_info(self, url, download=True, ie_key=None, extra_info=None,
61aa5ba3 1337 process=True, force_generic_extractor=False):
41d1cca3 1338 """
1339 Return a list with a dictionary for each video extracted.
1340
1341 Arguments:
1342 url -- URL to extract
1343
1344 Keyword arguments:
1345 download -- whether to download videos during extraction
1346 ie_key -- extractor key hint
1347 extra_info -- dictionary containing the extra values to add to each result
1348 process -- whether to resolve all unresolved references (URLs, playlist items),
1349 must be True for download to work.
1350 force_generic_extractor -- force using the generic extractor
1351 """
fe7e0c98 1352
409e1828 1353 if extra_info is None:
1354 extra_info = {}
1355
61aa5ba3 1356 if not ie_key and force_generic_extractor:
d22dec74
S
1357 ie_key = 'Generic'
1358
8222d8de 1359 if ie_key:
8b7491c8 1360 ies = {ie_key: self._get_info_extractor_class(ie_key)}
8222d8de
JMF
1361 else:
1362 ies = self._ies
1363
8b7491c8 1364 for ie_key, ie in ies.items():
8222d8de
JMF
1365 if not ie.suitable(url):
1366 continue
1367
1368 if not ie.working():
6febd1c1
PH
1369 self.report_warning('The program functionality for this site has been marked as broken, '
1370 'and will probably not work.')
8222d8de 1371
1151c407 1372 temp_id = ie.get_temp_id(url)
a0566bbf 1373 if temp_id is not None and self.in_download_archive({'id': temp_id, 'ie_key': ie_key}):
5e5be0c0 1374 self.to_screen(f'[{ie_key}] {temp_id}: has already been recorded in the archive')
1375 if self.params.get('break_on_existing', False):
1376 raise ExistingVideoReached()
a0566bbf 1377 break
8b7491c8 1378 return self.__extract_info(url, self.get_info_extractor(ie_key), download, extra_info, process)
a0566bbf 1379 else:
1380 self.report_error('no suitable InfoExtractor for URL %s' % url)
1381
8e5fecc8 1382 def __handle_extraction_exceptions(func):
b5ae35ee 1383 @functools.wraps(func)
a0566bbf 1384 def wrapper(self, *args, **kwargs):
6da22e7d 1385 while True:
1386 try:
1387 return func(self, *args, **kwargs)
1388 except (DownloadCancelled, LazyList.IndexError, PagedList.IndexError):
8222d8de 1389 raise
6da22e7d 1390 except ReExtractInfo as e:
1391 if e.expected:
1392 self.to_screen(f'{e}; Re-extracting data')
1393 else:
1394 self.to_stderr('\r')
1395 self.report_warning(f'{e}; Re-extracting data')
1396 continue
1397 except GeoRestrictedError as e:
1398 msg = e.msg
1399 if e.countries:
1400 msg += '\nThis video is available in %s.' % ', '.join(
1401 map(ISO3166Utils.short2full, e.countries))
1402 msg += '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1403 self.report_error(msg)
1404 except ExtractorError as e: # An error we somewhat expected
1405 self.report_error(str(e), e.format_traceback())
1406 except Exception as e:
1407 if self.params.get('ignoreerrors'):
1408 self.report_error(str(e), tb=encode_compat_str(traceback.format_exc()))
1409 else:
1410 raise
1411 break
a0566bbf 1412 return wrapper
1413
f2ebc5c7 1414 def _wait_for_video(self, ie_result):
1415 if (not self.params.get('wait_for_video')
1416 or ie_result.get('_type', 'video') != 'video'
1417 or ie_result.get('formats') or ie_result.get('url')):
1418 return
1419
1420 format_dur = lambda dur: '%02d:%02d:%02d' % timetuple_from_msec(dur * 1000)[:-1]
1421 last_msg = ''
1422
1423 def progress(msg):
1424 nonlocal last_msg
1425 self.to_screen(msg + ' ' * (len(last_msg) - len(msg)) + '\r', skip_eol=True)
1426 last_msg = msg
1427
1428 min_wait, max_wait = self.params.get('wait_for_video')
1429 diff = try_get(ie_result, lambda x: x['release_timestamp'] - time.time())
1430 if diff is None and ie_result.get('live_status') == 'is_upcoming':
38d79fd1 1431 diff = random.randrange(min_wait, max_wait) if (max_wait and min_wait) else (max_wait or min_wait)
f2ebc5c7 1432 self.report_warning('Release time of video is not known')
1433 elif (diff or 0) <= 0:
1434 self.report_warning('Video should already be available according to extracted info')
38d79fd1 1435 diff = min(max(diff or 0, min_wait or 0), max_wait or float('inf'))
f2ebc5c7 1436 self.to_screen(f'[wait] Waiting for {format_dur(diff)} - Press Ctrl+C to try now')
1437
1438 wait_till = time.time() + diff
1439 try:
1440 while True:
1441 diff = wait_till - time.time()
1442 if diff <= 0:
1443 progress('')
1444 raise ReExtractInfo('[wait] Wait period ended', expected=True)
1445 progress(f'[wait] Remaining time until next attempt: {self._format_screen(format_dur(diff), self.Styles.EMPHASIS)}')
1446 time.sleep(1)
1447 except KeyboardInterrupt:
1448 progress('')
1449 raise ReExtractInfo('[wait] Interrupted by user', expected=True)
1450 except BaseException as e:
1451 if not isinstance(e, ReExtractInfo):
1452 self.to_screen('')
1453 raise
1454
a0566bbf 1455 @__handle_extraction_exceptions
58f197b7 1456 def __extract_info(self, url, ie, download, extra_info, process):
a0566bbf 1457 ie_result = ie.extract(url)
1458 if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
1459 return
1460 if isinstance(ie_result, list):
1461 # Backwards compatibility: old IE result format
1462 ie_result = {
1463 '_type': 'compat_list',
1464 'entries': ie_result,
1465 }
e37d0efb 1466 if extra_info.get('original_url'):
1467 ie_result.setdefault('original_url', extra_info['original_url'])
a0566bbf 1468 self.add_default_extra_info(ie_result, ie, url)
1469 if process:
f2ebc5c7 1470 self._wait_for_video(ie_result)
a0566bbf 1471 return self.process_ie_result(ie_result, download, extra_info)
8222d8de 1472 else:
a0566bbf 1473 return ie_result
fe7e0c98 1474
ea38e55f 1475 def add_default_extra_info(self, ie_result, ie, url):
6033d980 1476 if url is not None:
1477 self.add_extra_info(ie_result, {
1478 'webpage_url': url,
1479 'original_url': url,
57ebfca3 1480 })
1481 webpage_url = ie_result.get('webpage_url')
1482 if webpage_url:
1483 self.add_extra_info(ie_result, {
1484 'webpage_url_basename': url_basename(webpage_url),
1485 'webpage_url_domain': get_domain(webpage_url),
6033d980 1486 })
1487 if ie is not None:
1488 self.add_extra_info(ie_result, {
1489 'extractor': ie.IE_NAME,
1490 'extractor_key': ie.ie_key(),
1491 })
ea38e55f 1492
58adec46 1493 def process_ie_result(self, ie_result, download=True, extra_info=None):
8222d8de
JMF
1494 """
1495 Take the result of the ie(may be modified) and resolve all unresolved
1496 references (URLs, playlist items).
1497
1498 It will also download the videos if 'download'.
1499 Returns the resolved ie_result.
1500 """
58adec46 1501 if extra_info is None:
1502 extra_info = {}
e8ee972c
PH
1503 result_type = ie_result.get('_type', 'video')
1504
057a5206 1505 if result_type in ('url', 'url_transparent'):
134c6ea8 1506 ie_result['url'] = sanitize_url(ie_result['url'])
e37d0efb 1507 if ie_result.get('original_url'):
1508 extra_info.setdefault('original_url', ie_result['original_url'])
1509
057a5206 1510 extract_flat = self.params.get('extract_flat', False)
3089bc74
S
1511 if ((extract_flat == 'in_playlist' and 'playlist' in extra_info)
1512 or extract_flat is True):
ecb54191 1513 info_copy = ie_result.copy()
6033d980 1514 ie = try_get(ie_result.get('ie_key'), self.get_info_extractor)
360167b9 1515 if ie and not ie_result.get('id'):
4614bc22 1516 info_copy['id'] = ie.get_temp_id(ie_result['url'])
6033d980 1517 self.add_default_extra_info(info_copy, ie, ie_result['url'])
4614bc22 1518 self.add_extra_info(info_copy, extra_info)
b5475f11 1519 info_copy, _ = self.pre_process(info_copy)
ecb54191 1520 self.__forced_printings(info_copy, self.prepare_filename(info_copy), incomplete=True)
4614bc22 1521 if self.params.get('force_write_download_archive', False):
1522 self.record_download_archive(info_copy)
e8ee972c
PH
1523 return ie_result
1524
8222d8de 1525 if result_type == 'video':
b6c45014 1526 self.add_extra_info(ie_result, extra_info)
9c2b75b5 1527 ie_result = self.process_video_result(ie_result, download=download)
28b0eb0f 1528 additional_urls = (ie_result or {}).get('additional_urls')
9c2b75b5 1529 if additional_urls:
e9f4ccd1 1530 # TODO: Improve MetadataParserPP to allow setting a list
9c2b75b5 1531 if isinstance(additional_urls, compat_str):
1532 additional_urls = [additional_urls]
1533 self.to_screen(
1534 '[info] %s: %d additional URL(s) requested' % (ie_result['id'], len(additional_urls)))
1535 self.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls))
1536 ie_result['additional_entries'] = [
1537 self.extract_info(
b69fd25c 1538 url, download, extra_info=extra_info,
9c2b75b5 1539 force_generic_extractor=self.params.get('force_generic_extractor'))
1540 for url in additional_urls
1541 ]
1542 return ie_result
8222d8de
JMF
1543 elif result_type == 'url':
1544 # We have to add extra_info to the results because it may be
1545 # contained in a playlist
07cce701 1546 return self.extract_info(
1547 ie_result['url'], download,
1548 ie_key=ie_result.get('ie_key'),
1549 extra_info=extra_info)
7fc3fa05
PH
1550 elif result_type == 'url_transparent':
1551 # Use the information from the embedding page
1552 info = self.extract_info(
1553 ie_result['url'], ie_key=ie_result.get('ie_key'),
1554 extra_info=extra_info, download=False, process=False)
1555
1640eb09
S
1556 # extract_info may return None when ignoreerrors is enabled and
1557 # extraction failed with an error, don't crash and return early
1558 # in this case
1559 if not info:
1560 return info
1561
412c617d
PH
1562 force_properties = dict(
1563 (k, v) for k, v in ie_result.items() if v is not None)
0396806f 1564 for f in ('_type', 'url', 'id', 'extractor', 'extractor_key', 'ie_key'):
412c617d
PH
1565 if f in force_properties:
1566 del force_properties[f]
1567 new_result = info.copy()
1568 new_result.update(force_properties)
7fc3fa05 1569
0563f7ac
S
1570 # Extracted info may not be a video result (i.e.
