]> jfr.im git - yt-dlp.git/blame - yt_dlp/YoutubeDL.py
[doc] Minor improvements
[yt-dlp.git] / yt_dlp / YoutubeDL.py
CommitLineData
cc52de43 1#!/usr/bin/env python3
26e63931 2import collections
31bd3925 3import contextlib
9d2ecdbc 4import datetime
c1c9a79c 5import errno
31bd3925 6import fileinput
b5ae35ee 7import functools
8222d8de 8import io
b82f815f 9import itertools
8694c600 10import json
62fec3b2 11import locale
083c9df9 12import operator
8222d8de 13import os
dca08720 14import platform
f8271158 15import random
8222d8de
JMF
16import re
17import shutil
dca08720 18import subprocess
8222d8de 19import sys
21cd8fae 20import tempfile
8222d8de 21import time
67134eab 22import tokenize
8222d8de 23import traceback
524e2e4f 24import unicodedata
f9934b96 25import urllib.request
961ea474
S
26from string import ascii_letters
27
f8271158 28from .cache import Cache
8c25f81b 29from .compat import (
003c69a8 30 compat_get_terminal_size,
e9c0cdd3 31 compat_os_name,
7d1eb38a 32 compat_shlex_quote,
ce02ed60
PH
33 compat_str,
34 compat_urllib_error,
35 compat_urllib_request,
819e0531 36 windows_enable_vt_mode,
8c25f81b 37)
982ee69a 38from .cookies import load_cookies
f8271158 39from .downloader import FFmpegFD, get_suitable_downloader, shorten_protocol_name
40from .downloader.rtmp import rtmpdump_version
41from .extractor import _LAZY_LOADER
42from .extractor import _PLUGIN_CLASSES as plugin_extractors
43from .extractor import gen_extractor_classes, get_info_extractor
44from .extractor.openload import PhantomJSwrapper
45from .minicurses import format_text
46from .postprocessor import _PLUGIN_CLASSES as plugin_postprocessors
47from .postprocessor import (
48 EmbedThumbnailPP,
49 FFmpegFixupDuplicateMoovPP,
50 FFmpegFixupDurationPP,
51 FFmpegFixupM3u8PP,
52 FFmpegFixupM4aPP,
53 FFmpegFixupStretchedPP,
54 FFmpegFixupTimestampPP,
55 FFmpegMergerPP,
56 FFmpegPostProcessor,
57 MoveFilesAfterDownloadPP,
58 get_postprocessor,
59)
60from .update import detect_variant
8c25f81b 61from .utils import (
f8271158 62 DEFAULT_OUTTMPL,
63 LINK_TEMPLATES,
64 NO_DEFAULT,
65 OUTTMPL_TYPES,
66 POSTPROCESS_WHEN,
67 STR_FORMAT_RE_TMPL,
68 STR_FORMAT_TYPES,
69 ContentTooShortError,
70 DateRange,
71 DownloadCancelled,
72 DownloadError,
73 EntryNotInPlaylist,
74 ExistingVideoReached,
75 ExtractorError,
76 GeoRestrictedError,
77 HEADRequest,
78 InAdvancePagedList,
79 ISO3166Utils,
80 LazyList,
81 MaxDownloadsReached,
19a03940 82 Namespace,
f8271158 83 PagedList,
84 PerRequestProxyHandler,
85 Popen,
86 PostProcessingError,
87 ReExtractInfo,
88 RejectedVideoReached,
89 SameFileError,
90 UnavailableVideoError,
91 YoutubeDLCookieProcessor,
92 YoutubeDLHandler,
93 YoutubeDLRedirectHandler,
eedb7ba5
S
94 age_restricted,
95 args_to_str,
ce02ed60 96 date_from_str,
ce02ed60 97 determine_ext,
b5559424 98 determine_protocol,
c0384f22 99 encode_compat_str,
ce02ed60 100 encodeFilename,
a06916d9 101 error_to_compat_str,
590bc6f6 102 expand_path,
90137ca4 103 filter_dict,
e29663c6 104 float_or_none,
02dbf93f 105 format_bytes,
e0fd9573 106 format_decimal_suffix,
f8271158 107 format_field,
525ef922 108 formatSeconds,
0bb322b9 109 get_domain,
c9969434 110 int_or_none,
732044af 111 iri_to_uri,
34921b43 112 join_nonempty,
ce02ed60 113 locked_file,
0202b52a 114 make_dir,
dca08720 115 make_HTTPS_handler,
8b7539d2 116 merge_headers,
3158150c 117 network_exceptions,
ec11a9f4 118 number_of_digits,
cd6fc19e 119 orderedSet,
083c9df9 120 parse_filesize,
dca08720 121 platform_name,
ce02ed60 122 preferredencoding,
eedb7ba5 123 prepend_extension,
51fb4995 124 register_socks_protocols,
3efb96a6 125 remove_terminal_sequences,
cfb56d1a 126 render_table,
eedb7ba5 127 replace_extension,
ce02ed60 128 sanitize_filename,
1bb5c511 129 sanitize_path,
dcf77cf1 130 sanitize_url,
67dda517 131 sanitized_Request,
e5660ee6 132 std_headers,
1211bb6d 133 str_or_none,
e29663c6 134 strftime_or_none,
ce02ed60 135 subtitles_filename,
819e0531 136 supports_terminal_sequences,
f2ebc5c7 137 timetuple_from_msec,
732044af 138 to_high_limit_path,
324ad820 139 traverse_obj,
6033d980 140 try_get,
29eb5174 141 url_basename,
7d1eb38a 142 variadic,
58b1f00d 143 version_tuple,
ce02ed60
PH
144 write_json_file,
145 write_string,
4f026faf 146)
f8271158 147from .version import RELEASE_GIT_HEAD, __version__
8222d8de 148
e9c0cdd3
YCH
149if compat_os_name == 'nt':
150 import ctypes
151
2459b6e1 152
86e5f3ed 153class YoutubeDL:
8222d8de
JMF
154 """YoutubeDL class.
155
156 YoutubeDL objects are the ones responsible of downloading the
157 actual video file and writing it to disk if the user has requested
158 it, among some other tasks. In most cases there should be one per
159 program. As, given a video URL, the downloader doesn't know how to
160 extract all the needed information, task that InfoExtractors do, it
161 has to pass the URL to one of them.
162
163 For this, YoutubeDL objects have a method that allows
164 InfoExtractors to be registered in a given order. When it is passed
165 a URL, the YoutubeDL object handles it to the first InfoExtractor it
166 finds that reports being able to handle it. The InfoExtractor extracts
167 all the information about the video or videos the URL refers to, and
168 YoutubeDL process the extracted information, possibly using a File
169 Downloader to download the video.
170
171 YoutubeDL objects accept a lot of parameters. In order not to saturate
172 the object constructor with arguments, it receives a dictionary of
173 options instead. These options are available through the params
174 attribute for the InfoExtractors to use. The YoutubeDL also
175 registers itself as the downloader in charge for the InfoExtractors
176 that are added to it, so this is a "mutual registration".
177
178 Available options:
179
180 username: Username for authentication purposes.
181 password: Password for authentication purposes.
180940e0 182 videopassword: Password for accessing a video.
1da50aa3
S
183 ap_mso: Adobe Pass multiple-system operator identifier.
184 ap_username: Multiple-system operator account username.
185 ap_password: Multiple-system operator account password.
8222d8de
JMF
186 usenetrc: Use netrc for authentication instead.
187 verbose: Print additional info to stdout.
188 quiet: Do not print messages to stdout.
ad8915b7 189 no_warnings: Do not print out anything for warnings.
bb66c247 190 forceprint: A dict with keys WHEN mapped to a list of templates to
191 print to stdout. The allowed keys are video or any of the
192 items in utils.POSTPROCESS_WHEN.
ca30f449 193 For compatibility, a single list is also accepted
bb66c247 194 print_to_file: A dict with keys WHEN (same as forceprint) mapped to
195 a list of tuples with (template, filename)
53c18592 196 forceurl: Force printing final URL. (Deprecated)
197 forcetitle: Force printing title. (Deprecated)
198 forceid: Force printing ID. (Deprecated)
199 forcethumbnail: Force printing thumbnail URL. (Deprecated)
200 forcedescription: Force printing description. (Deprecated)
201 forcefilename: Force printing final filename. (Deprecated)
202 forceduration: Force printing duration. (Deprecated)
8694c600 203 forcejson: Force printing info_dict as JSON.
63e0be34
PH
204 dump_single_json: Force printing the info_dict of the whole playlist
205 (or video) as a single JSON line.
c25228e5 206 force_write_download_archive: Force writing download archive regardless
207 of 'skip_download' or 'simulate'.
b7b04c78 208 simulate: Do not download the video files. If unset (or None),
209 simulate only if listsubtitles, listformats or list_thumbnails is used
eb8a4433 210 format: Video format code. see "FORMAT SELECTION" for more details.
093a1710 211 You can also pass a function. The function takes 'ctx' as
212 argument and returns the formats to download.
213 See "build_format_selector" for an implementation
63ad4d43 214 allow_unplayable_formats: Allow unplayable formats to be extracted and downloaded.
b7da73eb 215 ignore_no_formats_error: Ignore "No video formats" error. Usefull for
216 extracting metadata even if the video is not actually
217 available for download (experimental)
0930b11f 218 format_sort: A list of fields by which to sort the video formats.
219 See "Sorting Formats" for more details.
c25228e5 220 format_sort_force: Force the given format_sort. see "Sorting Formats"
221 for more details.
08d30158 222 prefer_free_formats: Whether to prefer video formats with free containers
223 over non-free ones of same quality.
c25228e5 224 allow_multiple_video_streams: Allow multiple video streams to be merged
225 into a single file
226 allow_multiple_audio_streams: Allow multiple audio streams to be merged
227 into a single file
0ba692ac 228 check_formats Whether to test if the formats are downloadable.
9f1a1c36 229 Can be True (check all), False (check none),
230 'selected' (check selected formats),
0ba692ac 231 or None (check only if requested by extractor)
4524baf0 232 paths: Dictionary of output paths. The allowed keys are 'home'
233 'temp' and the keys of OUTTMPL_TYPES (in utils.py)
de6000d9 234 outtmpl: Dictionary of templates for output names. Allowed keys
4524baf0 235 are 'default' and the keys of OUTTMPL_TYPES (in utils.py).
34488702 236 For compatibility with youtube-dl, a single string can also be used
a820dc72
RA
237 outtmpl_na_placeholder: Placeholder for unavailable meta fields.
238 restrictfilenames: Do not allow "&" and spaces in file names
239 trim_file_name: Limit length of filename (extension excluded)
4524baf0 240 windowsfilenames: Force the filenames to be windows compatible
b1940459 241 ignoreerrors: Do not stop on download/postprocessing errors.
242 Can be 'only_download' to ignore only download errors.
243 Default is 'only_download' for CLI, but False for API
26e2805c 244 skip_playlist_after_errors: Number of allowed failures until the rest of
245 the playlist is skipped
d22dec74 246 force_generic_extractor: Force downloader to use the generic extractor
0c3d0f51 247 overwrites: Overwrite all video and metadata files if True,
248 overwrite only non-video files if None
249 and don't overwrite any file if False
34488702 250 For compatibility with youtube-dl,
251 "nooverwrites" may also be used instead
8222d8de
JMF
252 playliststart: Playlist item to start at.
253 playlistend: Playlist item to end at.
c14e88f0 254 playlist_items: Specific indices of playlist to download.
ff815fe6 255 playlistreverse: Download playlist items in reverse order.
75822ca7 256 playlistrandom: Download playlist items in random order.
8222d8de
JMF
257 matchtitle: Download only matching titles.
258 rejecttitle: Reject downloads for matching titles.
8bf9319e 259 logger: Log messages to a logging.Logger instance.
8222d8de 260 logtostderr: Log messages to stderr instead of stdout.
819e0531 261 consoletitle: Display progress in console window's titlebar.
8222d8de
JMF
262 writedescription: Write the video description to a .description file
263 writeinfojson: Write the video description to a .info.json file
75d43ca0 264 clean_infojson: Remove private fields from the infojson
34488702 265 getcomments: Extract video comments. This will not be written to disk
06167fbb 266 unless writeinfojson is also given
1fb07d10 267 writeannotations: Write the video annotations to a .annotations.xml file
8222d8de 268 writethumbnail: Write the thumbnail image to a file
c25228e5 269 allow_playlist_files: Whether to write playlists' description, infojson etc
270 also to disk when using the 'write*' options
ec82d85a 271 write_all_thumbnails: Write all thumbnail formats to files
732044af 272 writelink: Write an internet shortcut file, depending on the
273 current platform (.url/.webloc/.desktop)
274 writeurllink: Write a Windows internet shortcut file (.url)
275 writewebloclink: Write a macOS internet shortcut file (.webloc)
276 writedesktoplink: Write a Linux internet shortcut file (.desktop)
8222d8de 277 writesubtitles: Write the video subtitles to a file
741dd8ea 278 writeautomaticsub: Write the automatically generated subtitles to a file
245524e6 279 allsubtitles: Deprecated - Use subtitleslangs = ['all']
c32b0aab 280 Downloads all the subtitles of the video
0b7f3118 281 (requires writesubtitles or writeautomaticsub)
8222d8de 282 listsubtitles: Lists all available subtitles for the video
a504ced0 283 subtitlesformat: The format code for subtitles
c32b0aab 284 subtitleslangs: List of languages of the subtitles to download (can be regex).
285 The list may contain "all" to refer to all the available
286 subtitles. The language can be prefixed with a "-" to
287 exclude it from the requested languages. Eg: ['all', '-live_chat']
8222d8de
JMF
288 keepvideo: Keep the video file after post-processing
289 daterange: A DateRange object, download only if the upload_date is in the range.
290 skip_download: Skip the actual download of the video file
c35f9e72 291 cachedir: Location of the cache files in the filesystem.
a0e07d31 292 False to disable filesystem cache.
47192f92 293 noplaylist: Download single video instead of a playlist if in doubt.
8dbe9899
PH
294 age_limit: An integer representing the user's age in years.
295 Unsuitable videos for the given age are skipped.
5fe18bdb
PH
296 min_views: An integer representing the minimum view count the video
297 must have in order to not be skipped.
298 Videos without view count information are always
299 downloaded. None for no limit.
300 max_views: An integer representing the maximum view count.
301 Videos that are more popular than that are not
302 downloaded.
303 Videos without view count information are always
304 downloaded. None for no limit.
305 download_archive: File name of a file where all downloads are recorded.
c1c9a79c
PH
306 Videos already present in the file are not downloaded
307 again.
8a51f564 308 break_on_existing: Stop the download process after attempting to download a
309 file that is in the archive.
310 break_on_reject: Stop the download process when encountering a video that
311 has been filtered out.
b222c271 312 break_per_url: Whether break_on_reject and break_on_existing
313 should act on each input URL as opposed to for the entire queue
8a51f564 314 cookiefile: File name where cookies should be read from and dumped to
f59f5ef8
MB
315 cookiesfrombrowser: A tuple containing the name of the browser, the profile
316 name/pathfrom where cookies are loaded, and the name of the
317 keyring. Eg: ('chrome', ) or ('vivaldi', 'default', 'BASICTEXT')
f81c62a6 318 legacyserverconnect: Explicitly allow HTTPS connection to servers that do not
319 support RFC 5746 secure renegotiation
f59f5ef8 320 nocheckcertificate: Do not verify SSL certificates
7e8c0af0
PH
321 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
322 At the moment, this is only supported by YouTube.
8b7539d2 323 http_headers: A dictionary of custom headers to be used for all requests
a1ee09e8 324 proxy: URL of the proxy server to use
38cce791 325 geo_verification_proxy: URL of the proxy to use for IP address verification
504f20dd 326 on geo-restricted sites.
e344693b 327 socket_timeout: Time to wait for unresponsive hosts, in seconds
0783b09b
PH
328 bidi_workaround: Work around buggy terminals without bidirectional text
329 support, using fridibi
a0ddb8a2 330 debug_printtraffic:Print out sent and received HTTP traffic
91f071af 331 include_ads: Download ads as well (deprecated)
04b4d394
PH
332 default_search: Prepend this string if an input url is not valid.
333 'auto' for elaborate guessing
62fec3b2 334 encoding: Use this encoding instead of the system-specified.
e8ee972c 335 extract_flat: Do not resolve URLs, return the immediate result.
057a5206
PH
336 Pass in 'in_playlist' to only show this behavior for
337 playlist items.
f2ebc5c7 338 wait_for_video: If given, wait for scheduled streams to become available.
339 The value should be a tuple containing the range
340 (min_secs, max_secs) to wait between retries
4f026faf 341 postprocessors: A list of dictionaries, each with an entry
71b640cc 342 * key: The name of the postprocessor. See
7a5c1cfe 343 yt_dlp/postprocessor/__init__.py for a list.
bb66c247 344 * when: When to run the postprocessor. Allowed values are
345 the entries of utils.POSTPROCESS_WHEN
56d868db 346 Assumed to be 'post_process' if not given
b5ae35ee 347 post_hooks: Deprecated - Register a custom postprocessor instead
348 A list of functions that get called as the final step
ab8e5e51
AM
349 for each video file, after all postprocessors have been
350 called. The filename will be passed as the only argument.
71b640cc
PH
351 progress_hooks: A list of functions that get called on download
352 progress, with a dictionary with the entries
5cda4eda 353 * status: One of "downloading", "error", or "finished".
ee69b99a 354 Check this first and ignore unknown values.
3ba7740d 355 * info_dict: The extracted info_dict
71b640cc 356
5cda4eda 357 If status is one of "downloading", or "finished", the
ee69b99a
PH
358 following properties may also be present:
359 * filename: The final filename (always present)
5cda4eda 360 * tmpfilename: The filename we're currently writing to
71b640cc
PH
361 * downloaded_bytes: Bytes on disk
362 * total_bytes: Size of the whole file, None if unknown
5cda4eda
PH
363 * total_bytes_estimate: Guess of the eventual file size,
364 None if unavailable.
365 * elapsed: The number of seconds since download started.
71b640cc
PH
366 * eta: The estimated time in seconds, None if unknown
367 * speed: The download speed in bytes/second, None if
368 unknown
5cda4eda
PH
369 * fragment_index: The counter of the currently
370 downloaded video fragment.
371 * fragment_count: The number of fragments (= individual
372 files that will be merged)
71b640cc
PH
373
374 Progress hooks are guaranteed to be called at least once
375 (with status "finished") if the download is successful.
819e0531 376 postprocessor_hooks: A list of functions that get called on postprocessing
377 progress, with a dictionary with the entries
378 * status: One of "started", "processing", or "finished".
379 Check this first and ignore unknown values.
380 * postprocessor: Name of the postprocessor
381 * info_dict: The extracted info_dict
382
383 Progress hooks are guaranteed to be called at least twice
384 (with status "started" and "finished") if the processing is successful.
45598f15 385 merge_output_format: Extension to use when merging formats.
6b591b29 386 final_ext: Expected final extension; used to detect when the file was
59a7a13e 387 already downloaded and converted
6271f1ca
PH
388 fixup: Automatically correct known faults of the file.
389 One of:
390 - "never": do nothing
391 - "warn": only emit a warning
392 - "detect_or_warn": check whether we can do anything
62cd676c 393 about it, warn otherwise (default)
504f20dd 394 source_address: Client-side IP address to bind to.