1571 # info.get('_type', 'video') != video) but rather an url or
1572 # url_transparent. In such cases outer metadata (from ie_result)
1573 # should be propagated to inner one (info). For this to happen
1574 # _type of info should be overridden with url_transparent. This
067aa17e 1575 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
0563f7ac
S
1576 if new_result.get('_type') == 'url':
1577 new_result['_type'] = 'url_transparent'
7fc3fa05
PH
1578
1579 return self.process_ie_result(
1580 new_result, download=download, extra_info=extra_info)
40fcba5e 1581 elif result_type in ('playlist', 'multi_video'):
30a074c2 1582 # Protect from infinite recursion due to recursively nested playlists
1583 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
1584 webpage_url = ie_result['webpage_url']
1585 if webpage_url in self._playlist_urls:
7e85e872 1586 self.to_screen(
30a074c2 1587 '[download] Skipping already downloaded playlist: %s'
1588 % ie_result.get('title') or ie_result.get('id'))
1589 return
7e85e872 1590
30a074c2 1591 self._playlist_level += 1
1592 self._playlist_urls.add(webpage_url)
03f83004 1593 self._fill_common_fields(ie_result, False)
bc516a3f 1594 self._sanitize_thumbnails(ie_result)
30a074c2 1595 try:
1596 return self.__process_playlist(ie_result, download)
1597 finally:
1598 self._playlist_level -= 1
1599 if not self._playlist_level:
1600 self._playlist_urls.clear()
8222d8de 1601 elif result_type == 'compat_list':
c9bf4114
PH
1602 self.report_warning(
1603 'Extractor %s returned a compat_list result. '
1604 'It needs to be updated.' % ie_result.get('extractor'))
5f6a1245 1605
8222d8de 1606 def _fixup(r):
b868936c 1607 self.add_extra_info(r, {
1608 'extractor': ie_result['extractor'],
1609 'webpage_url': ie_result['webpage_url'],
1610 'webpage_url_basename': url_basename(ie_result['webpage_url']),
0bb322b9 1611 'webpage_url_domain': get_domain(ie_result['webpage_url']),
b868936c 1612 'extractor_key': ie_result['extractor_key'],
1613 })
8222d8de
JMF
1614 return r
1615 ie_result['entries'] = [
b6c45014 1616 self.process_ie_result(_fixup(r), download, extra_info)
8222d8de
JMF
1617 for r in ie_result['entries']
1618 ]
1619 return ie_result
1620 else:
1621 raise Exception('Invalid result type: %s' % result_type)
1622
e92caff5 1623 def _ensure_dir_exists(self, path):
1624 return make_dir(path, self.report_error)
1625
3b603dbd 1626 @staticmethod
1627 def _playlist_infodict(ie_result, **kwargs):
1628 return {
1629 **ie_result,
1630 'playlist': ie_result.get('title') or ie_result.get('id'),
1631 'playlist_id': ie_result.get('id'),
1632 'playlist_title': ie_result.get('title'),
1633 'playlist_uploader': ie_result.get('uploader'),
1634 'playlist_uploader_id': ie_result.get('uploader_id'),
1635 'playlist_index': 0,
1636 **kwargs,
1637 }
1638
30a074c2 1639 def __process_playlist(self, ie_result, download):
1640 # We process each entry in the playlist
1641 playlist = ie_result.get('title') or ie_result.get('id')
1642 self.to_screen('[download] Downloading playlist: %s' % playlist)
1643
498f5606 1644 if 'entries' not in ie_result:
aa9369a2 1645 raise EntryNotInPlaylist('There are no entries')
7c7f7161 1646
1647 MissingEntry = object()
498f5606 1648 incomplete_entries = bool(ie_result.get('requested_entries'))
1649 if incomplete_entries:
bf5f605e 1650 def fill_missing_entries(entries, indices):
7c7f7161 1651 ret = [MissingEntry] * max(indices)
bf5f605e 1652 for i, entry in zip(indices, entries):
498f5606 1653 ret[i - 1] = entry
1654 return ret
1655 ie_result['entries'] = fill_missing_entries(ie_result['entries'], ie_result['requested_entries'])
02fd60d3 1656
30a074c2 1657 playlist_results = []
1658
56a8fb4f 1659 playliststart = self.params.get('playliststart', 1)
30a074c2 1660 playlistend = self.params.get('playlistend')
1661 # For backwards compatibility, interpret -1 as whole list
1662 if playlistend == -1:
1663 playlistend = None
1664
1665 playlistitems_str = self.params.get('playlist_items')
1666 playlistitems = None
1667 if playlistitems_str is not None:
1668 def iter_playlistitems(format):
1669 for string_segment in format.split(','):
1670 if '-' in string_segment:
1671 start, end = string_segment.split('-')
1672 for item in range(int(start), int(end) + 1):
1673 yield int(item)
1674 else:
1675 yield int(string_segment)
1676 playlistitems = orderedSet(iter_playlistitems(playlistitems_str))
1677
1678 ie_entries = ie_result['entries']
8e5fecc8 1679 if isinstance(ie_entries, list):
ed8d87f9 1680 playlist_count = len(ie_entries)
f0d785d3 1681 msg = f'Collected {playlist_count} videos; downloading %d of them'
1682 ie_result['playlist_count'] = ie_result.get('playlist_count') or playlist_count
1683
8e5fecc8 1684 def get_entry(i):
1685 return ie_entries[i - 1]
1686 else:
f0d785d3 1687 msg = 'Downloading %d videos'
c586f9e8 1688 if not isinstance(ie_entries, (PagedList, LazyList)):
8e5fecc8 1689 ie_entries = LazyList(ie_entries)
d37707bd 1690 elif isinstance(ie_entries, InAdvancePagedList):
1691 if ie_entries._pagesize == 1:
1692 playlist_count = ie_entries._pagecount
8e5fecc8 1693
1694 def get_entry(i):
1695 return YoutubeDL.__handle_extraction_exceptions(
1696 lambda self, i: ie_entries[i - 1]
1697 )(self, i)
50fed816 1698
f0d785d3 1699 entries, broken = [], False
ff1c7fc9 1700 items = playlistitems if playlistitems is not None else itertools.count(playliststart)
1701 for i in items:
1702 if i == 0:
1703 continue
56a8fb4f 1704 if playlistitems is None and playlistend is not None and playlistend < i:
1705 break
1706 entry = None
1707 try:
50fed816 1708 entry = get_entry(i)
7c7f7161 1709 if entry is MissingEntry:
498f5606 1710 raise EntryNotInPlaylist()
56a8fb4f 1711 except (IndexError, EntryNotInPlaylist):
1712 if incomplete_entries:
aa9369a2 1713 raise EntryNotInPlaylist(f'Entry {i} cannot be found')
56a8fb4f 1714 elif not playlistitems:
1715 break
1716 entries.append(entry)
120fe513 1717 try:
1718 if entry is not None:
1719 self._match_entry(entry, incomplete=True, silent=True)
1720 except (ExistingVideoReached, RejectedVideoReached):
f0d785d3 1721 broken = True
120fe513 1722 break
56a8fb4f 1723 ie_result['entries'] = entries
30a074c2 1724
56a8fb4f 1725 # Save playlist_index before re-ordering
1726 entries = [
9e598870 1727 ((playlistitems[i - 1] if playlistitems else i + playliststart - 1), entry)
56a8fb4f 1728 for i, entry in enumerate(entries, 1)
1729 if entry is not None]
1730 n_entries = len(entries)
498f5606 1731
f0d785d3 1732 if not (ie_result.get('playlist_count') or broken or playlistitems or playlistend):
1733 ie_result['playlist_count'] = n_entries
1734
e08a85d8 1735 if not playlistitems and (playliststart != 1 or playlistend):
56a8fb4f 1736 playlistitems = list(range(playliststart, playliststart + n_entries))
498f5606 1737 ie_result['requested_entries'] = playlistitems
1738
e08a85d8 1739 _infojson_written = False
0bfc53d0 1740 write_playlist_files = self.params.get('allow_playlist_files', True)
1741 if write_playlist_files and self.params.get('list_thumbnails'):
1742 self.list_thumbnails(ie_result)
1743 if write_playlist_files and not self.params.get('simulate'):
3b603dbd 1744 ie_copy = self._playlist_infodict(ie_result, n_entries=n_entries)
e08a85d8 1745 _infojson_written = self._write_info_json(
1746 'playlist', ie_result, self.prepare_filename(ie_copy, 'pl_infojson'))
1747 if _infojson_written is None:
80c03fa9 1748 return
1749 if self._write_description('playlist', ie_result,
1750 self.prepare_filename(ie_copy, 'pl_description')) is None:
1751 return
681de68e 1752 # TODO: This should be passed to ThumbnailsConvertor if necessary
80c03fa9 1753 self._write_thumbnails('playlist', ie_copy, self.prepare_filename(ie_copy, 'pl_thumbnail'))
30a074c2 1754
1755 if self.params.get('playlistreverse', False):
1756 entries = entries[::-1]
30a074c2 1757 if self.params.get('playlistrandom', False):
1758 random.shuffle(entries)
1759
1760 x_forwarded_for = ie_result.get('__x_forwarded_for_ip')
1761
56a8fb4f 1762 self.to_screen('[%s] playlist %s: %s' % (ie_result['extractor'], playlist, msg % n_entries))
26e2805c 1763 failures = 0
1764 max_failures = self.params.get('skip_playlist_after_errors') or float('inf')
71729754 1765 for i, entry_tuple in enumerate(entries, 1):
1766 playlist_index, entry = entry_tuple
81139999 1767 if 'playlist-index' in self.params.get('compat_opts', []):
1768 playlist_index = playlistitems[i - 1] if playlistitems else i + playliststart - 1
30a074c2 1769 self.to_screen('[download] Downloading video %s of %s' % (i, n_entries))
1770 # This __x_forwarded_for_ip thing is a bit ugly but requires
1771 # minimal changes
1772 if x_forwarded_for:
1773 entry['__x_forwarded_for_ip'] = x_forwarded_for
1774 extra = {
1775 'n_entries': n_entries,
f59ae581 1776 '_last_playlist_index': max(playlistitems) if playlistitems else (playlistend or n_entries),
f0d785d3 1777 'playlist_count': ie_result.get('playlist_count'),
71729754 1778 'playlist_index': playlist_index,
1779 'playlist_autonumber': i,
30a074c2 1780 'playlist': playlist,
1781 'playlist_id': ie_result.get('id'),
1782 'playlist_title': ie_result.get('title'),
1783 'playlist_uploader': ie_result.get('uploader'),
1784 'playlist_uploader_id': ie_result.get('uploader_id'),
30a074c2 1785 'extractor': ie_result['extractor'],
1786 'webpage_url': ie_result['webpage_url'],
1787 'webpage_url_basename': url_basename(ie_result['webpage_url']),
0bb322b9 1788 'webpage_url_domain': get_domain(ie_result['webpage_url']),
30a074c2 1789 'extractor_key': ie_result['extractor_key'],
1790 }
1791
1792 if self._match_entry(entry, incomplete=True) is not None:
1793 continue
1794
1795 entry_result = self.__process_iterable_entry(entry, download, extra)
26e2805c 1796 if not entry_result:
1797 failures += 1
1798 if failures >= max_failures:
1799 self.report_error(
1800 'Skipping the remaining entries in playlist "%s" since %d items failed extraction' % (playlist, failures))
1801 break
30a074c2 1802 playlist_results.append(entry_result)
1803 ie_result['entries'] = playlist_results
e08a85d8 1804
1805 # Write the updated info to json
1806 if _infojson_written and self._write_info_json(
1807 'updated playlist', ie_result,
1808 self.prepare_filename(ie_copy, 'pl_infojson'), overwrite=True) is None:
1809 return
ca30f449 1810
ed5835b4 1811 ie_result = self.run_all_pps('playlist', ie_result)
1812 self.to_screen(f'[download] Finished downloading playlist: {playlist}')
30a074c2 1813 return ie_result
1814
a0566bbf 1815 @__handle_extraction_exceptions
1816 def __process_iterable_entry(self, entry, download, extra_info):
1817 return self.process_ie_result(
1818 entry, download=download, extra_info=extra_info)
1819
67134eab
JMF
1820 def _build_format_filter(self, filter_spec):
1821 " Returns a function to filter the formats according to the filter_spec "
083c9df9
PH
1822
1823 OPERATORS = {
1824 '<': operator.lt,
1825 '<=': operator.le,
1826 '>': operator.gt,
1827 '>=': operator.ge,
1828 '=': operator.eq,
1829 '!=': operator.ne,
1830 }
67134eab 1831 operator_rex = re.compile(r'''(?x)\s*
187986a8 1832 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)\s*
1833 (?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1834 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)\s*
083c9df9 1835 ''' % '|'.join(map(re.escape, OPERATORS.keys())))
187986a8 1836 m = operator_rex.fullmatch(filter_spec)
9ddb6925
S
1837 if m:
1838 try:
1839 comparison_value = int(m.group('value'))
1840 except ValueError:
1841 comparison_value = parse_filesize(m.group('value'))
1842 if comparison_value is None:
1843 comparison_value = parse_filesize(m.group('value') + 'B')
1844 if comparison_value is None:
1845 raise ValueError(
1846 'Invalid value %r in format specification %r' % (
67134eab 1847 m.group('value'), filter_spec))
9ddb6925
S
1848 op = OPERATORS[m.group('op')]
1849
083c9df9 1850 if not m:
9ddb6925
S
1851 STR_OPERATORS = {
1852 '=': operator.eq,
10d33b34
YCH
1853 '^=': lambda attr, value: attr.startswith(value),
1854 '$=': lambda attr, value: attr.endswith(value),
1855 '*=': lambda attr, value: value in attr,
1ce9a3cb 1856 '~=': lambda attr, value: value.search(attr) is not None
9ddb6925 1857 }
187986a8 1858 str_operator_rex = re.compile(r'''(?x)\s*
1859 (?P<key>[a-zA-Z0-9._-]+)\s*
1ce9a3cb
LF
1860 (?P<negation>!\s*)?(?P<op>%s)\s*(?P<none_inclusive>\?\s*)?
1861 (?P<quote>["'])?
1862 (?P<value>(?(quote)(?:(?!(?P=quote))[^\\]|\\.)+|[\w.-]+))
1863 (?(quote)(?P=quote))\s*
9ddb6925 1864 ''' % '|'.join(map(re.escape, STR_OPERATORS.keys())))
187986a8 1865 m = str_operator_rex.fullmatch(filter_spec)
9ddb6925 1866 if m:
1ce9a3cb
LF
1867 if m.group('op') == '~=':
1868 comparison_value = re.compile(m.group('value'))
1869 else:
1870 comparison_value = re.sub(r'''\\([\\"'])''', r'\1', m.group('value'))
2cc779f4
S
1871 str_op = STR_OPERATORS[m.group('op')]
1872 if m.group('negation'):
e118a879 1873 op = lambda attr, value: not str_op(attr, value)
2cc779f4
S
1874 else:
1875 op = str_op
083c9df9 1876
9ddb6925 1877 if not m:
187986a8 1878 raise SyntaxError('Invalid filter specification %r' % filter_spec)
083c9df9
PH
1879
1880 def _filter(f):
1881 actual_value = f.get(m.group('key'))
1882 if actual_value is None:
1883 return m.group('none_inclusive')
1884 return op(actual_value, comparison_value)
67134eab
JMF
1885 return _filter
1886
9f1a1c36 1887 def _check_formats(self, formats):
1888 for f in formats:
1889 self.to_screen('[info] Testing format %s' % f['format_id'])
75689fe5 1890 path = self.get_output_path('temp')
1891 if not self._ensure_dir_exists(f'{path}/'):
1892 continue
1893 temp_file = tempfile.NamedTemporaryFile(suffix='.tmp', delete=False, dir=path or None)
9f1a1c36 1894 temp_file.close()
1895 try:
1896 success, _ = self.dl(temp_file.name, f, test=True)
1897 except (DownloadError, IOError, OSError, ValueError) + network_exceptions:
1898 success = False
1899 finally:
1900 if os.path.exists(temp_file.name):
1901 try:
1902 os.remove(temp_file.name)
1903 except OSError:
1904 self.report_warning('Unable to delete temporary file "%s"' % temp_file.name)
1905 if success:
1906 yield f
1907 else:
1908 self.to_screen('[info] Unable to download format %s. Skipping...' % f['format_id'])