6ec6cb4e 395 call_home: Boolean, true iff we are allowed to contact the
7a5c1cfe 396 yt-dlp servers for debugging. (BROKEN)
1cf376f5 397 sleep_interval_requests: Number of seconds to sleep between requests
398 during extraction
7aa589a5
S
399 sleep_interval: Number of seconds to sleep before each download when
400 used alone or a lower bound of a range for randomized
401 sleep before each download (minimum possible number
402 of seconds to sleep) when used along with
403 max_sleep_interval.
404 max_sleep_interval:Upper bound of a range for randomized sleep before each
405 download (maximum possible number of seconds to sleep).
406 Must only be used along with sleep_interval.
407 Actual sleep time will be a random float from range
408 [sleep_interval; max_sleep_interval].
1cf376f5 409 sleep_interval_subtitles: Number of seconds to sleep before each subtitle download
cfb56d1a
PH
410 listformats: Print an overview of available video formats and exit.
411 list_thumbnails: Print a table of all thumbnails and exit.
0a41f331 412 match_filter: A function that gets called for every video with the signature
413 (info_dict, *, incomplete: bool) -> Optional[str]
414 For backward compatibility with youtube-dl, the signature
415 (info_dict) -> Optional[str] is also allowed.
416 - If it returns a message, the video is ignored.
417 - If it returns None, the video is downloaded.
418 - If it returns utils.NO_DEFAULT, the user is interactively
419 asked whether to download the video.
347de493 420 match_filter_func in utils.py is one example for this.
7e5db8c9 421 no_color: Do not emit color codes in output.
0a840f58 422 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
504f20dd 423 HTTP header
0a840f58 424 geo_bypass_country:
773f291d
S
425 Two-letter ISO 3166-2 country code that will be used for
426 explicit geographic restriction bypassing via faking
504f20dd 427 X-Forwarded-For HTTP header
5f95927a
S
428 geo_bypass_ip_block:
429 IP range in CIDR notation that will be used similarly to
504f20dd 430 geo_bypass_country
71b640cc 431
85729c51 432 The following options determine which downloader is picked:
52a8a1e1 433 external_downloader: A dictionary of protocol keys and the executable of the
434 external downloader to use for it. The allowed protocols
435 are default|http|ftp|m3u8|dash|rtsp|rtmp|mms.
436 Set the value to 'native' to use the native downloader
437 hls_prefer_native: Deprecated - Use external_downloader = {'m3u8': 'native'}
438 or {'m3u8': 'ffmpeg'} instead.
439 Use the native HLS downloader instead of ffmpeg/avconv
bf09af3a
S
440 if True, otherwise use ffmpeg/avconv if False, otherwise
441 use downloader suggested by extractor if None.
53ed7066 442 compat_opts: Compatibility options. See "Differences in default behavior".
3acf6d38 443 The following options do not work when used through the API:
b5ae35ee 444 filename, abort-on-error, multistreams, no-live-chat, format-sort
dac5df5a 445 no-clean-infojson, no-playlist-metafiles, no-keep-subs, no-attach-info-json.
e4f02757 446 Refer __init__.py for their implementation
819e0531 447 progress_template: Dictionary of templates for progress outputs.
448 Allowed keys are 'download', 'postprocess',
449 'download-title' (console title) and 'postprocess-title'.
450 The template is mapped on a dictionary with keys 'progress' and 'info'
fe7e0c98 451
8222d8de 452 The following parameters are not used by YoutubeDL itself, they are used by
7a5c1cfe 453 the downloader (see yt_dlp/downloader/common.py):
51d9739f 454 nopart, updatetime, buffersize, ratelimit, throttledratelimit, min_filesize,
205a0654
EH
455 max_filesize, test, noresizebuffer, retries, file_access_retries, fragment_retries,
456 continuedl, noprogress, xattr_set_filesize, hls_use_mpegts, http_chunk_size,
59a7a13e 457 external_downloader_args, concurrent_fragment_downloads.
76b1bd67
JMF
458
459 The following options are used by the post processors:
d4a24f40 460 prefer_ffmpeg: If False, use avconv instead of ffmpeg if both are available,
e4172ac9 461 otherwise prefer ffmpeg. (avconv support is deprecated)
c0b7d117
S
462 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
463 to the binary or its containing directory.
43820c03 464 postprocessor_args: A dictionary of postprocessor/executable keys (in lower case)
34488702 465 and a list of additional command-line arguments for the
466 postprocessor/executable. The dict can also have "PP+EXE" keys
467 which are used when the given exe is used by the given PP.
468 Use 'default' as the name for arguments to passed to all PP
469 For compatibility with youtube-dl, a single list of args
470 can also be used
e409895f 471
472 The following options are used by the extractors:
62bff2c1 473 extractor_retries: Number of times to retry for known errors
474 dynamic_mpd: Whether to process dynamic DASH manifests (default: True)
e409895f 475 hls_split_discontinuity: Split HLS playlists to different formats at
62bff2c1 476 discontinuities such as ad breaks (default: False)
5d3a0e79 477 extractor_args: A dictionary of arguments to be passed to the extractors.
478 See "EXTRACTOR ARGUMENTS" for details.
479 Eg: {'youtube': {'skip': ['dash', 'hls']}}
88f23a18 480 mark_watched: Mark videos watched (even with --simulate). Only for YouTube
5d3a0e79 481 youtube_include_dash_manifest: Deprecated - Use extractor_args instead.
482 If True (default), DASH manifests and related
62bff2c1 483 data will be downloaded and processed by extractor.
484 You can reduce network I/O by disabling it if you don't
485 care about DASH. (only for youtube)
5d3a0e79 486 youtube_include_hls_manifest: Deprecated - Use extractor_args instead.
487 If True (default), HLS manifests and related
62bff2c1 488 data will be downloaded and processed by extractor.
489 You can reduce network I/O by disabling it if you don't
490 care about HLS. (only for youtube)
8222d8de
JMF
491 """
492
86e5f3ed 493 _NUMERIC_FIELDS = {
c9969434 494 'width', 'height', 'tbr', 'abr', 'asr', 'vbr', 'fps', 'filesize', 'filesize_approx',
e6f21b3d 495 'timestamp', 'release_timestamp',
c9969434
S
496 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
497 'average_rating', 'comment_count', 'age_limit',
498 'start_time', 'end_time',
499 'chapter_number', 'season_number', 'episode_number',
500 'track_number', 'disc_number', 'release_year',
86e5f3ed 501 }
c9969434 502
6db9c4d5 503 _format_fields = {
504 # NB: Keep in sync with the docstring of extractor/common.py
a44ca5a4 505 'url', 'manifest_url', 'manifest_stream_number', 'ext', 'format', 'format_id', 'format_note',
6db9c4d5 506 'width', 'height', 'resolution', 'dynamic_range', 'tbr', 'abr', 'acodec', 'asr',
507 'vbr', 'fps', 'vcodec', 'container', 'filesize', 'filesize_approx',
508 'player_url', 'protocol', 'fragment_base_url', 'fragments', 'is_from_start',
509 'preference', 'language', 'language_preference', 'quality', 'source_preference',
510 'http_headers', 'stretched_ratio', 'no_resume', 'has_drm', 'downloader_options',
511 'page_url', 'app', 'play_path', 'tc_url', 'flash_version', 'rtmp_live', 'rtmp_conn', 'rtmp_protocol', 'rtmp_real_time'
512 }
48ee10ee 513 _format_selection_exts = {
514 'audio': {'m4a', 'mp3', 'ogg', 'aac'},
515 'video': {'mp4', 'flv', 'webm', '3gp'},
516 'storyboards': {'mhtml'},
517 }
518
3511266b 519 def __init__(self, params=None, auto_init=True):
883d4b1e 520 """Create a FileDownloader object with the given options.
521 @param auto_init Whether to load the default extractors and print header (if verbose).
49a57e70 522 Set to 'no_verbose_header' to not print the header
883d4b1e 523 """
e9f9a10f
JMF
524 if params is None:
525 params = {}
592b7485 526 self.params = params
8b7491c8 527 self._ies = {}
56c73665 528 self._ies_instances = {}
1e43a6f7 529 self._pps = {k: [] for k in POSTPROCESS_WHEN}
b35496d8 530 self._printed_messages = set()
1cf376f5 531 self._first_webpage_request = True
ab8e5e51 532 self._post_hooks = []
933605d7 533 self._progress_hooks = []
819e0531 534 self._postprocessor_hooks = []
8222d8de
JMF
535 self._download_retcode = 0
536 self._num_downloads = 0
9c906919 537 self._num_videos = 0
592b7485 538 self._playlist_level = 0
539 self._playlist_urls = set()
a0e07d31 540 self.cache = Cache(self)
34308b30 541
819e0531 542 windows_enable_vt_mode()
cf4f42cb 543 self._out_files = {
544 'error': sys.stderr,
545 'print': sys.stderr if self.params.get('logtostderr') else sys.stdout,
546 'console': None if compat_os_name == 'nt' else next(
547 filter(supports_terminal_sequences, (sys.stderr, sys.stdout)), None)
548 }
549 self._out_files['screen'] = sys.stderr if self.params.get('quiet') else self._out_files['print']
ec11a9f4 550 self._allow_colors = {
cf4f42cb 551 type_: not self.params.get('no_color') and supports_terminal_sequences(self._out_files[type_])
552 for type_ in ('screen', 'error')
ec11a9f4 553 }
819e0531 554
a61f4b28 555 if sys.version_info < (3, 6):
556 self.report_warning(
0181adef 557 'Python version %d.%d is not supported! Please update to Python 3.6 or above' % sys.version_info[:2])
a61f4b28 558
88acdbc2 559 if self.params.get('allow_unplayable_formats'):
560 self.report_warning(
ec11a9f4 561 f'You have asked for {self._format_err("UNPLAYABLE", self.Styles.EMPHASIS)} formats to be listed/downloaded. '
819e0531 562 'This is a developer option intended for debugging. \n'
563 ' If you experience any issues while using this option, '
ec11a9f4 564 f'{self._format_err("DO NOT", self.Styles.ERROR)} open a bug report')
88acdbc2 565
be5df5ee
S
566 def check_deprecated(param, option, suggestion):
567 if self.params.get(param) is not None:
86e5f3ed 568 self.report_warning(f'{option} is deprecated. Use {suggestion} instead')
be5df5ee
S
569 return True
570 return False
571
572 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
38cce791
YCH
573 if self.params.get('geo_verification_proxy') is None:
574 self.params['geo_verification_proxy'] = self.params['cn_verification_proxy']
575
0d1bb027 576 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
577 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
53ed7066 578 check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"')
0d1bb027 579
49a57e70 580 for msg in self.params.get('_warnings', []):
0d1bb027 581 self.report_warning(msg)
ee8dd27a 582 for msg in self.params.get('_deprecation_warnings', []):
583 self.deprecation_warning(msg)
0d1bb027 584
ec11a9f4 585 if 'list-formats' in self.params.get('compat_opts', []):
586 self.params['listformats_table'] = False
587
b5ae35ee 588 if 'overwrites' not in self.params and self.params.get('nooverwrites') is not None:
b868936c 589 # nooverwrites was unnecessarily changed to overwrites
590 # in 0c3d0f51778b153f65c21906031c2e091fcfb641
591 # This ensures compatibility with both keys
592 self.params['overwrites'] = not self.params['nooverwrites']
b5ae35ee 593 elif self.params.get('overwrites') is None:
594 self.params.pop('overwrites', None)
b868936c 595 else:
596 self.params['nooverwrites'] = not self.params['overwrites']
b9d973be 597
455a15e2 598 self.params.setdefault('forceprint', {})
599 self.params.setdefault('print_to_file', {})
bb66c247 600
601 # Compatibility with older syntax
ca30f449 602 if not isinstance(params['forceprint'], dict):
455a15e2 603 self.params['forceprint'] = {'video': params['forceprint']}
ca30f449 604
455a15e2 605 if self.params.get('bidi_workaround', False):
1c088fa8
PH
606 try:
607 import pty
608 master, slave = pty.openpty()
003c69a8 609 width = compat_get_terminal_size().columns
1c088fa8
PH
610 if width is None:
611 width_args = []
612 else:
613 width_args = ['-w', str(width)]
5d681e96 614 sp_kwargs = dict(
1c088fa8
PH
615 stdin=subprocess.PIPE,
616 stdout=slave,
cf4f42cb 617 stderr=self._out_files['error'])
5d681e96 618 try:
d3c93ec2 619 self._output_process = Popen(['bidiv'] + width_args, **sp_kwargs)
5d681e96 620 except OSError:
d3c93ec2 621 self._output_process = Popen(['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
5d681e96 622 self._output_channel = os.fdopen(master, 'rb')
1c088fa8 623 except OSError as ose:
66e7ace1 624 if ose.errno == errno.ENOENT:
49a57e70 625 self.report_warning(
626 'Could not find fribidi executable, ignoring --bidi-workaround. '
627 'Make sure that fribidi is an executable file in one of the directories in your $PATH.')
1c088fa8
PH
628 else:
629 raise
0783b09b 630
97ec5bc5 631 if auto_init:
632 if auto_init != 'no_verbose_header':
633 self.print_debug_header()
634 self.add_default_info_extractors()
635
3089bc74
S
636 if (sys.platform != 'win32'
637 and sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
455a15e2 638 and not self.params.get('restrictfilenames', False)):
e9137224 639 # Unicode filesystem API will throw errors (#1474, #13027)
34308b30 640 self.report_warning(
6febd1c1 641 'Assuming --restrict-filenames since file system encoding '
1b725173 642 'cannot encode all characters. '
6febd1c1 643 'Set the LC_ALL environment variable to fix this.')
4a98cdbf 644 self.params['restrictfilenames'] = True
34308b30 645
de6000d9 646 self.outtmpl_dict = self.parse_outtmpl()
486dd09e 647
187986a8 648 # Creating format selector here allows us to catch syntax errors before the extraction
649 self.format_selector = (
fa9f30b8 650 self.params.get('format') if self.params.get('format') in (None, '-')
093a1710 651 else self.params['format'] if callable(self.params['format'])
187986a8 652 else self.build_format_selector(self.params['format']))
653
8b7539d2 654 # Set http_headers defaults according to std_headers
655 self.params['http_headers'] = merge_headers(std_headers, self.params.get('http_headers', {}))
656
013b50b7 657 hooks = {
658 'post_hooks': self.add_post_hook,
659 'progress_hooks': self.add_progress_hook,
660 'postprocessor_hooks': self.add_postprocessor_hook,
661 }
662 for opt, fn in hooks.items():
663 for ph in self.params.get(opt, []):
664 fn(ph)
71b640cc 665
5bfc8bee 666 for pp_def_raw in self.params.get('postprocessors', []):
667 pp_def = dict(pp_def_raw)
668 when = pp_def.pop('when', 'post_process')
669 self.add_post_processor(
f9934b96 670 get_postprocessor(pp_def.pop('key'))(self, **pp_def),
5bfc8bee 671 when=when)
672
97ec5bc5 673 self._setup_opener()
51fb4995
YCH
674 register_socks_protocols()
675
ed39cac5 676 def preload_download_archive(fn):
677 """Preload the archive, if any is specified"""
678 if fn is None:
679 return False
49a57e70 680 self.write_debug(f'Loading archive file {fn!r}')
ed39cac5 681 try:
682 with locked_file(fn, 'r', encoding='utf-8') as archive_file:
683 for line in archive_file:
684 self.archive.add(line.strip())
86e5f3ed 685 except OSError as ioe:
ed39cac5 686 if ioe.errno != errno.ENOENT:
687 raise
688 return False
689 return True
690
691 self.archive = set()
692 preload_download_archive(self.params.get('download_archive'))
693
7d4111ed
PH
694 def warn_if_short_id(self, argv):
695 # short YouTube ID starting with dash?
696 idxs = [
697 i for i, a in enumerate(argv)
698 if re.match(r'^-[0-9A-Za-z_-]{10}$', a)]
699 if idxs:
700 correct_argv = (
7a5c1cfe 701 ['yt-dlp']
3089bc74
S
702 + [a for i, a in enumerate(argv) if i not in idxs]
703 + ['--'] + [argv[i] for i in idxs]
7d4111ed
PH
704 )
705 self.report_warning(
706 'Long argument string detected. '
49a57e70 707 'Use -- to separate parameters and URLs, like this:\n%s' %
7d4111ed
PH
708 args_to_str(correct_argv))
709
8222d8de
JMF
710 def add_info_extractor(self, ie):
711 """Add an InfoExtractor object to the end of the list."""
8b7491c8 712 ie_key = ie.ie_key()
713 self._ies[ie_key] = ie
e52d7f85 714 if not isinstance(ie, type):
8b7491c8 715 self._ies_instances[ie_key] = ie
e52d7f85 716 ie.set_downloader(self)
8222d8de 717
8b7491c8 718 def _get_info_extractor_class(self, ie_key):
719 ie = self._ies.get(ie_key)
720 if ie is None:
721 ie = get_info_extractor(ie_key)
722 self.add_info_extractor(ie)
723 return ie
724
56c73665
JMF
725 def get_info_extractor(self, ie_key):
726 """
727 Get an instance of an IE with name ie_key, it will try to get one from
728 the _ies list, if there's no instance it will create a new one and add
729 it to the extractor list.
730 """
731 ie = self._ies_instances.get(ie_key)
732 if ie is None:
733 ie = get_info_extractor(ie_key)()
734 self.add_info_extractor(ie)
735 return ie
736
023fa8c4
JMF
737 def add_default_info_extractors(self):
738 """
739 Add the InfoExtractors returned by gen_extractors to the end of the list
740 """
e52d7f85 741 for ie in gen_extractor_classes():
023fa8c4
JMF
742 self.add_info_extractor(ie)
743
56d868db 744 def add_post_processor(self, pp, when='post_process'):
8222d8de 745 """Add a PostProcessor object to the end of the chain."""
5bfa4862 746 self._pps[when].append(pp)
8222d8de
JMF
747 pp.set_downloader(self)
748
ab8e5e51
AM
749 def add_post_hook(self, ph):
750 """Add the post hook"""
751 self._post_hooks.append(ph)
752
933605d7 753 def add_progress_hook(self, ph):
819e0531 754 """Add the download progress hook"""
933605d7 755 self._progress_hooks.append(ph)
8ab470f1 756
819e0531 757 def add_postprocessor_hook(self, ph):
758 """Add the postprocessing progress hook"""
759 self._postprocessor_hooks.append(ph)
5bfc8bee 760 for pps in self._pps.values():
761 for pp in pps:
762 pp.add_progress_hook(ph)
819e0531 763
1c088fa8 764 def _bidi_workaround(self, message):
5d681e96 765 if not hasattr(self, '_output_channel'):
1c088fa8
PH
766 return message
767
5d681e96 768 assert hasattr(self, '_output_process')
11b85ce6 769 assert isinstance(message, compat_str)
6febd1c1
PH
770 line_count = message.count('\n') + 1
771 self._output_process.stdin.write((message + '\n').encode('utf-8'))
5d681e96 772 self._output_process.stdin.flush()
6febd1c1 773 res = ''.join(self._output_channel.readline().decode('utf-8')
9e1a5b84 774 for _ in range(line_count))
6febd1c1 775 return res[:-len('\n')]
1c088fa8 776
b35496d8 777 def _write_string(self, message, out=None, only_once=False):
778 if only_once:
779 if message in self._printed_messages:
780 return
781 self._printed_messages.add(message)
782 write_string(message, out=out, encoding=self.params.get('encoding'))
734f90bb 783
cf4f42cb 784 def to_stdout(self, message, skip_eol=False, quiet=None):
0760b0a7 785 """Print message to stdout"""
cf4f42cb 786 if quiet is not None:
ae6a1b95 787 self.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument quiet. Use "YoutubeDL.to_screen" instead')
cf4f42cb 788 self._write_string(
789 '%s%s' % (self._bidi_workaround(message), ('' if skip_eol else '\n')),
790 self._out_files['print'])
791
792 def to_screen(self, message, skip_eol=False, quiet=None):
793 """Print message to screen if not in quiet mode"""
8bf9319e 794 if self.params.get('logger'):
43afe285 795 self.params['logger'].debug(message)
cf4f42cb 796 return
797 if (self.params.get('quiet') if quiet is None else quiet) and not self.params.get('verbose'):
798 return
799 self._write_string(
800 '%s%s' % (self._bidi_workaround(message), ('' if skip_eol else '\n')),
801 self._out_files['screen'])
8222d8de 802
b35496d8 803 def to_stderr(self, message, only_once=False):
0760b0a7 804 """Print message to stderr"""
11b85ce6 805 assert isinstance(message, compat_str)
8bf9319e 806 if self.params.get('logger'):
43afe285
IB
807 self.params['logger'].error(message)
808 else:
cf4f42cb 809 self._write_string('%s\n' % self._bidi_workaround(message), self._out_files['error'], only_once=only_once)
810
811 def _send_console_code(self, code):
812 if compat_os_name == 'nt' or not self._out_files['console']:
813 return
814 self._write_string(code, self._out_files['console'])
8222d8de 815
1e5b9a95
PH
816 def to_console_title(self, message):
817 if not self.params.get('consoletitle', False):
818 return
3efb96a6 819 message = remove_terminal_sequences(message)
4bede0d8
C
820 if compat_os_name == 'nt':
821 if ctypes.windll.kernel32.GetConsoleWindow():
822 # c_wchar_p() might not be necessary if `message` is
823 # already of type unicode()
824 ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
cf4f42cb 825 else:
826 self._send_console_code(f'\033]0;{message}\007')
1e5b9a95 827
bdde425c 828 def save_console_title(self):
cf4f42cb 829 if not self.params.get('consoletitle') or self.params.get('simulate'):
bdde425c 830 return
592b7485 831 self._send_console_code('\033[22;0t') # Save the title on stack
bdde425c
PH
832
833 def restore_console_title(self):
cf4f42cb 834 if not self.params.get('consoletitle') or self.params.get('simulate'):
bdde425c 835 return
592b7485 836 self._send_console_code('\033[23;0t') # Restore the title from stack
bdde425c
PH
837
838 def __enter__(self):
839 self.save_console_title()
840 return self
841
842 def __exit__(self, *args):
843 self.restore_console_title()
f89197d7 844
dca08720 845 if self.params.get('cookiefile') is not None:
1bab3437 846 self.cookiejar.save(ignore_discard=True, ignore_expires=True)
bdde425c 847
fa9f30b8 848 def trouble(self, message=None, tb=None, is_error=True):
8222d8de
JMF
849 """Determine action to take when a download problem appears.
850
851 Depending on if the downloader has been configured to ignore
852 download errors or not, this method may throw an exception or
853 not when errors are found, after printing the message.
854
fa9f30b8 855 @param tb If given, is additional traceback information
856 @param is_error Whether to raise error according to ignorerrors
8222d8de
JMF
857 """
858 if message is not None:
859 self.to_stderr(message)
860 if self.params.get('verbose'):
861 if tb is None:
862 if sys.exc_info()[0]: # if .trouble has been called from an except block
6febd1c1 863 tb = ''
8222d8de 864 if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
6febd1c1 865 tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
c0384f22 866 tb += encode_compat_str(traceback.format_exc())
8222d8de
JMF
867 else:
868 tb_data = traceback.format_list(traceback.extract_stack())
6febd1c1 869 tb = ''.join(tb_data)
c19bc311 870 if tb:
871 self.to_stderr(tb)
fa9f30b8 872 if not is_error:
873 return
b1940459 874 if not self.params.get('ignoreerrors'):
8222d8de
JMF
875 if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
876 exc_info = sys.exc_info()[1].exc_info
877 else:
878 exc_info = sys.exc_info()
879 raise DownloadError(message, exc_info)
880 self._download_retcode = 1
881
19a03940 882 Styles = Namespace(
883 HEADERS='yellow',
884 EMPHASIS='light blue',
492272fe 885 FILENAME='green',
19a03940 886 ID='green',
887 DELIM='blue',
888 ERROR='red',
889 WARNING='yellow',
890 SUPPRESS='light black',
891 )
ec11a9f4 892
7578d77d 893 def _format_text(self, handle, allow_colors, text, f, fallback=None, *, test_encoding=False):
e5a998f3 894 text = str(text)
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
7578d77d 902 return format_text(text, f) if allow_colors else text if fallback is None else fallback
ec11a9f4 903
904 def _format_screen(self, *args, **kwargs):
7578d77d 905 return self._format_text(
cf4f42cb 906 self._out_files['screen'], self._allow_colors['screen'], *args, **kwargs)
ec11a9f4 907
908 def _format_err(self, *args, **kwargs):
7578d77d 909 return self._format_text(
cf4f42cb 910 self._out_files['error'], self._allow_colors['error'], *args, **kwargs)
819e0531 911
c84aeac6 912 def report_warning(self, message, only_once=False):
8222d8de
JMF
913 '''
914 Print the message to stderr, it will be prefixed with 'WARNING:'
915 If stderr is a tty file the 'WARNING:' will be colored
916 '''
6d07ce01
JMF
917 if self.params.get('logger') is not None:
918 self.params['logger'].warning(message)
8222d8de 919 else:
ad8915b7
PH
920 if self.params.get('no_warnings'):
921 return
ec11a9f4 922 self.to_stderr(f'{self._format_err("WARNING:", self.Styles.WARNING)} {message}', only_once)
8222d8de 923
ee8dd27a 924 def deprecation_warning(self, message):
925 if self.params.get('logger') is not None:
a44ca5a4 926 self.params['logger'].warning(f'DeprecationWarning: {message}')
ee8dd27a 927 else:
928 self.to_stderr(f'{self._format_err("DeprecationWarning:", self.Styles.ERROR)} {message}', True)
929
fa9f30b8 930 def report_error(self, message, *args, **kwargs):
8222d8de
JMF
931 '''
932 Do the same as trouble, but prefixes the message with 'ERROR:', colored
933 in red if stderr is a tty file.
934 '''
fa9f30b8 935 self.trouble(f'{self._format_err("ERROR:", self.Styles.ERROR)} {message}', *args, **kwargs)
8222d8de 936
b35496d8 937 def write_debug(self, message, only_once=False):
0760b0a7 938 '''Log debug message or Print message to stderr'''
939 if not self.params.get('verbose', False):
940 return
941 message = '[debug] %s' % message
942 if self.params.get('logger'):
943 self.params['logger'].debug(message)
944 else:
b35496d8 945 self.to_stderr(message, only_once)
0760b0a7 946
8222d8de
JMF
947 def report_file_already_downloaded(self, file_name):
948 """Report file has already been fully downloaded."""
949 try:
6febd1c1 950 self.to_screen('[download] %s has already been downloaded' % file_name)
ce02ed60 951 except UnicodeEncodeError:
6febd1c1 952 self.to_screen('[download] The file has already been downloaded')
8222d8de 953
0c3d0f51 954 def report_file_delete(self, file_name):
955 """Report that existing file will be deleted."""
956 try:
c25228e5 957 self.to_screen('Deleting existing file %s' % file_name)
0c3d0f51 958 except UnicodeEncodeError:
c25228e5 959 self.to_screen('Deleting existing file')
0c3d0f51 960
319b6059 961 def raise_no_formats(self, info, forced=False, *, msg=None):
0a5a191a 962 has_drm = info.get('_has_drm')
319b6059 963 ignored, expected = self.params.get('ignore_no_formats_error'), bool(msg)
964 msg = msg or has_drm and 'This video is DRM protected' or 'No video formats found!'
965 if forced or not ignored:
1151c407 966 raise ExtractorError(msg, video_id=info['id'], ie=info['extractor'],
319b6059 967 expected=has_drm or ignored or expected)
88acdbc2 968 else:
969 self.report_warning(msg)
970
de6000d9 971 def parse_outtmpl(self):
972 outtmpl_dict = self.params.get('outtmpl', {})
973 if not isinstance(outtmpl_dict, dict):
974 outtmpl_dict = {'default': outtmpl_dict}
71ce444a 975 # Remove spaces in the default template
976 if self.params.get('restrictfilenames'):
977 sanitize = lambda x: x.replace(' - ', ' ').replace(' ', '-')
978 else:
979 sanitize = lambda x: x
de6000d9 980 outtmpl_dict.update({
71ce444a 981 k: sanitize(v) for k, v in DEFAULT_OUTTMPL.items()
80c03fa9 982 if outtmpl_dict.get(k) is None})
86e5f3ed 983 for _, val in outtmpl_dict.items():
de6000d9 984 if isinstance(val, bytes):
86e5f3ed 985 self.report_warning('Parameter outtmpl is bytes, but should be a unicode string')
de6000d9 986 return outtmpl_dict
987
21cd8fae 988 def get_output_path(self, dir_type='', filename=None):
989 paths = self.params.get('paths', {})
990 assert isinstance(paths, dict)
991 path = os.path.join(
992 expand_path(paths.get('home', '').strip()),
993 expand_path(paths.get(dir_type, '').strip()) if dir_type else '',
994 filename or '')
21cd8fae 995 return sanitize_path(path, force=self.params.get('windowsfilenames'))
996
76a264ac 997 @staticmethod
901130bb 998 def _outtmpl_expandpath(outtmpl):
999 # expand_path translates '%%' into '%' and '$$' into '$'
1000 # correspondingly that is not what we want since we need to keep
1001 # '%%' intact for template dict substitution step. Working around
1002 # with boundary-alike separator hack.
1003 sep = ''.join([random.choice(ascii_letters) for _ in range(32)])
86e5f3ed 1004 outtmpl = outtmpl.replace('%%', f'%{sep}%').replace('$$', f'${sep}$')
901130bb 1005
1006 # outtmpl should be expand_path'ed before template dict substitution
1007 # because meta fields may contain env variables we don't want to
1008 # be expanded. For example, for outtmpl "%(title)s.%(ext)s" and
1009 # title "Hello $PATH", we don't want `$PATH` to be expanded.
1010 return expand_path(outtmpl).replace(sep, '')
1011
1012 @staticmethod
1013 def escape_outtmpl(outtmpl):
1014 ''' Escape any remaining strings like %s, %abc% etc. '''
1015 return re.sub(
1016 STR_FORMAT_RE_TMPL.format('', '(?![%(\0])'),
1017 lambda mobj: ('' if mobj.group('has_key') else '%') + mobj.group(0),
1018 outtmpl)
1019
1020 @classmethod
1021 def validate_outtmpl(cls, outtmpl):
76a264ac 1022 ''' @return None or Exception object '''
7d1eb38a 1023 outtmpl = re.sub(
37893bb0 1024 STR_FORMAT_RE_TMPL.format('[^)]*', '[ljqBUDS]'),
7d1eb38a 1025 lambda mobj: f'{mobj.group(0)[:-1]}s',
1026 cls._outtmpl_expandpath(outtmpl))
76a264ac 1027 try:
7d1eb38a 1028 cls.escape_outtmpl(outtmpl) % collections.defaultdict(int)
76a264ac 1029 return None
1030 except ValueError as err:
1031 return err
1032
03b4de72 1033 @staticmethod
1034 def _copy_infodict(info_dict):
1035 info_dict = dict(info_dict)
09b49e1f 1036 info_dict.pop('__postprocessors', None)
03b4de72 1037 return info_dict
1038
e0fd9573 1039 def prepare_outtmpl(self, outtmpl, info_dict, sanitize=False):
1040 """ Make the outtmpl and info_dict suitable for substitution: ydl.escape_outtmpl(outtmpl) % info_dict
1041 @param sanitize Whether to sanitize the output as a filename.
1042 For backward compatibility, a function can also be passed
1043 """
1044
6e84b215 1045 info_dict.setdefault('epoch', int(time.time())) # keep epoch consistent once set
143db31d 1046
03b4de72 1047 info_dict = self._copy_infodict(info_dict)
752cda38 1048 info_dict['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
53c18592 1049 formatSeconds(info_dict['duration'], '-' if sanitize else ':')
143db31d 1050 if info_dict.get('duration', None) is not None
1051 else None)
752cda38 1052 info_dict['autonumber'] = self.params.get('autonumber_start', 1) - 1 + self._num_downloads
9c906919 1053 info_dict['video_autonumber'] = self._num_videos
752cda38 1054 if info_dict.get('resolution') is None:
1055 info_dict['resolution'] = self.format_resolution(info_dict, default=None)
143db31d 1056
e6f21b3d 1057 # For fields playlist_index, playlist_autonumber and autonumber convert all occurrences
143db31d 1058 # of %(field)s to %(field)0Nd for backward compatibility
1059 field_size_compat_map = {
0a5a191a 1060 'playlist_index': number_of_digits(info_dict.get('__last_playlist_index') or 0),
ec11a9f4 1061 'playlist_autonumber': number_of_digits(info_dict.get('n_entries') or 0),
752cda38 1062 'autonumber': self.params.get('autonumber_size') or 5,
143db31d 1063 }
752cda38 1064
385a27fa 1065 TMPL_DICT = {}
37893bb0 1066 EXTERNAL_FORMAT_RE = re.compile(STR_FORMAT_RE_TMPL.format('[^)]*', f'[{STR_FORMAT_TYPES}ljqBUDS]'))
385a27fa 1067 MATH_FUNCTIONS = {
1068 '+': float.__add__,
1069 '-': float.__sub__,
1070 }
e625be0d 1071 # Field is of the form key1.key2...
1072 # where keys (except first) can be string, int or slice
2b8a2973 1073 FIELD_RE = r'\w*(?:\.(?:\w+|{num}|{num}?(?::{num}?){{1,2}}))*'.format(num=r'(?:-?\d+)')
e0fd9573 1074 MATH_FIELD_RE = r'''(?:{field}|{num})'''.format(field=FIELD_RE, num=r'-?\d+(?:.\d+)?')
385a27fa 1075 MATH_OPERATORS_RE = r'(?:%s)' % '|'.join(map(re.escape, MATH_FUNCTIONS.keys()))
e625be0d 1076 INTERNAL_FORMAT_RE = re.compile(r'''(?x)
1077 (?P<negate>-)?
385a27fa 1078 (?P<fields>{field})
1079 (?P<maths>(?:{math_op}{math_field})*)
e625be0d 1080 (?:>(?P<strf_format>.+?))?
34baa9fd 1081 (?P<remaining>
1082 (?P<alternate>(?<!\\),[^|&)]+)?
1083 (?:&(?P<replacement>.*?))?
1084 (?:\|(?P<default>.*?))?
1085 )$'''.format(field=FIELD_RE, math_op=MATH_OPERATORS_RE, math_field=MATH_FIELD_RE))
752cda38 1086
2b8a2973 1087 def _traverse_infodict(k):
1088 k = k.split('.')