1909
0017d9ad 1910 def _default_format_spec(self, info_dict, download=True):
0017d9ad 1911
af0f7428
S
1912 def can_merge():
1913 merger = FFmpegMergerPP(self)
1914 return merger.available and merger.can_merge()
1915
91ebc640 1916 prefer_best = (
b7b04c78 1917 not self.params.get('simulate')
91ebc640 1918 and download
1919 and (
1920 not can_merge()
19807826 1921 or info_dict.get('is_live', False)
de6000d9 1922 or self.outtmpl_dict['default'] == '-'))
53ed7066 1923 compat = (
1924 prefer_best
1925 or self.params.get('allow_multiple_audio_streams', False)
1926 or 'format-spec' in self.params.get('compat_opts', []))
91ebc640 1927
1928 return (
53ed7066 1929 'best/bestvideo+bestaudio' if prefer_best
1930 else 'bestvideo*+bestaudio/best' if not compat
91ebc640 1931 else 'bestvideo+bestaudio/best')
0017d9ad 1932
67134eab
JMF
1933 def build_format_selector(self, format_spec):
1934 def syntax_error(note, start):
1935 message = (
1936 'Invalid format specification: '
1937 '{0}\n\t{1}\n\t{2}^'.format(note, format_spec, ' ' * start[1]))
1938 return SyntaxError(message)
1939
1940 PICKFIRST = 'PICKFIRST'
1941 MERGE = 'MERGE'
1942 SINGLE = 'SINGLE'
0130afb7 1943 GROUP = 'GROUP'
67134eab
JMF
1944 FormatSelector = collections.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
1945
91ebc640 1946 allow_multiple_streams = {'audio': self.params.get('allow_multiple_audio_streams', False),
1947 'video': self.params.get('allow_multiple_video_streams', False)}
909d24dd 1948
9f1a1c36 1949 check_formats = self.params.get('check_formats') == 'selected'
e8e73840 1950
67134eab
JMF
1951 def _parse_filter(tokens):
1952 filter_parts = []
1953 for type, string, start, _, _ in tokens:
1954 if type == tokenize.OP and string == ']':
1955 return ''.join(filter_parts)
1956 else:
1957 filter_parts.append(string)
1958
232541df 1959 def _remove_unused_ops(tokens):
17cc1534 1960 # Remove operators that we don't use and join them with the surrounding strings
232541df
JMF
1961 # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
1962 ALLOWED_OPS = ('/', '+', ',', '(', ')')
1963 last_string, last_start, last_end, last_line = None, None, None, None
1964 for type, string, start, end, line in tokens:
1965 if type == tokenize.OP and string == '[':
1966 if last_string:
1967 yield tokenize.NAME, last_string, last_start, last_end, last_line
1968 last_string = None
1969 yield type, string, start, end, line
1970 # everything inside brackets will be handled by _parse_filter
1971 for type, string, start, end, line in tokens:
1972 yield type, string, start, end, line
1973 if type == tokenize.OP and string == ']':
1974 break
1975 elif type == tokenize.OP and string in ALLOWED_OPS:
1976 if last_string:
1977 yield tokenize.NAME, last_string, last_start, last_end, last_line
1978 last_string = None
1979 yield type, string, start, end, line
1980 elif type in [tokenize.NAME, tokenize.NUMBER, tokenize.OP]:
1981 if not last_string:
1982 last_string = string
1983 last_start = start
1984 last_end = end
1985 else:
1986 last_string += string
1987 if last_string:
1988 yield tokenize.NAME, last_string, last_start, last_end, last_line
1989
cf2ac6df 1990 def _parse_format_selection(tokens, inside_merge=False, inside_choice=False, inside_group=False):
67134eab
JMF
1991 selectors = []
1992 current_selector = None
1993 for type, string, start, _, _ in tokens:
1994 # ENCODING is only defined in python 3.x
1995 if type == getattr(tokenize, 'ENCODING', None):
1996 continue
1997 elif type in [tokenize.NAME, tokenize.NUMBER]:
1998 current_selector = FormatSelector(SINGLE, string, [])
1999 elif type == tokenize.OP:
cf2ac6df
JMF
2000 if string == ')':
2001 if not inside_group:
2002 # ')' will be handled by the parentheses group
2003 tokens.restore_last_token()
67134eab 2004 break
cf2ac6df 2005 elif inside_merge and string in ['/', ',']:
0130afb7
JMF
2006 tokens.restore_last_token()
2007 break
cf2ac6df
JMF
2008 elif inside_choice and string == ',':
2009 tokens.restore_last_token()
2010 break
2011 elif string == ',':
0a31a350
JMF
2012 if not current_selector:
2013 raise syntax_error('"," must follow a format selector', start)
67134eab
JMF
2014 selectors.append(current_selector)
2015 current_selector = None
2016 elif string == '/':
d96d604e
JMF
2017 if not current_selector:
2018 raise syntax_error('"/" must follow a format selector', start)
67134eab 2019 first_choice = current_selector
cf2ac6df 2020 second_choice = _parse_format_selection(tokens, inside_choice=True)
f5f4a27a 2021 current_selector = FormatSelector(PICKFIRST, (first_choice, second_choice), [])
67134eab
JMF
2022 elif string == '[':
2023 if not current_selector:
2024 current_selector = FormatSelector(SINGLE, 'best', [])
2025 format_filter = _parse_filter(tokens)
2026 current_selector.filters.append(format_filter)
0130afb7
JMF
2027 elif string == '(':
2028 if current_selector:
2029 raise syntax_error('Unexpected "("', start)
cf2ac6df
JMF
2030 group = _parse_format_selection(tokens, inside_group=True)
2031 current_selector = FormatSelector(GROUP, group, [])
67134eab 2032 elif string == '+':
d03cfdce 2033 if not current_selector:
2034 raise syntax_error('Unexpected "+"', start)
2035 selector_1 = current_selector
2036 selector_2 = _parse_format_selection(tokens, inside_merge=True)
2037 if not selector_2:
2038 raise syntax_error('Expected a selector', start)
2039 current_selector = FormatSelector(MERGE, (selector_1, selector_2), [])
67134eab
JMF
2040 else:
2041 raise syntax_error('Operator not recognized: "{0}"'.format(string), start)
2042 elif type == tokenize.ENDMARKER:
2043 break
2044 if current_selector:
2045 selectors.append(current_selector)
2046 return selectors
2047
f8d4ad9a 2048 def _merge(formats_pair):
2049 format_1, format_2 = formats_pair
2050
2051 formats_info = []
2052 formats_info.extend(format_1.get('requested_formats', (format_1,)))
2053 formats_info.extend(format_2.get('requested_formats', (format_2,)))
2054
2055 if not allow_multiple_streams['video'] or not allow_multiple_streams['audio']:
551f9388 2056 get_no_more = {'video': False, 'audio': False}
f8d4ad9a 2057 for (i, fmt_info) in enumerate(formats_info):
551f9388 2058 if fmt_info.get('acodec') == fmt_info.get('vcodec') == 'none':
2059 formats_info.pop(i)
2060 continue
2061 for aud_vid in ['audio', 'video']:
f8d4ad9a 2062 if not allow_multiple_streams[aud_vid] and fmt_info.get(aud_vid[0] + 'codec') != 'none':
2063 if get_no_more[aud_vid]:
2064 formats_info.pop(i)
f5510afe 2065 break
f8d4ad9a 2066 get_no_more[aud_vid] = True
2067
2068 if len(formats_info) == 1:
2069 return formats_info[0]
2070
2071 video_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('vcodec') != 'none']
2072 audio_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('acodec') != 'none']
2073
2074 the_only_video = video_fmts[0] if len(video_fmts) == 1 else None
2075 the_only_audio = audio_fmts[0] if len(audio_fmts) == 1 else None
2076
2077 output_ext = self.params.get('merge_output_format')
2078 if not output_ext:
2079 if the_only_video:
2080 output_ext = the_only_video['ext']
2081 elif the_only_audio and not video_fmts:
2082 output_ext = the_only_audio['ext']
2083 else:
2084 output_ext = 'mkv'
2085
975a0d0d 2086 filtered = lambda *keys: filter(None, (traverse_obj(fmt, *keys) for fmt in formats_info))
2087
f8d4ad9a 2088 new_dict = {
2089 'requested_formats': formats_info,
975a0d0d 2090 'format': '+'.join(filtered('format')),
2091 'format_id': '+'.join(filtered('format_id')),
f8d4ad9a 2092 'ext': output_ext,
975a0d0d 2093 'protocol': '+'.join(map(determine_protocol, formats_info)),
093a1710 2094 'language': '+'.join(orderedSet(filtered('language'))) or None,
2095 'format_note': '+'.join(orderedSet(filtered('format_note'))) or None,
2096 'filesize_approx': sum(filtered('filesize', 'filesize_approx')) or None,
975a0d0d 2097 'tbr': sum(filtered('tbr', 'vbr', 'abr')),
f8d4ad9a 2098 }
2099
2100 if the_only_video:
2101 new_dict.update({
2102 'width': the_only_video.get('width'),
2103 'height': the_only_video.get('height'),
2104 'resolution': the_only_video.get('resolution') or self.format_resolution(the_only_video),
2105 'fps': the_only_video.get('fps'),
49a57e70 2106 'dynamic_range': the_only_video.get('dynamic_range'),
f8d4ad9a 2107 'vcodec': the_only_video.get('vcodec'),
2108 'vbr': the_only_video.get('vbr'),
2109 'stretched_ratio': the_only_video.get('stretched_ratio'),
2110 })
2111
2112 if the_only_audio:
2113 new_dict.update({
2114 'acodec': the_only_audio.get('acodec'),
2115 'abr': the_only_audio.get('abr'),
975a0d0d 2116 'asr': the_only_audio.get('asr'),
f8d4ad9a 2117 })
2118
2119 return new_dict
2120
e8e73840 2121 def _check_formats(formats):
981052c9 2122 if not check_formats:
2123 yield from formats
b5ac45b1 2124 return
9f1a1c36 2125 yield from self._check_formats(formats)
e8e73840 2126
67134eab 2127 def _build_selector_function(selector):
909d24dd 2128 if isinstance(selector, list): # ,
67134eab
JMF
2129 fs = [_build_selector_function(s) for s in selector]
2130
317f7ab6 2131 def selector_function(ctx):
67134eab 2132 for f in fs:
981052c9 2133 yield from f(ctx)
67134eab 2134 return selector_function
909d24dd 2135
2136 elif selector.type == GROUP: # ()
0130afb7 2137 selector_function = _build_selector_function(selector.selector)
909d24dd 2138
2139 elif selector.type == PICKFIRST: # /
67134eab
JMF
2140 fs = [_build_selector_function(s) for s in selector.selector]
2141
317f7ab6 2142 def selector_function(ctx):
67134eab 2143 for f in fs:
317f7ab6 2144 picked_formats = list(f(ctx))
67134eab
JMF
2145 if picked_formats:
2146 return picked_formats
2147 return []
67134eab 2148
981052c9 2149 elif selector.type == MERGE: # +
2150 selector_1, selector_2 = map(_build_selector_function, selector.selector)
2151
2152 def selector_function(ctx):
adbc4ec4 2153 for pair in itertools.product(selector_1(ctx), selector_2(ctx)):
981052c9 2154 yield _merge(pair)
2155
909d24dd 2156 elif selector.type == SINGLE: # atom
598d185d 2157 format_spec = selector.selector or 'best'
909d24dd 2158
f8d4ad9a 2159 # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
909d24dd 2160 if format_spec == 'all':
2161 def selector_function(ctx):
9222c381 2162 yield from _check_formats(ctx['formats'][::-1])
f8d4ad9a 2163 elif format_spec == 'mergeall':
2164 def selector_function(ctx):
dd2a987d 2165 formats = list(_check_formats(ctx['formats']))
e01d6aa4 2166 if not formats:
2167 return
921b76ca 2168 merged_format = formats[-1]
2169 for f in formats[-2::-1]:
f8d4ad9a 2170 merged_format = _merge((merged_format, f))
2171 yield merged_format
909d24dd 2172
2173 else:
e8e73840 2174 format_fallback, format_reverse, format_idx = False, True, 1
eff63539 2175 mobj = re.match(
2176 r'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
2177 format_spec)
2178 if mobj is not None:
2179 format_idx = int_or_none(mobj.group('n'), default=1)
e8e73840 2180 format_reverse = mobj.group('bw')[0] == 'b'
eff63539 2181 format_type = (mobj.group('type') or [None])[0]
2182 not_format_type = {'v': 'a', 'a': 'v'}.get(format_type)
2183 format_modified = mobj.group('mod') is not None
909d24dd 2184
2185 format_fallback = not format_type and not format_modified # for b, w
8326b00a 2186 _filter_f = (
eff63539 2187 (lambda f: f.get('%scodec' % format_type) != 'none')
2188 if format_type and format_modified # bv*, ba*, wv*, wa*
2189 else (lambda f: f.get('%scodec' % not_format_type) == 'none')
2190 if format_type # bv, ba, wv, wa
2191 else (lambda f: f.get('vcodec') != 'none' and f.get('acodec') != 'none')
2192 if not format_modified # b, w
8326b00a 2193 else lambda f: True) # b*, w*
2194 filter_f = lambda f: _filter_f(f) and (
2195 f.get('vcodec') != 'none' or f.get('acodec') != 'none')
67134eab 2196 else:
48ee10ee 2197 if format_spec in self._format_selection_exts['audio']:
b11c04a8 2198 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none'
48ee10ee 2199 elif format_spec in self._format_selection_exts['video']:
b11c04a8 2200 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none' and f.get('vcodec') != 'none'
48ee10ee 2201 elif format_spec in self._format_selection_exts['storyboards']:
b11c04a8 2202 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') == 'none' and f.get('vcodec') == 'none'
2203 else:
b5ae35ee 2204 filter_f = lambda f: f.get('format_id') == format_spec # id
909d24dd 2205
2206 def selector_function(ctx):
2207 formats = list(ctx['formats'])
909d24dd 2208 matches = list(filter(filter_f, formats)) if filter_f is not None else formats
e8e73840 2209 if format_fallback and ctx['incomplete_formats'] and not matches:
909d24dd 2210 # for extractors with incomplete formats (audio only (soundcloud)
2211 # or video only (imgur)) best/worst will fallback to
2212 # best/worst {video,audio}-only format
e8e73840 2213 matches = formats
981052c9 2214 matches = LazyList(_check_formats(matches[::-1 if format_reverse else 1]))
2215 try:
e8e73840 2216 yield matches[format_idx - 1]
981052c9 2217 except IndexError:
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:
232541df 2231 tokens = list(_remove_unused_ops(compat_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
2235 class TokenIterator(object):
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)
2340 try:
2341 upload_date = datetime.datetime.utcfromtimestamp(info_dict[ts_key])
2342 info_dict[date_key] = upload_date.strftime('%Y%m%d')
2343 except (ValueError, OverflowError, OSError):
2344 pass
2345
2346 live_keys = ('is_live', 'was_live')
2347 live_status = info_dict.get('live_status')
2348 if live_status is None:
2349 for key in live_keys:
2350 if info_dict.get(key) is False:
2351 continue
2352 if info_dict.get(key):
2353 live_status = key
2354 break
2355 if all(info_dict.get(key) is False for key in live_keys):
2356 live_status = 'not_live'
2357 if live_status:
2358 info_dict['live_status'] = live_status
2359 for key in live_keys:
2360 if info_dict.get(key) is None:
2361 info_dict[key] = (live_status == key)
2362
2363 # Auto generate title fields corresponding to the *_number fields when missing
2364 # in order to always have clean titles. This is very common for TV series.