1089 if k[0] == '':
1090 k.pop(0)
1091 return traverse_obj(info_dict, k, is_user_input=True, traverse_string=True)
76a264ac 1092
752cda38 1093 def get_value(mdict):
1094 # Object traversal
2b8a2973 1095 value = _traverse_infodict(mdict['fields'])
752cda38 1096 # Negative
1097 if mdict['negate']:
1098 value = float_or_none(value)
1099 if value is not None:
1100 value *= -1
1101 # Do maths
385a27fa 1102 offset_key = mdict['maths']
1103 if offset_key:
752cda38 1104 value = float_or_none(value)
1105 operator = None
385a27fa 1106 while offset_key:
1107 item = re.match(
1108 MATH_FIELD_RE if operator else MATH_OPERATORS_RE,
1109 offset_key).group(0)
1110 offset_key = offset_key[len(item):]
1111 if operator is None:
752cda38 1112 operator = MATH_FUNCTIONS[item]
385a27fa 1113 continue
1114 item, multiplier = (item[1:], -1) if item[0] == '-' else (item, 1)
1115 offset = float_or_none(item)
1116 if offset is None:
2b8a2973 1117 offset = float_or_none(_traverse_infodict(item))
385a27fa 1118 try:
1119 value = operator(value, multiplier * offset)
1120 except (TypeError, ZeroDivisionError):
1121 return None
1122 operator = None
752cda38 1123 # Datetime formatting
1124 if mdict['strf_format']:
7c37ff97 1125 value = strftime_or_none(value, mdict['strf_format'].replace('\\,', ','))
752cda38 1126
1127 return value
1128
b868936c 1129 na = self.params.get('outtmpl_na_placeholder', 'NA')
1130
e0fd9573 1131 def filename_sanitizer(key, value, restricted=self.params.get('restrictfilenames')):
5c3895ff 1132 return sanitize_filename(str(value), restricted=restricted, is_id=(
1133 bool(re.search(r'(^|[_.])id(\.|$)', key))
1134 if 'filename-sanitization' in self.params.get('compat_opts', [])
1135 else NO_DEFAULT))
e0fd9573 1136
1137 sanitizer = sanitize if callable(sanitize) else filename_sanitizer
1138 sanitize = bool(sanitize)
1139
6e84b215 1140 def _dumpjson_default(obj):
1141 if isinstance(obj, (set, LazyList)):
1142 return list(obj)
adbc4ec4 1143 return repr(obj)
6e84b215 1144
752cda38 1145 def create_key(outer_mobj):
1146 if not outer_mobj.group('has_key'):
b836dc94 1147 return outer_mobj.group(0)
752cda38 1148 key = outer_mobj.group('key')
752cda38 1149 mobj = re.match(INTERNAL_FORMAT_RE, key)
e0fd9573 1150 initial_field = mobj.group('fields') if mobj else ''
e978789f 1151 value, replacement, default = None, None, na
7c37ff97 1152 while mobj:
e625be0d 1153 mobj = mobj.groupdict()
7c37ff97 1154 default = mobj['default'] if mobj['default'] is not None else default
752cda38 1155 value = get_value(mobj)
e978789f 1156 replacement = mobj['replacement']
7c37ff97 1157 if value is None and mobj['alternate']:
34baa9fd 1158 mobj = re.match(INTERNAL_FORMAT_RE, mobj['remaining'][1:])
7c37ff97 1159 else:
1160 break
752cda38 1161
b868936c 1162 fmt = outer_mobj.group('format')
752cda38 1163 if fmt == 's' and value is not None and key in field_size_compat_map.keys():
86e5f3ed 1164 fmt = f'0{field_size_compat_map[key]:d}d'
752cda38 1165
e978789f 1166 value = default if value is None else value if replacement is None else replacement
752cda38 1167
4476d2c7 1168 flags = outer_mobj.group('conversion') or ''
7d1eb38a 1169 str_fmt = f'{fmt[:-1]}s'
524e2e4f 1170 if fmt[-1] == 'l': # list
4476d2c7 1171 delim = '\n' if '#' in flags else ', '
9e907ebd 1172 value, fmt = delim.join(map(str, variadic(value, allowed_types=(str, bytes)))), str_fmt
524e2e4f 1173 elif fmt[-1] == 'j': # json
4476d2c7 1174 value, fmt = json.dumps(value, default=_dumpjson_default, indent=4 if '#' in flags else None), str_fmt
524e2e4f 1175 elif fmt[-1] == 'q': # quoted
4476d2c7 1176 value = map(str, variadic(value) if '#' in flags else [value])
1177 value, fmt = ' '.join(map(compat_shlex_quote, value)), str_fmt
524e2e4f 1178 elif fmt[-1] == 'B': # bytes
86e5f3ed 1179 value = f'%{str_fmt}'.encode() % str(value).encode('utf-8')
f5aa5cfb 1180 value, fmt = value.decode('utf-8', 'ignore'), 's'
524e2e4f 1181 elif fmt[-1] == 'U': # unicode normalized
524e2e4f 1182 value, fmt = unicodedata.normalize(
1183 # "+" = compatibility equivalence, "#" = NFD
4476d2c7 1184 'NF%s%s' % ('K' if '+' in flags else '', 'D' if '#' in flags else 'C'),
524e2e4f 1185 value), str_fmt
e0fd9573 1186 elif fmt[-1] == 'D': # decimal suffix
abbeeebc 1187 num_fmt, fmt = fmt[:-1].replace('#', ''), 's'
1188 value = format_decimal_suffix(value, f'%{num_fmt}f%s' if num_fmt else '%d%s',
1189 factor=1024 if '#' in flags else 1000)
37893bb0 1190 elif fmt[-1] == 'S': # filename sanitization
e0fd9573 1191 value, fmt = filename_sanitizer(initial_field, value, restricted='#' in flags), str_fmt
7d1eb38a 1192 elif fmt[-1] == 'c':
524e2e4f 1193 if value:
1194 value = str(value)[0]
76a264ac 1195 else:
524e2e4f 1196 fmt = str_fmt
76a264ac 1197 elif fmt[-1] not in 'rs': # numeric
a439a3a4 1198 value = float_or_none(value)
752cda38 1199 if value is None:
1200 value, fmt = default, 's'
901130bb 1201
752cda38 1202 if sanitize:
1203 if fmt[-1] == 'r':
1204 # If value is an object, sanitize might convert it to a string
1205 # So we convert it to repr first
7d1eb38a 1206 value, fmt = repr(value), str_fmt
639f1cea 1207 if fmt[-1] in 'csr':
e0fd9573 1208 value = sanitizer(initial_field, value)
901130bb 1209
b868936c 1210 key = '%s\0%s' % (key.replace('%', '%\0'), outer_mobj.group('format'))
385a27fa 1211 TMPL_DICT[key] = value
b868936c 1212 return '{prefix}%({key}){fmt}'.format(key=key, fmt=fmt, prefix=outer_mobj.group('prefix'))
752cda38 1213
385a27fa 1214 return EXTERNAL_FORMAT_RE.sub(create_key, outtmpl), TMPL_DICT
143db31d 1215
819e0531 1216 def evaluate_outtmpl(self, outtmpl, info_dict, *args, **kwargs):
1217 outtmpl, info_dict = self.prepare_outtmpl(outtmpl, info_dict, *args, **kwargs)
1218 return self.escape_outtmpl(outtmpl) % info_dict
1219
5127e92a 1220 def _prepare_filename(self, info_dict, *, outtmpl=None, tmpl_type=None):
1221 assert None in (outtmpl, tmpl_type), 'outtmpl and tmpl_type are mutually exclusive'
1222 if outtmpl is None:
1223 outtmpl = self.outtmpl_dict.get(tmpl_type or 'default', self.outtmpl_dict['default'])
8222d8de 1224 try:
5127e92a 1225 outtmpl = self._outtmpl_expandpath(outtmpl)
e0fd9573 1226 filename = self.evaluate_outtmpl(outtmpl, info_dict, True)
6a0546e3 1227 if not filename:
1228 return None
15da37c7 1229
5127e92a 1230 if tmpl_type in ('', 'temp'):
6a0546e3 1231 final_ext, ext = self.params.get('final_ext'), info_dict.get('ext')
1232 if final_ext and ext and final_ext != ext and filename.endswith(f'.{final_ext}'):
1233 filename = replace_extension(filename, ext, final_ext)
5127e92a 1234 elif tmpl_type:
6a0546e3 1235 force_ext = OUTTMPL_TYPES[tmpl_type]
1236 if force_ext:
1237 filename = replace_extension(filename, force_ext, info_dict.get('ext'))
de6000d9 1238
bdc3fd2f
U
1239 # https://github.com/blackjack4494/youtube-dlc/issues/85
1240 trim_file_name = self.params.get('trim_file_name', False)
1241 if trim_file_name:
5c22c63d 1242 no_ext, *ext = filename.rsplit('.', 2)
1243 filename = join_nonempty(no_ext[:trim_file_name], *ext, delim='.')
bdc3fd2f 1244
0202b52a 1245 return filename
8222d8de 1246 except ValueError as err:
6febd1c1 1247 self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')')
8222d8de
JMF
1248 return None
1249
5127e92a 1250 def prepare_filename(self, info_dict, dir_type='', *, outtmpl=None, warn=False):
1251 """Generate the output filename"""
1252 if outtmpl:
1253 assert not dir_type, 'outtmpl and dir_type are mutually exclusive'
1254 dir_type = None
1255 filename = self._prepare_filename(info_dict, tmpl_type=dir_type, outtmpl=outtmpl)
80c03fa9 1256 if not filename and dir_type not in ('', 'temp'):
1257 return ''
de6000d9 1258
c84aeac6 1259 if warn:
21cd8fae 1260 if not self.params.get('paths'):
de6000d9 1261 pass
1262 elif filename == '-':
c84aeac6 1263 self.report_warning('--paths is ignored when an outputting to stdout', only_once=True)
de6000d9 1264 elif os.path.isabs(filename):
c84aeac6 1265 self.report_warning('--paths is ignored since an absolute path is given in output template', only_once=True)
de6000d9 1266 if filename == '-' or not filename:
1267 return filename
1268
21cd8fae 1269 return self.get_output_path(dir_type, filename)
0202b52a 1270
120fe513 1271 def _match_entry(self, info_dict, incomplete=False, silent=False):
ecdec191 1272 """ Returns None if the file should be downloaded """
8222d8de 1273
c77495e3 1274 video_title = info_dict.get('title', info_dict.get('id', 'video'))
1275
8b0d7497 1276 def check_filter():
8b0d7497 1277 if 'title' in info_dict:
1278 # This can happen when we're just evaluating the playlist
1279 title = info_dict['title']
1280 matchtitle = self.params.get('matchtitle', False)
1281 if matchtitle:
1282 if not re.search(matchtitle, title, re.IGNORECASE):
1283 return '"' + title + '" title did not match pattern "' + matchtitle + '"'
1284 rejecttitle = self.params.get('rejecttitle', False)
1285 if rejecttitle:
1286 if re.search(rejecttitle, title, re.IGNORECASE):
1287 return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
1288 date = info_dict.get('upload_date')
1289 if date is not None:
1290 dateRange = self.params.get('daterange', DateRange())
1291 if date not in dateRange:
86e5f3ed 1292 return f'{date_from_str(date).isoformat()} upload date is not in range {dateRange}'
8b0d7497 1293 view_count = info_dict.get('view_count')
1294 if view_count is not None:
1295 min_views = self.params.get('min_views')
1296 if min_views is not None and view_count < min_views:
1297 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
1298 max_views = self.params.get('max_views')
1299 if max_views is not None and view_count > max_views:
1300 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
1301 if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')):
1302 return 'Skipping "%s" because it is age restricted' % video_title
8b0d7497 1303
8f18aca8 1304 match_filter = self.params.get('match_filter')
1305 if match_filter is not None:
1306 try:
1307 ret = match_filter(info_dict, incomplete=incomplete)
1308 except TypeError:
1309 # For backward compatibility
1310 ret = None if incomplete else match_filter(info_dict)
492272fe 1311 if ret is NO_DEFAULT:
1312 while True:
1313 filename = self._format_screen(self.prepare_filename(info_dict), self.Styles.FILENAME)
1314 reply = input(self._format_screen(
1315 f'Download "{filename}"? (Y/n): ', self.Styles.EMPHASIS)).lower().strip()
1316 if reply in {'y', ''}:
1317 return None
1318 elif reply == 'n':
1319 return f'Skipping {video_title}'
1320 return True
1321 elif ret is not None:
8f18aca8 1322 return ret
8b0d7497 1323 return None
1324
c77495e3 1325 if self.in_download_archive(info_dict):
1326 reason = '%s has already been recorded in the archive' % video_title
1327 break_opt, break_err = 'break_on_existing', ExistingVideoReached
1328 else:
1329 reason = check_filter()
1330 break_opt, break_err = 'break_on_reject', RejectedVideoReached
8b0d7497 1331 if reason is not None:
120fe513 1332 if not silent:
1333 self.to_screen('[download] ' + reason)
c77495e3 1334 if self.params.get(break_opt, False):
1335 raise break_err()
8b0d7497 1336 return reason
fe7e0c98 1337
b6c45014
JMF
1338 @staticmethod
1339 def add_extra_info(info_dict, extra_info):
1340 '''Set the keys from extra_info in info dict if they are missing'''
1341 for key, value in extra_info.items():
1342 info_dict.setdefault(key, value)
1343
409e1828 1344 def extract_info(self, url, download=True, ie_key=None, extra_info=None,
61aa5ba3 1345 process=True, force_generic_extractor=False):
41d1cca3 1346 """
1347 Return a list with a dictionary for each video extracted.
1348
1349 Arguments:
1350 url -- URL to extract
1351
1352 Keyword arguments:
1353 download -- whether to download videos during extraction
1354 ie_key -- extractor key hint
1355 extra_info -- dictionary containing the extra values to add to each result
1356 process -- whether to resolve all unresolved references (URLs, playlist items),
1357 must be True for download to work.
1358 force_generic_extractor -- force using the generic extractor
1359 """
fe7e0c98 1360
409e1828 1361 if extra_info is None:
1362 extra_info = {}
1363
61aa5ba3 1364 if not ie_key and force_generic_extractor:
d22dec74
S
1365 ie_key = 'Generic'
1366
8222d8de 1367 if ie_key:
8b7491c8 1368 ies = {ie_key: self._get_info_extractor_class(ie_key)}
8222d8de
JMF
1369 else:
1370 ies = self._ies
1371
8b7491c8 1372 for ie_key, ie in ies.items():
8222d8de
JMF
1373 if not ie.suitable(url):
1374 continue
1375
1376 if not ie.working():
6febd1c1
PH
1377 self.report_warning('The program functionality for this site has been marked as broken, '
1378 'and will probably not work.')
8222d8de 1379
1151c407 1380 temp_id = ie.get_temp_id(url)
a0566bbf 1381 if temp_id is not None and self.in_download_archive({'id': temp_id, 'ie_key': ie_key}):
5e5be0c0 1382 self.to_screen(f'[{ie_key}] {temp_id}: has already been recorded in the archive')
1383 if self.params.get('break_on_existing', False):
1384 raise ExistingVideoReached()
a0566bbf 1385 break
8b7491c8 1386 return self.__extract_info(url, self.get_info_extractor(ie_key), download, extra_info, process)
a0566bbf 1387 else:
1388 self.report_error('no suitable InfoExtractor for URL %s' % url)
1389
8e5fecc8 1390 def __handle_extraction_exceptions(func):
b5ae35ee 1391 @functools.wraps(func)
a0566bbf 1392 def wrapper(self, *args, **kwargs):
6da22e7d 1393 while True:
1394 try:
1395 return func(self, *args, **kwargs)
1396 except (DownloadCancelled, LazyList.IndexError, PagedList.IndexError):
8222d8de 1397 raise
6da22e7d 1398 except ReExtractInfo as e:
1399 if e.expected:
1400 self.to_screen(f'{e}; Re-extracting data')
1401 else:
1402 self.to_stderr('\r')
1403 self.report_warning(f'{e}; Re-extracting data')
1404 continue
1405 except GeoRestrictedError as e:
1406 msg = e.msg
1407 if e.countries:
1408 msg += '\nThis video is available in %s.' % ', '.join(
1409 map(ISO3166Utils.short2full, e.countries))
1410 msg += '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1411 self.report_error(msg)
1412 except ExtractorError as e: # An error we somewhat expected
1413 self.report_error(str(e), e.format_traceback())
1414 except Exception as e:
1415 if self.params.get('ignoreerrors'):
1416 self.report_error(str(e), tb=encode_compat_str(traceback.format_exc()))
1417 else:
1418 raise
1419 break
a0566bbf 1420 return wrapper
1421
f2ebc5c7 1422 def _wait_for_video(self, ie_result):
1423 if (not self.params.get('wait_for_video')
1424 or ie_result.get('_type', 'video') != 'video'
1425 or ie_result.get('formats') or ie_result.get('url')):
1426 return
1427
1428 format_dur = lambda dur: '%02d:%02d:%02d' % timetuple_from_msec(dur * 1000)[:-1]
1429 last_msg = ''
1430
1431 def progress(msg):
1432 nonlocal last_msg
1433 self.to_screen(msg + ' ' * (len(last_msg) - len(msg)) + '\r', skip_eol=True)
1434 last_msg = msg
1435
1436 min_wait, max_wait = self.params.get('wait_for_video')
1437 diff = try_get(ie_result, lambda x: x['release_timestamp'] - time.time())
1438 if diff is None and ie_result.get('live_status') == 'is_upcoming':
16c620bc 1439 diff = round(random.uniform(min_wait, max_wait) if (max_wait and min_wait) else (max_wait or min_wait), 0)
f2ebc5c7 1440 self.report_warning('Release time of video is not known')
1441 elif (diff or 0) <= 0:
1442 self.report_warning('Video should already be available according to extracted info')
38d79fd1 1443 diff = min(max(diff or 0, min_wait or 0), max_wait or float('inf'))
f2ebc5c7 1444 self.to_screen(f'[wait] Waiting for {format_dur(diff)} - Press Ctrl+C to try now')
1445
1446 wait_till = time.time() + diff
1447 try:
1448 while True:
1449 diff = wait_till - time.time()
1450 if diff <= 0:
1451 progress('')
1452 raise ReExtractInfo('[wait] Wait period ended', expected=True)
1453 progress(f'[wait] Remaining time until next attempt: {self._format_screen(format_dur(diff), self.Styles.EMPHASIS)}')
1454 time.sleep(1)
1455 except KeyboardInterrupt:
1456 progress('')
1457 raise ReExtractInfo('[wait] Interrupted by user', expected=True)
1458 except BaseException as e:
1459 if not isinstance(e, ReExtractInfo):
1460 self.to_screen('')
1461 raise
1462
a0566bbf 1463 @__handle_extraction_exceptions
58f197b7 1464 def __extract_info(self, url, ie, download, extra_info, process):
a0566bbf 1465 ie_result = ie.extract(url)
1466 if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
1467 return
1468 if isinstance(ie_result, list):
1469 # Backwards compatibility: old IE result format
1470 ie_result = {
1471 '_type': 'compat_list',
1472 'entries': ie_result,
1473 }
e37d0efb 1474 if extra_info.get('original_url'):
1475 ie_result.setdefault('original_url', extra_info['original_url'])
a0566bbf 1476 self.add_default_extra_info(ie_result, ie, url)
1477 if process:
f2ebc5c7 1478 self._wait_for_video(ie_result)
a0566bbf 1479 return self.process_ie_result(ie_result, download, extra_info)
8222d8de 1480 else:
a0566bbf 1481 return ie_result
fe7e0c98 1482
ea38e55f 1483 def add_default_extra_info(self, ie_result, ie, url):
6033d980 1484 if url is not None:
1485 self.add_extra_info(ie_result, {
1486 'webpage_url': url,
1487 'original_url': url,
57ebfca3 1488 })
1489 webpage_url = ie_result.get('webpage_url')
1490 if webpage_url:
1491 self.add_extra_info(ie_result, {
1492 'webpage_url_basename': url_basename(webpage_url),
1493 'webpage_url_domain': get_domain(webpage_url),
6033d980 1494 })
1495 if ie is not None:
1496 self.add_extra_info(ie_result, {
1497 'extractor': ie.IE_NAME,
1498 'extractor_key': ie.ie_key(),
1499 })
ea38e55f 1500
58adec46 1501 def process_ie_result(self, ie_result, download=True, extra_info=None):
8222d8de
JMF
1502 """
1503 Take the result of the ie(may be modified) and resolve all unresolved
1504 references (URLs, playlist items).
1505
1506 It will also download the videos if 'download'.
1507 Returns the resolved ie_result.
1508 """
58adec46 1509 if extra_info is None:
1510 extra_info = {}
e8ee972c
PH
1511 result_type = ie_result.get('_type', 'video')
1512
057a5206 1513 if result_type in ('url', 'url_transparent'):
134c6ea8 1514 ie_result['url'] = sanitize_url(ie_result['url'])
e37d0efb 1515 if ie_result.get('original_url'):
1516 extra_info.setdefault('original_url', ie_result['original_url'])
1517
057a5206 1518 extract_flat = self.params.get('extract_flat', False)
3089bc74
S
1519 if ((extract_flat == 'in_playlist' and 'playlist' in extra_info)
1520 or extract_flat is True):
ecb54191 1521 info_copy = ie_result.copy()
6033d980 1522 ie = try_get(ie_result.get('ie_key'), self.get_info_extractor)
360167b9 1523 if ie and not ie_result.get('id'):
4614bc22 1524 info_copy['id'] = ie.get_temp_id(ie_result['url'])
6033d980 1525 self.add_default_extra_info(info_copy, ie, ie_result['url'])
4614bc22 1526 self.add_extra_info(info_copy, extra_info)
b5475f11 1527 info_copy, _ = self.pre_process(info_copy)
ecb54191 1528 self.__forced_printings(info_copy, self.prepare_filename(info_copy), incomplete=True)
4614bc22 1529 if self.params.get('force_write_download_archive', False):
1530 self.record_download_archive(info_copy)
e8ee972c
PH
1531 return ie_result
1532
8222d8de 1533 if result_type == 'video':
b6c45014 1534 self.add_extra_info(ie_result, extra_info)
9c2b75b5 1535 ie_result = self.process_video_result(ie_result, download=download)
28b0eb0f 1536 additional_urls = (ie_result or {}).get('additional_urls')
9c2b75b5 1537 if additional_urls:
e9f4ccd1 1538 # TODO: Improve MetadataParserPP to allow setting a list
9c2b75b5 1539 if isinstance(additional_urls, compat_str):
1540 additional_urls = [additional_urls]
1541 self.to_screen(
1542 '[info] %s: %d additional URL(s) requested' % (ie_result['id'], len(additional_urls)))
1543 self.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls))
1544 ie_result['additional_entries'] = [
1545 self.extract_info(
b69fd25c 1546 url, download, extra_info=extra_info,
9c2b75b5 1547 force_generic_extractor=self.params.get('force_generic_extractor'))
1548 for url in additional_urls
1549 ]
1550 return ie_result
8222d8de
JMF
1551 elif result_type == 'url':
1552 # We have to add extra_info to the results because it may be
1553 # contained in a playlist
07cce701 1554 return self.extract_info(
1555 ie_result['url'], download,
1556 ie_key=ie_result.get('ie_key'),
1557 extra_info=extra_info)
7fc3fa05
PH
1558 elif result_type == 'url_transparent':
1559 # Use the information from the embedding page
1560 info = self.extract_info(
1561 ie_result['url'], ie_key=ie_result.get('ie_key'),
1562 extra_info=extra_info, download=False, process=False)
1563
1640eb09
S
1564 # extract_info may return None when ignoreerrors is enabled and
1565 # extraction failed with an error, don't crash and return early
1566 # in this case
1567 if not info:
1568 return info
1569
412c617d 1570 new_result = info.copy()
90137ca4 1571 new_result.update(filter_dict(ie_result, lambda k, v: (
1572 v is not None and k not in {'_type', 'url', 'id', 'extractor', 'extractor_key', 'ie_key'})))
7fc3fa05 1573
0563f7ac
S
1574 # Extracted info may not be a video result (i.e.