2365 for field in ('chapter', 'season', 'episode'):
2366 if info_dict.get('%s_number' % field) is not None and not info_dict.get(field):
2367 info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field])
2368
dd82ffea
JMF
2369 def process_video_result(self, info_dict, download=True):
2370 assert info_dict.get('_type', 'video') == 'video'
9c906919 2371 self._num_videos += 1
dd82ffea 2372
bec1fad2 2373 if 'id' not in info_dict:
fc08bdd6 2374 raise ExtractorError('Missing "id" field in extractor result', ie=info_dict['extractor'])
2375 elif not info_dict.get('id'):
2376 raise ExtractorError('Extractor failed to obtain "id"', ie=info_dict['extractor'])
455a15e2 2377
c9969434
S
2378 def report_force_conversion(field, field_not, conversion):
2379 self.report_warning(
2380 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
2381 % (field, field_not, conversion))
2382
2383 def sanitize_string_field(info, string_field):
2384 field = info.get(string_field)
2385 if field is None or isinstance(field, compat_str):
2386 return
2387 report_force_conversion(string_field, 'a string', 'string')
2388 info[string_field] = compat_str(field)
2389
2390 def sanitize_numeric_fields(info):
2391 for numeric_field in self._NUMERIC_FIELDS:
2392 field = info.get(numeric_field)
2393 if field is None or isinstance(field, compat_numeric_types):
2394 continue
2395 report_force_conversion(numeric_field, 'numeric', 'int')
2396 info[numeric_field] = int_or_none(field)
2397
2398 sanitize_string_field(info_dict, 'id')
2399 sanitize_numeric_fields(info_dict)
4c3f8c3f 2400 if (info_dict.get('duration') or 0) <= 0 and info_dict.pop('duration', None):
50e93e03 2401 self.report_warning('"duration" field is negative, there is an error in extractor')
be6217b2 2402
dd82ffea
JMF
2403 if 'playlist' not in info_dict:
2404 # It isn't part of a playlist
2405 info_dict['playlist'] = None
2406 info_dict['playlist_index'] = None
2407
bc516a3f 2408 self._sanitize_thumbnails(info_dict)
d5519808 2409
536a55da 2410 thumbnail = info_dict.get('thumbnail')
bc516a3f 2411 thumbnails = info_dict.get('thumbnails')
536a55da
S
2412 if thumbnail:
2413 info_dict['thumbnail'] = sanitize_url(thumbnail)
2414 elif thumbnails:
d5519808
PH
2415 info_dict['thumbnail'] = thumbnails[-1]['url']
2416
ae30b840 2417 if info_dict.get('display_id') is None and 'id' in info_dict:
0afef30b
PH
2418 info_dict['display_id'] = info_dict['id']
2419
03f83004 2420 self._fill_common_fields(info_dict)
33d2fc2f 2421
05108a49
S
2422 for cc_kind in ('subtitles', 'automatic_captions'):
2423 cc = info_dict.get(cc_kind)
2424 if cc:
2425 for _, subtitle in cc.items():
2426 for subtitle_format in subtitle:
2427 if subtitle_format.get('url'):
2428 subtitle_format['url'] = sanitize_url(subtitle_format['url'])
2429 if subtitle_format.get('ext') is None:
2430 subtitle_format['ext'] = determine_ext(subtitle_format['url']).lower()
2431
2432 automatic_captions = info_dict.get('automatic_captions')
4bba3716 2433 subtitles = info_dict.get('subtitles')
4bba3716 2434
360e1ca5 2435 info_dict['requested_subtitles'] = self.process_subtitles(
05108a49 2436 info_dict['id'], subtitles, automatic_captions)
a504ced0 2437
dd82ffea
JMF
2438 if info_dict.get('formats') is None:
2439 # There's only one format available
2440 formats = [info_dict]
2441 else:
2442 formats = info_dict['formats']
2443
e0493e90 2444 info_dict['__has_drm'] = any(f.get('has_drm') for f in formats)
88acdbc2 2445 if not self.params.get('allow_unplayable_formats'):
2446 formats = [f for f in formats if not f.get('has_drm')]
88acdbc2 2447
319b6059 2448 get_from_start = not info_dict.get('is_live') or bool(self.params.get('live_from_start'))
2449 if not get_from_start:
2450 info_dict['title'] += ' ' + datetime.datetime.now().strftime('%Y-%m-%d %H:%M')
2451 if info_dict.get('is_live') and formats:
adbc4ec4 2452 formats = [f for f in formats if bool(f.get('is_from_start')) == get_from_start]
319b6059 2453 if get_from_start and not formats:
2454 self.raise_no_formats(info_dict, msg='--live-from-start is passed, but there are no formats that can be downloaded from the start. '
2455 'If you want to download from the current time, pass --no-live-from-start')
adbc4ec4 2456
db95dc13 2457 if not formats:
1151c407 2458 self.raise_no_formats(info_dict)
db95dc13 2459
73af5cc8
S
2460 def is_wellformed(f):
2461 url = f.get('url')
a5ac0c47 2462 if not url:
73af5cc8
S
2463 self.report_warning(
2464 '"url" field is missing or empty - skipping format, '
2465 'there is an error in extractor')
a5ac0c47
S
2466 return False
2467 if isinstance(url, bytes):
2468 sanitize_string_field(f, 'url')
2469 return True
73af5cc8
S
2470
2471 # Filter out malformed formats for better extraction robustness
2472 formats = list(filter(is_wellformed, formats))
2473
181c7053
S
2474 formats_dict = {}
2475
dd82ffea 2476 # We check that all the formats have the format and format_id fields
db95dc13 2477 for i, format in enumerate(formats):
c9969434
S
2478 sanitize_string_field(format, 'format_id')
2479 sanitize_numeric_fields(format)
dcf77cf1 2480 format['url'] = sanitize_url(format['url'])
e74e3b63 2481 if not format.get('format_id'):
8016c922 2482 format['format_id'] = compat_str(i)
e2effb08
S
2483 else:
2484 # Sanitize format_id from characters used in format selector expression
ec85ded8 2485 format['format_id'] = re.sub(r'[\s,/+\[\]()]', '_', format['format_id'])
181c7053
S
2486 format_id = format['format_id']
2487 if format_id not in formats_dict:
2488 formats_dict[format_id] = []
2489 formats_dict[format_id].append(format)
2490
2491 # Make sure all formats have unique format_id
03b4de72 2492 common_exts = set(itertools.chain(*self._format_selection_exts.values()))
181c7053 2493 for format_id, ambiguous_formats in formats_dict.items():
48ee10ee 2494 ambigious_id = len(ambiguous_formats) > 1
2495 for i, format in enumerate(ambiguous_formats):
2496 if ambigious_id:
181c7053 2497 format['format_id'] = '%s-%d' % (format_id, i)
48ee10ee 2498 if format.get('ext') is None:
2499 format['ext'] = determine_ext(format['url']).lower()
2500 # Ensure there is no conflict between id and ext in format selection
2501 # See https://github.com/yt-dlp/yt-dlp/issues/1282
2502 if format['format_id'] != format['ext'] and format['format_id'] in common_exts:
2503 format['format_id'] = 'f%s' % format['format_id']
181c7053
S
2504
2505 for i, format in enumerate(formats):
8c51aa65 2506 if format.get('format') is None:
6febd1c1 2507 format['format'] = '{id} - {res}{note}'.format(
8c51aa65
JMF
2508 id=format['format_id'],
2509 res=self.format_resolution(format),
b868936c 2510 note=format_field(format, 'format_note', ' (%s)'),
8c51aa65 2511 )
6f0be937 2512 if format.get('protocol') is None:
b5559424 2513 format['protocol'] = determine_protocol(format)
239df021 2514 if format.get('resolution') is None:
2515 format['resolution'] = self.format_resolution(format, default=None)
176f1866 2516 if format.get('dynamic_range') is None and format.get('vcodec') != 'none':
2517 format['dynamic_range'] = 'SDR'
f2fe69c7 2518 if (info_dict.get('duration') and format.get('tbr')
2519 and not format.get('filesize') and not format.get('filesize_approx')):
2520 format['filesize_approx'] = info_dict['duration'] * format['tbr'] * (1024 / 8)
2521
e5660ee6
JMF
2522 # Add HTTP headers, so that external programs can use them from the
2523 # json output
2524 full_format_info = info_dict.copy()
2525 full_format_info.update(format)
2526 format['http_headers'] = self._calc_headers(full_format_info)
0016b84e
S
2527 # Remove private housekeeping stuff
2528 if '__x_forwarded_for_ip' in info_dict:
2529 del info_dict['__x_forwarded_for_ip']
dd82ffea 2530
9f1a1c36 2531 if self.params.get('check_formats') is True:
282f5709 2532 formats = LazyList(self._check_formats(formats[::-1]), reverse=True)
9f1a1c36 2533
88acdbc2 2534 if not formats or formats[0] is not info_dict:
b3d9ef88
JMF
2535 # only set the 'formats' fields if the original info_dict list them
2536 # otherwise we end up with a circular reference, the first (and unique)
f89197d7 2537 # element in the 'formats' field in info_dict is info_dict itself,
dfb1b146 2538 # which can't be exported to json
b3d9ef88 2539 info_dict['formats'] = formats
4ec82a72 2540
2541 info_dict, _ = self.pre_process(info_dict)
2542
09b49e1f 2543 if self._match_entry(info_dict) is not None:
2544 return info_dict
2545
2546 self.post_extract(info_dict)
2547 info_dict, _ = self.pre_process(info_dict, 'after_filter')
2548
093a1710 2549 # The pre-processors may have modified the formats
2550 formats = info_dict.get('formats', [info_dict])
2551
fa9f30b8 2552 list_only = self.params.get('simulate') is None and (
2553 self.params.get('list_thumbnails') or self.params.get('listformats') or self.params.get('listsubtitles'))
2554 interactive_format_selection = not list_only and self.format_selector == '-'
b7b04c78 2555 if self.params.get('list_thumbnails'):
2556 self.list_thumbnails(info_dict)
b7b04c78 2557 if self.params.get('listsubtitles'):
2558 if 'automatic_captions' in info_dict:
2559 self.list_subtitles(
2560 info_dict['id'], automatic_captions, 'automatic captions')
2561 self.list_subtitles(info_dict['id'], subtitles, 'subtitles')
fa9f30b8 2562 if self.params.get('listformats') or interactive_format_selection:
b69fd25c 2563 self.list_formats(info_dict)
169dbde9 2564 if list_only:
b7b04c78 2565 # Without this printing, -F --print-json will not work
169dbde9 2566 self.__forced_printings(info_dict, self.prepare_filename(info_dict), incomplete=True)
bfaae0a7 2567 return
2568
187986a8 2569 format_selector = self.format_selector
2570 if format_selector is None:
0017d9ad 2571 req_format = self._default_format_spec(info_dict, download=download)
0760b0a7 2572 self.write_debug('Default format spec: %s' % req_format)
187986a8 2573 format_selector = self.build_format_selector(req_format)
317f7ab6 2574
fa9f30b8 2575 while True:
2576 if interactive_format_selection:
2577 req_format = input(
2578 self._format_screen('\nEnter format selector: ', self.Styles.EMPHASIS))
2579 try:
2580 format_selector = self.build_format_selector(req_format)
2581 except SyntaxError as err:
2582 self.report_error(err, tb=False, is_error=False)
2583 continue
2584
2585 # While in format selection we may need to have an access to the original
2586 # format set in order to calculate some metrics or do some processing.
2587 # For now we need to be able to guess whether original formats provided
2588 # by extractor are incomplete or not (i.e. whether extractor provides only
2589 # video-only or audio-only formats) for proper formats selection for
2590 # extractors with such incomplete formats (see
2591 # https://github.com/ytdl-org/youtube-dl/pull/5556).
2592 # Since formats may be filtered during format selection and may not match
2593 # the original formats the results may be incorrect. Thus original formats
2594 # or pre-calculated metrics should be passed to format selection routines
2595 # as well.
2596 # We will pass a context object containing all necessary additional data
2597 # instead of just formats.
2598 # This fixes incorrect format selection issue (see
2599 # https://github.com/ytdl-org/youtube-dl/issues/10083).
2600 incomplete_formats = (
2601 # All formats are video-only or
2602 all(f.get('vcodec') != 'none' and f.get('acodec') == 'none' for f in formats)
2603 # all formats are audio-only
2604 or all(f.get('vcodec') == 'none' and f.get('acodec') != 'none' for f in formats))
2605
2606 ctx = {
2607 'formats': formats,
2608 'incomplete_formats': incomplete_formats,
2609 }
2610
2611 formats_to_download = list(format_selector(ctx))
2612 if interactive_format_selection and not formats_to_download:
2613 self.report_error('Requested format is not available', tb=False, is_error=False)
2614 continue
2615 break
317f7ab6 2616
dd82ffea 2617 if not formats_to_download:
b7da73eb 2618 if not self.params.get('ignore_no_formats_error'):
1151c407 2619 raise ExtractorError('Requested format is not available', expected=True,
2620 video_id=info_dict['id'], ie=info_dict['extractor'])
b62fa6d7 2621 self.report_warning('Requested format is not available')
2622 # Process what we can, even without any available formats.