1575 # info.get('_type', 'video') != video) but rather an url or
1576 # url_transparent. In such cases outer metadata (from ie_result)
1577 # should be propagated to inner one (info). For this to happen
1578 # _type of info should be overridden with url_transparent. This
067aa17e 1579 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
0563f7ac
S
1580 if new_result.get('_type') == 'url':
1581 new_result['_type'] = 'url_transparent'
7fc3fa05
PH
1582
1583 return self.process_ie_result(
1584 new_result, download=download, extra_info=extra_info)
40fcba5e 1585 elif result_type in ('playlist', 'multi_video'):
30a074c2 1586 # Protect from infinite recursion due to recursively nested playlists
1587 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
1588 webpage_url = ie_result['webpage_url']
1589 if webpage_url in self._playlist_urls:
7e85e872 1590 self.to_screen(
30a074c2 1591 '[download] Skipping already downloaded playlist: %s'
1592 % ie_result.get('title') or ie_result.get('id'))
1593 return
7e85e872 1594
30a074c2 1595 self._playlist_level += 1
1596 self._playlist_urls.add(webpage_url)
03f83004 1597 self._fill_common_fields(ie_result, False)
bc516a3f 1598 self._sanitize_thumbnails(ie_result)
30a074c2 1599 try:
1600 return self.__process_playlist(ie_result, download)
1601 finally:
1602 self._playlist_level -= 1
1603 if not self._playlist_level:
1604 self._playlist_urls.clear()
8222d8de 1605 elif result_type == 'compat_list':
c9bf4114
PH
1606 self.report_warning(
1607 'Extractor %s returned a compat_list result. '
1608 'It needs to be updated.' % ie_result.get('extractor'))
5f6a1245 1609
8222d8de 1610 def _fixup(r):
b868936c 1611 self.add_extra_info(r, {
1612 'extractor': ie_result['extractor'],
1613 'webpage_url': ie_result['webpage_url'],
1614 'webpage_url_basename': url_basename(ie_result['webpage_url']),
0bb322b9 1615 'webpage_url_domain': get_domain(ie_result['webpage_url']),
b868936c 1616 'extractor_key': ie_result['extractor_key'],
1617 })
8222d8de
JMF
1618 return r
1619 ie_result['entries'] = [
b6c45014 1620 self.process_ie_result(_fixup(r), download, extra_info)
8222d8de
JMF
1621 for r in ie_result['entries']
1622 ]
1623 return ie_result
1624 else:
1625 raise Exception('Invalid result type: %s' % result_type)
1626
e92caff5 1627 def _ensure_dir_exists(self, path):
1628 return make_dir(path, self.report_error)
1629
3b603dbd 1630 @staticmethod
1631 def _playlist_infodict(ie_result, **kwargs):
1632 return {
1633 **ie_result,
1634 'playlist': ie_result.get('title') or ie_result.get('id'),
1635 'playlist_id': ie_result.get('id'),
1636 'playlist_title': ie_result.get('title'),
1637 'playlist_uploader': ie_result.get('uploader'),
1638 'playlist_uploader_id': ie_result.get('uploader_id'),
1639 'playlist_index': 0,
1640 **kwargs,
1641 }
1642
30a074c2 1643 def __process_playlist(self, ie_result, download):
1644 # We process each entry in the playlist
1645 playlist = ie_result.get('title') or ie_result.get('id')
1646 self.to_screen('[download] Downloading playlist: %s' % playlist)
1647
498f5606 1648 if 'entries' not in ie_result:
aa9369a2 1649 raise EntryNotInPlaylist('There are no entries')
7c7f7161 1650
1651 MissingEntry = object()
498f5606 1652 incomplete_entries = bool(ie_result.get('requested_entries'))
1653 if incomplete_entries:
bf5f605e 1654 def fill_missing_entries(entries, indices):
7c7f7161 1655 ret = [MissingEntry] * max(indices)
bf5f605e 1656 for i, entry in zip(indices, entries):
498f5606 1657 ret[i - 1] = entry
1658 return ret
1659 ie_result['entries'] = fill_missing_entries(ie_result['entries'], ie_result['requested_entries'])
02fd60d3 1660
30a074c2 1661 playlist_results = []
1662
56a8fb4f 1663 playliststart = self.params.get('playliststart', 1)
30a074c2 1664 playlistend = self.params.get('playlistend')
1665 # For backwards compatibility, interpret -1 as whole list
1666 if playlistend == -1:
1667 playlistend = None
1668
1669 playlistitems_str = self.params.get('playlist_items')
1670 playlistitems = None
1671 if playlistitems_str is not None:
1672 def iter_playlistitems(format):
1673 for string_segment in format.split(','):
1674 if '-' in string_segment:
1675 start, end = string_segment.split('-')
1676 for item in range(int(start), int(end) + 1):
1677 yield int(item)
1678 else:
1679 yield int(string_segment)
1680 playlistitems = orderedSet(iter_playlistitems(playlistitems_str))
1681
1682 ie_entries = ie_result['entries']
8e5fecc8 1683 if isinstance(ie_entries, list):
ed8d87f9 1684 playlist_count = len(ie_entries)
f0d785d3 1685 msg = f'Collected {playlist_count} videos; downloading %d of them'
1686 ie_result['playlist_count'] = ie_result.get('playlist_count') or playlist_count
1687
8e5fecc8 1688 def get_entry(i):
1689 return ie_entries[i - 1]
1690 else:
f0d785d3 1691 msg = 'Downloading %d videos'
c586f9e8 1692 if not isinstance(ie_entries, (PagedList, LazyList)):
8e5fecc8 1693 ie_entries = LazyList(ie_entries)
d37707bd 1694 elif isinstance(ie_entries, InAdvancePagedList):
1695 if ie_entries._pagesize == 1:
1696 playlist_count = ie_entries._pagecount
8e5fecc8 1697
1698 def get_entry(i):
1699 return YoutubeDL.__handle_extraction_exceptions(
1700 lambda self, i: ie_entries[i - 1]
1701 )(self, i)
50fed816 1702
f0d785d3 1703 entries, broken = [], False
ff1c7fc9 1704 items = playlistitems if playlistitems is not None else itertools.count(playliststart)
1705 for i in items:
1706 if i == 0:
1707 continue
56a8fb4f 1708 if playlistitems is None and playlistend is not None and playlistend < i:
1709 break
1710 entry = None
1711 try:
50fed816 1712 entry = get_entry(i)
7c7f7161 1713 if entry is MissingEntry:
498f5606 1714 raise EntryNotInPlaylist()
56a8fb4f 1715 except (IndexError, EntryNotInPlaylist):
1716 if incomplete_entries:
aa9369a2 1717 raise EntryNotInPlaylist(f'Entry {i} cannot be found')
56a8fb4f 1718 elif not playlistitems:
1719 break
1720 entries.append(entry)
120fe513 1721 try:
1722 if entry is not None:
e5a998f3 1723 # TODO: Add auto-generated fields
120fe513 1724 self._match_entry(entry, incomplete=True, silent=True)
1725 except (ExistingVideoReached, RejectedVideoReached):
f0d785d3 1726 broken = True
120fe513 1727 break
56a8fb4f 1728 ie_result['entries'] = entries
30a074c2 1729
56a8fb4f 1730 # Save playlist_index before re-ordering
1731 entries = [
9e598870 1732 ((playlistitems[i - 1] if playlistitems else i + playliststart - 1), entry)
56a8fb4f 1733 for i, entry in enumerate(entries, 1)
1734 if entry is not None]
1735 n_entries = len(entries)
498f5606 1736
f0d785d3 1737 if not (ie_result.get('playlist_count') or broken or playlistitems or playlistend):
1738 ie_result['playlist_count'] = n_entries
1739
e08a85d8 1740 if not playlistitems and (playliststart != 1 or playlistend):
56a8fb4f 1741 playlistitems = list(range(playliststart, playliststart + n_entries))
498f5606 1742 ie_result['requested_entries'] = playlistitems
1743
e08a85d8 1744 _infojson_written = False
0bfc53d0 1745 write_playlist_files = self.params.get('allow_playlist_files', True)
1746 if write_playlist_files and self.params.get('list_thumbnails'):
1747 self.list_thumbnails(ie_result)
1748 if write_playlist_files and not self.params.get('simulate'):
3b603dbd 1749 ie_copy = self._playlist_infodict(ie_result, n_entries=n_entries)
e08a85d8 1750 _infojson_written = self._write_info_json(
1751 'playlist', ie_result, self.prepare_filename(ie_copy, 'pl_infojson'))
1752 if _infojson_written is None:
80c03fa9 1753 return
1754 if self._write_description('playlist', ie_result,
1755 self.prepare_filename(ie_copy, 'pl_description')) is None:
1756 return
681de68e 1757 # TODO: This should be passed to ThumbnailsConvertor if necessary
80c03fa9 1758 self._write_thumbnails('playlist', ie_copy, self.prepare_filename(ie_copy, 'pl_thumbnail'))
30a074c2 1759
1760 if self.params.get('playlistreverse', False):
1761 entries = entries[::-1]
30a074c2 1762 if self.params.get('playlistrandom', False):
1763 random.shuffle(entries)
1764
1765 x_forwarded_for = ie_result.get('__x_forwarded_for_ip')
1766
86e5f3ed 1767 self.to_screen(f'[{ie_result["extractor"]}] playlist {playlist}: {msg % n_entries}')
26e2805c 1768 failures = 0
1769 max_failures = self.params.get('skip_playlist_after_errors') or float('inf')
71729754 1770 for i, entry_tuple in enumerate(entries, 1):
1771 playlist_index, entry = entry_tuple
81139999 1772 if 'playlist-index' in self.params.get('compat_opts', []):
1773 playlist_index = playlistitems[i - 1] if playlistitems else i + playliststart - 1
19a03940 1774 self.to_screen('[download] Downloading video %s of %s' % (
1775 self._format_screen(i, self.Styles.ID), self._format_screen(n_entries, self.Styles.EMPHASIS)))
30a074c2 1776 # This __x_forwarded_for_ip thing is a bit ugly but requires
1777 # minimal changes
1778 if x_forwarded_for:
1779 entry['__x_forwarded_for_ip'] = x_forwarded_for
1780 extra = {
1781 'n_entries': n_entries,
0a5a191a 1782 '__last_playlist_index': max(playlistitems) if playlistitems else (playlistend or n_entries),
f0d785d3 1783 'playlist_count': ie_result.get('playlist_count'),
71729754 1784 'playlist_index': playlist_index,
1785 'playlist_autonumber': i,
30a074c2 1786 'playlist': playlist,
1787 'playlist_id': ie_result.get('id'),
1788 'playlist_title': ie_result.get('title'),
1789 'playlist_uploader': ie_result.get('uploader'),
1790 'playlist_uploader_id': ie_result.get('uploader_id'),
30a074c2 1791 'extractor': ie_result['extractor'],
1792 'webpage_url': ie_result['webpage_url'],
1793 'webpage_url_basename': url_basename(ie_result['webpage_url']),
0bb322b9 1794 'webpage_url_domain': get_domain(ie_result['webpage_url']),
30a074c2 1795 'extractor_key': ie_result['extractor_key'],
1796 }
1797
1798 if self._match_entry(entry, incomplete=True) is not None:
1799 continue
1800
1801 entry_result = self.__process_iterable_entry(entry, download, extra)
26e2805c 1802 if not entry_result:
1803 failures += 1
1804 if failures >= max_failures:
1805 self.report_error(
1806 'Skipping the remaining entries in playlist "%s" since %d items failed extraction' % (playlist, failures))
1807 break
30a074c2 1808 playlist_results.append(entry_result)
1809 ie_result['entries'] = playlist_results
e08a85d8 1810
1811 # Write the updated info to json
cb96c5be 1812 if _infojson_written is True and self._write_info_json(
e08a85d8 1813 'updated playlist', ie_result,
1814 self.prepare_filename(ie_copy, 'pl_infojson'), overwrite=True) is None:
1815 return
ca30f449 1816
ed5835b4 1817 ie_result = self.run_all_pps('playlist', ie_result)
1818 self.to_screen(f'[download] Finished downloading playlist: {playlist}')
30a074c2 1819 return ie_result
1820
a0566bbf 1821 @__handle_extraction_exceptions
1822 def __process_iterable_entry(self, entry, download, extra_info):
1823 return self.process_ie_result(
1824 entry, download=download, extra_info=extra_info)
1825
67134eab
JMF
1826 def _build_format_filter(self, filter_spec):
1827 " Returns a function to filter the formats according to the filter_spec "
083c9df9
PH
1828
1829 OPERATORS = {
1830 '<': operator.lt,
1831 '<=': operator.le,
1832 '>': operator.gt,
1833 '>=': operator.ge,
1834 '=': operator.eq,
1835 '!=': operator.ne,
1836 }
67134eab 1837 operator_rex = re.compile(r'''(?x)\s*
187986a8 1838 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)\s*
1839 (?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1840 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)\s*
083c9df9 1841 ''' % '|'.join(map(re.escape, OPERATORS.keys())))
187986a8 1842 m = operator_rex.fullmatch(filter_spec)
9ddb6925
S
1843 if m:
1844 try:
1845 comparison_value = int(m.group('value'))
1846 except ValueError:
1847 comparison_value = parse_filesize(m.group('value'))
1848 if comparison_value is None:
1849 comparison_value = parse_filesize(m.group('value') + 'B')
1850 if comparison_value is None:
1851 raise ValueError(
1852 'Invalid value %r in format specification %r' % (
67134eab 1853 m.group('value'), filter_spec))
9ddb6925
S
1854 op = OPERATORS[m.group('op')]
1855
083c9df9 1856 if not m:
9ddb6925
S
1857 STR_OPERATORS = {
1858 '=': operator.eq,
10d33b34
YCH
1859 '^=': lambda attr, value: attr.startswith(value),
1860 '$=': lambda attr, value: attr.endswith(value),
1861 '*=': lambda attr, value: value in attr,
1ce9a3cb 1862 '~=': lambda attr, value: value.search(attr) is not None
9ddb6925 1863 }
187986a8 1864 str_operator_rex = re.compile(r'''(?x)\s*
1865 (?P<key>[a-zA-Z0-9._-]+)\s*
1ce9a3cb
LF
1866 (?P<negation>!\s*)?(?P<op>%s)\s*(?P<none_inclusive>\?\s*)?
1867 (?P<quote>["'])?