2623 formats_to_download = [{}]
a13e6848 2624
b62fa6d7 2625 best_format = formats_to_download[-1]
2626 if download:
2627 if best_format:
2628 self.to_screen(
2629 f'[info] {info_dict["id"]}: Downloading {len(formats_to_download)} format(s): '
2630 + ', '.join([f['format_id'] for f in formats_to_download]))
a13e6848 2631 max_downloads_reached = False
f46e2f9d 2632 for i, fmt in enumerate(formats_to_download):
09b49e1f 2633 formats_to_download[i] = new_info = self._copy_infodict(info_dict)
b7da73eb 2634 new_info.update(fmt)
a13e6848 2635 try:
2636 self.process_info(new_info)
2637 except MaxDownloadsReached:
2638 max_downloads_reached = True
f46e2f9d 2639 # Remove copied info
2640 for key, val in tuple(new_info.items()):
2641 if info_dict.get(key) == val:
2642 new_info.pop(key)
a13e6848 2643 if max_downloads_reached:
2644 break
ebed8b37 2645
9e907ebd 2646 write_archive = set(f.get('__write_download_archive', False) for f in formats_to_download)
a13e6848 2647 assert write_archive.issubset({True, False, 'ignore'})
2648 if True in write_archive and False not in write_archive:
2649 self.record_download_archive(info_dict)
be72c624 2650
2651 info_dict['requested_downloads'] = formats_to_download
ed5835b4 2652 info_dict = self.run_all_pps('after_video', info_dict)
a13e6848 2653 if max_downloads_reached:
2654 raise MaxDownloadsReached()
ebed8b37 2655
49a57e70 2656 # We update the info dict with the selected best quality format (backwards compatibility)
be72c624 2657 info_dict.update(best_format)
dd82ffea
JMF
2658 return info_dict
2659
98c70d6f 2660 def process_subtitles(self, video_id, normal_subtitles, automatic_captions):
a504ced0 2661 """Select the requested subtitles and their format"""
98c70d6f
JMF
2662 available_subs = {}
2663 if normal_subtitles and self.params.get('writesubtitles'):
2664 available_subs.update(normal_subtitles)
2665 if automatic_captions and self.params.get('writeautomaticsub'):
2666 for lang, cap_info in automatic_captions.items():
360e1ca5
JMF
2667 if lang not in available_subs:
2668 available_subs[lang] = cap_info
2669
4d171848
JMF
2670 if (not self.params.get('writesubtitles') and not
2671 self.params.get('writeautomaticsub') or not
2672 available_subs):
2673 return None
a504ced0 2674
c32b0aab 2675 all_sub_langs = available_subs.keys()
a504ced0 2676 if self.params.get('allsubtitles', False):
c32b0aab 2677 requested_langs = all_sub_langs
2678 elif self.params.get('subtitleslangs', False):
77c4a9ef 2679 # A list is used so that the order of languages will be the same as
2680 # given in subtitleslangs. See https://github.com/yt-dlp/yt-dlp/issues/1041
2681 requested_langs = []
2682 for lang_re in self.params.get('subtitleslangs'):
77c4a9ef 2683 discard = lang_re[0] == '-'
c32b0aab 2684 if discard:
77c4a9ef 2685 lang_re = lang_re[1:]
3aa91540 2686 if lang_re == 'all':
2687 if discard:
2688 requested_langs = []
2689 else:
2690 requested_langs.extend(all_sub_langs)
2691 continue
77c4a9ef 2692 current_langs = filter(re.compile(lang_re + '$').match, all_sub_langs)
c32b0aab 2693 if discard:
2694 for lang in current_langs:
77c4a9ef 2695 while lang in requested_langs:
2696 requested_langs.remove(lang)
c32b0aab 2697 else:
77c4a9ef 2698 requested_langs.extend(current_langs)
2699 requested_langs = orderedSet(requested_langs)
c32b0aab 2700 elif 'en' in available_subs:
2701 requested_langs = ['en']
a504ced0 2702 else:
c32b0aab 2703 requested_langs = [list(all_sub_langs)[0]]
ad3dc496 2704 if requested_langs:
2705 self.write_debug('Downloading subtitles: %s' % ', '.join(requested_langs))
a504ced0
JMF
2706
2707 formats_query = self.params.get('subtitlesformat', 'best')
2708 formats_preference = formats_query.split('/') if formats_query else []
2709 subs = {}
2710 for lang in requested_langs:
2711 formats = available_subs.get(lang)
2712 if formats is None:
2713 self.report_warning('%s subtitles not available for %s' % (lang, video_id))
2714 continue
a504ced0
JMF
2715 for ext in formats_preference:
2716 if ext == 'best':
2717 f = formats[-1]
2718 break
2719 matches = list(filter(lambda f: f['ext'] == ext, formats))
2720 if matches:
2721 f = matches[-1]
2722 break
2723 else:
2724 f = formats[-1]
2725 self.report_warning(
2726 'No subtitle format found matching "%s" for language %s, '
2727 'using %s' % (formats_query, lang, f['ext']))
2728 subs[lang] = f
2729 return subs
2730
bb66c247 2731 def _forceprint(self, key, info_dict):
2732 if info_dict is None:
2733 return
2734 info_copy = info_dict.copy()
2735 info_copy['formats_table'] = self.render_formats_table(info_dict)
2736 info_copy['thumbnails_table'] = self.render_thumbnails_table(info_dict)
2737 info_copy['subtitles_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('subtitles'))
2738 info_copy['automatic_captions_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('automatic_captions'))
2739
2740 def format_tmpl(tmpl):
2741 mobj = re.match(r'\w+(=?)$', tmpl)
2742 if mobj and mobj.group(1):
2743 return f'{tmpl[:-1]} = %({tmpl[:-1]})r'
2744 elif mobj:
2745 return f'%({tmpl})s'
2746 return tmpl
8130779d 2747
bb66c247 2748 for tmpl in self.params['forceprint'].get(key, []):
2749 self.to_stdout(self.evaluate_outtmpl(format_tmpl(tmpl), info_copy))
2750
2751 for tmpl, file_tmpl in self.params['print_to_file'].get(key, []):
2752 filename = self.evaluate_outtmpl(file_tmpl, info_dict)
2753 tmpl = format_tmpl(tmpl)
2754 self.to_screen(f'[info] Writing {tmpl!r} to: {filename}')
8d93e69d 2755 if self._ensure_dir_exists(filename):
2756 with io.open(filename, 'a', encoding='utf-8') as f:
2757 f.write(self.evaluate_outtmpl(tmpl, info_copy) + '\n')
ca30f449 2758
d06daf23 2759 def __forced_printings(self, info_dict, filename, incomplete):
53c18592 2760 def print_mandatory(field, actual_field=None):
2761 if actual_field is None:
2762 actual_field = field
d06daf23 2763 if (self.params.get('force%s' % field, False)
53c18592 2764 and (not incomplete or info_dict.get(actual_field) is not None)):
2765 self.to_stdout(info_dict[actual_field])
d06daf23
S
2766
2767 def print_optional(field):
2768 if (self.params.get('force%s' % field, False)
2769 and info_dict.get(field) is not None):
2770 self.to_stdout(info_dict[field])
2771
53c18592 2772 info_dict = info_dict.copy()
2773 if filename is not None:
2774 info_dict['filename'] = filename
2775 if info_dict.get('requested_formats') is not None:
2776 # For RTMP URLs, also include the playpath
2777 info_dict['urls'] = '\n'.join(f['url'] + f.get('play_path', '') for f in info_dict['requested_formats'])
10331a26 2778 elif info_dict.get('url'):
53c18592 2779 info_dict['urls'] = info_dict['url'] + info_dict.get('play_path', '')
2780
bb66c247 2781 if (self.params.get('forcejson')
2782 or self.params['forceprint'].get('video')
2783 or self.params['print_to_file'].get('video')):
2b8a2973 2784 self.post_extract(info_dict)
bb66c247 2785 self._forceprint('video', info_dict)
53c18592 2786
d06daf23
S
2787 print_mandatory('title')
2788 print_mandatory('id')
53c18592 2789 print_mandatory('url', 'urls')
d06daf23
S
2790 print_optional('thumbnail')
2791 print_optional('description')
53c18592 2792 print_optional('filename')
b868936c 2793 if self.params.get('forceduration') and info_dict.get('duration') is not None:
d06daf23
S
2794 self.to_stdout(formatSeconds(info_dict['duration']))
2795 print_mandatory('format')
53c18592 2796
2b8a2973 2797 if self.params.get('forcejson'):
6e84b215 2798 self.to_stdout(json.dumps(self.sanitize_info(info_dict)))
d06daf23 2799
e8e73840 2800 def dl(self, name, info, subtitle=False, test=False):
88acdbc2 2801 if not info.get('url'):
1151c407 2802 self.raise_no_formats(info, True)
e8e73840 2803
2804 if test:
2805 verbose = self.params.get('verbose')
2806 params = {
2807 'test': True,
a169858f 2808 'quiet': self.params.get('quiet') or not verbose,
e8e73840 2809 'verbose': verbose,
2810 'noprogress': not verbose,
2811 'nopart': True,
2812 'skip_unavailable_fragments': False,
2813 'keep_fragments': False,
2814 'overwrites': True,
2815 '_no_ytdl_file': True,
2816 }
2817 else:
2818 params = self.params
96fccc10 2819 fd = get_suitable_downloader(info, params, to_stdout=(name == '-'))(self, params)
e8e73840 2820 if not test:
2821 for ph in self._progress_hooks:
2822 fd.add_progress_hook(ph)
42676437
M
2823 urls = '", "'.join(
2824 (f['url'].split(',')[0] + ',<data>' if f['url'].startswith('data:') else f['url'])
2825 for f in info.get('requested_formats', []) or [info])
18e674b4 2826 self.write_debug('Invoking downloader on "%s"' % urls)
03b4de72 2827
adbc4ec4
THD
2828 # Note: Ideally info should be a deep-copied so that hooks cannot modify it.
2829 # But it may contain objects that are not deep-copyable
2830 new_info = self._copy_infodict(info)
e8e73840 2831 if new_info.get('http_headers') is None:
2832 new_info['http_headers'] = self._calc_headers(new_info)
2833 return fd.download(name, new_info, subtitle)
2834
e04938ab 2835 def existing_file(self, filepaths, *, default_overwrite=True):
2836 existing_files = list(filter(os.path.exists, orderedSet(filepaths)))
2837 if existing_files and not self.params.get('overwrites', default_overwrite):
2838 return existing_files[0]
2839
2840 for file in existing_files:
2841 self.report_file_delete(file)
2842 os.remove(file)
2843 return None
2844
8222d8de 2845 def process_info(self, info_dict):
09b49e1f 2846 """Process a single resolved IE result. (Modifies it in-place)"""
8222d8de
JMF
2847
2848 assert info_dict.get('_type', 'video') == 'video'
f46e2f9d 2849 original_infodict = info_dict
fd288278 2850
4513a41a 2851 if 'format' not in info_dict and 'ext' in info_dict:
8222d8de
JMF
2852 info_dict['format'] = info_dict['ext']
2853
09b49e1f 2854 # This is mostly just for backward compatibility of process_info
2855 # As a side-effect, this allows for format-specific filters
c77495e3 2856 if self._match_entry(info_dict) is not None:
9e907ebd 2857 info_dict['__write_download_archive'] = 'ignore'
8222d8de
JMF
2858 return
2859
09b49e1f 2860 # Does nothing under normal operation - for backward compatibility of process_info
277d6ff5 2861 self.post_extract(info_dict)
8222d8de 2862
dcf64d43 2863 # info_dict['_filename'] needs to be set for backward compatibility
de6000d9 2864 info_dict['_filename'] = full_filename = self.prepare_filename(info_dict, warn=True)
2865 temp_filename = self.prepare_filename(info_dict, 'temp')
0202b52a 2866 files_to_move = {}
8222d8de 2867
09b49e1f 2868 self._num_downloads += 1
2869
8222d8de 2870 # Forced printings
4513a41a 2871 self.__forced_printings(info_dict, full_filename, incomplete=('format' not in info_dict))
8222d8de 2872
b7b04c78 2873 if self.params.get('simulate'):
9e907ebd 2874 info_dict['__write_download_archive'] = self.params.get('force_write_download_archive')
8222d8de
JMF
2875 return
2876
de6000d9 2877 if full_filename is None:
8222d8de 2878 return
e92caff5 2879 if not self._ensure_dir_exists(encodeFilename(full_filename)):
0202b52a 2880 return
e92caff5 2881 if not self._ensure_dir_exists(encodeFilename(temp_filename)):
8222d8de
JMF
2882 return
2883
80c03fa9 2884 if self._write_description('video', info_dict,
2885 self.prepare_filename(info_dict, 'description')) is None:
2886 return
2887
2888 sub_files = self._write_subtitles(info_dict, temp_filename)
2889 if sub_files is None:
2890 return
2891 files_to_move.update(dict(sub_files))
2892
2893 thumb_files = self._write_thumbnails(
2894 'video', info_dict, temp_filename, self.prepare_filename(info_dict, 'thumbnail'))
2895 if thumb_files is None:
2896 return
2897 files_to_move.update(dict(thumb_files))
8222d8de 2898
80c03fa9 2899 infofn = self.prepare_filename(info_dict, 'infojson')
2900 _infojson_written = self._write_info_json('video', info_dict, infofn)
2901 if _infojson_written:
dac5df5a 2902 info_dict['infojson_filename'] = infofn
e75bb0d6 2903 # For backward compatibility, even though it was a private field
80c03fa9 2904 info_dict['__infojson_filename'] = infofn
2905 elif _infojson_written is None:
2906 return
2907
2908 # Note: Annotations are deprecated
2909 annofn = None
1fb07d10 2910 if self.params.get('writeannotations', False):
de6000d9 2911 annofn = self.prepare_filename(info_dict, 'annotation')
80c03fa9 2912 if annofn:
e92caff5 2913 if not self._ensure_dir_exists(encodeFilename(annofn)):
0202b52a 2914 return
0c3d0f51 2915 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(annofn)):
6febd1c1 2916 self.to_screen('[info] Video annotations are already present')
ffddb112
RA
2917 elif not info_dict.get('annotations'):
2918 self.report_warning('There are no annotations to write.')
7b6fefc9
PH
2919 else:
2920 try:
6febd1c1 2921 self.to_screen('[info] Writing video annotations to: ' + annofn)
7b6fefc9
PH
2922 with io.open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
2923 annofile.write(info_dict['annotations'])
2924 except (KeyError, TypeError):
6febd1c1 2925 self.report_warning('There are no annotations to write.')
7b6fefc9 2926 except (OSError, IOError):
6febd1c1 2927 self.report_error('Cannot write annotations file: ' + annofn)
7b6fefc9 2928 return
1fb07d10 2929
732044af 2930 # Write internet shortcut files
08438d2c 2931 def _write_link_file(link_type):
60f3e995 2932 url = try_get(info_dict['webpage_url'], iri_to_uri)
2933 if not url:
2934 self.report_warning(
2935 f'Cannot write internet shortcut file because the actual URL of "{info_dict["webpage_url"]}" is unknown')
2936 return True
08438d2c 2937 linkfn = replace_extension(self.prepare_filename(info_dict, 'link'), link_type, info_dict.get('ext'))
0e6b018a
Z
2938 if not self._ensure_dir_exists(encodeFilename(linkfn)):
2939 return False
10e3742e 2940 if self.params.get('overwrites', True) and os.path.exists(encodeFilename(linkfn)):
08438d2c 2941 self.to_screen(f'[info] Internet shortcut (.{link_type}) is already present')
2942 return True
2943 try:
2944 self.to_screen(f'[info] Writing internet shortcut (.{link_type}) to: {linkfn}')
2945 with io.open(encodeFilename(to_high_limit_path(linkfn)), 'w', encoding='utf-8',
2946 newline='\r\n' if link_type == 'url' else '\n') as linkfile:
60f3e995 2947 template_vars = {'url': url}
08438d2c 2948 if link_type == 'desktop':
2949 template_vars['filename'] = linkfn[:-(len(link_type) + 1)]
2950 linkfile.write(LINK_TEMPLATES[link_type] % template_vars)
2951 except (OSError, IOError):
2952 self.report_error(f'Cannot write internet shortcut {linkfn}')
2953 return False
732044af 2954 return True
2955
08438d2c 2956 write_links = {
2957 'url': self.params.get('writeurllink'),
2958 'webloc': self.params.get('writewebloclink'),
2959 'desktop': self.params.get('writedesktoplink'),
2960 }
2961 if self.params.get('writelink'):
2962 link_type = ('webloc' if sys.platform == 'darwin'
2963 else 'desktop' if sys.platform.startswith('linux')
2964 else 'url')
2965 write_links[link_type] = True
2966
2967 if any(should_write and not _write_link_file(link_type)
2968 for link_type, should_write in write_links.items()):
2969 return
732044af 2970
f46e2f9d 2971 def replace_info_dict(new_info):
2972 nonlocal info_dict
2973 if new_info == info_dict:
2974 return
2975 info_dict.clear()
2976 info_dict.update(new_info)
2977
56d868db 2978 try:
f46e2f9d 2979 new_info, files_to_move = self.pre_process(info_dict, 'before_dl', files_to_move)
2980 replace_info_dict(new_info)
56d868db 2981 except PostProcessingError as err:
2982 self.report_error('Preprocessing: %s' % str(err))
2983 return
2984
a13e6848 2985 if self.params.get('skip_download'):
56d868db 2986 info_dict['filepath'] = temp_filename
2987 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
2988 info_dict['__files_to_move'] = files_to_move
f46e2f9d 2989 replace_info_dict(self.run_pp(MoveFilesAfterDownloadPP(self, False), info_dict))
9e907ebd 2990 info_dict['__write_download_archive'] = self.params.get('force_write_download_archive')
56d868db 2991 else:
2992 # Download
b868936c 2993 info_dict.setdefault('__postprocessors', [])
4340deca 2994 try:
0202b52a 2995
e04938ab 2996 def existing_video_file(*filepaths):
6b591b29 2997 ext = info_dict.get('ext')
e04938ab 2998 converted = lambda file: replace_extension(file, self.params.get('final_ext') or ext, ext)
2999 file = self.existing_file(itertools.chain(*zip(map(converted, filepaths), filepaths)),
3000 default_overwrite=False)
3001 if file:
3002 info_dict['ext'] = os.path.splitext(file)[1][1:]
3003 return file
0202b52a 3004
3005 success = True
4340deca 3006 if info_dict.get('requested_formats') is not None:
81cd954a
S
3007
3008 def compatible_formats(formats):
d03cfdce 3009 # TODO: some formats actually allow this (mkv, webm, ogg, mp4), but not all of them.