1868 (?P<value>(?(quote)(?:(?!(?P=quote))[^\\]|\\.)+|[\w.-]+))
1869 (?(quote)(?P=quote))\s*
9ddb6925 1870 ''' % '|'.join(map(re.escape, STR_OPERATORS.keys())))
187986a8 1871 m = str_operator_rex.fullmatch(filter_spec)
9ddb6925 1872 if m:
1ce9a3cb
LF
1873 if m.group('op') == '~=':
1874 comparison_value = re.compile(m.group('value'))
1875 else:
1876 comparison_value = re.sub(r'''\\([\\"'])''', r'\1', m.group('value'))
2cc779f4
S
1877 str_op = STR_OPERATORS[m.group('op')]
1878 if m.group('negation'):
e118a879 1879 op = lambda attr, value: not str_op(attr, value)
2cc779f4
S
1880 else:
1881 op = str_op
083c9df9 1882
9ddb6925 1883 if not m:
187986a8 1884 raise SyntaxError('Invalid filter specification %r' % filter_spec)
083c9df9
PH
1885
1886 def _filter(f):
1887 actual_value = f.get(m.group('key'))
1888 if actual_value is None:
1889 return m.group('none_inclusive')
1890 return op(actual_value, comparison_value)
67134eab
JMF
1891 return _filter
1892
9f1a1c36 1893 def _check_formats(self, formats):
1894 for f in formats:
1895 self.to_screen('[info] Testing format %s' % f['format_id'])
75689fe5 1896 path = self.get_output_path('temp')
1897 if not self._ensure_dir_exists(f'{path}/'):
1898 continue
1899 temp_file = tempfile.NamedTemporaryFile(suffix='.tmp', delete=False, dir=path or None)
9f1a1c36 1900 temp_file.close()
1901 try:
1902 success, _ = self.dl(temp_file.name, f, test=True)
1903 except (DownloadError, IOError, OSError, ValueError) + network_exceptions:
1904 success = False
1905 finally:
1906 if os.path.exists(temp_file.name):
1907 try:
1908 os.remove(temp_file.name)
1909 except OSError:
1910 self.report_warning('Unable to delete temporary file "%s"' % temp_file.name)
1911 if success:
1912 yield f
1913 else:
1914 self.to_screen('[info] Unable to download format %s. Skipping...' % f['format_id'])
1915
0017d9ad 1916 def _default_format_spec(self, info_dict, download=True):
0017d9ad 1917
af0f7428
S
1918 def can_merge():
1919 merger = FFmpegMergerPP(self)
1920 return merger.available and merger.can_merge()
1921
91ebc640 1922 prefer_best = (
b7b04c78 1923 not self.params.get('simulate')
91ebc640 1924 and download
1925 and (
1926 not can_merge()
19807826 1927 or info_dict.get('is_live', False)
de6000d9 1928 or self.outtmpl_dict['default'] == '-'))
53ed7066 1929 compat = (
1930 prefer_best
1931 or self.params.get('allow_multiple_audio_streams', False)
1932 or 'format-spec' in self.params.get('compat_opts', []))
91ebc640 1933
1934 return (
53ed7066 1935 'best/bestvideo+bestaudio' if prefer_best
1936 else 'bestvideo*+bestaudio/best' if not compat
91ebc640 1937 else 'bestvideo+bestaudio/best')
0017d9ad 1938
67134eab
JMF
1939 def build_format_selector(self, format_spec):
1940 def syntax_error(note, start):
1941 message = (
1942 'Invalid format specification: '
86e5f3ed 1943 '{}\n\t{}\n\t{}^'.format(note, format_spec, ' ' * start[1]))
67134eab
JMF
1944 return SyntaxError(message)
1945
1946 PICKFIRST = 'PICKFIRST'
1947 MERGE = 'MERGE'
1948 SINGLE = 'SINGLE'
0130afb7 1949 GROUP = 'GROUP'
67134eab
JMF
1950 FormatSelector = collections.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
1951
91ebc640 1952 allow_multiple_streams = {'audio': self.params.get('allow_multiple_audio_streams', False),
1953 'video': self.params.get('allow_multiple_video_streams', False)}
909d24dd 1954
9f1a1c36 1955 check_formats = self.params.get('check_formats') == 'selected'
e8e73840 1956
67134eab
JMF
1957 def _parse_filter(tokens):
1958 filter_parts = []
1959 for type, string, start, _, _ in tokens:
1960 if type == tokenize.OP and string == ']':
1961 return ''.join(filter_parts)
1962 else:
1963 filter_parts.append(string)
1964
232541df 1965 def _remove_unused_ops(tokens):
17cc1534 1966 # Remove operators that we don't use and join them with the surrounding strings
232541df
JMF
1967 # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
1968 ALLOWED_OPS = ('/', '+', ',', '(', ')')
1969 last_string, last_start, last_end, last_line = None, None, None, None
1970 for type, string, start, end, line in tokens:
1971 if type == tokenize.OP and string == '[':
1972 if last_string:
1973 yield tokenize.NAME, last_string, last_start, last_end, last_line
1974 last_string = None
1975 yield type, string, start, end, line
1976 # everything inside brackets will be handled by _parse_filter
1977 for type, string, start, end, line in tokens:
1978 yield type, string, start, end, line
1979 if type == tokenize.OP and string == ']':
1980 break
1981 elif type == tokenize.OP and string in ALLOWED_OPS:
1982 if last_string:
1983 yield tokenize.NAME, last_string, last_start, last_end, last_line
1984 last_string = None
1985 yield type, string, start, end, line
1986 elif type in [tokenize.NAME, tokenize.NUMBER, tokenize.OP]:
1987 if not last_string:
1988 last_string = string
1989 last_start = start
1990 last_end = end
1991 else:
1992 last_string += string
1993 if last_string:
1994 yield tokenize.NAME, last_string, last_start, last_end, last_line
1995
cf2ac6df 1996 def _parse_format_selection(tokens, inside_merge=False, inside_choice=False, inside_group=False):
67134eab
JMF
1997 selectors = []
1998 current_selector = None
1999 for type, string, start, _, _ in tokens:
2000 # ENCODING is only defined in python 3.x
2001 if type == getattr(tokenize, 'ENCODING', None):
2002 continue
2003 elif type in [tokenize.NAME, tokenize.NUMBER]:
2004 current_selector = FormatSelector(SINGLE, string, [])
2005 elif type == tokenize.OP:
cf2ac6df
JMF
2006 if string == ')':
2007 if not inside_group:
2008 # ')' will be handled by the parentheses group
2009 tokens.restore_last_token()
67134eab 2010 break
cf2ac6df 2011 elif inside_merge and string in ['/', ',']:
0130afb7
JMF
2012 tokens.restore_last_token()
2013 break
cf2ac6df
JMF
2014 elif inside_choice and string == ',':
2015 tokens.restore_last_token()
2016 break
2017 elif string == ',':
0a31a350
JMF
2018 if not current_selector:
2019 raise syntax_error('"," must follow a format selector', start)
67134eab
JMF
2020 selectors.append(current_selector)
2021 current_selector = None
2022 elif string == '/':
d96d604e
JMF
2023 if not current_selector:
2024 raise syntax_error('"/" must follow a format selector', start)
67134eab 2025 first_choice = current_selector
cf2ac6df 2026 second_choice = _parse_format_selection(tokens, inside_choice=True)
f5f4a27a 2027 current_selector = FormatSelector(PICKFIRST, (first_choice, second_choice), [])
67134eab
JMF
2028 elif string == '[':
2029 if not current_selector:
2030 current_selector = FormatSelector(SINGLE, 'best', [])
2031 format_filter = _parse_filter(tokens)
2032 current_selector.filters.append(format_filter)
0130afb7
JMF
2033 elif string == '(':
2034 if current_selector:
2035 raise syntax_error('Unexpected "("', start)
cf2ac6df
JMF
2036 group = _parse_format_selection(tokens, inside_group=True)
2037 current_selector = FormatSelector(GROUP, group, [])
67134eab 2038 elif string == '+':
d03cfdce 2039 if not current_selector:
2040 raise syntax_error('Unexpected "+"', start)
2041 selector_1 = current_selector
2042 selector_2 = _parse_format_selection(tokens, inside_merge=True)
2043 if not selector_2:
2044 raise syntax_error('Expected a selector', start)
2045 current_selector = FormatSelector(MERGE, (selector_1, selector_2), [])
67134eab 2046 else:
86e5f3ed 2047 raise syntax_error(f'Operator not recognized: "{string}"', start)
67134eab
JMF
2048 elif type == tokenize.ENDMARKER:
2049 break
2050 if current_selector:
2051 selectors.append(current_selector)
2052 return selectors
2053
f8d4ad9a 2054 def _merge(formats_pair):
2055 format_1, format_2 = formats_pair
2056
2057 formats_info = []
2058 formats_info.extend(format_1.get('requested_formats', (format_1,)))
2059 formats_info.extend(format_2.get('requested_formats', (format_2,)))
2060
2061 if not allow_multiple_streams['video'] or not allow_multiple_streams['audio']:
551f9388 2062 get_no_more = {'video': False, 'audio': False}
f8d4ad9a 2063 for (i, fmt_info) in enumerate(formats_info):
551f9388 2064 if fmt_info.get('acodec') == fmt_info.get('vcodec') == 'none':
2065 formats_info.pop(i)
2066 continue
2067 for aud_vid in ['audio', 'video']:
f8d4ad9a 2068 if not allow_multiple_streams[aud_vid] and fmt_info.get(aud_vid[0] + 'codec') != 'none':
2069 if get_no_more[aud_vid]:
2070 formats_info.pop(i)
f5510afe 2071 break
f8d4ad9a 2072 get_no_more[aud_vid] = True
2073
2074 if len(formats_info) == 1:
2075 return formats_info[0]
2076
2077 video_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('vcodec') != 'none']
2078 audio_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('acodec') != 'none']
2079
2080 the_only_video = video_fmts[0] if len(video_fmts) == 1 else None
2081 the_only_audio = audio_fmts[0] if len(audio_fmts) == 1 else None
2082
2083 output_ext = self.params.get('merge_output_format')
2084 if not output_ext:
2085 if the_only_video:
2086 output_ext = the_only_video['ext']
2087 elif the_only_audio and not video_fmts:
2088 output_ext = the_only_audio['ext']
2089 else:
2090 output_ext = 'mkv'
2091
975a0d0d 2092 filtered = lambda *keys: filter(None, (traverse_obj(fmt, *keys) for fmt in formats_info))
2093
f8d4ad9a 2094 new_dict = {
2095 'requested_formats': formats_info,
975a0d0d 2096 'format': '+'.join(filtered('format')),
2097 'format_id': '+'.join(filtered('format_id')),
f8d4ad9a 2098 'ext': output_ext,
975a0d0d 2099 'protocol': '+'.join(map(determine_protocol, formats_info)),
093a1710 2100 'language': '+'.join(orderedSet(filtered('language'))) or None,
2101 'format_note': '+'.join(orderedSet(filtered('format_note'))) or None,
2102 'filesize_approx': sum(filtered('filesize', 'filesize_approx')) or None,
975a0d0d 2103 'tbr': sum(filtered('tbr', 'vbr', 'abr')),
f8d4ad9a 2104 }
2105
2106 if the_only_video:
2107 new_dict.update({
2108 'width': the_only_video.get('width'),
2109 'height': the_only_video.get('height'),
2110 'resolution': the_only_video.get('resolution') or self.format_resolution(the_only_video),
2111 'fps': the_only_video.get('fps'),
49a57e70 2112 'dynamic_range': the_only_video.get('dynamic_range'),
f8d4ad9a 2113 'vcodec': the_only_video.get('vcodec'),
2114 'vbr': the_only_video.get('vbr'),
2115 'stretched_ratio': the_only_video.get('stretched_ratio'),
2116 })
2117
2118 if the_only_audio:
2119 new_dict.update({
2120 'acodec': the_only_audio.get('acodec'),
2121 'abr': the_only_audio.get('abr'),
975a0d0d 2122 'asr': the_only_audio.get('asr'),
f8d4ad9a 2123 })
2124
2125 return new_dict
2126
e8e73840 2127 def _check_formats(formats):
981052c9 2128 if not check_formats:
2129 yield from formats
b5ac45b1 2130 return
9f1a1c36 2131 yield from self._check_formats(formats)
e8e73840 2132
67134eab 2133 def _build_selector_function(selector):
909d24dd 2134 if isinstance(selector, list): # ,
67134eab
JMF
2135 fs = [_build_selector_function(s) for s in selector]
2136
317f7ab6 2137 def selector_function(ctx):
67134eab 2138 for f in fs:
981052c9 2139 yield from f(ctx)
67134eab 2140 return selector_function
909d24dd 2141
2142 elif selector.type == GROUP: # ()
0130afb7 2143 selector_function = _build_selector_function(selector.selector)
909d24dd 2144
2145 elif selector.type == PICKFIRST: # /
67134eab
JMF
2146 fs = [_build_selector_function(s) for s in selector.selector]
2147
317f7ab6 2148 def selector_function(ctx):
67134eab 2149 for f in fs:
317f7ab6 2150 picked_formats = list(f(ctx))
67134eab
JMF
2151 if picked_formats:
2152 return picked_formats
2153 return []
67134eab 2154
981052c9 2155 elif selector.type == MERGE: # +
2156 selector_1, selector_2 = map(_build_selector_function, selector.selector)
2157
2158 def selector_function(ctx):
adbc4ec4 2159 for pair in itertools.product(selector_1(ctx), selector_2(ctx)):
981052c9 2160 yield _merge(pair)
2161
909d24dd 2162 elif selector.type == SINGLE: # atom
598d185d 2163 format_spec = selector.selector or 'best'
909d24dd 2164
f8d4ad9a 2165 # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
909d24dd 2166 if format_spec == 'all':
2167 def selector_function(ctx):
9222c381 2168 yield from _check_formats(ctx['formats'][::-1])
f8d4ad9a 2169 elif format_spec == 'mergeall':
2170 def selector_function(ctx):
316f2650 2171 formats = list(_check_formats(
2172 f for f in ctx['formats'] if f.get('vcodec') != 'none' or f.get('acodec') != 'none'))
e01d6aa4 2173 if not formats:
2174 return
921b76ca 2175 merged_format = formats[-1]
2176 for f in formats[-2::-1]:
f8d4ad9a 2177 merged_format = _merge((merged_format, f))
2178 yield merged_format
909d24dd 2179
2180 else:
85e801a9 2181 format_fallback, seperate_fallback, format_reverse, format_idx = False, None, True, 1
eff63539 2182 mobj = re.match(
2183 r'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
2184 format_spec)
2185 if mobj is not None:
2186 format_idx = int_or_none(mobj.group('n'), default=1)
e8e73840 2187 format_reverse = mobj.group('bw')[0] == 'b'
eff63539 2188 format_type = (mobj.group('type') or [None])[0]
2189 not_format_type = {'v': 'a', 'a': 'v'}.get(format_type)
2190 format_modified = mobj.group('mod') is not None
909d24dd 2191
2192 format_fallback = not format_type and not format_modified # for b, w
8326b00a 2193 _filter_f = (
eff63539 2194 (lambda f: f.get('%scodec' % format_type) != 'none')
2195 if format_type and format_modified # bv*, ba*, wv*, wa*
2196 else (lambda f: f.get('%scodec' % not_format_type) == 'none')
2197 if format_type # bv, ba, wv, wa
2198 else (lambda f: f.get('vcodec') != 'none' and f.get('acodec') != 'none')
2199 if not format_modified # b, w
8326b00a 2200 else lambda f: True) # b*, w*
2201 filter_f = lambda f: _filter_f(f) and (
2202 f.get('vcodec') != 'none' or f.get('acodec') != 'none')
67134eab 2203 else:
48ee10ee 2204 if format_spec in self._format_selection_exts['audio']:
b11c04a8 2205 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none'
48ee10ee 2206 elif format_spec in self._format_selection_exts['video']:
b11c04a8 2207 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none' and f.get('vcodec') != 'none'
85e801a9 2208 seperate_fallback = lambda f: f.get('ext') == format_spec and f.get('vcodec') != 'none'
48ee10ee 2209 elif format_spec in self._format_selection_exts['storyboards']:
b11c04a8 2210 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') == 'none' and f.get('vcodec') == 'none'
2211 else:
b5ae35ee 2212 filter_f = lambda f: f.get('format_id') == format_spec # id
909d24dd 2213
2214 def selector_function(ctx):
2215 formats = list(ctx['formats'])
909d24dd 2216 matches = list(filter(filter_f, formats)) if filter_f is not None else formats
85e801a9 2217 if not matches:
2218 if format_fallback and ctx['incomplete_formats']:
2219 # for extractors with incomplete formats (audio only (soundcloud)
2220 # or video only (imgur)) best/worst will fallback to
2221 # best/worst {video,audio}-only format
2222 matches = formats
2223 elif seperate_fallback and not ctx['has_merged_format']:
2224 # for compatibility with youtube-dl when there is no pre-merged format
2225 matches = list(filter(seperate_fallback, formats))
981052c9 2226 matches = LazyList(_check_formats(matches[::-1 if format_reverse else 1]))
2227 try:
e8e73840 2228 yield matches[format_idx - 1]
4abea8ca 2229 except LazyList.IndexError:
981052c9 2230 return
083c9df9 2231
67134eab 2232 filters = [self._build_format_filter(f) for f in selector.filters]
083c9df9 2233
317f7ab6 2234 def final_selector(ctx):
adbc4ec4 2235 ctx_copy = dict(ctx)
67134eab 2236 for _filter in filters:
317f7ab6
S
2237 ctx_copy['formats'] = list(filter(_filter, ctx_copy['formats']))
2238 return selector_function(ctx_copy)
67134eab 2239 return final_selector
083c9df9 2240
67134eab 2241 stream = io.BytesIO(format_spec.encode('utf-8'))
0130afb7 2242 try:
f9934b96 2243 tokens = list(_remove_unused_ops(tokenize.tokenize(stream.readline)))
0130afb7
JMF
2244 except tokenize.TokenError:
2245 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec)))
2246
86e5f3ed 2247 class TokenIterator:
0130afb7
JMF
2248 def __init__(self, tokens):
2249 self.tokens = tokens
2250 self.counter = 0
2251
2252 def __iter__(self):
2253 return self
2254
2255 def __next__(self):
2256 if self.counter >= len(self.tokens):
2257 raise StopIteration()
2258 value = self.tokens[self.counter]
2259 self.counter += 1
2260 return value
2261
2262 next = __next__
2263
2264 def restore_last_token(self):
2265 self.counter -= 1
2266
2267 parsed_selector = _parse_format_selection(iter(TokenIterator(tokens)))
67134eab 2268 return _build_selector_function(parsed_selector)
a9c58ad9 2269
e5660ee6 2270 def _calc_headers(self, info_dict):
8b7539d2 2271 res = merge_headers(self.params['http_headers'], info_dict.get('http_headers') or {})
e5660ee6
JMF
2272
2273 cookies = self._calc_cookies(info_dict)
2274 if cookies:
2275 res['Cookie'] = cookies
2276
0016b84e
S
2277 if 'X-Forwarded-For' not in res:
2278 x_forwarded_for_ip = info_dict.get('__x_forwarded_for_ip')
2279 if x_forwarded_for_ip:
2280 res['X-Forwarded-For'] = x_forwarded_for_ip
2281
e5660ee6
JMF
2282 return res
2283
2284 def _calc_cookies(self, info_dict):
5c2266df 2285 pr = sanitized_Request(info_dict['url'])
e5660ee6 2286 self.cookiejar.add_cookie_header(pr)
662435f7 2287 return pr.get_header('Cookie')
e5660ee6 2288
9f1a1c36 2289 def _sort_thumbnails(self, thumbnails):
2290 thumbnails.sort(key=lambda t: (
2291 t.get('preference') if t.get('preference') is not None else -1,
2292 t.get('width') if t.get('width') is not None else -1,
2293 t.get('height') if t.get('height') is not None else -1,
2294 t.get('id') if t.get('id') is not None else '',
2295 t.get('url')))
2296
b0249bca 2297 def _sanitize_thumbnails(self, info_dict):
bc516a3f 2298 thumbnails = info_dict.get('thumbnails')
2299 if thumbnails is None:
2300 thumbnail = info_dict.get('thumbnail')
2301 if thumbnail:
2302 info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}]
9f1a1c36 2303 if not thumbnails:
2304 return
2305
2306 def check_thumbnails(thumbnails):
2307 for t in thumbnails:
2308 self.to_screen(f'[info] Testing thumbnail {t["id"]}')
2309 try:
2310 self.urlopen(HEADRequest(t['url']))
2311 except network_exceptions as err:
2312 self.to_screen(f'[info] Unable to connect to thumbnail {t["id"]} URL {t["url"]!r} - {err}. Skipping...')
2313 continue
2314 yield t
2315
2316 self._sort_thumbnails(thumbnails)
2317 for i, t in enumerate(thumbnails):
2318 if t.get('id') is None:
2319 t['id'] = '%d' % i
2320 if t.get('width') and t.get('height'):
2321 t['resolution'] = '%dx%d' % (t['width'], t['height'])
2322 t['url'] = sanitize_url(t['url'])
2323
2324 if self.params.get('check_formats') is True:
282f5709 2325 info_dict['thumbnails'] = LazyList(check_thumbnails(thumbnails[::-1]), reverse=True)
9f1a1c36 2326 else:
2327 info_dict['thumbnails'] = thumbnails
bc516a3f 2328
03f83004
LNO
2329 def _fill_common_fields(self, info_dict, is_video=True):
2330 # TODO: move sanitization here
2331 if is_video:
2332 # playlists are allowed to lack "title"
2333 info_dict['fulltitle'] = info_dict.get('title')
2334 if 'title' not in info_dict:
2335 raise ExtractorError('Missing "title" field in extractor result',
2336 video_id=info_dict['id'], ie=info_dict['extractor'])
2337 elif not info_dict.get('title'):
2338 self.report_warning('Extractor failed to obtain "title". Creating a generic title instead')
2339 info_dict['title'] = f'{info_dict["extractor"]} video #{info_dict["id"]}'
2340
2341 if info_dict.get('duration') is not None:
2342 info_dict['duration_string'] = formatSeconds(info_dict['duration'])
2343
2344 for ts_key, date_key in (
2345 ('timestamp', 'upload_date'),
2346 ('release_timestamp', 'release_date'),
2347 ('modified_timestamp', 'modified_date'),
2348 ):
2349 if info_dict.get(date_key) is None and info_dict.get(ts_key) is not None:
2350 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
2351 # see http://bugs.python.org/issue1646728)
19a03940 2352 with contextlib.suppress(ValueError, OverflowError, OSError):
03f83004
LNO
2353 upload_date = datetime.datetime.utcfromtimestamp(info_dict[ts_key])
2354 info_dict[date_key] = upload_date.strftime('%Y%m%d')
03f83004
LNO
2355
2356 live_keys = ('is_live', 'was_live')
2357 live_status = info_dict.get('live_status')
2358 if live_status is None:
2359 for key in live_keys:
2360 if info_dict.get(key) is False:
2361 continue
2362 if info_dict.get(key):
2363 live_status = key
2364 break
2365 if all(info_dict.get(key) is False for key in live_keys):
2366 live_status = 'not_live'
2367 if live_status:
2368 info_dict['live_status'] = live_status
2369 for key in live_keys:
2370 if info_dict.get(key) is None:
2371 info_dict[key] = (live_status == key)
2372
2373 # Auto generate title fields corresponding to the *_number fields when missing
2374 # in order to always have clean titles. This is very common for TV series.