3010 video_formats = [format for format in formats if format.get('vcodec') != 'none']
3011 audio_formats = [format for format in formats if format.get('acodec') != 'none']
3012 if len(video_formats) > 2 or len(audio_formats) > 2:
3013 return False
3014
81cd954a 3015 # Check extension
d03cfdce 3016 exts = set(format.get('ext') for format in formats)
3017 COMPATIBLE_EXTS = (
3018 set(('mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma')),
3019 set(('webm',)),
3020 )
3021 for ext_sets in COMPATIBLE_EXTS:
3022 if ext_sets.issuperset(exts):
3023 return True
81cd954a
S
3024 # TODO: Check acodec/vcodec
3025 return False
3026
3027 requested_formats = info_dict['requested_formats']
0202b52a 3028 old_ext = info_dict['ext']
4e3b637d 3029 if self.params.get('merge_output_format') is None:
3030 if not compatible_formats(requested_formats):
3031 info_dict['ext'] = 'mkv'
3032 self.report_warning(
3033 'Requested formats are incompatible for merge and will be merged into mkv')
3034 if (info_dict['ext'] == 'webm'
3035 and info_dict.get('thumbnails')
3036 # check with type instead of pp_key, __name__, or isinstance
3037 # since we dont want any custom PPs to trigger this
3038 and any(type(pp) == EmbedThumbnailPP for pp in self._pps['post_process'])):
3039 info_dict['ext'] = 'mkv'
3040 self.report_warning(
3041 'webm doesn\'t support embedding a thumbnail, mkv will be used')
124bc071 3042 new_ext = info_dict['ext']
0202b52a 3043
124bc071 3044 def correct_ext(filename, ext=new_ext):
96fccc10 3045 if filename == '-':
3046 return filename
0202b52a 3047 filename_real_ext = os.path.splitext(filename)[1][1:]
3048 filename_wo_ext = (
3049 os.path.splitext(filename)[0]
124bc071 3050 if filename_real_ext in (old_ext, new_ext)
0202b52a 3051 else filename)
124bc071 3052 return '%s.%s' % (filename_wo_ext, ext)
0202b52a 3053
38c6902b 3054 # Ensure filename always has a correct extension for successful merge
0202b52a 3055 full_filename = correct_ext(full_filename)
3056 temp_filename = correct_ext(temp_filename)
e04938ab 3057 dl_filename = existing_video_file(full_filename, temp_filename)
1ea24129 3058 info_dict['__real_download'] = False
18e674b4 3059
adbc4ec4
THD
3060 downloaded = []
3061 merger = FFmpegMergerPP(self)
3062
3063 fd = get_suitable_downloader(info_dict, self.params, to_stdout=temp_filename == '-')
dbf5416a 3064 if dl_filename is not None:
6c7274ec 3065 self.report_file_already_downloaded(dl_filename)
adbc4ec4
THD
3066 elif fd:
3067 for f in requested_formats if fd != FFmpegFD else []:
3068 f['filepath'] = fname = prepend_extension(
3069 correct_ext(temp_filename, info_dict['ext']),
3070 'f%s' % f['format_id'], info_dict['ext'])
3071 downloaded.append(fname)
dbf5416a 3072 info_dict['url'] = '\n'.join(f['url'] for f in requested_formats)
3073 success, real_download = self.dl(temp_filename, info_dict)
3074 info_dict['__real_download'] = real_download
18e674b4 3075 else:
18e674b4 3076 if self.params.get('allow_unplayable_formats'):
3077 self.report_warning(
3078 'You have requested merging of multiple formats '
3079 'while also allowing unplayable formats to be downloaded. '
3080 'The formats won\'t be merged to prevent data corruption.')
3081 elif not merger.available:
e8969bda 3082 msg = 'You have requested merging of multiple formats but ffmpeg is not installed'
3083 if not self.params.get('ignoreerrors'):
3084 self.report_error(f'{msg}. Aborting due to --abort-on-error')
3085 return
3086 self.report_warning(f'{msg}. The formats won\'t be merged')
18e674b4 3087
96fccc10 3088 if temp_filename == '-':
adbc4ec4 3089 reason = ('using a downloader other than ffmpeg' if FFmpegFD.can_merge_formats(info_dict, self.params)
96fccc10 3090 else 'but the formats are incompatible for simultaneous download' if merger.available
3091 else 'but ffmpeg is not installed')
3092 self.report_warning(
3093 f'You have requested downloading multiple formats to stdout {reason}. '
3094 'The formats will be streamed one after the other')
3095 fname = temp_filename
dbf5416a 3096 for f in requested_formats:
3097 new_info = dict(info_dict)
3098 del new_info['requested_formats']
3099 new_info.update(f)
96fccc10 3100 if temp_filename != '-':
124bc071 3101 fname = prepend_extension(
3102 correct_ext(temp_filename, new_info['ext']),
3103 'f%s' % f['format_id'], new_info['ext'])
96fccc10 3104 if not self._ensure_dir_exists(fname):
3105 return
a21e0ab1 3106 f['filepath'] = fname
96fccc10 3107 downloaded.append(fname)
dbf5416a 3108 partial_success, real_download = self.dl(fname, new_info)
3109 info_dict['__real_download'] = info_dict['__real_download'] or real_download
3110 success = success and partial_success
adbc4ec4
THD
3111
3112 if downloaded and merger.available and not self.params.get('allow_unplayable_formats'):
3113 info_dict['__postprocessors'].append(merger)
3114 info_dict['__files_to_merge'] = downloaded
3115 # Even if there were no downloads, it is being merged only now
3116 info_dict['__real_download'] = True
3117 else:
3118 for file in downloaded:
3119 files_to_move[file] = None
4340deca
P
3120 else:
3121 # Just a single file
e04938ab 3122 dl_filename = existing_video_file(full_filename, temp_filename)
6c7274ec 3123 if dl_filename is None or dl_filename == temp_filename:
3124 # dl_filename == temp_filename could mean that the file was partially downloaded with --no-part.
3125 # So we should try to resume the download
e8e73840 3126 success, real_download = self.dl(temp_filename, info_dict)
0202b52a 3127 info_dict['__real_download'] = real_download
6c7274ec 3128 else:
3129 self.report_file_already_downloaded(dl_filename)
0202b52a 3130
0202b52a 3131 dl_filename = dl_filename or temp_filename
c571435f 3132 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
0202b52a 3133
3158150c 3134 except network_exceptions as err:
7960b056 3135 self.report_error('unable to download video data: %s' % error_to_compat_str(err))
4340deca
P
3136 return
3137 except (OSError, IOError) as err:
3138 raise UnavailableVideoError(err)
3139 except (ContentTooShortError, ) as err:
3140 self.report_error('content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded))
3141 return
8222d8de 3142
de6000d9 3143 if success and full_filename != '-':
f17f8651 3144
fd7cfb64 3145 def fixup():
3146 do_fixup = True
3147 fixup_policy = self.params.get('fixup')
3148 vid = info_dict['id']
3149
3150 if fixup_policy in ('ignore', 'never'):
3151 return
3152 elif fixup_policy == 'warn':
3153 do_fixup = False
f89b3e2d 3154 elif fixup_policy != 'force':
3155 assert fixup_policy in ('detect_or_warn', None)
3156 if not info_dict.get('__real_download'):
3157 do_fixup = False
fd7cfb64 3158
3159 def ffmpeg_fixup(cndn, msg, cls):
3160 if not cndn:
3161 return
3162 if not do_fixup:
3163 self.report_warning(f'{vid}: {msg}')
3164 return
3165 pp = cls(self)
3166 if pp.available:
3167 info_dict['__postprocessors'].append(pp)
3168 else:
3169 self.report_warning(f'{vid}: {msg}. Install ffmpeg to fix this automatically')
3170
3171 stretched_ratio = info_dict.get('stretched_ratio')
3172 ffmpeg_fixup(
3173 stretched_ratio not in (1, None),
3174 f'Non-uniform pixel ratio {stretched_ratio}',
3175 FFmpegFixupStretchedPP)
3176
3177 ffmpeg_fixup(
3178 (info_dict.get('requested_formats') is None
3179 and info_dict.get('container') == 'm4a_dash'
3180 and info_dict.get('ext') == 'm4a'),
3181 'writing DASH m4a. Only some players support this container',
3182 FFmpegFixupM4aPP)
3183
993191c0 3184 downloader = get_suitable_downloader(info_dict, self.params) if 'protocol' in info_dict else None
3185 downloader = downloader.__name__ if downloader else None
adbc4ec4
THD
3186
3187 if info_dict.get('requested_formats') is None: # Not necessary if doing merger
3188 ffmpeg_fixup(downloader == 'HlsFD',
3189 'Possible MPEG-TS in MP4 container or malformed AAC timestamps',
3190 FFmpegFixupM3u8PP)
3191 ffmpeg_fixup(info_dict.get('is_live') and downloader == 'DashSegmentsFD',
3192 'Possible duplicate MOOV atoms', FFmpegFixupDuplicateMoovPP)
3193
e04b003e 3194 ffmpeg_fixup(downloader == 'WebSocketFragmentFD', 'Malformed timestamps detected', FFmpegFixupTimestampPP)
3195 ffmpeg_fixup(downloader == 'WebSocketFragmentFD', 'Malformed duration detected', FFmpegFixupDurationPP)
fd7cfb64 3196
3197 fixup()
8222d8de 3198 try:
f46e2f9d 3199 replace_info_dict(self.post_process(dl_filename, info_dict, files_to_move))
af819c21 3200 except PostProcessingError as err:
3201 self.report_error('Postprocessing: %s' % str(err))
8222d8de 3202 return
ab8e5e51
AM
3203 try:
3204 for ph in self._post_hooks:
23c1a667 3205 ph(info_dict['filepath'])
ab8e5e51
AM
3206 except Exception as err:
3207 self.report_error('post hooks: %s' % str(err))
3208 return
9e907ebd 3209 info_dict['__write_download_archive'] = True
2d30509f 3210
a13e6848 3211 if self.params.get('force_write_download_archive'):
9e907ebd 3212 info_dict['__write_download_archive'] = True
a13e6848 3213
3214 # Make sure the info_dict was modified in-place
f46e2f9d 3215 assert info_dict is original_infodict
a13e6848 3216
c3e6ffba 3217 max_downloads = self.params.get('max_downloads')
3218 if max_downloads is not None and self._num_downloads >= int(max_downloads):
3219 raise MaxDownloadsReached()
8222d8de 3220
aa9369a2 3221 def __download_wrapper(self, func):
3222 @functools.wraps(func)
3223 def wrapper(*args, **kwargs):
3224 try:
3225 res = func(*args, **kwargs)
3226 except UnavailableVideoError as e:
3227 self.report_error(e)
b222c271 3228 except MaxDownloadsReached as e:
aa9369a2 3229 self.to_screen(f'[info] {e}')
3230 raise
b222c271 3231 except DownloadCancelled as e:
3232 self.to_screen(f'[info] {e}')
3233 if not self.params.get('break_per_url'):
3234 raise
aa9369a2 3235 else:
3236 if self.params.get('dump_single_json', False):
3237 self.post_extract(res)
3238 self.to_stdout(json.dumps(self.sanitize_info(res)))
3239 return wrapper
3240
8222d8de
JMF
3241 def download(self, url_list):
3242 """Download a given list of URLs."""
aa9369a2 3243 url_list = variadic(url_list) # Passing a single URL is a common mistake
de6000d9 3244 outtmpl = self.outtmpl_dict['default']
3089bc74
S
3245 if (len(url_list) > 1
3246 and outtmpl != '-'
3247 and '%' not in outtmpl
3248 and self.params.get('max_downloads') != 1):
acd69589 3249 raise SameFileError(outtmpl)
8222d8de
JMF
3250
3251 for url in url_list:
aa9369a2 3252 self.__download_wrapper(self.extract_info)(
3253 url, force_generic_extractor=self.params.get('force_generic_extractor', False))
8222d8de
JMF
3254
3255 return self._download_retcode
3256
1dcc4c0c 3257 def download_with_info_file(self, info_filename):
31bd3925
JMF
3258 with contextlib.closing(fileinput.FileInput(
3259 [info_filename], mode='r',
3260 openhook=fileinput.hook_encoded('utf-8'))) as f:
3261 # FileInput doesn't have a read method, we can't call json.load
8012d892 3262 info = self.sanitize_info(json.loads('\n'.join(f)), self.params.get('clean_infojson', True))
d4943898 3263 try:
aa9369a2 3264 self.__download_wrapper(self.process_ie_result)(info, download=True)
f2ebc5c7 3265 except (DownloadError, EntryNotInPlaylist, ReExtractInfo) as e:
bf5f605e 3266 if not isinstance(e, EntryNotInPlaylist):
3267 self.to_stderr('\r')
d4943898
JMF
3268 webpage_url = info.get('webpage_url')
3269 if webpage_url is not None:
aa9369a2 3270 self.report_warning(f'The info failed to download: {e}; trying with URL {webpage_url}')
d4943898
JMF
3271 return self.download([webpage_url])
3272 else:
3273 raise
3274 return self._download_retcode
1dcc4c0c 3275
cb202fd2 3276 @staticmethod
8012d892 3277 def sanitize_info(info_dict, remove_private_keys=False):
3278 ''' Sanitize the infodict for converting to json '''
3ad56b42 3279 if info_dict is None:
3280 return info_dict
6e84b215 3281 info_dict.setdefault('epoch', int(time.time()))
6a5a30f9 3282 info_dict.setdefault('_type', 'video')
09b49e1f 3283
8012d892 3284 if remove_private_keys:
09b49e1f 3285 reject = lambda k, v: v is None or (k.startswith('_') and k != '_type') or k in {
f46e2f9d 3286 'requested_downloads', 'requested_formats', 'requested_subtitles', 'requested_entries',
3287 'entries', 'filepath', 'infojson_filename', 'original_url', 'playlist_autonumber',
6e84b215 3288 }
ae8f99e6 3289 else:
09b49e1f 3290 reject = lambda k, v: False
adbc4ec4
THD
3291
3292 def filter_fn(obj):
3293 if isinstance(obj, dict):
3294 return {k: filter_fn(v) for k, v in obj.items() if not reject(k, v)}
3295 elif isinstance(obj, (list, tuple, set, LazyList)):
3296 return list(map(filter_fn, obj))
3297 elif obj is None or isinstance(obj, (str, int, float, bool)):
3298 return obj
3299 else:
3300 return repr(obj)
3301
5226731e 3302 return filter_fn(info_dict)
cb202fd2 3303
8012d892 3304 @staticmethod
3305 def filter_requested_info(info_dict, actually_filter=True):
3306 ''' Alias of sanitize_info for backward compatibility '''
3307 return YoutubeDL.sanitize_info(info_dict, actually_filter)
3308
ed5835b4 3309 @staticmethod
3310 def post_extract(info_dict):
3311 def actual_post_extract(info_dict):
3312 if info_dict.get('_type') in ('playlist', 'multi_video'):
3313 for video_dict in info_dict.get('entries', {}):
3314 actual_post_extract(video_dict or {})
3315 return
3316
09b49e1f 3317 post_extractor = info_dict.pop('__post_extractor', None) or (lambda: {})
3318 info_dict.update(post_extractor())
ed5835b4 3319
3320 actual_post_extract(info_dict or {})
3321
dcf64d43 3322 def run_pp(self, pp, infodict):
5bfa4862 3323 files_to_delete = []
dcf64d43 3324 if '__files_to_move' not in infodict:
3325 infodict['__files_to_move'] = {}
b1940459 3326 try:
3327 files_to_delete, infodict = pp.run(infodict)
3328 except PostProcessingError as e:
3329 # Must be True and not 'only_download'
3330 if self.params.get('ignoreerrors') is True:
3331 self.report_error(e)
3332 return infodict
3333 raise
3334
5bfa4862 3335 if not files_to_delete:
dcf64d43 3336 return infodict
5bfa4862 3337 if self.params.get('keepvideo', False):
3338 for f in files_to_delete:
dcf64d43 3339 infodict['__files_to_move'].setdefault(f, '')
5bfa4862 3340 else:
3341 for old_filename in set(files_to_delete):
3342 self.to_screen('Deleting original file %s (pass -k to keep)' % old_filename)
3343 try:
3344 os.remove(encodeFilename(old_filename))
3345 except (IOError, OSError):
3346 self.report_warning('Unable to remove downloaded original file')
dcf64d43 3347 if old_filename in infodict['__files_to_move']:
3348 del infodict['__files_to_move'][old_filename]
3349 return infodict
5bfa4862 3350
ed5835b4 3351 def run_all_pps(self, key, info, *, additional_pps=None):
bb66c247 3352 self._forceprint(key, info)
ed5835b4 3353 for pp in (additional_pps or []) + self._pps[key]:
dc5f409c 3354 info = self.run_pp(pp, info)
ed5835b4 3355 return info
277d6ff5 3356
56d868db 3357 def pre_process(self, ie_info, key='pre_process', files_to_move=None):
5bfa4862 3358 info = dict(ie_info)
56d868db 3359 info['__files_to_move'] = files_to_move or {}
ed5835b4 3360 info = self.run_all_pps(key, info)
56d868db 3361 return info, info.pop('__files_to_move', None)
5bfa4862 3362
f46e2f9d 3363 def post_process(self, filename, info, files_to_move=None):
8222d8de 3364 """Run all the postprocessors on the given file."""