2375 for field in ('chapter', 'season', 'episode'):
2376 if info_dict.get('%s_number' % field) is not None and not info_dict.get(field):
2377 info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field])
2378
dd82ffea
JMF
2379 def process_video_result(self, info_dict, download=True):
2380 assert info_dict.get('_type', 'video') == 'video'
9c906919 2381 self._num_videos += 1
dd82ffea 2382
bec1fad2 2383 if 'id' not in info_dict:
fc08bdd6 2384 raise ExtractorError('Missing "id" field in extractor result', ie=info_dict['extractor'])
2385 elif not info_dict.get('id'):
2386 raise ExtractorError('Extractor failed to obtain "id"', ie=info_dict['extractor'])
455a15e2 2387
c9969434
S
2388 def report_force_conversion(field, field_not, conversion):
2389 self.report_warning(
2390 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
2391 % (field, field_not, conversion))
2392
2393 def sanitize_string_field(info, string_field):
2394 field = info.get(string_field)
2395 if field is None or isinstance(field, compat_str):
2396 return
2397 report_force_conversion(string_field, 'a string', 'string')
2398 info[string_field] = compat_str(field)
2399
2400 def sanitize_numeric_fields(info):
2401 for numeric_field in self._NUMERIC_FIELDS:
2402 field = info.get(numeric_field)
f9934b96 2403 if field is None or isinstance(field, (int, float)):
c9969434
S
2404 continue
2405 report_force_conversion(numeric_field, 'numeric', 'int')
2406 info[numeric_field] = int_or_none(field)
2407
2408 sanitize_string_field(info_dict, 'id')
2409 sanitize_numeric_fields(info_dict)
4c3f8c3f 2410 if (info_dict.get('duration') or 0) <= 0 and info_dict.pop('duration', None):
50e93e03 2411 self.report_warning('"duration" field is negative, there is an error in extractor')
be6217b2 2412
dd82ffea
JMF
2413 if 'playlist' not in info_dict:
2414 # It isn't part of a playlist
2415 info_dict['playlist'] = None
2416 info_dict['playlist_index'] = None
2417
bc516a3f 2418 self._sanitize_thumbnails(info_dict)
d5519808 2419
536a55da 2420 thumbnail = info_dict.get('thumbnail')
bc516a3f 2421 thumbnails = info_dict.get('thumbnails')
536a55da
S
2422 if thumbnail:
2423 info_dict['thumbnail'] = sanitize_url(thumbnail)
2424 elif thumbnails:
d5519808
PH
2425 info_dict['thumbnail'] = thumbnails[-1]['url']
2426
ae30b840 2427 if info_dict.get('display_id') is None and 'id' in info_dict:
0afef30b
PH
2428 info_dict['display_id'] = info_dict['id']
2429
03f83004 2430 self._fill_common_fields(info_dict)
33d2fc2f 2431
05108a49
S
2432 for cc_kind in ('subtitles', 'automatic_captions'):
2433 cc = info_dict.get(cc_kind)
2434 if cc:
2435 for _, subtitle in cc.items():
2436 for subtitle_format in subtitle:
2437 if subtitle_format.get('url'):
2438 subtitle_format['url'] = sanitize_url(subtitle_format['url'])
2439 if subtitle_format.get('ext') is None:
2440 subtitle_format['ext'] = determine_ext(subtitle_format['url']).lower()
2441
2442 automatic_captions = info_dict.get('automatic_captions')
4bba3716 2443 subtitles = info_dict.get('subtitles')
4bba3716 2444
360e1ca5 2445 info_dict['requested_subtitles'] = self.process_subtitles(
05108a49 2446 info_dict['id'], subtitles, automatic_captions)
a504ced0 2447
dd82ffea
JMF
2448 if info_dict.get('formats') is None:
2449 # There's only one format available
2450 formats = [info_dict]
2451 else:
2452 formats = info_dict['formats']
2453
0a5a191a 2454 # or None ensures --clean-infojson removes it
2455 info_dict['_has_drm'] = any(f.get('has_drm') for f in formats) or None
88acdbc2 2456 if not self.params.get('allow_unplayable_formats'):
2457 formats = [f for f in formats if not f.get('has_drm')]
0a5a191a 2458 if info_dict['_has_drm'] and all(
c0b6e5c7 2459 f.get('acodec') == f.get('vcodec') == 'none' for f in formats):
2460 self.report_warning(
2461 'This video is DRM protected and only images are available for download. '
2462 'Use --list-formats to see them')
88acdbc2 2463
319b6059 2464 get_from_start = not info_dict.get('is_live') or bool(self.params.get('live_from_start'))
2465 if not get_from_start:
2466 info_dict['title'] += ' ' + datetime.datetime.now().strftime('%Y-%m-%d %H:%M')
2467 if info_dict.get('is_live') and formats:
adbc4ec4 2468 formats = [f for f in formats if bool(f.get('is_from_start')) == get_from_start]
319b6059 2469 if get_from_start and not formats:
a44ca5a4 2470 self.raise_no_formats(info_dict, msg=(
2471 '--live-from-start is passed, but there are no formats that can be downloaded from the start. '
2472 'If you want to download from the current time, use --no-live-from-start'))
adbc4ec4 2473
db95dc13 2474 if not formats:
1151c407 2475 self.raise_no_formats(info_dict)
db95dc13 2476
73af5cc8
S
2477 def is_wellformed(f):
2478 url = f.get('url')
a5ac0c47 2479 if not url:
73af5cc8
S
2480 self.report_warning(
2481 '"url" field is missing or empty - skipping format, '
2482 'there is an error in extractor')
a5ac0c47
S
2483 return False
2484 if isinstance(url, bytes):
2485 sanitize_string_field(f, 'url')
2486 return True
73af5cc8
S
2487
2488 # Filter out malformed formats for better extraction robustness
2489 formats = list(filter(is_wellformed, formats))
2490
181c7053
S
2491 formats_dict = {}
2492
dd82ffea 2493 # We check that all the formats have the format and format_id fields
db95dc13 2494 for i, format in enumerate(formats):
c9969434
S
2495 sanitize_string_field(format, 'format_id')
2496 sanitize_numeric_fields(format)
dcf77cf1 2497 format['url'] = sanitize_url(format['url'])
e74e3b63 2498 if not format.get('format_id'):
8016c922 2499 format['format_id'] = compat_str(i)
e2effb08
S
2500 else:
2501 # Sanitize format_id from characters used in format selector expression
ec85ded8 2502 format['format_id'] = re.sub(r'[\s,/+\[\]()]', '_', format['format_id'])
181c7053
S
2503 format_id = format['format_id']
2504 if format_id not in formats_dict:
2505 formats_dict[format_id] = []
2506 formats_dict[format_id].append(format)
2507
2508 # Make sure all formats have unique format_id
03b4de72 2509 common_exts = set(itertools.chain(*self._format_selection_exts.values()))
181c7053 2510 for format_id, ambiguous_formats in formats_dict.items():
48ee10ee 2511 ambigious_id = len(ambiguous_formats) > 1
2512 for i, format in enumerate(ambiguous_formats):
2513 if ambigious_id:
181c7053 2514 format['format_id'] = '%s-%d' % (format_id, i)
48ee10ee 2515 if format.get('ext') is None:
2516 format['ext'] = determine_ext(format['url']).lower()
2517 # Ensure there is no conflict between id and ext in format selection
2518 # See https://github.com/yt-dlp/yt-dlp/issues/1282
2519 if format['format_id'] != format['ext'] and format['format_id'] in common_exts:
2520 format['format_id'] = 'f%s' % format['format_id']
181c7053
S
2521
2522 for i, format in enumerate(formats):
8c51aa65 2523 if format.get('format') is None:
6febd1c1 2524 format['format'] = '{id} - {res}{note}'.format(
8c51aa65
JMF
2525 id=format['format_id'],
2526 res=self.format_resolution(format),
b868936c 2527 note=format_field(format, 'format_note', ' (%s)'),
8c51aa65 2528 )
6f0be937 2529 if format.get('protocol') is None:
b5559424 2530 format['protocol'] = determine_protocol(format)
239df021 2531 if format.get('resolution') is None:
2532 format['resolution'] = self.format_resolution(format, default=None)
176f1866 2533 if format.get('dynamic_range') is None and format.get('vcodec') != 'none':
2534 format['dynamic_range'] = 'SDR'
f2fe69c7 2535 if (info_dict.get('duration') and format.get('tbr')
2536 and not format.get('filesize') and not format.get('filesize_approx')):
2537 format['filesize_approx'] = info_dict['duration'] * format['tbr'] * (1024 / 8)
2538
e5660ee6
JMF
2539 # Add HTTP headers, so that external programs can use them from the
2540 # json output
2541 full_format_info = info_dict.copy()
2542 full_format_info.update(format)
2543 format['http_headers'] = self._calc_headers(full_format_info)
0016b84e
S
2544 # Remove private housekeeping stuff
2545 if '__x_forwarded_for_ip' in info_dict:
2546 del info_dict['__x_forwarded_for_ip']
dd82ffea 2547
9f1a1c36 2548 if self.params.get('check_formats') is True:
282f5709 2549 formats = LazyList(self._check_formats(formats[::-1]), reverse=True)
9f1a1c36 2550
88acdbc2 2551 if not formats or formats[0] is not info_dict:
b3d9ef88
JMF
2552 # only set the 'formats' fields if the original info_dict list them
2553 # otherwise we end up with a circular reference, the first (and unique)
f89197d7 2554 # element in the 'formats' field in info_dict is info_dict itself,
dfb1b146 2555 # which can't be exported to json
b3d9ef88 2556 info_dict['formats'] = formats
4ec82a72 2557
2558 info_dict, _ = self.pre_process(info_dict)
2559
6db9c4d5 2560 if self._match_entry(info_dict, incomplete=self._format_fields) is not None:
09b49e1f 2561 return info_dict
2562
2563 self.post_extract(info_dict)
2564 info_dict, _ = self.pre_process(info_dict, 'after_filter')
2565
093a1710 2566 # The pre-processors may have modified the formats
2567 formats = info_dict.get('formats', [info_dict])
2568
fa9f30b8 2569 list_only = self.params.get('simulate') is None and (
2570 self.params.get('list_thumbnails') or self.params.get('listformats') or self.params.get('listsubtitles'))
2571 interactive_format_selection = not list_only and self.format_selector == '-'
b7b04c78 2572 if self.params.get('list_thumbnails'):
2573 self.list_thumbnails(info_dict)
b7b04c78 2574 if self.params.get('listsubtitles'):
2575 if 'automatic_captions' in info_dict:
2576 self.list_subtitles(
2577 info_dict['id'], automatic_captions, 'automatic captions')
2578 self.list_subtitles(info_dict['id'], subtitles, 'subtitles')
fa9f30b8 2579 if self.params.get('listformats') or interactive_format_selection:
b69fd25c 2580 self.list_formats(info_dict)
169dbde9 2581 if list_only:
b7b04c78 2582 # Without this printing, -F --print-json will not work
169dbde9 2583 self.__forced_printings(info_dict, self.prepare_filename(info_dict), incomplete=True)
bfaae0a7 2584 return
2585
187986a8 2586 format_selector = self.format_selector
2587 if format_selector is None:
0017d9ad 2588 req_format = self._default_format_spec(info_dict, download=download)
0760b0a7 2589 self.write_debug('Default format spec: %s' % req_format)
187986a8 2590 format_selector = self.build_format_selector(req_format)
317f7ab6 2591
fa9f30b8 2592 while True:
2593 if interactive_format_selection:
2594 req_format = input(
2595 self._format_screen('\nEnter format selector: ', self.Styles.EMPHASIS))
2596 try:
2597 format_selector = self.build_format_selector(req_format)
2598 except SyntaxError as err:
2599 self.report_error(err, tb=False, is_error=False)
2600 continue
2601
85e801a9 2602 formats_to_download = list(format_selector({
fa9f30b8 2603 'formats': formats,
85e801a9 2604 'has_merged_format': any('none' not in (f.get('acodec'), f.get('vcodec')) for f in formats),
2605 'incomplete_formats': (
2606 # All formats are video-only or
2607 all(f.get('vcodec') != 'none' and f.get('acodec') == 'none' for f in formats)
2608 # all formats are audio-only
2609 or all(f.get('vcodec') == 'none' and f.get('acodec') != 'none' for f in formats)),
2610 }))
fa9f30b8 2611 if interactive_format_selection and not formats_to_download:
2612 self.report_error('Requested format is not available', tb=False, is_error=False)
2613 continue
2614 break
317f7ab6 2615
dd82ffea 2616 if not formats_to_download:
b7da73eb 2617 if not self.params.get('ignore_no_formats_error'):
c0b6e5c7 2618 raise ExtractorError(
2619 'Requested format is not available. Use --list-formats for a list of available formats',
2620 expected=True, 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
86e5f3ed 2646 write_archive = {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"""
d8a58ddc 2662 available_subs, normal_sub_langs = {}, []
98c70d6f
JMF
2663 if normal_subtitles and self.params.get('writesubtitles'):
2664 available_subs.update(normal_subtitles)
d8a58ddc 2665 normal_sub_langs = tuple(normal_subtitles.keys())
98c70d6f
JMF
2666 if automatic_captions and self.params.get('writeautomaticsub'):
2667 for lang, cap_info in automatic_captions.items():
360e1ca5
JMF
2668 if lang not in available_subs:
2669 available_subs[lang] = cap_info
2670
4d171848
JMF
2671 if (not self.params.get('writesubtitles') and not
2672 self.params.get('writeautomaticsub') or not
2673 available_subs):
2674 return None
a504ced0 2675
d8a58ddc 2676 all_sub_langs = tuple(available_subs.keys())
a504ced0 2677 if self.params.get('allsubtitles', False):
c32b0aab 2678 requested_langs = all_sub_langs
2679 elif self.params.get('subtitleslangs', False):
77c4a9ef 2680 # A list is used so that the order of languages will be the same as
2681 # given in subtitleslangs. See https://github.com/yt-dlp/yt-dlp/issues/1041
2682 requested_langs = []
2683 for lang_re in self.params.get('subtitleslangs'):
77c4a9ef 2684 discard = lang_re[0] == '-'
c32b0aab 2685 if discard:
77c4a9ef 2686 lang_re = lang_re[1:]
3aa91540 2687 if lang_re == 'all':
2688 if discard:
2689 requested_langs = []
2690 else:
2691 requested_langs.extend(all_sub_langs)
2692 continue
77c4a9ef 2693 current_langs = filter(re.compile(lang_re + '$').match, all_sub_langs)
c32b0aab 2694 if discard:
2695 for lang in current_langs:
77c4a9ef 2696 while lang in requested_langs:
2697 requested_langs.remove(lang)
c32b0aab 2698 else:
77c4a9ef 2699 requested_langs.extend(current_langs)
2700 requested_langs = orderedSet(requested_langs)
d8a58ddc 2701 elif normal_sub_langs:
2702 requested_langs = ['en'] if 'en' in normal_sub_langs else normal_sub_langs[:1]
a504ced0 2703 else:
d8a58ddc 2704 requested_langs = ['en'] if 'en' in all_sub_langs else all_sub_langs[:1]
ad3dc496 2705 if requested_langs:
2706 self.write_debug('Downloading subtitles: %s' % ', '.join(requested_langs))
a504ced0
JMF
2707
2708 formats_query = self.params.get('subtitlesformat', 'best')
2709 formats_preference = formats_query.split('/') if formats_query else []
2710 subs = {}
2711 for lang in requested_langs:
2712 formats = available_subs.get(lang)
2713 if formats is None:
86e5f3ed 2714 self.report_warning(f'{lang} subtitles not available for {video_id}')
a504ced0 2715 continue
a504ced0
JMF
2716 for ext in formats_preference:
2717 if ext == 'best':
2718 f = formats[-1]
2719 break
2720 matches = list(filter(lambda f: f['ext'] == ext, formats))
2721 if matches:
2722 f = matches[-1]
2723 break
2724 else:
2725 f = formats[-1]
2726 self.report_warning(
2727 'No subtitle format found matching "%s" for language %s, '
2728 'using %s' % (formats_query, lang, f['ext']))
2729 subs[lang] = f
2730 return subs
2731
bb66c247 2732 def _forceprint(self, key, info_dict):
2733 if info_dict is None:
2734 return
2735 info_copy = info_dict.copy()
2736 info_copy['formats_table'] = self.render_formats_table(info_dict)
2737 info_copy['thumbnails_table'] = self.render_thumbnails_table(info_dict)
2738 info_copy['subtitles_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('subtitles'))
2739 info_copy['automatic_captions_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('automatic_captions'))
2740
2741 def format_tmpl(tmpl):
2742 mobj = re.match(r'\w+(=?)$', tmpl)
2743 if mobj and mobj.group(1):
2744 return f'{tmpl[:-1]} = %({tmpl[:-1]})r'
2745 elif mobj:
2746 return f'%({tmpl})s'
2747 return tmpl
8130779d 2748
bb66c247 2749 for tmpl in self.params['forceprint'].get(key, []):
2750 self.to_stdout(self.evaluate_outtmpl(format_tmpl(tmpl), info_copy))
2751
2752 for tmpl, file_tmpl in self.params['print_to_file'].get(key, []):
5127e92a 2753 filename = self.prepare_filename(info_dict, outtmpl=file_tmpl)
bb66c247 2754 tmpl = format_tmpl(tmpl)
2755 self.to_screen(f'[info] Writing {tmpl!r} to: {filename}')
8d93e69d 2756 if self._ensure_dir_exists(filename):
86e5f3ed 2757 with open(filename, 'a', encoding='utf-8') as f:
8d93e69d 2758 f.write(self.evaluate_outtmpl(tmpl, info_copy) + '\n')
ca30f449 2759
d06daf23 2760 def __forced_printings(self, info_dict, filename, incomplete):
53c18592 2761 def print_mandatory(field, actual_field=None):
2762 if actual_field is None:
2763 actual_field = field
d06daf23 2764 if (self.params.get('force%s' % field, False)
53c18592 2765 and (not incomplete or info_dict.get(actual_field) is not None)):
2766 self.to_stdout(info_dict[actual_field])
d06daf23
S
2767
2768 def print_optional(field):
2769 if (self.params.get('force%s' % field, False)
2770 and info_dict.get(field) is not None):
2771 self.to_stdout(info_dict[field])
2772
53c18592 2773 info_dict = info_dict.copy()
2774 if filename is not None:
2775 info_dict['filename'] = filename
2776 if info_dict.get('requested_formats') is not None:
2777 # For RTMP URLs, also include the playpath
2778 info_dict['urls'] = '\n'.join(f['url'] + f.get('play_path', '') for f in info_dict['requested_formats'])
10331a26 2779 elif info_dict.get('url'):
53c18592 2780 info_dict['urls'] = info_dict['url'] + info_dict.get('play_path', '')
2781
bb66c247 2782 if (self.params.get('forcejson')
2783 or self.params['forceprint'].get('video')
2784 or self.params['print_to_file'].get('video')):
2b8a2973 2785 self.post_extract(info_dict)
bb66c247 2786 self._forceprint('video', info_dict)
53c18592 2787
d06daf23
S
2788 print_mandatory('title')
2789 print_mandatory('id')
53c18592 2790 print_mandatory('url', 'urls')
d06daf23
S
2791 print_optional('thumbnail')
2792 print_optional('description')
53c18592 2793 print_optional('filename')
b868936c 2794 if self.params.get('forceduration') and info_dict.get('duration') is not None:
d06daf23
S
2795 self.to_stdout(formatSeconds(info_dict['duration']))
2796 print_mandatory('format')
53c18592 2797
2b8a2973 2798 if self.params.get('forcejson'):
6e84b215 2799 self.to_stdout(json.dumps(self.sanitize_info(info_dict)))
d06daf23 2800
e8e73840 2801 def dl(self, name, info, subtitle=False, test=False):
88acdbc2 2802 if not info.get('url'):
1151c407 2803 self.raise_no_formats(info, True)
e8e73840 2804
2805 if test:
2806 verbose = self.params.get('verbose')
2807 params = {
2808 'test': True,
a169858f 2809 'quiet': self.params.get('quiet') or not verbose,
e8e73840 2810 'verbose': verbose,
2811 'noprogress': not verbose,
2812 'nopart': True,
2813 'skip_unavailable_fragments': False,
2814 'keep_fragments': False,
2815 'overwrites': True,
2816 '_no_ytdl_file': True,
2817 }
2818 else:
2819 params = self.params
96fccc10 2820 fd = get_suitable_downloader(info, params, to_stdout=(name == '-'))(self, params)
e8e73840 2821 if not test:
2822 for ph in self._progress_hooks:
2823 fd.add_progress_hook(ph)
42676437
M
2824 urls = '", "'.join(
2825 (f['url'].split(',')[0] + ',<data>' if f['url'].startswith('data:') else f['url'])
2826 for f in info.get('requested_formats', []) or [info])
18e674b4 2827 self.write_debug('Invoking downloader on "%s"' % urls)
03b4de72 2828
adbc4ec4
THD
2829 # Note: Ideally info should be a deep-copied so that hooks cannot modify it.