8222d8de 3365 info['filepath'] = filename
dcf64d43 3366 info['__files_to_move'] = files_to_move or {}
ed5835b4 3367 info = self.run_all_pps('post_process', info, additional_pps=info.get('__postprocessors'))
dcf64d43 3368 info = self.run_pp(MoveFilesAfterDownloadPP(self), info)
3369 del info['__files_to_move']
ed5835b4 3370 return self.run_all_pps('after_move', info)
c1c9a79c 3371
5db07df6 3372 def _make_archive_id(self, info_dict):
e9fef7ee
S
3373 video_id = info_dict.get('id')
3374 if not video_id:
3375 return
5db07df6
PH
3376 # Future-proof against any change in case
3377 # and backwards compatibility with prior versions
e9fef7ee 3378 extractor = info_dict.get('extractor_key') or info_dict.get('ie_key') # key in a playlist
7012b23c 3379 if extractor is None:
1211bb6d
S
3380 url = str_or_none(info_dict.get('url'))
3381 if not url:
3382 return
e9fef7ee 3383 # Try to find matching extractor for the URL and take its ie_key
8b7491c8 3384 for ie_key, ie in self._ies.items():
1211bb6d 3385 if ie.suitable(url):
8b7491c8 3386 extractor = ie_key
e9fef7ee
S
3387 break
3388 else:
3389 return
d0757229 3390 return '%s %s' % (extractor.lower(), video_id)
5db07df6
PH
3391
3392 def in_download_archive(self, info_dict):
3393 fn = self.params.get('download_archive')
3394 if fn is None:
3395 return False
3396
3397 vid_id = self._make_archive_id(info_dict)
e9fef7ee 3398 if not vid_id:
7012b23c 3399 return False # Incomplete video information
5db07df6 3400
a45e8619 3401 return vid_id in self.archive
c1c9a79c
PH
3402
3403 def record_download_archive(self, info_dict):
3404 fn = self.params.get('download_archive')
3405 if fn is None:
3406 return
5db07df6
PH
3407 vid_id = self._make_archive_id(info_dict)
3408 assert vid_id
a13e6848 3409 self.write_debug(f'Adding to archive: {vid_id}')
c1c9a79c 3410 with locked_file(fn, 'a', encoding='utf-8') as archive_file:
6febd1c1 3411 archive_file.write(vid_id + '\n')
a45e8619 3412 self.archive.add(vid_id)
dd82ffea 3413
8c51aa65 3414 @staticmethod
8abeeb94 3415 def format_resolution(format, default='unknown'):
9359f3d4 3416 if format.get('vcodec') == 'none' and format.get('acodec') != 'none':
fb04e403 3417 return 'audio only'
f49d89ee
PH
3418 if format.get('resolution') is not None:
3419 return format['resolution']
35615307 3420 if format.get('width') and format.get('height'):
ff51ed58 3421 return '%dx%d' % (format['width'], format['height'])
35615307 3422 elif format.get('height'):
ff51ed58 3423 return '%sp' % format['height']
35615307 3424 elif format.get('width'):
ff51ed58 3425 return '%dx?' % format['width']
3426 return default
8c51aa65 3427
8130779d 3428 def _list_format_headers(self, *headers):
3429 if self.params.get('listformats_table', True) is not False:
3430 return [self._format_screen(header, self.Styles.HEADERS) for header in headers]
3431 return headers
3432
c57f7757
PH
3433 def _format_note(self, fdict):
3434 res = ''
3435 if fdict.get('ext') in ['f4f', 'f4m']:
f304da8a 3436 res += '(unsupported)'
32f90364
PH
3437 if fdict.get('language'):
3438 if res:
3439 res += ' '
f304da8a 3440 res += '[%s]' % fdict['language']
c57f7757 3441 if fdict.get('format_note') is not None:
f304da8a 3442 if res:
3443 res += ' '
3444 res += fdict['format_note']
c57f7757 3445 if fdict.get('tbr') is not None:
f304da8a 3446 if res:
3447 res += ', '
3448 res += '%4dk' % fdict['tbr']
c57f7757
PH
3449 if fdict.get('container') is not None:
3450 if res:
3451 res += ', '
3452 res += '%s container' % fdict['container']
3089bc74
S
3453 if (fdict.get('vcodec') is not None
3454 and fdict.get('vcodec') != 'none'):
c57f7757
PH
3455 if res:
3456 res += ', '
3457 res += fdict['vcodec']
91c7271a 3458 if fdict.get('vbr') is not None:
c57f7757
PH
3459 res += '@'
3460 elif fdict.get('vbr') is not None and fdict.get('abr') is not None:
3461 res += 'video@'
3462 if fdict.get('vbr') is not None:
3463 res += '%4dk' % fdict['vbr']
fbb21cf5 3464 if fdict.get('fps') is not None:
5d583bdf
S
3465 if res:
3466 res += ', '
3467 res += '%sfps' % fdict['fps']
c57f7757
PH
3468 if fdict.get('acodec') is not None:
3469 if res:
3470 res += ', '
3471 if fdict['acodec'] == 'none':
3472 res += 'video only'
3473 else:
3474 res += '%-5s' % fdict['acodec']
3475 elif fdict.get('abr') is not None:
3476 if res:
3477 res += ', '
3478 res += 'audio'
3479 if fdict.get('abr') is not None:
3480 res += '@%3dk' % fdict['abr']
3481 if fdict.get('asr') is not None:
3482 res += ' (%5dHz)' % fdict['asr']
3483 if fdict.get('filesize') is not None:
3484 if res:
3485 res += ', '
3486 res += format_bytes(fdict['filesize'])
9732d77e
PH
3487 elif fdict.get('filesize_approx') is not None:
3488 if res:
3489 res += ', '
3490 res += '~' + format_bytes(fdict['filesize_approx'])
c57f7757 3491 return res
91c7271a 3492
8130779d 3493 def render_formats_table(self, info_dict):
b69fd25c 3494 if not info_dict.get('formats') and not info_dict.get('url'):
8130779d 3495 return None
b69fd25c 3496
94badb25 3497 formats = info_dict.get('formats', [info_dict])
8130779d 3498 if not self.params.get('listformats_table', True) is not False:
76d321f6 3499 table = [
3500 [
3501 format_field(f, 'format_id'),
3502 format_field(f, 'ext'),
3503 self.format_resolution(f),
8130779d 3504 self._format_note(f)
3505 ] for f in formats if f.get('preference') is None or f['preference'] >= -1000]
3506 return render_table(['format code', 'extension', 'resolution', 'note'], table, extra_gap=1)
3507
3508 delim = self._format_screen('\u2502', self.Styles.DELIM, '|', test_encoding=True)
3509 table = [
3510 [
3511 self._format_screen(format_field(f, 'format_id'), self.Styles.ID),
3512 format_field(f, 'ext'),
3513 format_field(f, func=self.format_resolution, ignore=('audio only', 'images')),
3514 format_field(f, 'fps', '\t%d'),
3515 format_field(f, 'dynamic_range', '%s', ignore=(None, 'SDR')).replace('HDR', ''),
3516 delim,
3517 format_field(f, 'filesize', ' \t%s', func=format_bytes) + format_field(f, 'filesize_approx', '~\t%s', func=format_bytes),
3518 format_field(f, 'tbr', '\t%dk'),
3519 shorten_protocol_name(f.get('protocol', '')),
3520 delim,
3521 format_field(f, 'vcodec', default='unknown').replace(
3522 'none', 'images' if f.get('acodec') == 'none'
3523 else self._format_screen('audio only', self.Styles.SUPPRESS)),
3524 format_field(f, 'vbr', '\t%dk'),
3525 format_field(f, 'acodec', default='unknown').replace(
3526 'none', '' if f.get('vcodec') == 'none'
3527 else self._format_screen('video only', self.Styles.SUPPRESS)),
3528 format_field(f, 'abr', '\t%dk'),
3529 format_field(f, 'asr', '\t%dHz'),
3530 join_nonempty(
3531 self._format_screen('UNSUPPORTED', 'light red') if f.get('ext') in ('f4f', 'f4m') else None,
3532 format_field(f, 'language', '[%s]'),
3533 join_nonempty(format_field(f, 'format_note'),
3534 format_field(f, 'container', ignore=(None, f.get('ext'))),
3535 delim=', '),
3536 delim=' '),
3537 ] for f in formats if f.get('preference') is None or f['preference'] >= -1000]
3538 header_line = self._list_format_headers(
3539 'ID', 'EXT', 'RESOLUTION', '\tFPS', 'HDR', delim, '\tFILESIZE', '\tTBR', 'PROTO',
3540 delim, 'VCODEC', '\tVBR', 'ACODEC', '\tABR', '\tASR', 'MORE INFO')
3541
3542 return render_table(
3543 header_line, table, hide_empty=True,
3544 delim=self._format_screen('\u2500', self.Styles.DELIM, '-', test_encoding=True))
3545
3546 def render_thumbnails_table(self, info_dict):
88f23a18 3547 thumbnails = list(info_dict.get('thumbnails') or [])
cfb56d1a 3548 if not thumbnails:
8130779d 3549 return None
3550 return render_table(
ec11a9f4 3551 self._list_format_headers('ID', 'Width', 'Height', 'URL'),
6970b600 3552 [[t.get('id'), t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails])
2412044c 3553
8130779d 3554 def render_subtitles_table(self, video_id, subtitles):
2412044c 3555 def _row(lang, formats):
49c258e1 3556 exts, names = zip(*((f['ext'], f.get('name') or 'unknown') for f in reversed(formats)))
2412044c 3557 if len(set(names)) == 1:
7aee40c1 3558 names = [] if names[0] == 'unknown' else names[:1]
2412044c 3559 return [lang, ', '.join(names), ', '.join(exts)]
3560
8130779d 3561 if not subtitles:
3562 return None
3563 return render_table(
ec11a9f4 3564 self._list_format_headers('Language', 'Name', 'Formats'),
2412044c 3565 [_row(lang, formats) for lang, formats in subtitles.items()],
8130779d 3566 hide_empty=True)
3567
3568 def __list_table(self, video_id, name, func, *args):
3569 table = func(*args)
3570 if not table:
3571 self.to_screen(f'{video_id} has no {name}')
3572 return
3573 self.to_screen(f'[info] Available {name} for {video_id}:')
3574 self.to_stdout(table)
3575
3576 def list_formats(self, info_dict):
3577 self.__list_table(info_dict['id'], 'formats', self.render_formats_table, info_dict)
3578
3579 def list_thumbnails(self, info_dict):
3580 self.__list_table(info_dict['id'], 'thumbnails', self.render_thumbnails_table, info_dict)
3581
3582 def list_subtitles(self, video_id, subtitles, name='subtitles'):
3583 self.__list_table(video_id, name, self.render_subtitles_table, video_id, subtitles)
a504ced0 3584
dca08720
PH
3585 def urlopen(self, req):
3586 """ Start an HTTP download """
82d8a8b6 3587 if isinstance(req, compat_basestring):
67dda517 3588 req = sanitized_Request(req)
19a41fc6 3589 return self._opener.open(req, timeout=self._socket_timeout)
dca08720
PH
3590
3591 def print_debug_header(self):
3592 if not self.params.get('verbose'):
3593 return
49a57e70 3594
3595 def get_encoding(stream):
2a938746 3596 ret = str(getattr(stream, 'encoding', 'missing (%s)' % type(stream).__name__))
49a57e70 3597 if not supports_terminal_sequences(stream):
e3c7d495 3598 from .compat import WINDOWS_VT_MODE
3599 ret += ' (No VT)' if WINDOWS_VT_MODE is False else ' (No ANSI)'
49a57e70 3600 return ret
3601
3602 encoding_str = 'Encodings: locale %s, fs %s, out %s, err %s, pref %s' % (
3603 locale.getpreferredencoding(),
3604 sys.getfilesystemencoding(),
cf4f42cb 3605 get_encoding(self._out_files['screen']), get_encoding(self._out_files['error']),
49a57e70 3606 self.get_encoding())
883d4b1e 3607
3608 logger = self.params.get('logger')
3609 if logger:
3610 write_debug = lambda msg: logger.debug(f'[debug] {msg}')
3611 write_debug(encoding_str)
3612 else:
96565c7e 3613 write_string(f'[debug] {encoding_str}\n', encoding=None)
49a57e70 3614 write_debug = lambda msg: self._write_string(f'[debug] {msg}\n')
734f90bb 3615
4c88ff87 3616 source = detect_variant()
36eaf303 3617 write_debug(join_nonempty(
3618 'yt-dlp version', __version__,
3619 f'[{RELEASE_GIT_HEAD}]' if RELEASE_GIT_HEAD else '',
3620 '' if source == 'unknown' else f'({source})',
3621 delim=' '))
6e21fdd2 3622 if not _LAZY_LOADER:
3623 if os.environ.get('YTDLP_NO_LAZY_EXTRACTORS'):
49a57e70 3624 write_debug('Lazy loading extractors is forcibly disabled')
6e21fdd2 3625 else:
49a57e70 3626 write_debug('Lazy loading extractors is disabled')
3ae5e797 3627 if plugin_extractors or plugin_postprocessors:
49a57e70 3628 write_debug('Plugins: %s' % [
3ae5e797 3629 '%s%s' % (klass.__name__, '' if klass.__name__ == name else f' as {name}')
3630 for name, klass in itertools.chain(plugin_extractors.items(), plugin_postprocessors.items())])
53ed7066 3631 if self.params.get('compat_opts'):
49a57e70 3632 write_debug('Compatibility options: %s' % ', '.join(self.params.get('compat_opts')))
36eaf303 3633
3634 if source == 'source':
dca08720 3635 try:
36eaf303 3636 sp = Popen(
3637 ['git', 'rev-parse', '--short', 'HEAD'],
3638 stdout=subprocess.PIPE, stderr=subprocess.PIPE,
3639 cwd=os.path.dirname(os.path.abspath(__file__)))
3640 out, err = sp.communicate_or_kill()
3641 out = out.decode().strip()
3642 if re.match('[0-9a-f]+', out):
3643 write_debug('Git HEAD: %s' % out)
70a1165b 3644 except Exception:
36eaf303 3645 try:
3646 sys.exc_clear()
3647 except Exception:
3648 pass
b300cda4
S
3649
3650 def python_implementation():
3651 impl_name = platform.python_implementation()
3652 if impl_name == 'PyPy' and hasattr(sys, 'pypy_version_info'):
3653 return impl_name + ' version %d.%d.%d' % sys.pypy_version_info[:3]
3654 return impl_name
3655
49a57e70 3656 write_debug('Python version %s (%s %s) - %s' % (
e5813e53 3657 platform.python_version(),
3658 python_implementation(),
3659 platform.architecture()[0],
b300cda4 3660 platform_name()))
d28b5171 3661
8913ef74 3662 exe_versions, ffmpeg_features = FFmpegPostProcessor.get_versions_and_features(self)
3663 ffmpeg_features = {key for key, val in ffmpeg_features.items() if val}
3664 if ffmpeg_features:
a4211baf 3665 exe_versions['ffmpeg'] += ' (%s)' % ','.join(ffmpeg_features)
8913ef74 3666
4c83c967 3667 exe_versions['rtmpdump'] = rtmpdump_version()
feee8d32 3668 exe_versions['phantomjs'] = PhantomJSwrapper._version()
d28b5171 3669 exe_str = ', '.join(
2831b468 3670 f'{exe} {v}' for exe, v in sorted(exe_versions.items()) if v
3671 ) or 'none'
49a57e70 3672 write_debug('exe versions: %s' % exe_str)
dca08720 3673
2831b468 3674 from .downloader.websocket import has_websockets
3675 from .postprocessor.embedthumbnail import has_mutagen
f59f5ef8 3676 from .cookies import SQLITE_AVAILABLE, SECRETSTORAGE_AVAILABLE
2831b468 3677
c586f9e8 3678 lib_str = join_nonempty(
4390d5ec 3679 compat_brotli and compat_brotli.__name__,
edf65256 3680 compat_pycrypto_AES and compat_pycrypto_AES.__name__.split('.')[0],
f59f5ef8 3681 SECRETSTORAGE_AVAILABLE and 'secretstorage',
2831b468 3682 has_mutagen and 'mutagen',
3683 SQLITE_AVAILABLE and 'sqlite',
c586f9e8 3684 has_websockets and 'websockets',
3685 delim=', ') or 'none'
49a57e70 3686 write_debug('Optional libraries: %s' % lib_str)
2831b468 3687
dca08720
PH
3688 proxy_map = {}
3689 for handler in self._opener.handlers:
3690 if hasattr(handler, 'proxies'):
3691 proxy_map.update(handler.proxies)
49a57e70 3692 write_debug(f'Proxy map: {proxy_map}')
dca08720 3693
49a57e70 3694 # Not implemented
3695 if False and self.params.get('call_home'):
58b1f00d 3696 ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode('utf-8')
49a57e70 3697 write_debug('Public IP address: %s' % ipaddr)
58b1f00d
PH
3698 latest_version = self.urlopen(
3699 'https://yt-dl.org/latest/version').read().decode('utf-8')
3700 if version_tuple(latest_version) > version_tuple(__version__):
3701 self.report_warning(
3702 'You are using an outdated version (newest version: %s)! '
3703 'See https://yt-dl.org/update if you need help updating.' %
3704 latest_version)
3705
e344693b 3706 def _setup_opener(self):
6ad14cab 3707 timeout_val = self.params.get('socket_timeout')
17bddf3e 3708 self._socket_timeout = 20 if timeout_val is None else float(timeout_val)
6ad14cab 3709
982ee69a 3710 opts_cookiesfrombrowser = self.params.get('cookiesfrombrowser')
dca08720
PH
3711 opts_cookiefile = self.params.get('cookiefile')
3712 opts_proxy = self.params.get('proxy')
3713
982ee69a 3714 self.cookiejar = load_cookies(opts_cookiefile, opts_cookiesfrombrowser, self)
dca08720 3715
6a3f4c3f 3716 cookie_processor = YoutubeDLCookieProcessor(self.cookiejar)
dca08720
PH
3717 if opts_proxy is not None:
3718 if opts_proxy == '':
3719 proxies = {}
3720 else:
3721 proxies = {'http': opts_proxy, 'https': opts_proxy}
3722 else:
3723 proxies = compat_urllib_request.getproxies()
067aa17e 3724 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
dca08720
PH
3725 if 'http' in proxies and 'https' not in proxies:
3726 proxies['https'] = proxies['http']
91410c9b 3727 proxy_handler = PerRequestProxyHandler(proxies)
a0ddb8a2
PH
3728
3729 debuglevel = 1 if self.params.get('debug_printtraffic') else 0
be4a824d
PH
3730 https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel)
3731 ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel)
fca6dba8 3732 redirect_handler = YoutubeDLRedirectHandler()
8b172c2e 3733 data_handler = compat_urllib_request_DataHandler()
6240b0a2
JMF
3734
3735 # When passing our own FileHandler instance, build_opener won't add the
3736 # default FileHandler and allows us to disable the file protocol, which
3737 # can be used for malicious purposes (see
067aa17e 3738 # https://github.com/ytdl-org/youtube-dl/issues/8227)
6240b0a2
JMF
3739 file_handler = compat_urllib_request.FileHandler()
3740
3741 def file_open(*args, **kwargs):
7a5c1cfe 3742 raise compat_urllib_error.URLError('file:// scheme is explicitly disabled in yt-dlp for security reasons')
6240b0a2
JMF
3743 file_handler.file_open = file_open
3744
3745 opener = compat_urllib_request.build_opener(
fca6dba8 3746 proxy_handler, https_handler, cookie_processor, ydlh, redirect_handler, data_handler, file_handler)
2461f79d 3747
dca08720
PH
3748 # Delete the default user-agent header, which would otherwise apply in
3749 # cases where our custom HTTP handler doesn't come into play
067aa17e 3750 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
dca08720
PH
3751 opener.addheaders = []
3752 self._opener = opener
62fec3b2
PH
3753
3754 def encode(self, s):
3755 if isinstance(s, bytes):
3756 return s # Already encoded
3757
3758 try:
3759 return s.encode(self.get_encoding())
3760 except UnicodeEncodeError as err:
3761 err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.'
3762 raise
3763
3764 def get_encoding(self):
3765 encoding = self.params.get('encoding')
3766 if encoding is None:
3767 encoding = preferredencoding()
3768 return encoding
ec82d85a 3769
e08a85d8 3770 def _write_info_json(self, label, ie_result, infofn, overwrite=None):
80c03fa9 3771 ''' Write infojson and returns True = written, False = skip, None = error '''
e08a85d8 3772 if overwrite is None:
3773 overwrite = self.params.get('overwrites', True)
80c03fa9 3774 if not self.params.get('writeinfojson'):
3775 return False
3776 elif not infofn:
3777 self.write_debug(f'Skipping writing {label} infojson')
3778 return False
3779 elif not self._ensure_dir_exists(infofn):
3780 return None
e08a85d8 3781 elif not overwrite and os.path.exists(infofn):
80c03fa9 3782 self.to_screen(f'[info] {label.title()} metadata is already present')
3783 else:
3784 self.to_screen(f'[info] Writing {label} metadata as JSON to: {infofn}')
3785 try:
3786 write_json_file(self.sanitize_info(ie_result, self.params.get('clean_infojson', True)), infofn)
3787 except (OSError, IOError):
3788 self.report_error(f'Cannot write {label} metadata to JSON file {infofn}')
3789 return None
3790 return True
3791
3792 def _write_description(self, label, ie_result, descfn):
3793 ''' Write description and returns True = written, False = skip, None = error '''
3794 if not self.params.get('writedescription'):
3795 return False
3796 elif not descfn:
3797 self.write_debug(f'Skipping writing {label} description')
3798 return False
3799 elif not self._ensure_dir_exists(descfn):
3800 return None
3801 elif not self.params.get('overwrites', True) and os.path.exists(descfn):
3802 self.to_screen(f'[info] {label.title()} description is already present')
3803 elif ie_result.get('description') is None:
3804 self.report_warning(f'There\'s no {label} description to write')
3805 return False
3806 else:
3807 try:
3808 self.to_screen(f'[info] Writing {label} description to: {descfn}')
3809 with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
3810 descfile.write(ie_result['description'])
3811 except (OSError, IOError):
3812 self.report_error(f'Cannot write {label} description file {descfn}')
3813 return None
3814 return True
3815
3816 def _write_subtitles(self, info_dict, filename):
3817 ''' Write subtitles to file and return list of (sub_filename, final_sub_filename); or None if error'''
3818 ret = []
3819 subtitles = info_dict.get('requested_subtitles')
3820 if not subtitles or not (self.params.get('writesubtitles') or self.params.get('writeautomaticsub')):
3821 # subtitles download errors are already managed as troubles in relevant IE
3822 # that way it will silently go on when used with unsupporting IE
3823 return ret
3824
3825 sub_filename_base = self.prepare_filename(info_dict, 'subtitle')
3826 if not sub_filename_base:
3827 self.to_screen('[info] Skipping writing video subtitles')
3828 return ret
3829 for sub_lang, sub_info in subtitles.items():
3830 sub_format = sub_info['ext']
3831 sub_filename = subtitles_filename(filename, sub_lang, sub_format, info_dict.get('ext'))
3832 sub_filename_final = subtitles_filename(sub_filename_base, sub_lang, sub_format, info_dict.get('ext'))
e04938ab 3833 existing_sub = self.existing_file((sub_filename_final, sub_filename))
3834 if existing_sub:
80c03fa9 3835 self.to_screen(f'[info] Video subtitle {sub_lang}.{sub_format} is already present')
e04938ab 3836 sub_info['filepath'] = existing_sub
3837 ret.append((existing_sub, sub_filename_final))
80c03fa9 3838 continue
3839
3840 self.to_screen(f'[info] Writing video subtitles to: {sub_filename}')
3841 if sub_info.get('data') is not None:
3842 try:
3843 # Use newline='' to prevent conversion of newline characters
3844 # See https://github.com/ytdl-org/youtube-dl/issues/10268
3845 with io.open(sub_filename, 'w', encoding='utf-8', newline='') as subfile:
3846 subfile.write(sub_info['data'])
3847 sub_info['filepath'] = sub_filename
3848 ret.append((sub_filename, sub_filename_final))
3849 continue
3850 except (OSError, IOError):
3851 self.report_error(f'Cannot write video subtitles file {sub_filename}')
3852 return None
3853
3854 try:
3855 sub_copy = sub_info.copy()
3856 sub_copy.setdefault('http_headers', info_dict.get('http_headers'))
3857 self.dl(sub_filename, sub_copy, subtitle=True)
3858 sub_info['filepath'] = sub_filename
3859 ret.append((sub_filename, sub_filename_final))
6020e05d 3860 except (DownloadError, ExtractorError, IOError, OSError, ValueError) + network_exceptions as err:
3861 if self.params.get('ignoreerrors') is not True: # False or 'only_download'
3862 raise DownloadError(f'Unable to download video subtitles for {sub_lang!r}: {err}', err)
80c03fa9 3863 self.report_warning(f'Unable to download video subtitles for {sub_lang!r}: {err}')
519804a9 3864 return ret
80c03fa9 3865
3866 def _write_thumbnails(self, label, info_dict, filename, thumb_filename_base=None):
3867 ''' Write thumbnails to file and return list of (thumb_filename, final_thumb_filename) '''
6c4fd172 3868 write_all = self.params.get('write_all_thumbnails', False)
80c03fa9 3869 thumbnails, ret = [], []
6c4fd172 3870 if write_all or self.params.get('writethumbnail', False):
0202b52a 3871 thumbnails = info_dict.get('thumbnails') or []
6c4fd172 3872 multiple = write_all and len(thumbnails) > 1
ec82d85a 3873
80c03fa9 3874 if thumb_filename_base is None:
3875 thumb_filename_base = filename
3876 if thumbnails and not thumb_filename_base:
3877 self.write_debug(f'Skipping writing {label} thumbnail')
3878 return ret
3879
dd0228ce 3880 for idx, t in list(enumerate(thumbnails))[::-1]:
80c03fa9 3881 thumb_ext = (f'{t["id"]}.' if multiple else '') + determine_ext(t['url'], 'jpg')
aa9369a2 3882 thumb_display_id = f'{label} thumbnail {t["id"]}'
80c03fa9 3883 thumb_filename = replace_extension(filename, thumb_ext, info_dict.get('ext'))
3884 thumb_filename_final = replace_extension(thumb_filename_base, thumb_ext, info_dict.get('ext'))
ec82d85a 3885
e04938ab 3886 existing_thumb = self.existing_file((thumb_filename_final, thumb_filename))
3887 if existing_thumb:
aa9369a2 3888 self.to_screen('[info] %s is already present' % (
3889 thumb_display_id if multiple else f'{label} thumbnail').capitalize())
e04938ab 3890 t['filepath'] = existing_thumb
3891 ret.append((existing_thumb, thumb_filename_final))
ec82d85a 3892 else:
80c03fa9 3893 self.to_screen(f'[info] Downloading {thumb_display_id} ...')
ec82d85a 3894 try:
297e9952 3895 uf = self.urlopen(sanitized_Request(t['url'], headers=t.get('http_headers', {})))
80c03fa9 3896 self.to_screen(f'[info] Writing {thumb_display_id} to: {thumb_filename}')
d3d89c32 3897 with open(encodeFilename(thumb_filename), 'wb') as thumbf:
ec82d85a 3898 shutil.copyfileobj(uf, thumbf)
80c03fa9 3899 ret.append((thumb_filename, thumb_filename_final))
885cc0b7 3900 t['filepath'] = thumb_filename
3158150c 3901 except network_exceptions as err:
dd0228ce 3902 thumbnails.pop(idx)
80c03fa9 3903 self.report_warning(f'Unable to download {thumb_display_id}: {err}')
6c4fd172 3904 if ret and not write_all:
3905 break
0202b52a 3906 return ret