2830 # But it may contain objects that are not deep-copyable
2831 new_info = self._copy_infodict(info)
e8e73840 2832 if new_info.get('http_headers') is None:
2833 new_info['http_headers'] = self._calc_headers(new_info)
2834 return fd.download(name, new_info, subtitle)
2835
e04938ab 2836 def existing_file(self, filepaths, *, default_overwrite=True):
2837 existing_files = list(filter(os.path.exists, orderedSet(filepaths)))
2838 if existing_files and not self.params.get('overwrites', default_overwrite):
2839 return existing_files[0]
2840
2841 for file in existing_files:
2842 self.report_file_delete(file)
2843 os.remove(file)
2844 return None
2845
8222d8de 2846 def process_info(self, info_dict):
09b49e1f 2847 """Process a single resolved IE result. (Modifies it in-place)"""
8222d8de
JMF
2848
2849 assert info_dict.get('_type', 'video') == 'video'
f46e2f9d 2850 original_infodict = info_dict
fd288278 2851
4513a41a 2852 if 'format' not in info_dict and 'ext' in info_dict:
8222d8de
JMF
2853 info_dict['format'] = info_dict['ext']
2854
09b49e1f 2855 # This is mostly just for backward compatibility of process_info
2856 # As a side-effect, this allows for format-specific filters
c77495e3 2857 if self._match_entry(info_dict) is not None:
9e907ebd 2858 info_dict['__write_download_archive'] = 'ignore'
8222d8de
JMF
2859 return
2860
09b49e1f 2861 # Does nothing under normal operation - for backward compatibility of process_info
277d6ff5 2862 self.post_extract(info_dict)
0c14d66a 2863 self._num_downloads += 1
8222d8de 2864
dcf64d43 2865 # info_dict['_filename'] needs to be set for backward compatibility
de6000d9 2866 info_dict['_filename'] = full_filename = self.prepare_filename(info_dict, warn=True)
2867 temp_filename = self.prepare_filename(info_dict, 'temp')
0202b52a 2868 files_to_move = {}
8222d8de
JMF
2869
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)
86e5f3ed 2922 with open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
7b6fefc9
PH
2923 annofile.write(info_dict['annotations'])
2924 except (KeyError, TypeError):
6febd1c1 2925 self.report_warning('There are no annotations to write.')
86e5f3ed 2926 except OSError:
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}')
86e5f3ed 2945 with 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)
86e5f3ed 2951 except OSError:
08438d2c 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
86e5f3ed 3016 exts = {format.get('ext') for format in formats}
d03cfdce 3017 COMPATIBLE_EXTS = (
86e5f3ed 3018 {'mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma'},
3019 {'webm'},
d03cfdce 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)
86e5f3ed 3052 return f'{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 3136 return
86e5f3ed 3137 except OSError as err:
4340deca
P
3138 raise UnavailableVideoError(err)
3139 except (ContentTooShortError, ) as err:
86e5f3ed 3140 self.report_error(f'content too short (expected {err.expected} bytes and served {err.downloaded})')
4340deca 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:
0a5a191a 3285 reject = lambda k, v: v is None or k.startswith('__') or k in {
f46e2f9d 3286 'requested_downloads', 'requested_formats', 'requested_subtitles', 'requested_entries',
0a5a191a 3287 'entries', 'filepath', '_filename', '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))
86e5f3ed 3345 except OSError:
5bfa4862 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
86e5f3ed 3390 return f'{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 """
f9934b96 3587 if isinstance(req, str):
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):
59f943cd 3598 from .compat import WINDOWS_VT_MODE # Must be imported locally
e3c7d495 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:
19a03940 3645 with contextlib.suppress(Exception):
36eaf303 3646 sys.exc_clear()
b300cda4
S
3647
3648 def python_implementation():
3649 impl_name = platform.python_implementation()
3650 if impl_name == 'PyPy' and hasattr(sys, 'pypy_version_info'):
3651 return impl_name + ' version %d.%d.%d' % sys.pypy_version_info[:3]
3652 return impl_name
3653
49a57e70 3654 write_debug('Python version %s (%s %s) - %s' % (
e5813e53 3655 platform.python_version(),
3656 python_implementation(),
3657 platform.architecture()[0],
b300cda4 3658 platform_name()))
d28b5171 3659
8913ef74 3660 exe_versions, ffmpeg_features = FFmpegPostProcessor.get_versions_and_features(self)
3661 ffmpeg_features = {key for key, val in ffmpeg_features.items() if val}
3662 if ffmpeg_features:
19a03940 3663 exe_versions['ffmpeg'] += ' (%s)' % ','.join(sorted(ffmpeg_features))
8913ef74 3664
4c83c967 3665 exe_versions['rtmpdump'] = rtmpdump_version()
feee8d32 3666 exe_versions['phantomjs'] = PhantomJSwrapper._version()
d28b5171 3667 exe_str = ', '.join(
2831b468 3668 f'{exe} {v}' for exe, v in sorted(exe_versions.items()) if v
3669 ) or 'none'
49a57e70 3670 write_debug('exe versions: %s' % exe_str)
dca08720 3671
9b8ee23b 3672 from .dependencies import available_dependencies
3673
3674 write_debug('Optional libraries: %s' % (', '.join(sorted({
3675 module.__name__.split('.')[0] for module in available_dependencies.values()
3676 })) or 'none'))
2831b468 3677
97ec5bc5 3678 self._setup_opener()
dca08720
PH
3679 proxy_map = {}
3680 for handler in self._opener.handlers:
3681 if hasattr(handler, 'proxies'):
3682 proxy_map.update(handler.proxies)
49a57e70 3683 write_debug(f'Proxy map: {proxy_map}')
dca08720 3684
49a57e70 3685 # Not implemented
3686 if False and self.params.get('call_home'):
58b1f00d 3687 ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode('utf-8')
49a57e70 3688 write_debug('Public IP address: %s' % ipaddr)
58b1f00d
PH
3689 latest_version = self.urlopen(
3690 'https://yt-dl.org/latest/version').read().decode('utf-8')
3691 if version_tuple(latest_version) > version_tuple(__version__):
3692 self.report_warning(
3693 'You are using an outdated version (newest version: %s)! '
3694 'See https://yt-dl.org/update if you need help updating.' %
3695 latest_version)
3696
e344693b 3697 def _setup_opener(self):
97ec5bc5 3698 if hasattr(self, '_opener'):
3699 return
6ad14cab 3700 timeout_val = self.params.get('socket_timeout')
17bddf3e 3701 self._socket_timeout = 20 if timeout_val is None else float(timeout_val)
6ad14cab 3702
982ee69a 3703 opts_cookiesfrombrowser = self.params.get('cookiesfrombrowser')
dca08720
PH
3704 opts_cookiefile = self.params.get('cookiefile')
3705 opts_proxy = self.params.get('proxy')
3706
982ee69a 3707 self.cookiejar = load_cookies(opts_cookiefile, opts_cookiesfrombrowser, self)
dca08720 3708
6a3f4c3f 3709 cookie_processor = YoutubeDLCookieProcessor(self.cookiejar)
dca08720
PH
3710 if opts_proxy is not None:
3711 if opts_proxy == '':
3712 proxies = {}
3713 else:
3714 proxies = {'http': opts_proxy, 'https': opts_proxy}
3715 else:
3716 proxies = compat_urllib_request.getproxies()
067aa17e 3717 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
dca08720
PH
3718 if 'http' in proxies and 'https' not in proxies:
3719 proxies['https'] = proxies['http']
91410c9b 3720 proxy_handler = PerRequestProxyHandler(proxies)
a0ddb8a2
PH
3721
3722 debuglevel = 1 if self.params.get('debug_printtraffic') else 0
be4a824d
PH
3723 https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel)
3724 ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel)
fca6dba8 3725 redirect_handler = YoutubeDLRedirectHandler()
f9934b96 3726 data_handler = urllib.request.DataHandler()
6240b0a2
JMF
3727
3728 # When passing our own FileHandler instance, build_opener won't add the
3729 # default FileHandler and allows us to disable the file protocol, which
3730 # can be used for malicious purposes (see
067aa17e 3731 # https://github.com/ytdl-org/youtube-dl/issues/8227)
6240b0a2
JMF
3732 file_handler = compat_urllib_request.FileHandler()
3733
3734 def file_open(*args, **kwargs):
7a5c1cfe 3735 raise compat_urllib_error.URLError('file:// scheme is explicitly disabled in yt-dlp for security reasons')
6240b0a2
JMF
3736 file_handler.file_open = file_open
3737
3738 opener = compat_urllib_request.build_opener(
fca6dba8 3739 proxy_handler, https_handler, cookie_processor, ydlh, redirect_handler, data_handler, file_handler)
2461f79d 3740
dca08720
PH
3741 # Delete the default user-agent header, which would otherwise apply in
3742 # cases where our custom HTTP handler doesn't come into play
067aa17e 3743 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
dca08720
PH
3744 opener.addheaders = []
3745 self._opener = opener
62fec3b2
PH
3746
3747 def encode(self, s):
3748 if isinstance(s, bytes):
3749 return s # Already encoded
3750
3751 try:
3752 return s.encode(self.get_encoding())
3753 except UnicodeEncodeError as err:
3754 err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.'
3755 raise
3756
3757 def get_encoding(self):
3758 encoding = self.params.get('encoding')
3759 if encoding is None:
3760 encoding = preferredencoding()
3761 return encoding
ec82d85a 3762
e08a85d8 3763 def _write_info_json(self, label, ie_result, infofn, overwrite=None):
cb96c5be 3764 ''' Write infojson and returns True = written, 'exists' = Already exists, False = skip, None = error '''
e08a85d8 3765 if overwrite is None:
3766 overwrite = self.params.get('overwrites', True)
80c03fa9 3767 if not self.params.get('writeinfojson'):
3768 return False
3769 elif not infofn:
3770 self.write_debug(f'Skipping writing {label} infojson')
3771 return False
3772 elif not self._ensure_dir_exists(infofn):
3773 return None
e08a85d8 3774 elif not overwrite and os.path.exists(infofn):
80c03fa9 3775 self.to_screen(f'[info] {label.title()} metadata is already present')
cb96c5be 3776 return 'exists'
3777
3778 self.to_screen(f'[info] Writing {label} metadata as JSON to: {infofn}')
3779 try:
3780 write_json_file(self.sanitize_info(ie_result, self.params.get('clean_infojson', True)), infofn)
3781 return True
86e5f3ed 3782 except OSError:
cb96c5be 3783 self.report_error(f'Cannot write {label} metadata to JSON file {infofn}')
3784 return None
80c03fa9 3785
3786 def _write_description(self, label, ie_result, descfn):
3787 ''' Write description and returns True = written, False = skip, None = error '''
3788 if not self.params.get('writedescription'):
3789 return False
3790 elif not descfn:
3791 self.write_debug(f'Skipping writing {label} description')
3792 return False
3793 elif not self._ensure_dir_exists(descfn):
3794 return None
3795 elif not self.params.get('overwrites', True) and os.path.exists(descfn):
3796 self.to_screen(f'[info] {label.title()} description is already present')
3797 elif ie_result.get('description') is None:
3798 self.report_warning(f'There\'s no {label} description to write')
3799 return False
3800 else:
3801 try:
3802 self.to_screen(f'[info] Writing {label} description to: {descfn}')
86e5f3ed 3803 with open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
80c03fa9 3804 descfile.write(ie_result['description'])
86e5f3ed 3805 except OSError:
80c03fa9 3806 self.report_error(f'Cannot write {label} description file {descfn}')
3807 return None
3808 return True
3809
3810 def _write_subtitles(self, info_dict, filename):
3811 ''' Write subtitles to file and return list of (sub_filename, final_sub_filename); or None if error'''
3812 ret = []
3813 subtitles = info_dict.get('requested_subtitles')
3814 if not subtitles or not (self.params.get('writesubtitles') or self.params.get('writeautomaticsub')):
3815 # subtitles download errors are already managed as troubles in relevant IE
3816 # that way it will silently go on when used with unsupporting IE
3817 return ret
3818
3819 sub_filename_base = self.prepare_filename(info_dict, 'subtitle')
3820 if not sub_filename_base:
3821 self.to_screen('[info] Skipping writing video subtitles')
3822 return ret
3823 for sub_lang, sub_info in subtitles.items():
3824 sub_format = sub_info['ext']
3825 sub_filename = subtitles_filename(filename, sub_lang, sub_format, info_dict.get('ext'))
3826 sub_filename_final = subtitles_filename(sub_filename_base, sub_lang, sub_format, info_dict.get('ext'))
e04938ab 3827 existing_sub = self.existing_file((sub_filename_final, sub_filename))
3828 if existing_sub:
80c03fa9 3829 self.to_screen(f'[info] Video subtitle {sub_lang}.{sub_format} is already present')
e04938ab 3830 sub_info['filepath'] = existing_sub
3831 ret.append((existing_sub, sub_filename_final))
80c03fa9 3832 continue
3833
3834 self.to_screen(f'[info] Writing video subtitles to: {sub_filename}')
3835 if sub_info.get('data') is not None:
3836 try:
3837 # Use newline='' to prevent conversion of newline characters
3838 # See https://github.com/ytdl-org/youtube-dl/issues/10268
86e5f3ed 3839 with open(sub_filename, 'w', encoding='utf-8', newline='') as subfile:
80c03fa9 3840 subfile.write(sub_info['data'])
3841 sub_info['filepath'] = sub_filename
3842 ret.append((sub_filename, sub_filename_final))
3843 continue
86e5f3ed 3844 except OSError:
80c03fa9 3845 self.report_error(f'Cannot write video subtitles file {sub_filename}')
3846 return None
3847
3848 try:
3849 sub_copy = sub_info.copy()
3850 sub_copy.setdefault('http_headers', info_dict.get('http_headers'))
3851 self.dl(sub_filename, sub_copy, subtitle=True)
3852 sub_info['filepath'] = sub_filename
3853 ret.append((sub_filename, sub_filename_final))
6020e05d 3854 except (DownloadError, ExtractorError, IOError, OSError, ValueError) + network_exceptions as err:
c70c418d 3855 msg = f'Unable to download video subtitles for {sub_lang!r}: {err}'
6020e05d 3856 if self.params.get('ignoreerrors') is not True: # False or 'only_download'
c70c418d 3857 if not self.params.get('ignoreerrors'):
3858 self.report_error(msg)
3859 raise DownloadError(msg)
3860 self.report_warning(msg)
519804a9 3861 return ret
80c03fa9 3862
3863 def _write_thumbnails(self, label, info_dict, filename, thumb_filename_base=None):
3864 ''' Write thumbnails to file and return list of (thumb_filename, final_thumb_filename) '''
6c4fd172 3865 write_all = self.params.get('write_all_thumbnails', False)
80c03fa9 3866 thumbnails, ret = [], []
6c4fd172 3867 if write_all or self.params.get('writethumbnail', False):
0202b52a 3868 thumbnails = info_dict.get('thumbnails') or []
6c4fd172 3869 multiple = write_all and len(thumbnails) > 1
ec82d85a 3870
80c03fa9 3871 if thumb_filename_base is None:
3872 thumb_filename_base = filename
3873 if thumbnails and not thumb_filename_base:
3874 self.write_debug(f'Skipping writing {label} thumbnail')
3875 return ret
3876
dd0228ce 3877 for idx, t in list(enumerate(thumbnails))[::-1]:
80c03fa9 3878 thumb_ext = (f'{t["id"]}.' if multiple else '') + determine_ext(t['url'], 'jpg')
aa9369a2 3879 thumb_display_id = f'{label} thumbnail {t["id"]}'
80c03fa9 3880 thumb_filename = replace_extension(filename, thumb_ext, info_dict.get('ext'))
3881 thumb_filename_final = replace_extension(thumb_filename_base, thumb_ext, info_dict.get('ext'))
ec82d85a 3882
e04938ab 3883 existing_thumb = self.existing_file((thumb_filename_final, thumb_filename))
3884 if existing_thumb:
aa9369a2 3885 self.to_screen('[info] %s is already present' % (
3886 thumb_display_id if multiple else f'{label} thumbnail').capitalize())
e04938ab 3887 t['filepath'] = existing_thumb
3888 ret.append((existing_thumb, thumb_filename_final))
ec82d85a 3889 else:
80c03fa9 3890 self.to_screen(f'[info] Downloading {thumb_display_id} ...')
ec82d85a 3891 try:
297e9952 3892 uf = self.urlopen(sanitized_Request(t['url'], headers=t.get('http_headers', {})))
80c03fa9 3893 self.to_screen(f'[info] Writing {thumb_display_id} to: {thumb_filename}')
d3d89c32 3894 with open(encodeFilename(thumb_filename), 'wb') as thumbf:
ec82d85a 3895 shutil.copyfileobj(uf, thumbf)
80c03fa9 3896 ret.append((thumb_filename, thumb_filename_final))
885cc0b7 3897 t['filepath'] = thumb_filename
3158150c 3898 except network_exceptions as err:
dd0228ce 3899 thumbnails.pop(idx)
80c03fa9 3900 self.report_warning(f'Unable to download {thumb_display_id}: {err}')
6c4fd172 3901 if ret and not write_all:
3902 break
0202b52a 3903 return ret